diff --git a/CMakeLists.txt b/CMakeLists.txt index 9b13398f24..a69195d496 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -1,9 +1,9 @@ cmake_minimum_required( VERSION 2.4 ) -IF( NOT CMAKE_BUILD_TYPE ) - SET( CMAKE_BUILD_TYPE Debug CACHE STRING - "Choose the type of build, options are: Debug Release RelWithDebInfo MinSizeRel." - FORCE ) +IF( NOT CMAKE_BUILD_TYPE ) + SET( CMAKE_BUILD_TYPE Debug CACHE STRING + "Choose the type of build, options are: Debug Release RelWithDebInfo MinSizeRel." + FORCE ) ENDIF( NOT CMAKE_BUILD_TYPE ) set( ZDOOM_OUTPUT_DIR ${CMAKE_BINARY_DIR} CACHE PATH "Directory where zdoom.pk3 and the executable will be created" ) @@ -79,6 +79,11 @@ else( JPEG_FOUND ) set( JPEG_LIBRARY jpeg ) endif( JPEG_FOUND ) +set( BZIP2_INCLUDE_DIR "${CMAKE_CURRENT_SOURCE_DIR}/bzip2" ) +set( LZMA_INCLUDE_DIR "${CMAKE_CURRENT_SOURCE_DIR}/lzma/C" ) + +add_subdirectory( bzip2 ) +add_subdirectory( lzma ) add_subdirectory( tools ) add_subdirectory( snes_spc ) add_subdirectory( dumb ) diff --git a/bzip2/CHANGES b/bzip2/CHANGES new file mode 100644 index 0000000000..6e4f65e2e0 --- /dev/null +++ b/bzip2/CHANGES @@ -0,0 +1,319 @@ + ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ + + +0.9.0 +~~~~~ +First version. + + +0.9.0a +~~~~~~ +Removed 'ranlib' from Makefile, since most modern Unix-es +don't need it, or even know about it. + + +0.9.0b +~~~~~~ +Fixed a problem with error reporting in bzip2.c. This does not effect +the library in any way. Problem is: versions 0.9.0 and 0.9.0a (of the +program proper) compress and decompress correctly, but give misleading +error messages (internal panics) when an I/O error occurs, instead of +reporting the problem correctly. This shouldn't give any data loss +(as far as I can see), but is confusing. + +Made the inline declarations disappear for non-GCC compilers. + + +0.9.0c +~~~~~~ +Fixed some problems in the library pertaining to some boundary cases. +This makes the library behave more correctly in those situations. The +fixes apply only to features (calls and parameters) not used by +bzip2.c, so the non-fixedness of them in previous versions has no +effect on reliability of bzip2.c. + +In bzlib.c: + * made zero-length BZ_FLUSH work correctly in bzCompress(). + * fixed bzWrite/bzRead to ignore zero-length requests. + * fixed bzread to correctly handle read requests after EOF. + * wrong parameter order in call to bzDecompressInit in + bzBuffToBuffDecompress. Fixed. + +In compress.c: + * changed setting of nGroups in sendMTFValues() so as to + do a bit better on small files. This _does_ effect + bzip2.c. + + +0.9.5a +~~~~~~ +Major change: add a fallback sorting algorithm (blocksort.c) +to give reasonable behaviour even for very repetitive inputs. +Nuked --repetitive-best and --repetitive-fast since they are +no longer useful. + +Minor changes: mostly a whole bunch of small changes/ +bugfixes in the driver (bzip2.c). Changes pertaining to the +user interface are: + + allow decompression of symlink'd files to stdout + decompress/test files even without .bz2 extension + give more accurate error messages for I/O errors + when compressing/decompressing to stdout, don't catch control-C + read flags from BZIP2 and BZIP environment variables + decline to break hard links to a file unless forced with -f + allow -c flag even with no filenames + preserve file ownerships as far as possible + make -s -1 give the expected block size (100k) + add a flag -q --quiet to suppress nonessential warnings + stop decoding flags after --, so files beginning in - can be handled + resolved inconsistent naming: bzcat or bz2cat ? + bzip2 --help now returns 0 + +Programming-level changes are: + + fixed syntax error in GET_LL4 for Borland C++ 5.02 + let bzBuffToBuffDecompress return BZ_DATA_ERROR{_MAGIC} + fix overshoot of mode-string end in bzopen_or_bzdopen + wrapped bzlib.h in #ifdef __cplusplus ... extern "C" { ... } + close file handles under all error conditions + added minor mods so it compiles with DJGPP out of the box + fixed Makefile so it doesn't give problems with BSD make + fix uninitialised memory reads in dlltest.c + +0.9.5b +~~~~~~ +Open stdin/stdout in binary mode for DJGPP. + +0.9.5c +~~~~~~ +Changed BZ_N_OVERSHOOT to be ... + 2 instead of ... + 1. The + 1 +version could cause the sorted order to be wrong in some extremely +obscure cases. Also changed setting of quadrant in blocksort.c. + +0.9.5d +~~~~~~ +The only functional change is to make bzlibVersion() in the library +return the correct string. This has no effect whatsoever on the +functioning of the bzip2 program or library. Added a couple of casts +so the library compiles without warnings at level 3 in MS Visual +Studio 6.0. Included a Y2K statement in the file Y2K_INFO. All other +changes are minor documentation changes. + +1.0 +~~~ +Several minor bugfixes and enhancements: + +* Large file support. The library uses 64-bit counters to + count the volume of data passing through it. bzip2.c + is now compiled with -D_FILE_OFFSET_BITS=64 to get large + file support from the C library. -v correctly prints out + file sizes greater than 4 gigabytes. All these changes have + been made without assuming a 64-bit platform or a C compiler + which supports 64-bit ints, so, except for the C library + aspect, they are fully portable. + +* Decompression robustness. The library/program should be + robust to any corruption of compressed data, detecting and + handling _all_ corruption, instead of merely relying on + the CRCs. What this means is that the program should + never crash, given corrupted data, and the library should + always return BZ_DATA_ERROR. + +* Fixed an obscure race-condition bug only ever observed on + Solaris, in which, if you were very unlucky and issued + control-C at exactly the wrong time, both input and output + files would be deleted. + +* Don't run out of file handles on test/decompression when + large numbers of files have invalid magic numbers. + +* Avoid library namespace pollution. Prefix all exported + symbols with BZ2_. + +* Minor sorting enhancements from my DCC2000 paper. + +* Advance the version number to 1.0, so as to counteract the + (false-in-this-case) impression some people have that programs + with version numbers less than 1.0 are in some way, experimental, + pre-release versions. + +* Create an initial Makefile-libbz2_so to build a shared library. + Yes, I know I should really use libtool et al ... + +* Make the program exit with 2 instead of 0 when decompression + fails due to a bad magic number (ie, an invalid bzip2 header). + Also exit with 1 (as the manual claims :-) whenever a diagnostic + message would have been printed AND the corresponding operation + is aborted, for example + bzip2: Output file xx already exists. + When a diagnostic message is printed but the operation is not + aborted, for example + bzip2: Can't guess original name for wurble -- using wurble.out + then the exit value 0 is returned, unless some other problem is + also detected. + + I think it corresponds more closely to what the manual claims now. + + +1.0.1 +~~~~~ +* Modified dlltest.c so it uses the new BZ2_ naming scheme. +* Modified makefile-msc to fix minor build probs on Win2k. +* Updated README.COMPILATION.PROBLEMS. + +There are no functionality changes or bug fixes relative to version +1.0.0. This is just a documentation update + a fix for minor Win32 +build problems. For almost everyone, upgrading from 1.0.0 to 1.0.1 is +utterly pointless. Don't bother. + + +1.0.2 +~~~~~ +A bug fix release, addressing various minor issues which have appeared +in the 18 or so months since 1.0.1 was released. Most of the fixes +are to do with file-handling or documentation bugs. To the best of my +knowledge, there have been no data-loss-causing bugs reported in the +compression/decompression engine of 1.0.0 or 1.0.1. + +Note that this release does not improve the rather crude build system +for Unix platforms. The general plan here is to autoconfiscate/ +libtoolise 1.0.2 soon after release, and release the result as 1.1.0 +or perhaps 1.2.0. That, however, is still just a plan at this point. + +Here are the changes in 1.0.2. Bug-reporters and/or patch-senders in +parentheses. + +* Fix an infinite segfault loop in 1.0.1 when a directory is + encountered in -f (force) mode. + (Trond Eivind Glomsrod, Nicholas Nethercote, Volker Schmidt) + +* Avoid double fclose() of output file on certain I/O error paths. + (Solar Designer) + +* Don't fail with internal error 1007 when fed a long stream (> 48MB) + of byte 251. Also print useful message suggesting that 1007s may be + caused by bad memory. + (noticed by Juan Pedro Vallejo, fixed by me) + +* Fix uninitialised variable silly bug in demo prog dlltest.c. + (Jorj Bauer) + +* Remove 512-MB limitation on recovered file size for bzip2recover + on selected platforms which support 64-bit ints. At the moment + all GCC supported platforms, and Win32. + (me, Alson van der Meulen) + +* Hard-code header byte values, to give correct operation on platforms + using EBCDIC as their native character set (IBM's OS/390). + (Leland Lucius) + +* Copy file access times correctly. + (Marty Leisner) + +* Add distclean and check targets to Makefile. + (Michael Carmack) + +* Parameterise use of ar and ranlib in Makefile. Also add $(LDFLAGS). + (Rich Ireland, Bo Thorsen) + +* Pass -p (create parent dirs as needed) to mkdir during make install. + (Jeremy Fusco) + +* Dereference symlinks when copying file permissions in -f mode. + (Volker Schmidt) + +* Majorly simplify implementation of uInt64_qrm10. + (Bo Lindbergh) + +* Check the input file still exists before deleting the output one, + when aborting in cleanUpAndFail(). + (Joerg Prante, Robert Linden, Matthias Krings) + +Also a bunch of patches courtesy of Philippe Troin, the Debian maintainer +of bzip2: + +* Wrapper scripts (with manpages): bzdiff, bzgrep, bzmore. + +* Spelling changes and minor enhancements in bzip2.1. + +* Avoid race condition between creating the output file and setting its + interim permissions safely, by using fopen_output_safely(). + No changes to bzip2recover since there is no issue with file + permissions there. + +* do not print senseless report with -v when compressing an empty + file. + +* bzcat -f works on non-bzip2 files. + +* do not try to escape shell meta-characters on unix (the shell takes + care of these). + +* added --fast and --best aliases for -1 -9 for gzip compatibility. + + +1.0.3 (15 Feb 05) +~~~~~~~~~~~~~~~~~ +Fixes some minor bugs since the last version, 1.0.2. + +* Further robustification against corrupted compressed data. + There are currently no known bitstreams which can cause the + decompressor to crash, loop or access memory which does not + belong to it. If you are using bzip2 or the library to + decompress bitstreams from untrusted sources, an upgrade + to 1.0.3 is recommended. This fixes CAN-2005-1260. + +* The documentation has been converted to XML, from which html + and pdf can be derived. + +* Various minor bugs in the documentation have been fixed. + +* Fixes for various compilation warnings with newer versions of + gcc, and on 64-bit platforms. + +* The BZ_NO_STDIO cpp symbol was not properly observed in 1.0.2. + This has been fixed. + + +1.0.4 (20 Dec 06) +~~~~~~~~~~~~~~~~~ +Fixes some minor bugs since the last version, 1.0.3. + +* Fix file permissions race problem (CAN-2005-0953). + +* Avoid possible segfault in BZ2_bzclose. From Coverity's NetBSD + scan. + +* 'const'/prototype cleanups in the C code. + +* Change default install location to /usr/local, and handle multiple + 'make install's without error. + +* Sanitise file names more carefully in bzgrep. Fixes CAN-2005-0758 + to the extent that applies to bzgrep. + +* Use 'mktemp' rather than 'tempfile' in bzdiff. + +* Tighten up a couple of assertions in blocksort.c following automated + analysis. + +* Fix minor doc/comment bugs. + + +1.0.5 (10 Dec 07) +~~~~~~~~~~~~~~~~~ +Security fix only. Fixes CERT-FI 20469 as it applies to bzip2. + diff --git a/bzip2/CMakeLists.txt b/bzip2/CMakeLists.txt new file mode 100644 index 0000000000..54b3849bc9 --- /dev/null +++ b/bzip2/CMakeLists.txt @@ -0,0 +1,16 @@ +cmake_minimum_required( VERSION 2.4 ) + +if( CMAKE_COMPILER_IS_GNUC ) + set( CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Wall -fomit-frame-pointer" ) +endif( CMAKE_COMPILER_IS_GNUC ) + +add_definitions( -DBZ_NO_STDIO ) +add_library( bzip2 + blocksort.c + bzlib.c + compress.c + crctable.c + decompress.c + huffman.c + randtable.c ) +target_link_libraries( bzip2 ) diff --git a/bzip2/LICENSE b/bzip2/LICENSE new file mode 100644 index 0000000000..f420cffb67 --- /dev/null +++ b/bzip2/LICENSE @@ -0,0 +1,42 @@ + +-------------------------------------------------------------------------- + +This program, "bzip2", the associated library "libbzip2", and all +documentation, are copyright (C) 1996-2007 Julian R Seward. All +rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions +are met: + +1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + +2. The origin of this software must not be misrepresented; you must + not claim that you wrote the original software. If you use this + software in a product, an acknowledgment in the product + documentation would be appreciated but is not required. + +3. Altered source versions must be plainly marked as such, and must + not be misrepresented as being the original software. + +4. The name of the author may not be used to endorse or promote + products derived from this software without specific prior written + permission. + +THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS +OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED +WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY +DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE +GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, +WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING +NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS +SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +Julian Seward, jseward@bzip.org +bzip2/libbzip2 version 1.0.5 of 10 December 2007 + +-------------------------------------------------------------------------- diff --git a/bzip2/README b/bzip2/README new file mode 100644 index 0000000000..e17a84e049 --- /dev/null +++ b/bzip2/README @@ -0,0 +1,210 @@ + +This is the README for bzip2/libzip2. +This version is fully compatible with the previous public releases. + +------------------------------------------------------------------ +This file is part of bzip2/libbzip2, a program and library for +lossless, block-sorting data compression. + +bzip2/libbzip2 version 1.0.5 of 10 December 2007 +Copyright (C) 1996-2007 Julian Seward + +Please read the WARNING, DISCLAIMER and PATENTS sections in this file. + +This program is released under the terms of the license contained +in the file LICENSE. +------------------------------------------------------------------ + +Complete documentation is available in Postscript form (manual.ps), +PDF (manual.pdf) or html (manual.html). A plain-text version of the +manual page is available as bzip2.txt. + + +HOW TO BUILD -- UNIX + +Type 'make'. This builds the library libbz2.a and then the programs +bzip2 and bzip2recover. Six self-tests are run. If the self-tests +complete ok, carry on to installation: + +To install in /usr/local/bin, /usr/local/lib, /usr/local/man and +/usr/local/include, type + + make install + +To install somewhere else, eg, /xxx/yyy/{bin,lib,man,include}, type + + make install PREFIX=/xxx/yyy + +If you are (justifiably) paranoid and want to see what 'make install' +is going to do, you can first do + + make -n install or + make -n install PREFIX=/xxx/yyy respectively. + +The -n instructs make to show the commands it would execute, but not +actually execute them. + + +HOW TO BUILD -- UNIX, shared library libbz2.so. + +Do 'make -f Makefile-libbz2_so'. This Makefile seems to work for +Linux-ELF (RedHat 7.2 on an x86 box), with gcc. I make no claims +that it works for any other platform, though I suspect it probably +will work for most platforms employing both ELF and gcc. + +bzip2-shared, a client of the shared library, is also built, but not +self-tested. So I suggest you also build using the normal Makefile, +since that conducts a self-test. A second reason to prefer the +version statically linked to the library is that, on x86 platforms, +building shared objects makes a valuable register (%ebx) unavailable +to gcc, resulting in a slowdown of 10%-20%, at least for bzip2. + +Important note for people upgrading .so's from 0.9.0/0.9.5 to version +1.0.X. All the functions in the library have been renamed, from (eg) +bzCompress to BZ2_bzCompress, to avoid namespace pollution. +Unfortunately this means that the libbz2.so created by +Makefile-libbz2_so will not work with any program which used an older +version of the library. I do encourage library clients to make the +effort to upgrade to use version 1.0, since it is both faster and more +robust than previous versions. + + +HOW TO BUILD -- Windows 95, NT, DOS, Mac, etc. + +It's difficult for me to support compilation on all these platforms. +My approach is to collect binaries for these platforms, and put them +on the master web site (http://www.bzip.org). Look there. However +(FWIW), bzip2-1.0.X is very standard ANSI C and should compile +unmodified with MS Visual C. If you have difficulties building, you +might want to read README.COMPILATION.PROBLEMS. + +At least using MS Visual C++ 6, you can build from the unmodified +sources by issuing, in a command shell: + + nmake -f makefile.msc + +(you may need to first run the MSVC-provided script VCVARS32.BAT + so as to set up paths to the MSVC tools correctly). + + +VALIDATION + +Correct operation, in the sense that a compressed file can always be +decompressed to reproduce the original, is obviously of paramount +importance. To validate bzip2, I used a modified version of Mark +Nelson's churn program. Churn is an automated test driver which +recursively traverses a directory structure, using bzip2 to compress +and then decompress each file it encounters, and checking that the +decompressed data is the same as the original. + + + +Please read and be aware of the following: + +WARNING: + + This program and library (attempts to) compress data by + performing several non-trivial transformations on it. + Unless you are 100% familiar with *all* the algorithms + contained herein, and with the consequences of modifying them, + you should NOT meddle with the compression or decompression + machinery. Incorrect changes can and very likely *will* + lead to disastrous loss of data. + + +DISCLAIMER: + + I TAKE NO RESPONSIBILITY FOR ANY LOSS OF DATA ARISING FROM THE + USE OF THIS PROGRAM/LIBRARY, HOWSOEVER CAUSED. + + Every compression of a file implies an assumption that the + compressed file can be decompressed to reproduce the original. + Great efforts in design, coding and testing have been made to + ensure that this program works correctly. However, the complexity + of the algorithms, and, in particular, the presence of various + special cases in the code which occur with very low but non-zero + probability make it impossible to rule out the possibility of bugs + remaining in the program. DO NOT COMPRESS ANY DATA WITH THIS + PROGRAM UNLESS YOU ARE PREPARED TO ACCEPT THE POSSIBILITY, HOWEVER + SMALL, THAT THE DATA WILL NOT BE RECOVERABLE. + + That is not to say this program is inherently unreliable. + Indeed, I very much hope the opposite is true. bzip2/libbzip2 + has been carefully constructed and extensively tested. + + +PATENTS: + + To the best of my knowledge, bzip2/libbzip2 does not use any + patented algorithms. However, I do not have the resources + to carry out a patent search. Therefore I cannot give any + guarantee of the above statement. + + + +WHAT'S NEW IN 0.9.0 (as compared to 0.1pl2) ? + + * Approx 10% faster compression, 30% faster decompression + * -t (test mode) is a lot quicker + * Can decompress concatenated compressed files + * Programming interface, so programs can directly read/write .bz2 files + * Less restrictive (BSD-style) licensing + * Flag handling more compatible with GNU gzip + * Much more documentation, i.e., a proper user manual + * Hopefully, improved portability (at least of the library) + +WHAT'S NEW IN 0.9.5 ? + + * Compression speed is much less sensitive to the input + data than in previous versions. Specifically, the very + slow performance caused by repetitive data is fixed. + * Many small improvements in file and flag handling. + * A Y2K statement. + +WHAT'S NEW IN 1.0.0 ? + + See the CHANGES file. + +WHAT'S NEW IN 1.0.2 ? + + See the CHANGES file. + +WHAT'S NEW IN 1.0.3 ? + + See the CHANGES file. + +WHAT'S NEW IN 1.0.4 ? + + See the CHANGES file. + +WHAT'S NEW IN 1.0.5 ? + + See the CHANGES file. + + +I hope you find bzip2 useful. Feel free to contact me at + jseward@bzip.org +if you have any suggestions or queries. Many people mailed me with +comments, suggestions and patches after the releases of bzip-0.15, +bzip-0.21, and bzip2 versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0, 1.0.1, +1.0.2 and 1.0.3, and the changes in bzip2 are largely a result of this +feedback. I thank you for your comments. + +bzip2's "home" is http://www.bzip.org/ + +Julian Seward +jseward@bzip.org +Cambridge, UK. + +18 July 1996 (version 0.15) +25 August 1996 (version 0.21) + 7 August 1997 (bzip2, version 0.1) +29 August 1997 (bzip2, version 0.1pl2) +23 August 1998 (bzip2, version 0.9.0) + 8 June 1999 (bzip2, version 0.9.5) + 4 Sept 1999 (bzip2, version 0.9.5d) + 5 May 2000 (bzip2, version 1.0pre8) +30 December 2001 (bzip2, version 1.0.2pre1) +15 February 2005 (bzip2, version 1.0.3) +20 December 2006 (bzip2, version 1.0.4) +10 December 2007 (bzip2, version 1.0.5) diff --git a/bzip2/blocksort.c b/bzip2/blocksort.c new file mode 100644 index 0000000000..bd2dec157f --- /dev/null +++ b/bzip2/blocksort.c @@ -0,0 +1,1094 @@ + +/*-------------------------------------------------------------*/ +/*--- Block sorting machinery ---*/ +/*--- blocksort.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#include "bzlib_private.h" + +/*---------------------------------------------*/ +/*--- Fallback O(N log(N)^2) sorting ---*/ +/*--- algorithm, for repetitive blocks ---*/ +/*---------------------------------------------*/ + +/*---------------------------------------------*/ +static +__inline__ +void fallbackSimpleSort ( UInt32* fmap, + UInt32* eclass, + Int32 lo, + Int32 hi ) +{ + Int32 i, j, tmp; + UInt32 ec_tmp; + + if (lo == hi) return; + + if (hi - lo > 3) { + for ( i = hi-4; i >= lo; i-- ) { + tmp = fmap[i]; + ec_tmp = eclass[tmp]; + for ( j = i+4; j <= hi && ec_tmp > eclass[fmap[j]]; j += 4 ) + fmap[j-4] = fmap[j]; + fmap[j-4] = tmp; + } + } + + for ( i = hi-1; i >= lo; i-- ) { + tmp = fmap[i]; + ec_tmp = eclass[tmp]; + for ( j = i+1; j <= hi && ec_tmp > eclass[fmap[j]]; j++ ) + fmap[j-1] = fmap[j]; + fmap[j-1] = tmp; + } +} + + +/*---------------------------------------------*/ +#define fswap(zz1, zz2) \ + { Int32 zztmp = zz1; zz1 = zz2; zz2 = zztmp; } + +#define fvswap(zzp1, zzp2, zzn) \ +{ \ + Int32 yyp1 = (zzp1); \ + Int32 yyp2 = (zzp2); \ + Int32 yyn = (zzn); \ + while (yyn > 0) { \ + fswap(fmap[yyp1], fmap[yyp2]); \ + yyp1++; yyp2++; yyn--; \ + } \ +} + + +#define fmin(a,b) ((a) < (b)) ? (a) : (b) + +#define fpush(lz,hz) { stackLo[sp] = lz; \ + stackHi[sp] = hz; \ + sp++; } + +#define fpop(lz,hz) { sp--; \ + lz = stackLo[sp]; \ + hz = stackHi[sp]; } + +#define FALLBACK_QSORT_SMALL_THRESH 10 +#define FALLBACK_QSORT_STACK_SIZE 100 + + +static +void fallbackQSort3 ( UInt32* fmap, + UInt32* eclass, + Int32 loSt, + Int32 hiSt ) +{ + Int32 unLo, unHi, ltLo, gtHi, n, m; + Int32 sp, lo, hi; + UInt32 med, r, r3; + Int32 stackLo[FALLBACK_QSORT_STACK_SIZE]; + Int32 stackHi[FALLBACK_QSORT_STACK_SIZE]; + + r = 0; + + sp = 0; + fpush ( loSt, hiSt ); + + while (sp > 0) { + + AssertH ( sp < FALLBACK_QSORT_STACK_SIZE - 1, 1004 ); + + fpop ( lo, hi ); + if (hi - lo < FALLBACK_QSORT_SMALL_THRESH) { + fallbackSimpleSort ( fmap, eclass, lo, hi ); + continue; + } + + /* Random partitioning. Median of 3 sometimes fails to + avoid bad cases. Median of 9 seems to help but + looks rather expensive. This too seems to work but + is cheaper. Guidance for the magic constants + 7621 and 32768 is taken from Sedgewick's algorithms + book, chapter 35. + */ + r = ((r * 7621) + 1) % 32768; + r3 = r % 3; + if (r3 == 0) med = eclass[fmap[lo]]; else + if (r3 == 1) med = eclass[fmap[(lo+hi)>>1]]; else + med = eclass[fmap[hi]]; + + unLo = ltLo = lo; + unHi = gtHi = hi; + + while (1) { + while (1) { + if (unLo > unHi) break; + n = (Int32)eclass[fmap[unLo]] - (Int32)med; + if (n == 0) { + fswap(fmap[unLo], fmap[ltLo]); + ltLo++; unLo++; + continue; + }; + if (n > 0) break; + unLo++; + } + while (1) { + if (unLo > unHi) break; + n = (Int32)eclass[fmap[unHi]] - (Int32)med; + if (n == 0) { + fswap(fmap[unHi], fmap[gtHi]); + gtHi--; unHi--; + continue; + }; + if (n < 0) break; + unHi--; + } + if (unLo > unHi) break; + fswap(fmap[unLo], fmap[unHi]); unLo++; unHi--; + } + + AssertD ( unHi == unLo-1, "fallbackQSort3(2)" ); + + if (gtHi < ltLo) continue; + + n = fmin(ltLo-lo, unLo-ltLo); fvswap(lo, unLo-n, n); + m = fmin(hi-gtHi, gtHi-unHi); fvswap(unLo, hi-m+1, m); + + n = lo + unLo - ltLo - 1; + m = hi - (gtHi - unHi) + 1; + + if (n - lo > hi - m) { + fpush ( lo, n ); + fpush ( m, hi ); + } else { + fpush ( m, hi ); + fpush ( lo, n ); + } + } +} + +#undef fmin +#undef fpush +#undef fpop +#undef fswap +#undef fvswap +#undef FALLBACK_QSORT_SMALL_THRESH +#undef FALLBACK_QSORT_STACK_SIZE + + +/*---------------------------------------------*/ +/* Pre: + nblock > 0 + eclass exists for [0 .. nblock-1] + ((UChar*)eclass) [0 .. nblock-1] holds block + ptr exists for [0 .. nblock-1] + + Post: + ((UChar*)eclass) [0 .. nblock-1] holds block + All other areas of eclass destroyed + fmap [0 .. nblock-1] holds sorted order + bhtab [ 0 .. 2+(nblock/32) ] destroyed +*/ + +#define SET_BH(zz) bhtab[(zz) >> 5] |= (1 << ((zz) & 31)) +#define CLEAR_BH(zz) bhtab[(zz) >> 5] &= ~(1 << ((zz) & 31)) +#define ISSET_BH(zz) (bhtab[(zz) >> 5] & (1 << ((zz) & 31))) +#define WORD_BH(zz) bhtab[(zz) >> 5] +#define UNALIGNED_BH(zz) ((zz) & 0x01f) + +static +void fallbackSort ( UInt32* fmap, + UInt32* eclass, + UInt32* bhtab, + Int32 nblock, + Int32 verb ) +{ + Int32 ftab[257]; + Int32 ftabCopy[256]; + Int32 H, i, j, k, l, r, cc, cc1; + Int32 nNotDone; + Int32 nBhtab; + UChar* eclass8 = (UChar*)eclass; + + /*-- + Initial 1-char radix sort to generate + initial fmap and initial BH bits. + --*/ + if (verb >= 4) + VPrintf0 ( " bucket sorting ...\n" ); + for (i = 0; i < 257; i++) ftab[i] = 0; + for (i = 0; i < nblock; i++) ftab[eclass8[i]]++; + for (i = 0; i < 256; i++) ftabCopy[i] = ftab[i]; + for (i = 1; i < 257; i++) ftab[i] += ftab[i-1]; + + for (i = 0; i < nblock; i++) { + j = eclass8[i]; + k = ftab[j] - 1; + ftab[j] = k; + fmap[k] = i; + } + + nBhtab = 2 + (nblock / 32); + for (i = 0; i < nBhtab; i++) bhtab[i] = 0; + for (i = 0; i < 256; i++) SET_BH(ftab[i]); + + /*-- + Inductively refine the buckets. Kind-of an + "exponential radix sort" (!), inspired by the + Manber-Myers suffix array construction algorithm. + --*/ + + /*-- set sentinel bits for block-end detection --*/ + for (i = 0; i < 32; i++) { + SET_BH(nblock + 2*i); + CLEAR_BH(nblock + 2*i + 1); + } + + /*-- the log(N) loop --*/ + H = 1; + while (1) { + + if (verb >= 4) + VPrintf1 ( " depth %6d has ", H ); + + j = 0; + for (i = 0; i < nblock; i++) { + if (ISSET_BH(i)) j = i; + k = fmap[i] - H; if (k < 0) k += nblock; + eclass[k] = j; + } + + nNotDone = 0; + r = -1; + while (1) { + + /*-- find the next non-singleton bucket --*/ + k = r + 1; + while (ISSET_BH(k) && UNALIGNED_BH(k)) k++; + if (ISSET_BH(k)) { + while (WORD_BH(k) == 0xffffffff) k += 32; + while (ISSET_BH(k)) k++; + } + l = k - 1; + if (l >= nblock) break; + while (!ISSET_BH(k) && UNALIGNED_BH(k)) k++; + if (!ISSET_BH(k)) { + while (WORD_BH(k) == 0x00000000) k += 32; + while (!ISSET_BH(k)) k++; + } + r = k - 1; + if (r >= nblock) break; + + /*-- now [l, r] bracket current bucket --*/ + if (r > l) { + nNotDone += (r - l + 1); + fallbackQSort3 ( fmap, eclass, l, r ); + + /*-- scan bucket and generate header bits-- */ + cc = -1; + for (i = l; i <= r; i++) { + cc1 = eclass[fmap[i]]; + if (cc != cc1) { SET_BH(i); cc = cc1; }; + } + } + } + + if (verb >= 4) + VPrintf1 ( "%6d unresolved strings\n", nNotDone ); + + H *= 2; + if (H > nblock || nNotDone == 0) break; + } + + /*-- + Reconstruct the original block in + eclass8 [0 .. nblock-1], since the + previous phase destroyed it. + --*/ + if (verb >= 4) + VPrintf0 ( " reconstructing block ...\n" ); + j = 0; + for (i = 0; i < nblock; i++) { + while (ftabCopy[j] == 0) j++; + ftabCopy[j]--; + eclass8[fmap[i]] = (UChar)j; + } + AssertH ( j < 256, 1005 ); +} + +#undef SET_BH +#undef CLEAR_BH +#undef ISSET_BH +#undef WORD_BH +#undef UNALIGNED_BH + + +/*---------------------------------------------*/ +/*--- The main, O(N^2 log(N)) sorting ---*/ +/*--- algorithm. Faster for "normal" ---*/ +/*--- non-repetitive blocks. ---*/ +/*---------------------------------------------*/ + +/*---------------------------------------------*/ +static +__inline__ +Bool mainGtU ( UInt32 i1, + UInt32 i2, + UChar* block, + UInt16* quadrant, + UInt32 nblock, + Int32* budget ) +{ + Int32 k; + UChar c1, c2; + UInt16 s1, s2; + + AssertD ( i1 != i2, "mainGtU" ); + /* 1 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 2 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 3 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 4 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 5 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 6 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 7 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 8 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 9 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 10 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 11 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + /* 12 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + i1++; i2++; + + k = nblock + 8; + + do { + /* 1 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 2 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 3 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 4 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 5 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 6 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 7 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + /* 8 */ + c1 = block[i1]; c2 = block[i2]; + if (c1 != c2) return (c1 > c2); + s1 = quadrant[i1]; s2 = quadrant[i2]; + if (s1 != s2) return (s1 > s2); + i1++; i2++; + + if (i1 >= nblock) i1 -= nblock; + if (i2 >= nblock) i2 -= nblock; + + k -= 8; + (*budget)--; + } + while (k >= 0); + + return False; +} + + +/*---------------------------------------------*/ +/*-- + Knuth's increments seem to work better + than Incerpi-Sedgewick here. Possibly + because the number of elems to sort is + usually small, typically <= 20. +--*/ +static +Int32 incs[14] = { 1, 4, 13, 40, 121, 364, 1093, 3280, + 9841, 29524, 88573, 265720, + 797161, 2391484 }; + +static +void mainSimpleSort ( UInt32* ptr, + UChar* block, + UInt16* quadrant, + Int32 nblock, + Int32 lo, + Int32 hi, + Int32 d, + Int32* budget ) +{ + Int32 i, j, h, bigN, hp; + UInt32 v; + + bigN = hi - lo + 1; + if (bigN < 2) return; + + hp = 0; + while (incs[hp] < bigN) hp++; + hp--; + + for (; hp >= 0; hp--) { + h = incs[hp]; + + i = lo + h; + while (True) { + + /*-- copy 1 --*/ + if (i > hi) break; + v = ptr[i]; + j = i; + while ( mainGtU ( + ptr[j-h]+d, v+d, block, quadrant, nblock, budget + ) ) { + ptr[j] = ptr[j-h]; + j = j - h; + if (j <= (lo + h - 1)) break; + } + ptr[j] = v; + i++; + + /*-- copy 2 --*/ + if (i > hi) break; + v = ptr[i]; + j = i; + while ( mainGtU ( + ptr[j-h]+d, v+d, block, quadrant, nblock, budget + ) ) { + ptr[j] = ptr[j-h]; + j = j - h; + if (j <= (lo + h - 1)) break; + } + ptr[j] = v; + i++; + + /*-- copy 3 --*/ + if (i > hi) break; + v = ptr[i]; + j = i; + while ( mainGtU ( + ptr[j-h]+d, v+d, block, quadrant, nblock, budget + ) ) { + ptr[j] = ptr[j-h]; + j = j - h; + if (j <= (lo + h - 1)) break; + } + ptr[j] = v; + i++; + + if (*budget < 0) return; + } + } +} + + +/*---------------------------------------------*/ +/*-- + The following is an implementation of + an elegant 3-way quicksort for strings, + described in a paper "Fast Algorithms for + Sorting and Searching Strings", by Robert + Sedgewick and Jon L. Bentley. +--*/ + +#define mswap(zz1, zz2) \ + { Int32 zztmp = zz1; zz1 = zz2; zz2 = zztmp; } + +#define mvswap(zzp1, zzp2, zzn) \ +{ \ + Int32 yyp1 = (zzp1); \ + Int32 yyp2 = (zzp2); \ + Int32 yyn = (zzn); \ + while (yyn > 0) { \ + mswap(ptr[yyp1], ptr[yyp2]); \ + yyp1++; yyp2++; yyn--; \ + } \ +} + +static +__inline__ +UChar mmed3 ( UChar a, UChar b, UChar c ) +{ + UChar t; + if (a > b) { t = a; a = b; b = t; }; + if (b > c) { + b = c; + if (a > b) b = a; + } + return b; +} + +#define mmin(a,b) ((a) < (b)) ? (a) : (b) + +#define mpush(lz,hz,dz) { stackLo[sp] = lz; \ + stackHi[sp] = hz; \ + stackD [sp] = dz; \ + sp++; } + +#define mpop(lz,hz,dz) { sp--; \ + lz = stackLo[sp]; \ + hz = stackHi[sp]; \ + dz = stackD [sp]; } + + +#define mnextsize(az) (nextHi[az]-nextLo[az]) + +#define mnextswap(az,bz) \ + { Int32 tz; \ + tz = nextLo[az]; nextLo[az] = nextLo[bz]; nextLo[bz] = tz; \ + tz = nextHi[az]; nextHi[az] = nextHi[bz]; nextHi[bz] = tz; \ + tz = nextD [az]; nextD [az] = nextD [bz]; nextD [bz] = tz; } + + +#define MAIN_QSORT_SMALL_THRESH 20 +#define MAIN_QSORT_DEPTH_THRESH (BZ_N_RADIX + BZ_N_QSORT) +#define MAIN_QSORT_STACK_SIZE 100 + +static +void mainQSort3 ( UInt32* ptr, + UChar* block, + UInt16* quadrant, + Int32 nblock, + Int32 loSt, + Int32 hiSt, + Int32 dSt, + Int32* budget ) +{ + Int32 unLo, unHi, ltLo, gtHi, n, m, med; + Int32 sp, lo, hi, d; + + Int32 stackLo[MAIN_QSORT_STACK_SIZE]; + Int32 stackHi[MAIN_QSORT_STACK_SIZE]; + Int32 stackD [MAIN_QSORT_STACK_SIZE]; + + Int32 nextLo[3]; + Int32 nextHi[3]; + Int32 nextD [3]; + + sp = 0; + mpush ( loSt, hiSt, dSt ); + + while (sp > 0) { + + AssertH ( sp < MAIN_QSORT_STACK_SIZE - 2, 1001 ); + + mpop ( lo, hi, d ); + if (hi - lo < MAIN_QSORT_SMALL_THRESH || + d > MAIN_QSORT_DEPTH_THRESH) { + mainSimpleSort ( ptr, block, quadrant, nblock, lo, hi, d, budget ); + if (*budget < 0) return; + continue; + } + + med = (Int32) + mmed3 ( block[ptr[ lo ]+d], + block[ptr[ hi ]+d], + block[ptr[ (lo+hi)>>1 ]+d] ); + + unLo = ltLo = lo; + unHi = gtHi = hi; + + while (True) { + while (True) { + if (unLo > unHi) break; + n = ((Int32)block[ptr[unLo]+d]) - med; + if (n == 0) { + mswap(ptr[unLo], ptr[ltLo]); + ltLo++; unLo++; continue; + }; + if (n > 0) break; + unLo++; + } + while (True) { + if (unLo > unHi) break; + n = ((Int32)block[ptr[unHi]+d]) - med; + if (n == 0) { + mswap(ptr[unHi], ptr[gtHi]); + gtHi--; unHi--; continue; + }; + if (n < 0) break; + unHi--; + } + if (unLo > unHi) break; + mswap(ptr[unLo], ptr[unHi]); unLo++; unHi--; + } + + AssertD ( unHi == unLo-1, "mainQSort3(2)" ); + + if (gtHi < ltLo) { + mpush(lo, hi, d+1 ); + continue; + } + + n = mmin(ltLo-lo, unLo-ltLo); mvswap(lo, unLo-n, n); + m = mmin(hi-gtHi, gtHi-unHi); mvswap(unLo, hi-m+1, m); + + n = lo + unLo - ltLo - 1; + m = hi - (gtHi - unHi) + 1; + + nextLo[0] = lo; nextHi[0] = n; nextD[0] = d; + nextLo[1] = m; nextHi[1] = hi; nextD[1] = d; + nextLo[2] = n+1; nextHi[2] = m-1; nextD[2] = d+1; + + if (mnextsize(0) < mnextsize(1)) mnextswap(0,1); + if (mnextsize(1) < mnextsize(2)) mnextswap(1,2); + if (mnextsize(0) < mnextsize(1)) mnextswap(0,1); + + AssertD (mnextsize(0) >= mnextsize(1), "mainQSort3(8)" ); + AssertD (mnextsize(1) >= mnextsize(2), "mainQSort3(9)" ); + + mpush (nextLo[0], nextHi[0], nextD[0]); + mpush (nextLo[1], nextHi[1], nextD[1]); + mpush (nextLo[2], nextHi[2], nextD[2]); + } +} + +#undef mswap +#undef mvswap +#undef mpush +#undef mpop +#undef mmin +#undef mnextsize +#undef mnextswap +#undef MAIN_QSORT_SMALL_THRESH +#undef MAIN_QSORT_DEPTH_THRESH +#undef MAIN_QSORT_STACK_SIZE + + +/*---------------------------------------------*/ +/* Pre: + nblock > N_OVERSHOOT + block32 exists for [0 .. nblock-1 +N_OVERSHOOT] + ((UChar*)block32) [0 .. nblock-1] holds block + ptr exists for [0 .. nblock-1] + + Post: + ((UChar*)block32) [0 .. nblock-1] holds block + All other areas of block32 destroyed + ftab [0 .. 65536 ] destroyed + ptr [0 .. nblock-1] holds sorted order + if (*budget < 0), sorting was abandoned +*/ + +#define BIGFREQ(b) (ftab[((b)+1) << 8] - ftab[(b) << 8]) +#define SETMASK (1 << 21) +#define CLEARMASK (~(SETMASK)) + +static +void mainSort ( UInt32* ptr, + UChar* block, + UInt16* quadrant, + UInt32* ftab, + Int32 nblock, + Int32 verb, + Int32* budget ) +{ + Int32 i, j, k, ss, sb; + Int32 runningOrder[256]; + Bool bigDone[256]; + Int32 copyStart[256]; + Int32 copyEnd [256]; + UChar c1; + Int32 numQSorted; + UInt16 s; + if (verb >= 4) VPrintf0 ( " main sort initialise ...\n" ); + + /*-- set up the 2-byte frequency table --*/ + for (i = 65536; i >= 0; i--) ftab[i] = 0; + + j = block[0] << 8; + i = nblock-1; + for (; i >= 3; i -= 4) { + quadrant[i] = 0; + j = (j >> 8) | ( ((UInt16)block[i]) << 8); + ftab[j]++; + quadrant[i-1] = 0; + j = (j >> 8) | ( ((UInt16)block[i-1]) << 8); + ftab[j]++; + quadrant[i-2] = 0; + j = (j >> 8) | ( ((UInt16)block[i-2]) << 8); + ftab[j]++; + quadrant[i-3] = 0; + j = (j >> 8) | ( ((UInt16)block[i-3]) << 8); + ftab[j]++; + } + for (; i >= 0; i--) { + quadrant[i] = 0; + j = (j >> 8) | ( ((UInt16)block[i]) << 8); + ftab[j]++; + } + + /*-- (emphasises close relationship of block & quadrant) --*/ + for (i = 0; i < BZ_N_OVERSHOOT; i++) { + block [nblock+i] = block[i]; + quadrant[nblock+i] = 0; + } + + if (verb >= 4) VPrintf0 ( " bucket sorting ...\n" ); + + /*-- Complete the initial radix sort --*/ + for (i = 1; i <= 65536; i++) ftab[i] += ftab[i-1]; + + s = block[0] << 8; + i = nblock-1; + for (; i >= 3; i -= 4) { + s = (s >> 8) | (block[i] << 8); + j = ftab[s] -1; + ftab[s] = j; + ptr[j] = i; + s = (s >> 8) | (block[i-1] << 8); + j = ftab[s] -1; + ftab[s] = j; + ptr[j] = i-1; + s = (s >> 8) | (block[i-2] << 8); + j = ftab[s] -1; + ftab[s] = j; + ptr[j] = i-2; + s = (s >> 8) | (block[i-3] << 8); + j = ftab[s] -1; + ftab[s] = j; + ptr[j] = i-3; + } + for (; i >= 0; i--) { + s = (s >> 8) | (block[i] << 8); + j = ftab[s] -1; + ftab[s] = j; + ptr[j] = i; + } + + /*-- + Now ftab contains the first loc of every small bucket. + Calculate the running order, from smallest to largest + big bucket. + --*/ + for (i = 0; i <= 255; i++) { + bigDone [i] = False; + runningOrder[i] = i; + } + + { + Int32 vv; + Int32 h = 1; + do h = 3 * h + 1; while (h <= 256); + do { + h = h / 3; + for (i = h; i <= 255; i++) { + vv = runningOrder[i]; + j = i; + while ( BIGFREQ(runningOrder[j-h]) > BIGFREQ(vv) ) { + runningOrder[j] = runningOrder[j-h]; + j = j - h; + if (j <= (h - 1)) goto zero; + } + zero: + runningOrder[j] = vv; + } + } while (h != 1); + } + + /*-- + The main sorting loop. + --*/ + + numQSorted = 0; + + for (i = 0; i <= 255; i++) { + + /*-- + Process big buckets, starting with the least full. + Basically this is a 3-step process in which we call + mainQSort3 to sort the small buckets [ss, j], but + also make a big effort to avoid the calls if we can. + --*/ + ss = runningOrder[i]; + + /*-- + Step 1: + Complete the big bucket [ss] by quicksorting + any unsorted small buckets [ss, j], for j != ss. + Hopefully previous pointer-scanning phases have already + completed many of the small buckets [ss, j], so + we don't have to sort them at all. + --*/ + for (j = 0; j <= 255; j++) { + if (j != ss) { + sb = (ss << 8) + j; + if ( ! (ftab[sb] & SETMASK) ) { + Int32 lo = ftab[sb] & CLEARMASK; + Int32 hi = (ftab[sb+1] & CLEARMASK) - 1; + if (hi > lo) { + if (verb >= 4) + VPrintf4 ( " qsort [0x%x, 0x%x] " + "done %d this %d\n", + ss, j, numQSorted, hi - lo + 1 ); + mainQSort3 ( + ptr, block, quadrant, nblock, + lo, hi, BZ_N_RADIX, budget + ); + numQSorted += (hi - lo + 1); + if (*budget < 0) return; + } + } + ftab[sb] |= SETMASK; + } + } + + AssertH ( !bigDone[ss], 1006 ); + + /*-- + Step 2: + Now scan this big bucket [ss] so as to synthesise the + sorted order for small buckets [t, ss] for all t, + including, magically, the bucket [ss,ss] too. + This will avoid doing Real Work in subsequent Step 1's. + --*/ + { + for (j = 0; j <= 255; j++) { + copyStart[j] = ftab[(j << 8) + ss] & CLEARMASK; + copyEnd [j] = (ftab[(j << 8) + ss + 1] & CLEARMASK) - 1; + } + for (j = ftab[ss << 8] & CLEARMASK; j < copyStart[ss]; j++) { + k = ptr[j]-1; if (k < 0) k += nblock; + c1 = block[k]; + if (!bigDone[c1]) + ptr[ copyStart[c1]++ ] = k; + } + for (j = (ftab[(ss+1) << 8] & CLEARMASK) - 1; j > copyEnd[ss]; j--) { + k = ptr[j]-1; if (k < 0) k += nblock; + c1 = block[k]; + if (!bigDone[c1]) + ptr[ copyEnd[c1]-- ] = k; + } + } + + AssertH ( (copyStart[ss]-1 == copyEnd[ss]) + || + /* Extremely rare case missing in bzip2-1.0.0 and 1.0.1. + Necessity for this case is demonstrated by compressing + a sequence of approximately 48.5 million of character + 251; 1.0.0/1.0.1 will then die here. */ + (copyStart[ss] == 0 && copyEnd[ss] == nblock-1), + 1007 ) + + for (j = 0; j <= 255; j++) ftab[(j << 8) + ss] |= SETMASK; + + /*-- + Step 3: + The [ss] big bucket is now done. Record this fact, + and update the quadrant descriptors. Remember to + update quadrants in the overshoot area too, if + necessary. The "if (i < 255)" test merely skips + this updating for the last bucket processed, since + updating for the last bucket is pointless. + + The quadrant array provides a way to incrementally + cache sort orderings, as they appear, so as to + make subsequent comparisons in fullGtU() complete + faster. For repetitive blocks this makes a big + difference (but not big enough to be able to avoid + the fallback sorting mechanism, exponential radix sort). + + The precise meaning is: at all times: + + for 0 <= i < nblock and 0 <= j <= nblock + + if block[i] != block[j], + + then the relative values of quadrant[i] and + quadrant[j] are meaningless. + + else { + if quadrant[i] < quadrant[j] + then the string starting at i lexicographically + precedes the string starting at j + + else if quadrant[i] > quadrant[j] + then the string starting at j lexicographically + precedes the string starting at i + + else + the relative ordering of the strings starting + at i and j has not yet been determined. + } + --*/ + bigDone[ss] = True; + + if (i < 255) { + Int32 bbStart = ftab[ss << 8] & CLEARMASK; + Int32 bbSize = (ftab[(ss+1) << 8] & CLEARMASK) - bbStart; + Int32 shifts = 0; + + while ((bbSize >> shifts) > 65534) shifts++; + + for (j = bbSize-1; j >= 0; j--) { + Int32 a2update = ptr[bbStart + j]; + UInt16 qVal = (UInt16)(j >> shifts); + quadrant[a2update] = qVal; + if (a2update < BZ_N_OVERSHOOT) + quadrant[a2update + nblock] = qVal; + } + AssertH ( ((bbSize-1) >> shifts) <= 65535, 1002 ); + } + + } + + if (verb >= 4) + VPrintf3 ( " %d pointers, %d sorted, %d scanned\n", + nblock, numQSorted, nblock - numQSorted ); +} + +#undef BIGFREQ +#undef SETMASK +#undef CLEARMASK + + +/*---------------------------------------------*/ +/* Pre: + nblock > 0 + arr2 exists for [0 .. nblock-1 +N_OVERSHOOT] + ((UChar*)arr2) [0 .. nblock-1] holds block + arr1 exists for [0 .. nblock-1] + + Post: + ((UChar*)arr2) [0 .. nblock-1] holds block + All other areas of block destroyed + ftab [ 0 .. 65536 ] destroyed + arr1 [0 .. nblock-1] holds sorted order +*/ +void BZ2_blockSort ( EState* s ) +{ + UInt32* ptr = s->ptr; + UChar* block = s->block; + UInt32* ftab = s->ftab; + Int32 nblock = s->nblock; + Int32 verb = s->verbosity; + Int32 wfact = s->workFactor; + UInt16* quadrant; + Int32 budget; + Int32 budgetInit; + Int32 i; + + if (nblock < 10000) { + fallbackSort ( s->arr1, s->arr2, ftab, nblock, verb ); + } else { + /* Calculate the location for quadrant, remembering to get + the alignment right. Assumes that &(block[0]) is at least + 2-byte aligned -- this should be ok since block is really + the first section of arr2. + */ + i = nblock+BZ_N_OVERSHOOT; + if (i & 1) i++; + quadrant = (UInt16*)(&(block[i])); + + /* (wfact-1) / 3 puts the default-factor-30 + transition point at very roughly the same place as + with v0.1 and v0.9.0. + Not that it particularly matters any more, since the + resulting compressed stream is now the same regardless + of whether or not we use the main sort or fallback sort. + */ + if (wfact < 1 ) wfact = 1; + if (wfact > 100) wfact = 100; + budgetInit = nblock * ((wfact-1) / 3); + budget = budgetInit; + + mainSort ( ptr, block, quadrant, ftab, nblock, verb, &budget ); + if (verb >= 3) + VPrintf3 ( " %d work, %d block, ratio %5.2f\n", + budgetInit - budget, + nblock, + (float)(budgetInit - budget) / + (float)(nblock==0 ? 1 : nblock) ); + if (budget < 0) { + if (verb >= 2) + VPrintf0 ( " too repetitive; using fallback" + " sorting algorithm\n" ); + fallbackSort ( s->arr1, s->arr2, ftab, nblock, verb ); + } + } + + s->origPtr = -1; + for (i = 0; i < s->nblock; i++) + if (ptr[i] == 0) + { s->origPtr = i; break; }; + + AssertH( s->origPtr != -1, 1003 ); +} + + +/*-------------------------------------------------------------*/ +/*--- end blocksort.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/bzip2/bzip2.vcproj b/bzip2/bzip2.vcproj new file mode 100644 index 0000000000..72510949d0 --- /dev/null +++ b/bzip2/bzip2.vcproj @@ -0,0 +1,329 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + diff --git a/bzip2/bzlib.c b/bzip2/bzlib.c new file mode 100644 index 0000000000..82385bfcfc --- /dev/null +++ b/bzip2/bzlib.c @@ -0,0 +1,1572 @@ + +/*-------------------------------------------------------------*/ +/*--- Library top-level functions. ---*/ +/*--- bzlib.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + +/* CHANGES + 0.9.0 -- original version. + 0.9.0a/b -- no changes in this file. + 0.9.0c -- made zero-length BZ_FLUSH work correctly in bzCompress(). + fixed bzWrite/bzRead to ignore zero-length requests. + fixed bzread to correctly handle read requests after EOF. + wrong parameter order in call to bzDecompressInit in + bzBuffToBuffDecompress. Fixed. +*/ + +#include "bzlib_private.h" + + +/*---------------------------------------------------*/ +/*--- Compression stuff ---*/ +/*---------------------------------------------------*/ + + +/*---------------------------------------------------*/ +#ifndef BZ_NO_STDIO +void BZ2_bz__AssertH__fail ( int errcode ) +{ + fprintf(stderr, + "\n\nbzip2/libbzip2: internal error number %d.\n" + "This is a bug in bzip2/libbzip2, %s.\n" + "Please report it to me at: jseward@bzip.org. If this happened\n" + "when you were using some program which uses libbzip2 as a\n" + "component, you should also report this bug to the author(s)\n" + "of that program. Please make an effort to report this bug;\n" + "timely and accurate bug reports eventually lead to higher\n" + "quality software. Thanks. Julian Seward, 10 December 2007.\n\n", + errcode, + BZ2_bzlibVersion() + ); + + if (errcode == 1007) { + fprintf(stderr, + "\n*** A special note about internal error number 1007 ***\n" + "\n" + "Experience suggests that a common cause of i.e. 1007\n" + "is unreliable memory or other hardware. The 1007 assertion\n" + "just happens to cross-check the results of huge numbers of\n" + "memory reads/writes, and so acts (unintendedly) as a stress\n" + "test of your memory system.\n" + "\n" + "I suggest the following: try compressing the file again,\n" + "possibly monitoring progress in detail with the -vv flag.\n" + "\n" + "* If the error cannot be reproduced, and/or happens at different\n" + " points in compression, you may have a flaky memory system.\n" + " Try a memory-test program. I have used Memtest86\n" + " (www.memtest86.com). At the time of writing it is free (GPLd).\n" + " Memtest86 tests memory much more thorougly than your BIOSs\n" + " power-on test, and may find failures that the BIOS doesn't.\n" + "\n" + "* If the error can be repeatably reproduced, this is a bug in\n" + " bzip2, and I would very much like to hear about it. Please\n" + " let me know, and, ideally, save a copy of the file causing the\n" + " problem -- without which I will be unable to investigate it.\n" + "\n" + ); + } + + exit(3); +} +#endif + + +/*---------------------------------------------------*/ +static +int bz_config_ok ( void ) +{ + if (sizeof(int) != 4) return 0; + if (sizeof(short) != 2) return 0; + if (sizeof(char) != 1) return 0; + return 1; +} + + +/*---------------------------------------------------*/ +static +void* default_bzalloc ( void* opaque, Int32 items, Int32 size ) +{ + void* v = malloc ( items * size ); + return v; +} + +static +void default_bzfree ( void* opaque, void* addr ) +{ + if (addr != NULL) free ( addr ); +} + + +/*---------------------------------------------------*/ +static +void prepare_new_block ( EState* s ) +{ + Int32 i; + s->nblock = 0; + s->numZ = 0; + s->state_out_pos = 0; + BZ_INITIALISE_CRC ( s->blockCRC ); + for (i = 0; i < 256; i++) s->inUse[i] = False; + s->blockNo++; +} + + +/*---------------------------------------------------*/ +static +void init_RL ( EState* s ) +{ + s->state_in_ch = 256; + s->state_in_len = 0; +} + + +static +Bool isempty_RL ( EState* s ) +{ + if (s->state_in_ch < 256 && s->state_in_len > 0) + return False; else + return True; +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzCompressInit) + ( bz_stream* strm, + int blockSize100k, + int verbosity, + int workFactor ) +{ + Int32 n; + EState* s; + + if (!bz_config_ok()) return BZ_CONFIG_ERROR; + + if (strm == NULL || + blockSize100k < 1 || blockSize100k > 9 || + workFactor < 0 || workFactor > 250) + return BZ_PARAM_ERROR; + + if (workFactor == 0) workFactor = 30; + if (strm->bzalloc == NULL) strm->bzalloc = default_bzalloc; + if (strm->bzfree == NULL) strm->bzfree = default_bzfree; + + s = BZALLOC( sizeof(EState) ); + if (s == NULL) return BZ_MEM_ERROR; + s->strm = strm; + + s->arr1 = NULL; + s->arr2 = NULL; + s->ftab = NULL; + + n = 100000 * blockSize100k; + s->arr1 = BZALLOC( n * sizeof(UInt32) ); + s->arr2 = BZALLOC( (n+BZ_N_OVERSHOOT) * sizeof(UInt32) ); + s->ftab = BZALLOC( 65537 * sizeof(UInt32) ); + + if (s->arr1 == NULL || s->arr2 == NULL || s->ftab == NULL) { + if (s->arr1 != NULL) BZFREE(s->arr1); + if (s->arr2 != NULL) BZFREE(s->arr2); + if (s->ftab != NULL) BZFREE(s->ftab); + if (s != NULL) BZFREE(s); + return BZ_MEM_ERROR; + } + + s->blockNo = 0; + s->state = BZ_S_INPUT; + s->mode = BZ_M_RUNNING; + s->combinedCRC = 0; + s->blockSize100k = blockSize100k; + s->nblockMAX = 100000 * blockSize100k - 19; + s->verbosity = verbosity; + s->workFactor = workFactor; + + s->block = (UChar*)s->arr2; + s->mtfv = (UInt16*)s->arr1; + s->zbits = NULL; + s->ptr = (UInt32*)s->arr1; + + strm->state = s; + strm->total_in_lo32 = 0; + strm->total_in_hi32 = 0; + strm->total_out_lo32 = 0; + strm->total_out_hi32 = 0; + init_RL ( s ); + prepare_new_block ( s ); + return BZ_OK; +} + + +/*---------------------------------------------------*/ +static +void add_pair_to_block ( EState* s ) +{ + Int32 i; + UChar ch = (UChar)(s->state_in_ch); + for (i = 0; i < s->state_in_len; i++) { + BZ_UPDATE_CRC( s->blockCRC, ch ); + } + s->inUse[s->state_in_ch] = True; + switch (s->state_in_len) { + case 1: + s->block[s->nblock] = (UChar)ch; s->nblock++; + break; + case 2: + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = (UChar)ch; s->nblock++; + break; + case 3: + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = (UChar)ch; s->nblock++; + break; + default: + s->inUse[s->state_in_len-4] = True; + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = (UChar)ch; s->nblock++; + s->block[s->nblock] = ((UChar)(s->state_in_len-4)); + s->nblock++; + break; + } +} + + +/*---------------------------------------------------*/ +static +void flush_RL ( EState* s ) +{ + if (s->state_in_ch < 256) add_pair_to_block ( s ); + init_RL ( s ); +} + + +/*---------------------------------------------------*/ +#define ADD_CHAR_TO_BLOCK(zs,zchh0) \ +{ \ + UInt32 zchh = (UInt32)(zchh0); \ + /*-- fast track the common case --*/ \ + if (zchh != zs->state_in_ch && \ + zs->state_in_len == 1) { \ + UChar ch = (UChar)(zs->state_in_ch); \ + BZ_UPDATE_CRC( zs->blockCRC, ch ); \ + zs->inUse[zs->state_in_ch] = True; \ + zs->block[zs->nblock] = (UChar)ch; \ + zs->nblock++; \ + zs->state_in_ch = zchh; \ + } \ + else \ + /*-- general, uncommon cases --*/ \ + if (zchh != zs->state_in_ch || \ + zs->state_in_len == 255) { \ + if (zs->state_in_ch < 256) \ + add_pair_to_block ( zs ); \ + zs->state_in_ch = zchh; \ + zs->state_in_len = 1; \ + } else { \ + zs->state_in_len++; \ + } \ +} + + +/*---------------------------------------------------*/ +static +Bool copy_input_until_stop ( EState* s ) +{ + Bool progress_in = False; + + if (s->mode == BZ_M_RUNNING) { + + /*-- fast track the common case --*/ + while (True) { + /*-- block full? --*/ + if (s->nblock >= s->nblockMAX) break; + /*-- no input? --*/ + if (s->strm->avail_in == 0) break; + progress_in = True; + ADD_CHAR_TO_BLOCK ( s, (UInt32)(*((UChar*)(s->strm->next_in))) ); + s->strm->next_in++; + s->strm->avail_in--; + s->strm->total_in_lo32++; + if (s->strm->total_in_lo32 == 0) s->strm->total_in_hi32++; + } + + } else { + + /*-- general, uncommon case --*/ + while (True) { + /*-- block full? --*/ + if (s->nblock >= s->nblockMAX) break; + /*-- no input? --*/ + if (s->strm->avail_in == 0) break; + /*-- flush/finish end? --*/ + if (s->avail_in_expect == 0) break; + progress_in = True; + ADD_CHAR_TO_BLOCK ( s, (UInt32)(*((UChar*)(s->strm->next_in))) ); + s->strm->next_in++; + s->strm->avail_in--; + s->strm->total_in_lo32++; + if (s->strm->total_in_lo32 == 0) s->strm->total_in_hi32++; + s->avail_in_expect--; + } + } + return progress_in; +} + + +/*---------------------------------------------------*/ +static +Bool copy_output_until_stop ( EState* s ) +{ + Bool progress_out = False; + + while (True) { + + /*-- no output space? --*/ + if (s->strm->avail_out == 0) break; + + /*-- block done? --*/ + if (s->state_out_pos >= s->numZ) break; + + progress_out = True; + *(s->strm->next_out) = s->zbits[s->state_out_pos]; + s->state_out_pos++; + s->strm->avail_out--; + s->strm->next_out++; + s->strm->total_out_lo32++; + if (s->strm->total_out_lo32 == 0) s->strm->total_out_hi32++; + } + + return progress_out; +} + + +/*---------------------------------------------------*/ +static +Bool handle_compress ( bz_stream* strm ) +{ + Bool progress_in = False; + Bool progress_out = False; + EState* s = strm->state; + + while (True) { + + if (s->state == BZ_S_OUTPUT) { + progress_out |= copy_output_until_stop ( s ); + if (s->state_out_pos < s->numZ) break; + if (s->mode == BZ_M_FINISHING && + s->avail_in_expect == 0 && + isempty_RL(s)) break; + prepare_new_block ( s ); + s->state = BZ_S_INPUT; + if (s->mode == BZ_M_FLUSHING && + s->avail_in_expect == 0 && + isempty_RL(s)) break; + } + + if (s->state == BZ_S_INPUT) { + progress_in |= copy_input_until_stop ( s ); + if (s->mode != BZ_M_RUNNING && s->avail_in_expect == 0) { + flush_RL ( s ); + BZ2_compressBlock ( s, (Bool)(s->mode == BZ_M_FINISHING) ); + s->state = BZ_S_OUTPUT; + } + else + if (s->nblock >= s->nblockMAX) { + BZ2_compressBlock ( s, False ); + s->state = BZ_S_OUTPUT; + } + else + if (s->strm->avail_in == 0) { + break; + } + } + + } + + return progress_in || progress_out; +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzCompress) ( bz_stream *strm, int action ) +{ + Bool progress; + EState* s; + if (strm == NULL) return BZ_PARAM_ERROR; + s = strm->state; + if (s == NULL) return BZ_PARAM_ERROR; + if (s->strm != strm) return BZ_PARAM_ERROR; + + preswitch: + switch (s->mode) { + + case BZ_M_IDLE: + return BZ_SEQUENCE_ERROR; + + case BZ_M_RUNNING: + if (action == BZ_RUN) { + progress = handle_compress ( strm ); + return progress ? BZ_RUN_OK : BZ_PARAM_ERROR; + } + else + if (action == BZ_FLUSH) { + s->avail_in_expect = strm->avail_in; + s->mode = BZ_M_FLUSHING; + goto preswitch; + } + else + if (action == BZ_FINISH) { + s->avail_in_expect = strm->avail_in; + s->mode = BZ_M_FINISHING; + goto preswitch; + } + else + return BZ_PARAM_ERROR; + + case BZ_M_FLUSHING: + if (action != BZ_FLUSH) return BZ_SEQUENCE_ERROR; + if (s->avail_in_expect != s->strm->avail_in) + return BZ_SEQUENCE_ERROR; + progress = handle_compress ( strm ); + if (s->avail_in_expect > 0 || !isempty_RL(s) || + s->state_out_pos < s->numZ) return BZ_FLUSH_OK; + s->mode = BZ_M_RUNNING; + return BZ_RUN_OK; + + case BZ_M_FINISHING: + if (action != BZ_FINISH) return BZ_SEQUENCE_ERROR; + if (s->avail_in_expect != s->strm->avail_in) + return BZ_SEQUENCE_ERROR; + progress = handle_compress ( strm ); + if (!progress) return BZ_SEQUENCE_ERROR; + if (s->avail_in_expect > 0 || !isempty_RL(s) || + s->state_out_pos < s->numZ) return BZ_FINISH_OK; + s->mode = BZ_M_IDLE; + return BZ_STREAM_END; + } + return BZ_OK; /*--not reached--*/ +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzCompressEnd) ( bz_stream *strm ) +{ + EState* s; + if (strm == NULL) return BZ_PARAM_ERROR; + s = strm->state; + if (s == NULL) return BZ_PARAM_ERROR; + if (s->strm != strm) return BZ_PARAM_ERROR; + + if (s->arr1 != NULL) BZFREE(s->arr1); + if (s->arr2 != NULL) BZFREE(s->arr2); + if (s->ftab != NULL) BZFREE(s->ftab); + BZFREE(strm->state); + + strm->state = NULL; + + return BZ_OK; +} + + +/*---------------------------------------------------*/ +/*--- Decompression stuff ---*/ +/*---------------------------------------------------*/ + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzDecompressInit) + ( bz_stream* strm, + int verbosity, + int small ) +{ + DState* s; + + if (!bz_config_ok()) return BZ_CONFIG_ERROR; + + if (strm == NULL) return BZ_PARAM_ERROR; + if (small != 0 && small != 1) return BZ_PARAM_ERROR; + if (verbosity < 0 || verbosity > 4) return BZ_PARAM_ERROR; + + if (strm->bzalloc == NULL) strm->bzalloc = default_bzalloc; + if (strm->bzfree == NULL) strm->bzfree = default_bzfree; + + s = BZALLOC( sizeof(DState) ); + if (s == NULL) return BZ_MEM_ERROR; + s->strm = strm; + strm->state = s; + s->state = BZ_X_MAGIC_1; + s->bsLive = 0; + s->bsBuff = 0; + s->calculatedCombinedCRC = 0; + strm->total_in_lo32 = 0; + strm->total_in_hi32 = 0; + strm->total_out_lo32 = 0; + strm->total_out_hi32 = 0; + s->smallDecompress = (Bool)small; + s->ll4 = NULL; + s->ll16 = NULL; + s->tt = NULL; + s->currBlockNo = 0; + s->verbosity = verbosity; + + return BZ_OK; +} + + +/*---------------------------------------------------*/ +/* Return True iff data corruption is discovered. + Returns False if there is no problem. +*/ +static +Bool unRLE_obuf_to_output_FAST ( DState* s ) +{ + UChar k1; + + if (s->blockRandomised) { + + while (True) { + /* try to finish existing run */ + while (True) { + if (s->strm->avail_out == 0) return False; + if (s->state_out_len == 0) break; + *( (UChar*)(s->strm->next_out) ) = s->state_out_ch; + BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch ); + s->state_out_len--; + s->strm->next_out++; + s->strm->avail_out--; + s->strm->total_out_lo32++; + if (s->strm->total_out_lo32 == 0) s->strm->total_out_hi32++; + } + + /* can a new run be started? */ + if (s->nblock_used == s->save_nblock+1) return False; + + /* Only caused by corrupt data stream? */ + if (s->nblock_used > s->save_nblock+1) + return True; + + s->state_out_len = 1; + s->state_out_ch = s->k0; + BZ_GET_FAST(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + s->state_out_len = 2; + BZ_GET_FAST(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + s->state_out_len = 3; + BZ_GET_FAST(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + BZ_GET_FAST(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + s->state_out_len = ((Int32)k1) + 4; + BZ_GET_FAST(s->k0); BZ_RAND_UPD_MASK; + s->k0 ^= BZ_RAND_MASK; s->nblock_used++; + } + + } else { + + /* restore */ + UInt32 c_calculatedBlockCRC = s->calculatedBlockCRC; + UChar c_state_out_ch = s->state_out_ch; + Int32 c_state_out_len = s->state_out_len; + Int32 c_nblock_used = s->nblock_used; + Int32 c_k0 = s->k0; + UInt32* c_tt = s->tt; + UInt32 c_tPos = s->tPos; + char* cs_next_out = s->strm->next_out; + unsigned int cs_avail_out = s->strm->avail_out; + Int32 ro_blockSize100k = s->blockSize100k; + /* end restore */ + + UInt32 avail_out_INIT = cs_avail_out; + Int32 s_save_nblockPP = s->save_nblock+1; + unsigned int total_out_lo32_old; + + while (True) { + + /* try to finish existing run */ + if (c_state_out_len > 0) { + while (True) { + if (cs_avail_out == 0) goto return_notr; + if (c_state_out_len == 1) break; + *( (UChar*)(cs_next_out) ) = c_state_out_ch; + BZ_UPDATE_CRC ( c_calculatedBlockCRC, c_state_out_ch ); + c_state_out_len--; + cs_next_out++; + cs_avail_out--; + } + s_state_out_len_eq_one: + { + if (cs_avail_out == 0) { + c_state_out_len = 1; goto return_notr; + }; + *( (UChar*)(cs_next_out) ) = c_state_out_ch; + BZ_UPDATE_CRC ( c_calculatedBlockCRC, c_state_out_ch ); + cs_next_out++; + cs_avail_out--; + } + } + /* Only caused by corrupt data stream? */ + if (c_nblock_used > s_save_nblockPP) + return True; + + /* can a new run be started? */ + if (c_nblock_used == s_save_nblockPP) { + c_state_out_len = 0; goto return_notr; + }; + c_state_out_ch = c_k0; + BZ_GET_FAST_C(k1); c_nblock_used++; + if (k1 != c_k0) { + c_k0 = k1; goto s_state_out_len_eq_one; + }; + if (c_nblock_used == s_save_nblockPP) + goto s_state_out_len_eq_one; + + c_state_out_len = 2; + BZ_GET_FAST_C(k1); c_nblock_used++; + if (c_nblock_used == s_save_nblockPP) continue; + if (k1 != c_k0) { c_k0 = k1; continue; }; + + c_state_out_len = 3; + BZ_GET_FAST_C(k1); c_nblock_used++; + if (c_nblock_used == s_save_nblockPP) continue; + if (k1 != c_k0) { c_k0 = k1; continue; }; + + BZ_GET_FAST_C(k1); c_nblock_used++; + c_state_out_len = ((Int32)k1) + 4; + BZ_GET_FAST_C(c_k0); c_nblock_used++; + } + + return_notr: + total_out_lo32_old = s->strm->total_out_lo32; + s->strm->total_out_lo32 += (avail_out_INIT - cs_avail_out); + if (s->strm->total_out_lo32 < total_out_lo32_old) + s->strm->total_out_hi32++; + + /* save */ + s->calculatedBlockCRC = c_calculatedBlockCRC; + s->state_out_ch = c_state_out_ch; + s->state_out_len = c_state_out_len; + s->nblock_used = c_nblock_used; + s->k0 = c_k0; + s->tt = c_tt; + s->tPos = c_tPos; + s->strm->next_out = cs_next_out; + s->strm->avail_out = cs_avail_out; + /* end save */ + } + return False; +} + + + +/*---------------------------------------------------*/ +__inline__ Int32 BZ2_indexIntoF ( Int32 indx, Int32 *cftab ) +{ + Int32 nb, na, mid; + nb = 0; + na = 256; + do { + mid = (nb + na) >> 1; + if (indx >= cftab[mid]) nb = mid; else na = mid; + } + while (na - nb != 1); + return nb; +} + + +/*---------------------------------------------------*/ +/* Return True iff data corruption is discovered. + Returns False if there is no problem. +*/ +static +Bool unRLE_obuf_to_output_SMALL ( DState* s ) +{ + UChar k1; + + if (s->blockRandomised) { + + while (True) { + /* try to finish existing run */ + while (True) { + if (s->strm->avail_out == 0) return False; + if (s->state_out_len == 0) break; + *( (UChar*)(s->strm->next_out) ) = s->state_out_ch; + BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch ); + s->state_out_len--; + s->strm->next_out++; + s->strm->avail_out--; + s->strm->total_out_lo32++; + if (s->strm->total_out_lo32 == 0) s->strm->total_out_hi32++; + } + + /* can a new run be started? */ + if (s->nblock_used == s->save_nblock+1) return False; + + /* Only caused by corrupt data stream? */ + if (s->nblock_used > s->save_nblock+1) + return True; + + s->state_out_len = 1; + s->state_out_ch = s->k0; + BZ_GET_SMALL(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + s->state_out_len = 2; + BZ_GET_SMALL(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + s->state_out_len = 3; + BZ_GET_SMALL(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + BZ_GET_SMALL(k1); BZ_RAND_UPD_MASK; + k1 ^= BZ_RAND_MASK; s->nblock_used++; + s->state_out_len = ((Int32)k1) + 4; + BZ_GET_SMALL(s->k0); BZ_RAND_UPD_MASK; + s->k0 ^= BZ_RAND_MASK; s->nblock_used++; + } + + } else { + + while (True) { + /* try to finish existing run */ + while (True) { + if (s->strm->avail_out == 0) return False; + if (s->state_out_len == 0) break; + *( (UChar*)(s->strm->next_out) ) = s->state_out_ch; + BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch ); + s->state_out_len--; + s->strm->next_out++; + s->strm->avail_out--; + s->strm->total_out_lo32++; + if (s->strm->total_out_lo32 == 0) s->strm->total_out_hi32++; + } + + /* can a new run be started? */ + if (s->nblock_used == s->save_nblock+1) return False; + + /* Only caused by corrupt data stream? */ + if (s->nblock_used > s->save_nblock+1) + return True; + + s->state_out_len = 1; + s->state_out_ch = s->k0; + BZ_GET_SMALL(k1); s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + s->state_out_len = 2; + BZ_GET_SMALL(k1); s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + s->state_out_len = 3; + BZ_GET_SMALL(k1); s->nblock_used++; + if (s->nblock_used == s->save_nblock+1) continue; + if (k1 != s->k0) { s->k0 = k1; continue; }; + + BZ_GET_SMALL(k1); s->nblock_used++; + s->state_out_len = ((Int32)k1) + 4; + BZ_GET_SMALL(s->k0); s->nblock_used++; + } + + } +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzDecompress) ( bz_stream *strm ) +{ + Bool corrupt; + DState* s; + if (strm == NULL) return BZ_PARAM_ERROR; + s = strm->state; + if (s == NULL) return BZ_PARAM_ERROR; + if (s->strm != strm) return BZ_PARAM_ERROR; + + while (True) { + if (s->state == BZ_X_IDLE) return BZ_SEQUENCE_ERROR; + if (s->state == BZ_X_OUTPUT) { + if (s->smallDecompress) + corrupt = unRLE_obuf_to_output_SMALL ( s ); else + corrupt = unRLE_obuf_to_output_FAST ( s ); + if (corrupt) return BZ_DATA_ERROR; + if (s->nblock_used == s->save_nblock+1 && s->state_out_len == 0) { + BZ_FINALISE_CRC ( s->calculatedBlockCRC ); + if (s->verbosity >= 3) + VPrintf2 ( " {0x%08x, 0x%08x}", s->storedBlockCRC, + s->calculatedBlockCRC ); + if (s->verbosity >= 2) VPrintf0 ( "]" ); + if (s->calculatedBlockCRC != s->storedBlockCRC) + return BZ_DATA_ERROR; + s->calculatedCombinedCRC + = (s->calculatedCombinedCRC << 1) | + (s->calculatedCombinedCRC >> 31); + s->calculatedCombinedCRC ^= s->calculatedBlockCRC; + s->state = BZ_X_BLKHDR_1; + } else { + return BZ_OK; + } + } + if (s->state >= BZ_X_MAGIC_1) { + Int32 r = BZ2_decompress ( s ); + if (r == BZ_STREAM_END) { + if (s->verbosity >= 3) + VPrintf2 ( "\n combined CRCs: stored = 0x%08x, computed = 0x%08x", + s->storedCombinedCRC, s->calculatedCombinedCRC ); + if (s->calculatedCombinedCRC != s->storedCombinedCRC) + return BZ_DATA_ERROR; + return r; + } + if (s->state != BZ_X_OUTPUT) return r; + } + } + + AssertH ( 0, 6001 ); + + return 0; /*NOTREACHED*/ +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzDecompressEnd) ( bz_stream *strm ) +{ + DState* s; + if (strm == NULL) return BZ_PARAM_ERROR; + s = strm->state; + if (s == NULL) return BZ_PARAM_ERROR; + if (s->strm != strm) return BZ_PARAM_ERROR; + + if (s->tt != NULL) BZFREE(s->tt); + if (s->ll16 != NULL) BZFREE(s->ll16); + if (s->ll4 != NULL) BZFREE(s->ll4); + + BZFREE(strm->state); + strm->state = NULL; + + return BZ_OK; +} + + +#ifndef BZ_NO_STDIO +/*---------------------------------------------------*/ +/*--- File I/O stuff ---*/ +/*---------------------------------------------------*/ + +#define BZ_SETERR(eee) \ +{ \ + if (bzerror != NULL) *bzerror = eee; \ + if (bzf != NULL) bzf->lastErr = eee; \ +} + +typedef + struct { + FILE* handle; + Char buf[BZ_MAX_UNUSED]; + Int32 bufN; + Bool writing; + bz_stream strm; + Int32 lastErr; + Bool initialisedOk; + } + bzFile; + + +/*---------------------------------------------*/ +static Bool myfeof ( FILE* f ) +{ + Int32 c = fgetc ( f ); + if (c == EOF) return True; + ungetc ( c, f ); + return False; +} + + +/*---------------------------------------------------*/ +BZFILE* BZ_API(BZ2_bzWriteOpen) + ( int* bzerror, + FILE* f, + int blockSize100k, + int verbosity, + int workFactor ) +{ + Int32 ret; + bzFile* bzf = NULL; + + BZ_SETERR(BZ_OK); + + if (f == NULL || + (blockSize100k < 1 || blockSize100k > 9) || + (workFactor < 0 || workFactor > 250) || + (verbosity < 0 || verbosity > 4)) + { BZ_SETERR(BZ_PARAM_ERROR); return NULL; }; + + if (ferror(f)) + { BZ_SETERR(BZ_IO_ERROR); return NULL; }; + + bzf = malloc ( sizeof(bzFile) ); + if (bzf == NULL) + { BZ_SETERR(BZ_MEM_ERROR); return NULL; }; + + BZ_SETERR(BZ_OK); + bzf->initialisedOk = False; + bzf->bufN = 0; + bzf->handle = f; + bzf->writing = True; + bzf->strm.bzalloc = NULL; + bzf->strm.bzfree = NULL; + bzf->strm.opaque = NULL; + + if (workFactor == 0) workFactor = 30; + ret = BZ2_bzCompressInit ( &(bzf->strm), blockSize100k, + verbosity, workFactor ); + if (ret != BZ_OK) + { BZ_SETERR(ret); free(bzf); return NULL; }; + + bzf->strm.avail_in = 0; + bzf->initialisedOk = True; + return bzf; +} + + + +/*---------------------------------------------------*/ +void BZ_API(BZ2_bzWrite) + ( int* bzerror, + BZFILE* b, + void* buf, + int len ) +{ + Int32 n, n2, ret; + bzFile* bzf = (bzFile*)b; + + BZ_SETERR(BZ_OK); + if (bzf == NULL || buf == NULL || len < 0) + { BZ_SETERR(BZ_PARAM_ERROR); return; }; + if (!(bzf->writing)) + { BZ_SETERR(BZ_SEQUENCE_ERROR); return; }; + if (ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return; }; + + if (len == 0) + { BZ_SETERR(BZ_OK); return; }; + + bzf->strm.avail_in = len; + bzf->strm.next_in = buf; + + while (True) { + bzf->strm.avail_out = BZ_MAX_UNUSED; + bzf->strm.next_out = bzf->buf; + ret = BZ2_bzCompress ( &(bzf->strm), BZ_RUN ); + if (ret != BZ_RUN_OK) + { BZ_SETERR(ret); return; }; + + if (bzf->strm.avail_out < BZ_MAX_UNUSED) { + n = BZ_MAX_UNUSED - bzf->strm.avail_out; + n2 = fwrite ( (void*)(bzf->buf), sizeof(UChar), + n, bzf->handle ); + if (n != n2 || ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return; }; + } + + if (bzf->strm.avail_in == 0) + { BZ_SETERR(BZ_OK); return; }; + } +} + + +/*---------------------------------------------------*/ +void BZ_API(BZ2_bzWriteClose) + ( int* bzerror, + BZFILE* b, + int abandon, + unsigned int* nbytes_in, + unsigned int* nbytes_out ) +{ + BZ2_bzWriteClose64 ( bzerror, b, abandon, + nbytes_in, NULL, nbytes_out, NULL ); +} + + +void BZ_API(BZ2_bzWriteClose64) + ( int* bzerror, + BZFILE* b, + int abandon, + unsigned int* nbytes_in_lo32, + unsigned int* nbytes_in_hi32, + unsigned int* nbytes_out_lo32, + unsigned int* nbytes_out_hi32 ) +{ + Int32 n, n2, ret; + bzFile* bzf = (bzFile*)b; + + if (bzf == NULL) + { BZ_SETERR(BZ_OK); return; }; + if (!(bzf->writing)) + { BZ_SETERR(BZ_SEQUENCE_ERROR); return; }; + if (ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return; }; + + if (nbytes_in_lo32 != NULL) *nbytes_in_lo32 = 0; + if (nbytes_in_hi32 != NULL) *nbytes_in_hi32 = 0; + if (nbytes_out_lo32 != NULL) *nbytes_out_lo32 = 0; + if (nbytes_out_hi32 != NULL) *nbytes_out_hi32 = 0; + + if ((!abandon) && bzf->lastErr == BZ_OK) { + while (True) { + bzf->strm.avail_out = BZ_MAX_UNUSED; + bzf->strm.next_out = bzf->buf; + ret = BZ2_bzCompress ( &(bzf->strm), BZ_FINISH ); + if (ret != BZ_FINISH_OK && ret != BZ_STREAM_END) + { BZ_SETERR(ret); return; }; + + if (bzf->strm.avail_out < BZ_MAX_UNUSED) { + n = BZ_MAX_UNUSED - bzf->strm.avail_out; + n2 = fwrite ( (void*)(bzf->buf), sizeof(UChar), + n, bzf->handle ); + if (n != n2 || ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return; }; + } + + if (ret == BZ_STREAM_END) break; + } + } + + if ( !abandon && !ferror ( bzf->handle ) ) { + fflush ( bzf->handle ); + if (ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return; }; + } + + if (nbytes_in_lo32 != NULL) + *nbytes_in_lo32 = bzf->strm.total_in_lo32; + if (nbytes_in_hi32 != NULL) + *nbytes_in_hi32 = bzf->strm.total_in_hi32; + if (nbytes_out_lo32 != NULL) + *nbytes_out_lo32 = bzf->strm.total_out_lo32; + if (nbytes_out_hi32 != NULL) + *nbytes_out_hi32 = bzf->strm.total_out_hi32; + + BZ_SETERR(BZ_OK); + BZ2_bzCompressEnd ( &(bzf->strm) ); + free ( bzf ); +} + + +/*---------------------------------------------------*/ +BZFILE* BZ_API(BZ2_bzReadOpen) + ( int* bzerror, + FILE* f, + int verbosity, + int small, + void* unused, + int nUnused ) +{ + bzFile* bzf = NULL; + int ret; + + BZ_SETERR(BZ_OK); + + if (f == NULL || + (small != 0 && small != 1) || + (verbosity < 0 || verbosity > 4) || + (unused == NULL && nUnused != 0) || + (unused != NULL && (nUnused < 0 || nUnused > BZ_MAX_UNUSED))) + { BZ_SETERR(BZ_PARAM_ERROR); return NULL; }; + + if (ferror(f)) + { BZ_SETERR(BZ_IO_ERROR); return NULL; }; + + bzf = malloc ( sizeof(bzFile) ); + if (bzf == NULL) + { BZ_SETERR(BZ_MEM_ERROR); return NULL; }; + + BZ_SETERR(BZ_OK); + + bzf->initialisedOk = False; + bzf->handle = f; + bzf->bufN = 0; + bzf->writing = False; + bzf->strm.bzalloc = NULL; + bzf->strm.bzfree = NULL; + bzf->strm.opaque = NULL; + + while (nUnused > 0) { + bzf->buf[bzf->bufN] = *((UChar*)(unused)); bzf->bufN++; + unused = ((void*)( 1 + ((UChar*)(unused)) )); + nUnused--; + } + + ret = BZ2_bzDecompressInit ( &(bzf->strm), verbosity, small ); + if (ret != BZ_OK) + { BZ_SETERR(ret); free(bzf); return NULL; }; + + bzf->strm.avail_in = bzf->bufN; + bzf->strm.next_in = bzf->buf; + + bzf->initialisedOk = True; + return bzf; +} + + +/*---------------------------------------------------*/ +void BZ_API(BZ2_bzReadClose) ( int *bzerror, BZFILE *b ) +{ + bzFile* bzf = (bzFile*)b; + + BZ_SETERR(BZ_OK); + if (bzf == NULL) + { BZ_SETERR(BZ_OK); return; }; + + if (bzf->writing) + { BZ_SETERR(BZ_SEQUENCE_ERROR); return; }; + + if (bzf->initialisedOk) + (void)BZ2_bzDecompressEnd ( &(bzf->strm) ); + free ( bzf ); +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzRead) + ( int* bzerror, + BZFILE* b, + void* buf, + int len ) +{ + Int32 n, ret; + bzFile* bzf = (bzFile*)b; + + BZ_SETERR(BZ_OK); + + if (bzf == NULL || buf == NULL || len < 0) + { BZ_SETERR(BZ_PARAM_ERROR); return 0; }; + + if (bzf->writing) + { BZ_SETERR(BZ_SEQUENCE_ERROR); return 0; }; + + if (len == 0) + { BZ_SETERR(BZ_OK); return 0; }; + + bzf->strm.avail_out = len; + bzf->strm.next_out = buf; + + while (True) { + + if (ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return 0; }; + + if (bzf->strm.avail_in == 0 && !myfeof(bzf->handle)) { + n = fread ( bzf->buf, sizeof(UChar), + BZ_MAX_UNUSED, bzf->handle ); + if (ferror(bzf->handle)) + { BZ_SETERR(BZ_IO_ERROR); return 0; }; + bzf->bufN = n; + bzf->strm.avail_in = bzf->bufN; + bzf->strm.next_in = bzf->buf; + } + + ret = BZ2_bzDecompress ( &(bzf->strm) ); + + if (ret != BZ_OK && ret != BZ_STREAM_END) + { BZ_SETERR(ret); return 0; }; + + if (ret == BZ_OK && myfeof(bzf->handle) && + bzf->strm.avail_in == 0 && bzf->strm.avail_out > 0) + { BZ_SETERR(BZ_UNEXPECTED_EOF); return 0; }; + + if (ret == BZ_STREAM_END) + { BZ_SETERR(BZ_STREAM_END); + return len - bzf->strm.avail_out; }; + if (bzf->strm.avail_out == 0) + { BZ_SETERR(BZ_OK); return len; }; + + } + + return 0; /*not reached*/ +} + + +/*---------------------------------------------------*/ +void BZ_API(BZ2_bzReadGetUnused) + ( int* bzerror, + BZFILE* b, + void** unused, + int* nUnused ) +{ + bzFile* bzf = (bzFile*)b; + if (bzf == NULL) + { BZ_SETERR(BZ_PARAM_ERROR); return; }; + if (bzf->lastErr != BZ_STREAM_END) + { BZ_SETERR(BZ_SEQUENCE_ERROR); return; }; + if (unused == NULL || nUnused == NULL) + { BZ_SETERR(BZ_PARAM_ERROR); return; }; + + BZ_SETERR(BZ_OK); + *nUnused = bzf->strm.avail_in; + *unused = (void **)bzf->strm.next_in; +} +#endif + + +/*---------------------------------------------------*/ +/*--- Misc convenience stuff ---*/ +/*---------------------------------------------------*/ + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzBuffToBuffCompress) + ( char* dest, + unsigned int* destLen, + const char* source, + unsigned int sourceLen, + int blockSize100k, + int verbosity, + int workFactor ) +{ + bz_stream strm; + int ret; + + if (dest == NULL || destLen == NULL || + source == NULL || + blockSize100k < 1 || blockSize100k > 9 || + verbosity < 0 || verbosity > 4 || + workFactor < 0 || workFactor > 250) + return BZ_PARAM_ERROR; + + if (workFactor == 0) workFactor = 30; + strm.bzalloc = NULL; + strm.bzfree = NULL; + strm.opaque = NULL; + ret = BZ2_bzCompressInit ( &strm, blockSize100k, + verbosity, workFactor ); + if (ret != BZ_OK) return ret; + + strm.next_in = source; + strm.next_out = dest; + strm.avail_in = sourceLen; + strm.avail_out = *destLen; + + ret = BZ2_bzCompress ( &strm, BZ_FINISH ); + if (ret == BZ_FINISH_OK) goto output_overflow; + if (ret != BZ_STREAM_END) goto errhandler; + + /* normal termination */ + *destLen -= strm.avail_out; + BZ2_bzCompressEnd ( &strm ); + return BZ_OK; + + output_overflow: + BZ2_bzCompressEnd ( &strm ); + return BZ_OUTBUFF_FULL; + + errhandler: + BZ2_bzCompressEnd ( &strm ); + return ret; +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzBuffToBuffDecompress) + ( char* dest, + unsigned int* destLen, + const char* source, + unsigned int sourceLen, + int small, + int verbosity ) +{ + bz_stream strm; + int ret; + + if (dest == NULL || destLen == NULL || + source == NULL || + (small != 0 && small != 1) || + verbosity < 0 || verbosity > 4) + return BZ_PARAM_ERROR; + + strm.bzalloc = NULL; + strm.bzfree = NULL; + strm.opaque = NULL; + ret = BZ2_bzDecompressInit ( &strm, verbosity, small ); + if (ret != BZ_OK) return ret; + + strm.next_in = source; + strm.next_out = dest; + strm.avail_in = sourceLen; + strm.avail_out = *destLen; + + ret = BZ2_bzDecompress ( &strm ); + if (ret == BZ_OK) goto output_overflow_or_eof; + if (ret != BZ_STREAM_END) goto errhandler; + + /* normal termination */ + *destLen -= strm.avail_out; + BZ2_bzDecompressEnd ( &strm ); + return BZ_OK; + + output_overflow_or_eof: + if (strm.avail_out > 0) { + BZ2_bzDecompressEnd ( &strm ); + return BZ_UNEXPECTED_EOF; + } else { + BZ2_bzDecompressEnd ( &strm ); + return BZ_OUTBUFF_FULL; + }; + + errhandler: + BZ2_bzDecompressEnd ( &strm ); + return ret; +} + + +/*---------------------------------------------------*/ +/*-- + Code contributed by Yoshioka Tsuneo (tsuneo@rr.iij4u.or.jp) + to support better zlib compatibility. + This code is not _officially_ part of libbzip2 (yet); + I haven't tested it, documented it, or considered the + threading-safeness of it. + If this code breaks, please contact both Yoshioka and me. +--*/ +/*---------------------------------------------------*/ + +/*---------------------------------------------------*/ +/*-- + return version like "0.9.5d, 4-Sept-1999". +--*/ +const char * BZ_API(BZ2_bzlibVersion)(void) +{ + return BZ_VERSION; +} + + +#ifndef BZ_NO_STDIO +/*---------------------------------------------------*/ + +#if defined(_WIN32) || defined(OS2) || defined(MSDOS) +# include +# include +# define SET_BINARY_MODE(file) setmode(fileno(file),O_BINARY) +#else +# define SET_BINARY_MODE(file) +#endif +static +BZFILE * bzopen_or_bzdopen + ( const char *path, /* no use when bzdopen */ + int fd, /* no use when bzdopen */ + const char *mode, + int open_mode) /* bzopen: 0, bzdopen:1 */ +{ + int bzerr; + char unused[BZ_MAX_UNUSED]; + int blockSize100k = 9; + int writing = 0; + char mode2[10] = ""; + FILE *fp = NULL; + BZFILE *bzfp = NULL; + int verbosity = 0; + int workFactor = 30; + int smallMode = 0; + int nUnused = 0; + + if (mode == NULL) return NULL; + while (*mode) { + switch (*mode) { + case 'r': + writing = 0; break; + case 'w': + writing = 1; break; + case 's': + smallMode = 1; break; + default: + if (isdigit((int)(*mode))) { + blockSize100k = *mode-BZ_HDR_0; + } + } + mode++; + } + strcat(mode2, writing ? "w" : "r" ); + strcat(mode2,"b"); /* binary mode */ + + if (open_mode==0) { + if (path==NULL || strcmp(path,"")==0) { + fp = (writing ? stdout : stdin); + SET_BINARY_MODE(fp); + } else { + fp = fopen(path,mode2); + } + } else { +#ifdef BZ_STRICT_ANSI + fp = NULL; +#else + fp = fdopen(fd,mode2); +#endif + } + if (fp == NULL) return NULL; + + if (writing) { + /* Guard against total chaos and anarchy -- JRS */ + if (blockSize100k < 1) blockSize100k = 1; + if (blockSize100k > 9) blockSize100k = 9; + bzfp = BZ2_bzWriteOpen(&bzerr,fp,blockSize100k, + verbosity,workFactor); + } else { + bzfp = BZ2_bzReadOpen(&bzerr,fp,verbosity,smallMode, + unused,nUnused); + } + if (bzfp == NULL) { + if (fp != stdin && fp != stdout) fclose(fp); + return NULL; + } + return bzfp; +} + + +/*---------------------------------------------------*/ +/*-- + open file for read or write. + ex) bzopen("file","w9") + case path="" or NULL => use stdin or stdout. +--*/ +BZFILE * BZ_API(BZ2_bzopen) + ( const char *path, + const char *mode ) +{ + return bzopen_or_bzdopen(path,-1,mode,/*bzopen*/0); +} + + +/*---------------------------------------------------*/ +BZFILE * BZ_API(BZ2_bzdopen) + ( int fd, + const char *mode ) +{ + return bzopen_or_bzdopen(NULL,fd,mode,/*bzdopen*/1); +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzread) (BZFILE* b, void* buf, int len ) +{ + int bzerr, nread; + if (((bzFile*)b)->lastErr == BZ_STREAM_END) return 0; + nread = BZ2_bzRead(&bzerr,b,buf,len); + if (bzerr == BZ_OK || bzerr == BZ_STREAM_END) { + return nread; + } else { + return -1; + } +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzwrite) (BZFILE* b, void* buf, int len ) +{ + int bzerr; + + BZ2_bzWrite(&bzerr,b,buf,len); + if(bzerr == BZ_OK){ + return len; + }else{ + return -1; + } +} + + +/*---------------------------------------------------*/ +int BZ_API(BZ2_bzflush) (BZFILE *b) +{ + /* do nothing now... */ + return 0; +} + + +/*---------------------------------------------------*/ +void BZ_API(BZ2_bzclose) (BZFILE* b) +{ + int bzerr; + FILE *fp; + + if (b==NULL) {return;} + fp = ((bzFile *)b)->handle; + if(((bzFile*)b)->writing){ + BZ2_bzWriteClose(&bzerr,b,0,NULL,NULL); + if(bzerr != BZ_OK){ + BZ2_bzWriteClose(NULL,b,1,NULL,NULL); + } + }else{ + BZ2_bzReadClose(&bzerr,b); + } + if(fp!=stdin && fp!=stdout){ + fclose(fp); + } +} + + +/*---------------------------------------------------*/ +/*-- + return last error code +--*/ +static const char *bzerrorstrings[] = { + "OK" + ,"SEQUENCE_ERROR" + ,"PARAM_ERROR" + ,"MEM_ERROR" + ,"DATA_ERROR" + ,"DATA_ERROR_MAGIC" + ,"IO_ERROR" + ,"UNEXPECTED_EOF" + ,"OUTBUFF_FULL" + ,"CONFIG_ERROR" + ,"???" /* for future */ + ,"???" /* for future */ + ,"???" /* for future */ + ,"???" /* for future */ + ,"???" /* for future */ + ,"???" /* for future */ +}; + + +const char * BZ_API(BZ2_bzerror) (BZFILE *b, int *errnum) +{ + int err = ((bzFile *)b)->lastErr; + + if(err>0) err = 0; + *errnum = err; + return bzerrorstrings[err*-1]; +} +#endif + + +/*-------------------------------------------------------------*/ +/*--- end bzlib.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/bzip2/bzlib.h b/bzip2/bzlib.h new file mode 100644 index 0000000000..a42a1a30a3 --- /dev/null +++ b/bzip2/bzlib.h @@ -0,0 +1,266 @@ + +/*-------------------------------------------------------------*/ +/*--- Public header file for the library. ---*/ +/*--- bzlib.h ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#ifndef _BZLIB_H +#define _BZLIB_H + +#ifdef __cplusplus +extern "C" { +#endif + +#define BZ_RUN 0 +#define BZ_FLUSH 1 +#define BZ_FINISH 2 + +#define BZ_OK 0 +#define BZ_RUN_OK 1 +#define BZ_FLUSH_OK 2 +#define BZ_FINISH_OK 3 +#define BZ_STREAM_END 4 +#define BZ_SEQUENCE_ERROR (-1) +#define BZ_PARAM_ERROR (-2) +#define BZ_MEM_ERROR (-3) +#define BZ_DATA_ERROR (-4) +#define BZ_DATA_ERROR_MAGIC (-5) +#define BZ_IO_ERROR (-6) +#define BZ_UNEXPECTED_EOF (-7) +#define BZ_OUTBUFF_FULL (-8) +#define BZ_CONFIG_ERROR (-9) + +typedef + struct { + const char *next_in; + unsigned int avail_in; + unsigned int total_in_lo32; + unsigned int total_in_hi32; + + char *next_out; + unsigned int avail_out; + unsigned int total_out_lo32; + unsigned int total_out_hi32; + + void *state; + + void *(*bzalloc)(void *,int,int); + void (*bzfree)(void *,void *); + void *opaque; + } + bz_stream; + + +#ifndef BZ_IMPORT +#define BZ_EXPORT +#endif + +#ifndef BZ_NO_STDIO +/* Need a definitition for FILE */ +#include +#endif + +#define BZ_API(func) func +#define BZ_EXTERN extern + + +/*-- Core (low-level) library functions --*/ + +BZ_EXTERN int BZ_API(BZ2_bzCompressInit) ( + bz_stream* strm, + int blockSize100k, + int verbosity, + int workFactor + ); + +BZ_EXTERN int BZ_API(BZ2_bzCompress) ( + bz_stream* strm, + int action + ); + +BZ_EXTERN int BZ_API(BZ2_bzCompressEnd) ( + bz_stream* strm + ); + +BZ_EXTERN int BZ_API(BZ2_bzDecompressInit) ( + bz_stream *strm, + int verbosity, + int lowmem + ); + +BZ_EXTERN int BZ_API(BZ2_bzDecompress) ( + bz_stream* strm + ); + +BZ_EXTERN int BZ_API(BZ2_bzDecompressEnd) ( + bz_stream *strm + ); + + + +/*-- High(er) level library functions --*/ + +#ifndef BZ_NO_STDIO +#define BZ_MAX_UNUSED 5000 + +typedef void BZFILE; + +BZ_EXTERN BZFILE* BZ_API(BZ2_bzReadOpen) ( + int* bzerror, + FILE* f, + int verbosity, + int lowmem, + void* unused, + int nUnused + ); + +BZ_EXTERN void BZ_API(BZ2_bzReadClose) ( + int* bzerror, + BZFILE* b + ); + +BZ_EXTERN void BZ_API(BZ2_bzReadGetUnused) ( + int* bzerror, + BZFILE* b, + void** unused, + int* nUnused + ); + +BZ_EXTERN int BZ_API(BZ2_bzRead) ( + int* bzerror, + BZFILE* b, + void* buf, + int len + ); + +BZ_EXTERN BZFILE* BZ_API(BZ2_bzWriteOpen) ( + int* bzerror, + FILE* f, + int blockSize100k, + int verbosity, + int workFactor + ); + +BZ_EXTERN void BZ_API(BZ2_bzWrite) ( + int* bzerror, + BZFILE* b, + void* buf, + int len + ); + +BZ_EXTERN void BZ_API(BZ2_bzWriteClose) ( + int* bzerror, + BZFILE* b, + int abandon, + unsigned int* nbytes_in, + unsigned int* nbytes_out + ); + +BZ_EXTERN void BZ_API(BZ2_bzWriteClose64) ( + int* bzerror, + BZFILE* b, + int abandon, + unsigned int* nbytes_in_lo32, + unsigned int* nbytes_in_hi32, + unsigned int* nbytes_out_lo32, + unsigned int* nbytes_out_hi32 + ); +#endif + + +/*-- Utility functions --*/ + +BZ_EXTERN int BZ_API(BZ2_bzBuffToBuffCompress) ( + char* dest, + unsigned int* destLen, + const char* source, + unsigned int sourceLen, + int blockSize100k, + int verbosity, + int workFactor + ); + +BZ_EXTERN int BZ_API(BZ2_bzBuffToBuffDecompress) ( + char* dest, + unsigned int* destLen, + const char* source, + unsigned int sourceLen, + int lowmem, + int verbosity + ); + + +/*-- + Code contributed by Yoshioka Tsuneo (tsuneo@rr.iij4u.or.jp) + to support better zlib compatibility. + This code is not _officially_ part of libbzip2 (yet); + I haven't tested it, documented it, or considered the + threading-safeness of it. + If this code breaks, please contact both Yoshioka and me. +--*/ + +BZ_EXTERN const char * BZ_API(BZ2_bzlibVersion) ( + void + ); + +#ifndef BZ_NO_STDIO +BZ_EXTERN BZFILE * BZ_API(BZ2_bzopen) ( + const char *path, + const char *mode + ); + +BZ_EXTERN BZFILE * BZ_API(BZ2_bzdopen) ( + int fd, + const char *mode + ); + +BZ_EXTERN int BZ_API(BZ2_bzread) ( + BZFILE* b, + void* buf, + int len + ); + +BZ_EXTERN int BZ_API(BZ2_bzwrite) ( + BZFILE* b, + void* buf, + int len + ); + +BZ_EXTERN int BZ_API(BZ2_bzflush) ( + BZFILE* b + ); + +BZ_EXTERN void BZ_API(BZ2_bzclose) ( + BZFILE* b + ); + +BZ_EXTERN const char * BZ_API(BZ2_bzerror) ( + BZFILE *b, + int *errnum + ); +#endif + +#ifdef __cplusplus +} +#endif + +#endif + +/*-------------------------------------------------------------*/ +/*--- end bzlib.h ---*/ +/*-------------------------------------------------------------*/ diff --git a/bzip2/bzlib_private.h b/bzip2/bzlib_private.h new file mode 100644 index 0000000000..23427879b1 --- /dev/null +++ b/bzip2/bzlib_private.h @@ -0,0 +1,509 @@ + +/*-------------------------------------------------------------*/ +/*--- Private header file for the library. ---*/ +/*--- bzlib_private.h ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#ifndef _BZLIB_PRIVATE_H +#define _BZLIB_PRIVATE_H + +#include + +#ifndef BZ_NO_STDIO +#include +#include +#include +#endif + +#include "bzlib.h" + + + +/*-- General stuff. --*/ + +#define BZ_VERSION "1.0.5, 10-Dec-2007" + +typedef char Char; +typedef unsigned char Bool; +typedef unsigned char UChar; +typedef int Int32; +typedef unsigned int UInt32; +typedef short Int16; +typedef unsigned short UInt16; + +#define True ((Bool)1) +#define False ((Bool)0) + +#ifndef __GNUC__ +#define __inline__ /* */ +#endif + +#ifndef BZ_NO_STDIO + +extern void BZ2_bz__AssertH__fail ( int errcode ); +#define AssertH(cond,errcode) \ + { if (!(cond)) BZ2_bz__AssertH__fail ( errcode ); } + +#if BZ_DEBUG +#define AssertD(cond,msg) \ + { if (!(cond)) { \ + fprintf ( stderr, \ + "\n\nlibbzip2(debug build): internal error\n\t%s\n", msg );\ + exit(1); \ + }} +#else +#define AssertD(cond,msg) /* */ +#endif + +#define VPrintf0(zf) \ + fprintf(stderr,zf) +#define VPrintf1(zf,za1) \ + fprintf(stderr,zf,za1) +#define VPrintf2(zf,za1,za2) \ + fprintf(stderr,zf,za1,za2) +#define VPrintf3(zf,za1,za2,za3) \ + fprintf(stderr,zf,za1,za2,za3) +#define VPrintf4(zf,za1,za2,za3,za4) \ + fprintf(stderr,zf,za1,za2,za3,za4) +#define VPrintf5(zf,za1,za2,za3,za4,za5) \ + fprintf(stderr,zf,za1,za2,za3,za4,za5) + +#else + +extern void bz_internal_error ( int errcode ); +#define AssertH(cond,errcode) \ + { if (!(cond)) bz_internal_error ( errcode ); } +#define AssertD(cond,msg) do { } while (0) +#define VPrintf0(zf) do { } while (0) +#define VPrintf1(zf,za1) do { } while (0) +#define VPrintf2(zf,za1,za2) do { } while (0) +#define VPrintf3(zf,za1,za2,za3) do { } while (0) +#define VPrintf4(zf,za1,za2,za3,za4) do { } while (0) +#define VPrintf5(zf,za1,za2,za3,za4,za5) do { } while (0) + +#endif + + +#define BZALLOC(nnn) (strm->bzalloc)(strm->opaque,(nnn),1) +#define BZFREE(ppp) (strm->bzfree)(strm->opaque,(ppp)) + + +/*-- Header bytes. --*/ + +#define BZ_HDR_B 0x42 /* 'B' */ +#define BZ_HDR_Z 0x5a /* 'Z' */ +#define BZ_HDR_h 0x68 /* 'h' */ +#define BZ_HDR_0 0x30 /* '0' */ + +/*-- Constants for the back end. --*/ + +#define BZ_MAX_ALPHA_SIZE 258 +#define BZ_MAX_CODE_LEN 23 + +#define BZ_RUNA 0 +#define BZ_RUNB 1 + +#define BZ_N_GROUPS 6 +#define BZ_G_SIZE 50 +#define BZ_N_ITERS 4 + +#define BZ_MAX_SELECTORS (2 + (900000 / BZ_G_SIZE)) + + + +/*-- Stuff for randomising repetitive blocks. --*/ + +extern Int32 BZ2_rNums[512]; + +#define BZ_RAND_DECLS \ + Int32 rNToGo; \ + Int32 rTPos \ + +#define BZ_RAND_INIT_MASK \ + s->rNToGo = 0; \ + s->rTPos = 0 \ + +#define BZ_RAND_MASK ((s->rNToGo == 1) ? 1 : 0) + +#define BZ_RAND_UPD_MASK \ + if (s->rNToGo == 0) { \ + s->rNToGo = BZ2_rNums[s->rTPos]; \ + s->rTPos++; \ + if (s->rTPos == 512) s->rTPos = 0; \ + } \ + s->rNToGo--; + + + +/*-- Stuff for doing CRCs. --*/ + +extern UInt32 BZ2_crc32Table[256]; + +#define BZ_INITIALISE_CRC(crcVar) \ +{ \ + crcVar = 0xffffffffL; \ +} + +#define BZ_FINALISE_CRC(crcVar) \ +{ \ + crcVar = ~(crcVar); \ +} + +#define BZ_UPDATE_CRC(crcVar,cha) \ +{ \ + crcVar = (crcVar << 8) ^ \ + BZ2_crc32Table[(crcVar >> 24) ^ \ + ((UChar)cha)]; \ +} + + + +/*-- States and modes for compression. --*/ + +#define BZ_M_IDLE 1 +#define BZ_M_RUNNING 2 +#define BZ_M_FLUSHING 3 +#define BZ_M_FINISHING 4 + +#define BZ_S_OUTPUT 1 +#define BZ_S_INPUT 2 + +#define BZ_N_RADIX 2 +#define BZ_N_QSORT 12 +#define BZ_N_SHELL 18 +#define BZ_N_OVERSHOOT (BZ_N_RADIX + BZ_N_QSORT + BZ_N_SHELL + 2) + + + + +/*-- Structure holding all the compression-side stuff. --*/ + +typedef + struct { + /* pointer back to the struct bz_stream */ + bz_stream* strm; + + /* mode this stream is in, and whether inputting */ + /* or outputting data */ + Int32 mode; + Int32 state; + + /* remembers avail_in when flush/finish requested */ + UInt32 avail_in_expect; + + /* for doing the block sorting */ + UInt32* arr1; + UInt32* arr2; + UInt32* ftab; + Int32 origPtr; + + /* aliases for arr1 and arr2 */ + UInt32* ptr; + UChar* block; + UInt16* mtfv; + UChar* zbits; + + /* for deciding when to use the fallback sorting algorithm */ + Int32 workFactor; + + /* run-length-encoding of the input */ + UInt32 state_in_ch; + Int32 state_in_len; + BZ_RAND_DECLS; + + /* input and output limits and current posns */ + Int32 nblock; + Int32 nblockMAX; + Int32 numZ; + Int32 state_out_pos; + + /* map of bytes used in block */ + Int32 nInUse; + Bool inUse[256]; + UChar unseqToSeq[256]; + + /* the buffer for bit stream creation */ + UInt32 bsBuff; + Int32 bsLive; + + /* block and combined CRCs */ + UInt32 blockCRC; + UInt32 combinedCRC; + + /* misc administratium */ + Int32 verbosity; + Int32 blockNo; + Int32 blockSize100k; + + /* stuff for coding the MTF values */ + Int32 nMTF; + Int32 mtfFreq [BZ_MAX_ALPHA_SIZE]; + UChar selector [BZ_MAX_SELECTORS]; + UChar selectorMtf[BZ_MAX_SELECTORS]; + + UChar len [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + Int32 code [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + Int32 rfreq [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + /* second dimension: only 3 needed; 4 makes index calculations faster */ + UInt32 len_pack[BZ_MAX_ALPHA_SIZE][4]; + + } + EState; + + + +/*-- externs for compression. --*/ + +extern void +BZ2_blockSort ( EState* ); + +extern void +BZ2_compressBlock ( EState*, Bool ); + +extern void +BZ2_bsInitWrite ( EState* ); + +extern void +BZ2_hbAssignCodes ( Int32*, UChar*, Int32, Int32, Int32 ); + +extern void +BZ2_hbMakeCodeLengths ( UChar*, Int32*, Int32, Int32 ); + + + +/*-- states for decompression. --*/ + +#define BZ_X_IDLE 1 +#define BZ_X_OUTPUT 2 + +#define BZ_X_MAGIC_1 10 +#define BZ_X_MAGIC_2 11 +#define BZ_X_MAGIC_3 12 +#define BZ_X_MAGIC_4 13 +#define BZ_X_BLKHDR_1 14 +#define BZ_X_BLKHDR_2 15 +#define BZ_X_BLKHDR_3 16 +#define BZ_X_BLKHDR_4 17 +#define BZ_X_BLKHDR_5 18 +#define BZ_X_BLKHDR_6 19 +#define BZ_X_BCRC_1 20 +#define BZ_X_BCRC_2 21 +#define BZ_X_BCRC_3 22 +#define BZ_X_BCRC_4 23 +#define BZ_X_RANDBIT 24 +#define BZ_X_ORIGPTR_1 25 +#define BZ_X_ORIGPTR_2 26 +#define BZ_X_ORIGPTR_3 27 +#define BZ_X_MAPPING_1 28 +#define BZ_X_MAPPING_2 29 +#define BZ_X_SELECTOR_1 30 +#define BZ_X_SELECTOR_2 31 +#define BZ_X_SELECTOR_3 32 +#define BZ_X_CODING_1 33 +#define BZ_X_CODING_2 34 +#define BZ_X_CODING_3 35 +#define BZ_X_MTF_1 36 +#define BZ_X_MTF_2 37 +#define BZ_X_MTF_3 38 +#define BZ_X_MTF_4 39 +#define BZ_X_MTF_5 40 +#define BZ_X_MTF_6 41 +#define BZ_X_ENDHDR_2 42 +#define BZ_X_ENDHDR_3 43 +#define BZ_X_ENDHDR_4 44 +#define BZ_X_ENDHDR_5 45 +#define BZ_X_ENDHDR_6 46 +#define BZ_X_CCRC_1 47 +#define BZ_X_CCRC_2 48 +#define BZ_X_CCRC_3 49 +#define BZ_X_CCRC_4 50 + + + +/*-- Constants for the fast MTF decoder. --*/ + +#define MTFA_SIZE 4096 +#define MTFL_SIZE 16 + + + +/*-- Structure holding all the decompression-side stuff. --*/ + +typedef + struct { + /* pointer back to the struct bz_stream */ + bz_stream* strm; + + /* state indicator for this stream */ + Int32 state; + + /* for doing the final run-length decoding */ + UChar state_out_ch; + Int32 state_out_len; + Bool blockRandomised; + BZ_RAND_DECLS; + + /* the buffer for bit stream reading */ + UInt32 bsBuff; + Int32 bsLive; + + /* misc administratium */ + Int32 blockSize100k; + Bool smallDecompress; + Int32 currBlockNo; + Int32 verbosity; + + /* for undoing the Burrows-Wheeler transform */ + Int32 origPtr; + UInt32 tPos; + Int32 k0; + Int32 unzftab[256]; + Int32 nblock_used; + Int32 cftab[257]; + Int32 cftabCopy[257]; + + /* for undoing the Burrows-Wheeler transform (FAST) */ + UInt32 *tt; + + /* for undoing the Burrows-Wheeler transform (SMALL) */ + UInt16 *ll16; + UChar *ll4; + + /* stored and calculated CRCs */ + UInt32 storedBlockCRC; + UInt32 storedCombinedCRC; + UInt32 calculatedBlockCRC; + UInt32 calculatedCombinedCRC; + + /* map of bytes used in block */ + Int32 nInUse; + Bool inUse[256]; + Bool inUse16[16]; + UChar seqToUnseq[256]; + + /* for decoding the MTF values */ + UChar mtfa [MTFA_SIZE]; + Int32 mtfbase[256 / MTFL_SIZE]; + UChar selector [BZ_MAX_SELECTORS]; + UChar selectorMtf[BZ_MAX_SELECTORS]; + UChar len [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + + Int32 limit [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + Int32 base [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + Int32 perm [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + Int32 minLens[BZ_N_GROUPS]; + + /* save area for scalars in the main decompress code */ + Int32 save_i; + Int32 save_j; + Int32 save_t; + Int32 save_alphaSize; + Int32 save_nGroups; + Int32 save_nSelectors; + Int32 save_EOB; + Int32 save_groupNo; + Int32 save_groupPos; + Int32 save_nextSym; + Int32 save_nblockMAX; + Int32 save_nblock; + Int32 save_es; + Int32 save_N; + Int32 save_curr; + Int32 save_zt; + Int32 save_zn; + Int32 save_zvec; + Int32 save_zj; + Int32 save_gSel; + Int32 save_gMinlen; + Int32* save_gLimit; + Int32* save_gBase; + Int32* save_gPerm; + + } + DState; + + + +/*-- Macros for decompression. --*/ + +#define BZ_GET_FAST(cccc) \ + /* c_tPos is unsigned, hence test < 0 is pointless. */ \ + if (s->tPos >= (UInt32)100000 * (UInt32)s->blockSize100k) return True; \ + s->tPos = s->tt[s->tPos]; \ + cccc = (UChar)(s->tPos & 0xff); \ + s->tPos >>= 8; + +#define BZ_GET_FAST_C(cccc) \ + /* c_tPos is unsigned, hence test < 0 is pointless. */ \ + if (c_tPos >= (UInt32)100000 * (UInt32)ro_blockSize100k) return True; \ + c_tPos = c_tt[c_tPos]; \ + cccc = (UChar)(c_tPos & 0xff); \ + c_tPos >>= 8; + +#define SET_LL4(i,n) \ + { if (((i) & 0x1) == 0) \ + s->ll4[(i) >> 1] = (s->ll4[(i) >> 1] & 0xf0) | (n); else \ + s->ll4[(i) >> 1] = (s->ll4[(i) >> 1] & 0x0f) | ((n) << 4); \ + } + +#define GET_LL4(i) \ + ((((UInt32)(s->ll4[(i) >> 1])) >> (((i) << 2) & 0x4)) & 0xF) + +#define SET_LL(i,n) \ + { s->ll16[i] = (UInt16)(n & 0x0000ffff); \ + SET_LL4(i, n >> 16); \ + } + +#define GET_LL(i) \ + (((UInt32)s->ll16[i]) | (GET_LL4(i) << 16)) + +#define BZ_GET_SMALL(cccc) \ + /* c_tPos is unsigned, hence test < 0 is pointless. */ \ + if (s->tPos >= (UInt32)100000 * (UInt32)s->blockSize100k) return True; \ + cccc = BZ2_indexIntoF ( s->tPos, s->cftab ); \ + s->tPos = GET_LL(s->tPos); + + +/*-- externs for decompression. --*/ + +extern Int32 +BZ2_indexIntoF ( Int32, Int32* ); + +extern Int32 +BZ2_decompress ( DState* ); + +extern void +BZ2_hbCreateDecodeTables ( Int32*, Int32*, Int32*, UChar*, + Int32, Int32, Int32 ); + + +#endif + + +/*-- BZ_NO_STDIO seems to make NULL disappear on some platforms. --*/ + +#ifdef BZ_NO_STDIO +#ifndef NULL +#define NULL 0 +#endif +#endif + + +/*-------------------------------------------------------------*/ +/*--- end bzlib_private.h ---*/ +/*-------------------------------------------------------------*/ diff --git a/bzip2/compress.c b/bzip2/compress.c new file mode 100644 index 0000000000..b5df41b127 --- /dev/null +++ b/bzip2/compress.c @@ -0,0 +1,672 @@ + +/*-------------------------------------------------------------*/ +/*--- Compression machinery (not incl block sorting) ---*/ +/*--- compress.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +/* CHANGES + 0.9.0 -- original version. + 0.9.0a/b -- no changes in this file. + 0.9.0c -- changed setting of nGroups in sendMTFValues() + so as to do a bit better on small files +*/ + +#include "bzlib_private.h" + + +/*---------------------------------------------------*/ +/*--- Bit stream I/O ---*/ +/*---------------------------------------------------*/ + +/*---------------------------------------------------*/ +void BZ2_bsInitWrite ( EState* s ) +{ + s->bsLive = 0; + s->bsBuff = 0; +} + + +/*---------------------------------------------------*/ +static +void bsFinishWrite ( EState* s ) +{ + while (s->bsLive > 0) { + s->zbits[s->numZ] = (UChar)(s->bsBuff >> 24); + s->numZ++; + s->bsBuff <<= 8; + s->bsLive -= 8; + } +} + + +/*---------------------------------------------------*/ +#define bsNEEDW(nz) \ +{ \ + while (s->bsLive >= 8) { \ + s->zbits[s->numZ] \ + = (UChar)(s->bsBuff >> 24); \ + s->numZ++; \ + s->bsBuff <<= 8; \ + s->bsLive -= 8; \ + } \ +} + + +/*---------------------------------------------------*/ +static +__inline__ +void bsW ( EState* s, Int32 n, UInt32 v ) +{ + bsNEEDW ( n ); + s->bsBuff |= (v << (32 - s->bsLive - n)); + s->bsLive += n; +} + + +/*---------------------------------------------------*/ +static +void bsPutUInt32 ( EState* s, UInt32 u ) +{ + bsW ( s, 8, (u >> 24) & 0xffL ); + bsW ( s, 8, (u >> 16) & 0xffL ); + bsW ( s, 8, (u >> 8) & 0xffL ); + bsW ( s, 8, u & 0xffL ); +} + + +/*---------------------------------------------------*/ +static +void bsPutUChar ( EState* s, UChar c ) +{ + bsW( s, 8, (UInt32)c ); +} + + +/*---------------------------------------------------*/ +/*--- The back end proper ---*/ +/*---------------------------------------------------*/ + +/*---------------------------------------------------*/ +static +void makeMaps_e ( EState* s ) +{ + Int32 i; + s->nInUse = 0; + for (i = 0; i < 256; i++) + if (s->inUse[i]) { + s->unseqToSeq[i] = s->nInUse; + s->nInUse++; + } +} + + +/*---------------------------------------------------*/ +static +void generateMTFValues ( EState* s ) +{ + UChar yy[256]; + Int32 i, j; + Int32 zPend; + Int32 wr; + Int32 EOB; + + /* + After sorting (eg, here), + s->arr1 [ 0 .. s->nblock-1 ] holds sorted order, + and + ((UChar*)s->arr2) [ 0 .. s->nblock-1 ] + holds the original block data. + + The first thing to do is generate the MTF values, + and put them in + ((UInt16*)s->arr1) [ 0 .. s->nblock-1 ]. + Because there are strictly fewer or equal MTF values + than block values, ptr values in this area are overwritten + with MTF values only when they are no longer needed. + + The final compressed bitstream is generated into the + area starting at + (UChar*) (&((UChar*)s->arr2)[s->nblock]) + + These storage aliases are set up in bzCompressInit(), + except for the last one, which is arranged in + compressBlock(). + */ + UInt32* ptr = s->ptr; + UChar* block = s->block; + UInt16* mtfv = s->mtfv; + + makeMaps_e ( s ); + EOB = s->nInUse+1; + + for (i = 0; i <= EOB; i++) s->mtfFreq[i] = 0; + + wr = 0; + zPend = 0; + for (i = 0; i < s->nInUse; i++) yy[i] = (UChar) i; + + for (i = 0; i < s->nblock; i++) { + UChar ll_i; + AssertD ( wr <= i, "generateMTFValues(1)" ); + j = ptr[i]-1; if (j < 0) j += s->nblock; + ll_i = s->unseqToSeq[block[j]]; + AssertD ( ll_i < s->nInUse, "generateMTFValues(2a)" ); + + if (yy[0] == ll_i) { + zPend++; + } else { + + if (zPend > 0) { + zPend--; + while (True) { + if (zPend & 1) { + mtfv[wr] = BZ_RUNB; wr++; + s->mtfFreq[BZ_RUNB]++; + } else { + mtfv[wr] = BZ_RUNA; wr++; + s->mtfFreq[BZ_RUNA]++; + } + if (zPend < 2) break; + zPend = (zPend - 2) / 2; + }; + zPend = 0; + } + { + register UChar rtmp; + register UChar* ryy_j; + register UChar rll_i; + rtmp = yy[1]; + yy[1] = yy[0]; + ryy_j = &(yy[1]); + rll_i = ll_i; + while ( rll_i != rtmp ) { + register UChar rtmp2; + ryy_j++; + rtmp2 = rtmp; + rtmp = *ryy_j; + *ryy_j = rtmp2; + }; + yy[0] = rtmp; + j = (Int32)(ryy_j - &(yy[0])); + mtfv[wr] = j+1; wr++; s->mtfFreq[j+1]++; + } + + } + } + + if (zPend > 0) { + zPend--; + while (True) { + if (zPend & 1) { + mtfv[wr] = BZ_RUNB; wr++; + s->mtfFreq[BZ_RUNB]++; + } else { + mtfv[wr] = BZ_RUNA; wr++; + s->mtfFreq[BZ_RUNA]++; + } + if (zPend < 2) break; + zPend = (zPend - 2) / 2; + }; + zPend = 0; + } + + mtfv[wr] = EOB; wr++; s->mtfFreq[EOB]++; + + s->nMTF = wr; +} + + +/*---------------------------------------------------*/ +#define BZ_LESSER_ICOST 0 +#define BZ_GREATER_ICOST 15 + +static +void sendMTFValues ( EState* s ) +{ + Int32 v, t, i, j, gs, ge, totc, bt, bc, iter; + Int32 nSelectors, alphaSize, minLen, maxLen, selCtr; + Int32 nGroups, nBytes; + + /*-- + UChar len [BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + is a global since the decoder also needs it. + + Int32 code[BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + Int32 rfreq[BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE]; + are also globals only used in this proc. + Made global to keep stack frame size small. + --*/ + + + UInt16 cost[BZ_N_GROUPS]; + Int32 fave[BZ_N_GROUPS]; + + UInt16* mtfv = s->mtfv; + + if (s->verbosity >= 3) + VPrintf3( " %d in block, %d after MTF & 1-2 coding, " + "%d+2 syms in use\n", + s->nblock, s->nMTF, s->nInUse ); + + alphaSize = s->nInUse+2; + for (t = 0; t < BZ_N_GROUPS; t++) + for (v = 0; v < alphaSize; v++) + s->len[t][v] = BZ_GREATER_ICOST; + + /*--- Decide how many coding tables to use ---*/ + AssertH ( s->nMTF > 0, 3001 ); + if (s->nMTF < 200) nGroups = 2; else + if (s->nMTF < 600) nGroups = 3; else + if (s->nMTF < 1200) nGroups = 4; else + if (s->nMTF < 2400) nGroups = 5; else + nGroups = 6; + + /*--- Generate an initial set of coding tables ---*/ + { + Int32 nPart, remF, tFreq, aFreq; + + nPart = nGroups; + remF = s->nMTF; + gs = 0; + while (nPart > 0) { + tFreq = remF / nPart; + ge = gs-1; + aFreq = 0; + while (aFreq < tFreq && ge < alphaSize-1) { + ge++; + aFreq += s->mtfFreq[ge]; + } + + if (ge > gs + && nPart != nGroups && nPart != 1 + && ((nGroups-nPart) % 2 == 1)) { + aFreq -= s->mtfFreq[ge]; + ge--; + } + + if (s->verbosity >= 3) + VPrintf5( " initial group %d, [%d .. %d], " + "has %d syms (%4.1f%%)\n", + nPart, gs, ge, aFreq, + (100.0 * (float)aFreq) / (float)(s->nMTF) ); + + for (v = 0; v < alphaSize; v++) + if (v >= gs && v <= ge) + s->len[nPart-1][v] = BZ_LESSER_ICOST; else + s->len[nPart-1][v] = BZ_GREATER_ICOST; + + nPart--; + gs = ge+1; + remF -= aFreq; + } + } + + /*--- + Iterate up to BZ_N_ITERS times to improve the tables. + ---*/ + for (iter = 0; iter < BZ_N_ITERS; iter++) { + + for (t = 0; t < nGroups; t++) fave[t] = 0; + + for (t = 0; t < nGroups; t++) + for (v = 0; v < alphaSize; v++) + s->rfreq[t][v] = 0; + + /*--- + Set up an auxiliary length table which is used to fast-track + the common case (nGroups == 6). + ---*/ + if (nGroups == 6) { + for (v = 0; v < alphaSize; v++) { + s->len_pack[v][0] = (s->len[1][v] << 16) | s->len[0][v]; + s->len_pack[v][1] = (s->len[3][v] << 16) | s->len[2][v]; + s->len_pack[v][2] = (s->len[5][v] << 16) | s->len[4][v]; + } + } + + nSelectors = 0; + totc = 0; + gs = 0; + while (True) { + + /*--- Set group start & end marks. --*/ + if (gs >= s->nMTF) break; + ge = gs + BZ_G_SIZE - 1; + if (ge >= s->nMTF) ge = s->nMTF-1; + + /*-- + Calculate the cost of this group as coded + by each of the coding tables. + --*/ + for (t = 0; t < nGroups; t++) cost[t] = 0; + + if (nGroups == 6 && 50 == ge-gs+1) { + /*--- fast track the common case ---*/ + register UInt32 cost01, cost23, cost45; + register UInt16 icv; + cost01 = cost23 = cost45 = 0; + +# define BZ_ITER(nn) \ + icv = mtfv[gs+(nn)]; \ + cost01 += s->len_pack[icv][0]; \ + cost23 += s->len_pack[icv][1]; \ + cost45 += s->len_pack[icv][2]; \ + + BZ_ITER(0); BZ_ITER(1); BZ_ITER(2); BZ_ITER(3); BZ_ITER(4); + BZ_ITER(5); BZ_ITER(6); BZ_ITER(7); BZ_ITER(8); BZ_ITER(9); + BZ_ITER(10); BZ_ITER(11); BZ_ITER(12); BZ_ITER(13); BZ_ITER(14); + BZ_ITER(15); BZ_ITER(16); BZ_ITER(17); BZ_ITER(18); BZ_ITER(19); + BZ_ITER(20); BZ_ITER(21); BZ_ITER(22); BZ_ITER(23); BZ_ITER(24); + BZ_ITER(25); BZ_ITER(26); BZ_ITER(27); BZ_ITER(28); BZ_ITER(29); + BZ_ITER(30); BZ_ITER(31); BZ_ITER(32); BZ_ITER(33); BZ_ITER(34); + BZ_ITER(35); BZ_ITER(36); BZ_ITER(37); BZ_ITER(38); BZ_ITER(39); + BZ_ITER(40); BZ_ITER(41); BZ_ITER(42); BZ_ITER(43); BZ_ITER(44); + BZ_ITER(45); BZ_ITER(46); BZ_ITER(47); BZ_ITER(48); BZ_ITER(49); + +# undef BZ_ITER + + cost[0] = cost01 & 0xffff; cost[1] = cost01 >> 16; + cost[2] = cost23 & 0xffff; cost[3] = cost23 >> 16; + cost[4] = cost45 & 0xffff; cost[5] = cost45 >> 16; + + } else { + /*--- slow version which correctly handles all situations ---*/ + for (i = gs; i <= ge; i++) { + UInt16 icv = mtfv[i]; + for (t = 0; t < nGroups; t++) cost[t] += s->len[t][icv]; + } + } + + /*-- + Find the coding table which is best for this group, + and record its identity in the selector table. + --*/ + bc = 999999999; bt = -1; + for (t = 0; t < nGroups; t++) + if (cost[t] < bc) { bc = cost[t]; bt = t; }; + totc += bc; + fave[bt]++; + s->selector[nSelectors] = bt; + nSelectors++; + + /*-- + Increment the symbol frequencies for the selected table. + --*/ + if (nGroups == 6 && 50 == ge-gs+1) { + /*--- fast track the common case ---*/ + +# define BZ_ITUR(nn) s->rfreq[bt][ mtfv[gs+(nn)] ]++ + + BZ_ITUR(0); BZ_ITUR(1); BZ_ITUR(2); BZ_ITUR(3); BZ_ITUR(4); + BZ_ITUR(5); BZ_ITUR(6); BZ_ITUR(7); BZ_ITUR(8); BZ_ITUR(9); + BZ_ITUR(10); BZ_ITUR(11); BZ_ITUR(12); BZ_ITUR(13); BZ_ITUR(14); + BZ_ITUR(15); BZ_ITUR(16); BZ_ITUR(17); BZ_ITUR(18); BZ_ITUR(19); + BZ_ITUR(20); BZ_ITUR(21); BZ_ITUR(22); BZ_ITUR(23); BZ_ITUR(24); + BZ_ITUR(25); BZ_ITUR(26); BZ_ITUR(27); BZ_ITUR(28); BZ_ITUR(29); + BZ_ITUR(30); BZ_ITUR(31); BZ_ITUR(32); BZ_ITUR(33); BZ_ITUR(34); + BZ_ITUR(35); BZ_ITUR(36); BZ_ITUR(37); BZ_ITUR(38); BZ_ITUR(39); + BZ_ITUR(40); BZ_ITUR(41); BZ_ITUR(42); BZ_ITUR(43); BZ_ITUR(44); + BZ_ITUR(45); BZ_ITUR(46); BZ_ITUR(47); BZ_ITUR(48); BZ_ITUR(49); + +# undef BZ_ITUR + + } else { + /*--- slow version which correctly handles all situations ---*/ + for (i = gs; i <= ge; i++) + s->rfreq[bt][ mtfv[i] ]++; + } + + gs = ge+1; + } + if (s->verbosity >= 3) { + VPrintf2 ( " pass %d: size is %d, grp uses are ", + iter+1, totc/8 ); + for (t = 0; t < nGroups; t++) + VPrintf1 ( "%d ", fave[t] ); + VPrintf0 ( "\n" ); + } + + /*-- + Recompute the tables based on the accumulated frequencies. + --*/ + /* maxLen was changed from 20 to 17 in bzip2-1.0.3. See + comment in huffman.c for details. */ + for (t = 0; t < nGroups; t++) + BZ2_hbMakeCodeLengths ( &(s->len[t][0]), &(s->rfreq[t][0]), + alphaSize, 17 /*20*/ ); + } + + + AssertH( nGroups < 8, 3002 ); + AssertH( nSelectors < 32768 && + nSelectors <= (2 + (900000 / BZ_G_SIZE)), + 3003 ); + + + /*--- Compute MTF values for the selectors. ---*/ + { + UChar pos[BZ_N_GROUPS], ll_i, tmp2, tmp; + for (i = 0; i < nGroups; i++) pos[i] = i; + for (i = 0; i < nSelectors; i++) { + ll_i = s->selector[i]; + j = 0; + tmp = pos[j]; + while ( ll_i != tmp ) { + j++; + tmp2 = tmp; + tmp = pos[j]; + pos[j] = tmp2; + }; + pos[0] = tmp; + s->selectorMtf[i] = j; + } + }; + + /*--- Assign actual codes for the tables. --*/ + for (t = 0; t < nGroups; t++) { + minLen = 32; + maxLen = 0; + for (i = 0; i < alphaSize; i++) { + if (s->len[t][i] > maxLen) maxLen = s->len[t][i]; + if (s->len[t][i] < minLen) minLen = s->len[t][i]; + } + AssertH ( !(maxLen > 17 /*20*/ ), 3004 ); + AssertH ( !(minLen < 1), 3005 ); + BZ2_hbAssignCodes ( &(s->code[t][0]), &(s->len[t][0]), + minLen, maxLen, alphaSize ); + } + + /*--- Transmit the mapping table. ---*/ + { + Bool inUse16[16]; + for (i = 0; i < 16; i++) { + inUse16[i] = False; + for (j = 0; j < 16; j++) + if (s->inUse[i * 16 + j]) inUse16[i] = True; + } + + nBytes = s->numZ; + for (i = 0; i < 16; i++) + if (inUse16[i]) bsW(s,1,1); else bsW(s,1,0); + + for (i = 0; i < 16; i++) + if (inUse16[i]) + for (j = 0; j < 16; j++) { + if (s->inUse[i * 16 + j]) bsW(s,1,1); else bsW(s,1,0); + } + + if (s->verbosity >= 3) + VPrintf1( " bytes: mapping %d, ", s->numZ-nBytes ); + } + + /*--- Now the selectors. ---*/ + nBytes = s->numZ; + bsW ( s, 3, nGroups ); + bsW ( s, 15, nSelectors ); + for (i = 0; i < nSelectors; i++) { + for (j = 0; j < s->selectorMtf[i]; j++) bsW(s,1,1); + bsW(s,1,0); + } + if (s->verbosity >= 3) + VPrintf1( "selectors %d, ", s->numZ-nBytes ); + + /*--- Now the coding tables. ---*/ + nBytes = s->numZ; + + for (t = 0; t < nGroups; t++) { + Int32 curr = s->len[t][0]; + bsW ( s, 5, curr ); + for (i = 0; i < alphaSize; i++) { + while (curr < s->len[t][i]) { bsW(s,2,2); curr++; /* 10 */ }; + while (curr > s->len[t][i]) { bsW(s,2,3); curr--; /* 11 */ }; + bsW ( s, 1, 0 ); + } + } + + if (s->verbosity >= 3) + VPrintf1 ( "code lengths %d, ", s->numZ-nBytes ); + + /*--- And finally, the block data proper ---*/ + nBytes = s->numZ; + selCtr = 0; + gs = 0; + while (True) { + if (gs >= s->nMTF) break; + ge = gs + BZ_G_SIZE - 1; + if (ge >= s->nMTF) ge = s->nMTF-1; + AssertH ( s->selector[selCtr] < nGroups, 3006 ); + + if (nGroups == 6 && 50 == ge-gs+1) { + /*--- fast track the common case ---*/ + UInt16 mtfv_i; + UChar* s_len_sel_selCtr + = &(s->len[s->selector[selCtr]][0]); + Int32* s_code_sel_selCtr + = &(s->code[s->selector[selCtr]][0]); + +# define BZ_ITAH(nn) \ + mtfv_i = mtfv[gs+(nn)]; \ + bsW ( s, \ + s_len_sel_selCtr[mtfv_i], \ + s_code_sel_selCtr[mtfv_i] ) + + BZ_ITAH(0); BZ_ITAH(1); BZ_ITAH(2); BZ_ITAH(3); BZ_ITAH(4); + BZ_ITAH(5); BZ_ITAH(6); BZ_ITAH(7); BZ_ITAH(8); BZ_ITAH(9); + BZ_ITAH(10); BZ_ITAH(11); BZ_ITAH(12); BZ_ITAH(13); BZ_ITAH(14); + BZ_ITAH(15); BZ_ITAH(16); BZ_ITAH(17); BZ_ITAH(18); BZ_ITAH(19); + BZ_ITAH(20); BZ_ITAH(21); BZ_ITAH(22); BZ_ITAH(23); BZ_ITAH(24); + BZ_ITAH(25); BZ_ITAH(26); BZ_ITAH(27); BZ_ITAH(28); BZ_ITAH(29); + BZ_ITAH(30); BZ_ITAH(31); BZ_ITAH(32); BZ_ITAH(33); BZ_ITAH(34); + BZ_ITAH(35); BZ_ITAH(36); BZ_ITAH(37); BZ_ITAH(38); BZ_ITAH(39); + BZ_ITAH(40); BZ_ITAH(41); BZ_ITAH(42); BZ_ITAH(43); BZ_ITAH(44); + BZ_ITAH(45); BZ_ITAH(46); BZ_ITAH(47); BZ_ITAH(48); BZ_ITAH(49); + +# undef BZ_ITAH + + } else { + /*--- slow version which correctly handles all situations ---*/ + for (i = gs; i <= ge; i++) { + bsW ( s, + s->len [s->selector[selCtr]] [mtfv[i]], + s->code [s->selector[selCtr]] [mtfv[i]] ); + } + } + + + gs = ge+1; + selCtr++; + } + AssertH( selCtr == nSelectors, 3007 ); + + if (s->verbosity >= 3) + VPrintf1( "codes %d\n", s->numZ-nBytes ); +} + + +/*---------------------------------------------------*/ +void BZ2_compressBlock ( EState* s, Bool is_last_block ) +{ + if (s->nblock > 0) { + + BZ_FINALISE_CRC ( s->blockCRC ); + s->combinedCRC = (s->combinedCRC << 1) | (s->combinedCRC >> 31); + s->combinedCRC ^= s->blockCRC; + if (s->blockNo > 1) s->numZ = 0; + + if (s->verbosity >= 2) + VPrintf4( " block %d: crc = 0x%08x, " + "combined CRC = 0x%08x, size = %d\n", + s->blockNo, s->blockCRC, s->combinedCRC, s->nblock ); + + BZ2_blockSort ( s ); + } + + s->zbits = (UChar*) (&((UChar*)s->arr2)[s->nblock]); + + /*-- If this is the first block, create the stream header. --*/ + if (s->blockNo == 1) { + BZ2_bsInitWrite ( s ); + bsPutUChar ( s, BZ_HDR_B ); + bsPutUChar ( s, BZ_HDR_Z ); + bsPutUChar ( s, BZ_HDR_h ); + bsPutUChar ( s, (UChar)(BZ_HDR_0 + s->blockSize100k) ); + } + + if (s->nblock > 0) { + + bsPutUChar ( s, 0x31 ); bsPutUChar ( s, 0x41 ); + bsPutUChar ( s, 0x59 ); bsPutUChar ( s, 0x26 ); + bsPutUChar ( s, 0x53 ); bsPutUChar ( s, 0x59 ); + + /*-- Now the block's CRC, so it is in a known place. --*/ + bsPutUInt32 ( s, s->blockCRC ); + + /*-- + Now a single bit indicating (non-)randomisation. + As of version 0.9.5, we use a better sorting algorithm + which makes randomisation unnecessary. So always set + the randomised bit to 'no'. Of course, the decoder + still needs to be able to handle randomised blocks + so as to maintain backwards compatibility with + older versions of bzip2. + --*/ + bsW(s,1,0); + + bsW ( s, 24, s->origPtr ); + generateMTFValues ( s ); + sendMTFValues ( s ); + } + + + /*-- If this is the last block, add the stream trailer. --*/ + if (is_last_block) { + + bsPutUChar ( s, 0x17 ); bsPutUChar ( s, 0x72 ); + bsPutUChar ( s, 0x45 ); bsPutUChar ( s, 0x38 ); + bsPutUChar ( s, 0x50 ); bsPutUChar ( s, 0x90 ); + bsPutUInt32 ( s, s->combinedCRC ); + if (s->verbosity >= 2) + VPrintf1( " final combined CRC = 0x%08x\n ", s->combinedCRC ); + bsFinishWrite ( s ); + } +} + + +/*-------------------------------------------------------------*/ +/*--- end compress.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/bzip2/crctable.c b/bzip2/crctable.c new file mode 100644 index 0000000000..215687b2c0 --- /dev/null +++ b/bzip2/crctable.c @@ -0,0 +1,104 @@ + +/*-------------------------------------------------------------*/ +/*--- Table for doing CRCs ---*/ +/*--- crctable.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#include "bzlib_private.h" + +/*-- + I think this is an implementation of the AUTODIN-II, + Ethernet & FDDI 32-bit CRC standard. Vaguely derived + from code by Rob Warnock, in Section 51 of the + comp.compression FAQ. +--*/ + +UInt32 BZ2_crc32Table[256] = { + + /*-- Ugly, innit? --*/ + + 0x00000000L, 0x04c11db7L, 0x09823b6eL, 0x0d4326d9L, + 0x130476dcL, 0x17c56b6bL, 0x1a864db2L, 0x1e475005L, + 0x2608edb8L, 0x22c9f00fL, 0x2f8ad6d6L, 0x2b4bcb61L, + 0x350c9b64L, 0x31cd86d3L, 0x3c8ea00aL, 0x384fbdbdL, + 0x4c11db70L, 0x48d0c6c7L, 0x4593e01eL, 0x4152fda9L, + 0x5f15adacL, 0x5bd4b01bL, 0x569796c2L, 0x52568b75L, + 0x6a1936c8L, 0x6ed82b7fL, 0x639b0da6L, 0x675a1011L, + 0x791d4014L, 0x7ddc5da3L, 0x709f7b7aL, 0x745e66cdL, + 0x9823b6e0L, 0x9ce2ab57L, 0x91a18d8eL, 0x95609039L, + 0x8b27c03cL, 0x8fe6dd8bL, 0x82a5fb52L, 0x8664e6e5L, + 0xbe2b5b58L, 0xbaea46efL, 0xb7a96036L, 0xb3687d81L, + 0xad2f2d84L, 0xa9ee3033L, 0xa4ad16eaL, 0xa06c0b5dL, + 0xd4326d90L, 0xd0f37027L, 0xddb056feL, 0xd9714b49L, + 0xc7361b4cL, 0xc3f706fbL, 0xceb42022L, 0xca753d95L, + 0xf23a8028L, 0xf6fb9d9fL, 0xfbb8bb46L, 0xff79a6f1L, + 0xe13ef6f4L, 0xe5ffeb43L, 0xe8bccd9aL, 0xec7dd02dL, + 0x34867077L, 0x30476dc0L, 0x3d044b19L, 0x39c556aeL, + 0x278206abL, 0x23431b1cL, 0x2e003dc5L, 0x2ac12072L, + 0x128e9dcfL, 0x164f8078L, 0x1b0ca6a1L, 0x1fcdbb16L, + 0x018aeb13L, 0x054bf6a4L, 0x0808d07dL, 0x0cc9cdcaL, + 0x7897ab07L, 0x7c56b6b0L, 0x71159069L, 0x75d48ddeL, + 0x6b93dddbL, 0x6f52c06cL, 0x6211e6b5L, 0x66d0fb02L, + 0x5e9f46bfL, 0x5a5e5b08L, 0x571d7dd1L, 0x53dc6066L, + 0x4d9b3063L, 0x495a2dd4L, 0x44190b0dL, 0x40d816baL, + 0xaca5c697L, 0xa864db20L, 0xa527fdf9L, 0xa1e6e04eL, + 0xbfa1b04bL, 0xbb60adfcL, 0xb6238b25L, 0xb2e29692L, + 0x8aad2b2fL, 0x8e6c3698L, 0x832f1041L, 0x87ee0df6L, + 0x99a95df3L, 0x9d684044L, 0x902b669dL, 0x94ea7b2aL, + 0xe0b41de7L, 0xe4750050L, 0xe9362689L, 0xedf73b3eL, + 0xf3b06b3bL, 0xf771768cL, 0xfa325055L, 0xfef34de2L, + 0xc6bcf05fL, 0xc27dede8L, 0xcf3ecb31L, 0xcbffd686L, + 0xd5b88683L, 0xd1799b34L, 0xdc3abdedL, 0xd8fba05aL, + 0x690ce0eeL, 0x6dcdfd59L, 0x608edb80L, 0x644fc637L, + 0x7a089632L, 0x7ec98b85L, 0x738aad5cL, 0x774bb0ebL, + 0x4f040d56L, 0x4bc510e1L, 0x46863638L, 0x42472b8fL, + 0x5c007b8aL, 0x58c1663dL, 0x558240e4L, 0x51435d53L, + 0x251d3b9eL, 0x21dc2629L, 0x2c9f00f0L, 0x285e1d47L, + 0x36194d42L, 0x32d850f5L, 0x3f9b762cL, 0x3b5a6b9bL, + 0x0315d626L, 0x07d4cb91L, 0x0a97ed48L, 0x0e56f0ffL, + 0x1011a0faL, 0x14d0bd4dL, 0x19939b94L, 0x1d528623L, + 0xf12f560eL, 0xf5ee4bb9L, 0xf8ad6d60L, 0xfc6c70d7L, + 0xe22b20d2L, 0xe6ea3d65L, 0xeba91bbcL, 0xef68060bL, + 0xd727bbb6L, 0xd3e6a601L, 0xdea580d8L, 0xda649d6fL, + 0xc423cd6aL, 0xc0e2d0ddL, 0xcda1f604L, 0xc960ebb3L, + 0xbd3e8d7eL, 0xb9ff90c9L, 0xb4bcb610L, 0xb07daba7L, + 0xae3afba2L, 0xaafbe615L, 0xa7b8c0ccL, 0xa379dd7bL, + 0x9b3660c6L, 0x9ff77d71L, 0x92b45ba8L, 0x9675461fL, + 0x8832161aL, 0x8cf30badL, 0x81b02d74L, 0x857130c3L, + 0x5d8a9099L, 0x594b8d2eL, 0x5408abf7L, 0x50c9b640L, + 0x4e8ee645L, 0x4a4ffbf2L, 0x470cdd2bL, 0x43cdc09cL, + 0x7b827d21L, 0x7f436096L, 0x7200464fL, 0x76c15bf8L, + 0x68860bfdL, 0x6c47164aL, 0x61043093L, 0x65c52d24L, + 0x119b4be9L, 0x155a565eL, 0x18197087L, 0x1cd86d30L, + 0x029f3d35L, 0x065e2082L, 0x0b1d065bL, 0x0fdc1becL, + 0x3793a651L, 0x3352bbe6L, 0x3e119d3fL, 0x3ad08088L, + 0x2497d08dL, 0x2056cd3aL, 0x2d15ebe3L, 0x29d4f654L, + 0xc5a92679L, 0xc1683bceL, 0xcc2b1d17L, 0xc8ea00a0L, + 0xd6ad50a5L, 0xd26c4d12L, 0xdf2f6bcbL, 0xdbee767cL, + 0xe3a1cbc1L, 0xe760d676L, 0xea23f0afL, 0xeee2ed18L, + 0xf0a5bd1dL, 0xf464a0aaL, 0xf9278673L, 0xfde69bc4L, + 0x89b8fd09L, 0x8d79e0beL, 0x803ac667L, 0x84fbdbd0L, + 0x9abc8bd5L, 0x9e7d9662L, 0x933eb0bbL, 0x97ffad0cL, + 0xafb010b1L, 0xab710d06L, 0xa6322bdfL, 0xa2f33668L, + 0xbcb4666dL, 0xb8757bdaL, 0xb5365d03L, 0xb1f740b4L +}; + + +/*-------------------------------------------------------------*/ +/*--- end crctable.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/bzip2/decompress.c b/bzip2/decompress.c new file mode 100644 index 0000000000..bba5e0fa36 --- /dev/null +++ b/bzip2/decompress.c @@ -0,0 +1,626 @@ + +/*-------------------------------------------------------------*/ +/*--- Decompression machinery ---*/ +/*--- decompress.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#include "bzlib_private.h" + + +/*---------------------------------------------------*/ +static +void makeMaps_d ( DState* s ) +{ + Int32 i; + s->nInUse = 0; + for (i = 0; i < 256; i++) + if (s->inUse[i]) { + s->seqToUnseq[s->nInUse] = i; + s->nInUse++; + } +} + + +/*---------------------------------------------------*/ +#define RETURN(rrr) \ + { retVal = rrr; goto save_state_and_return; }; + +#define GET_BITS(lll,vvv,nnn) \ + case lll: s->state = lll; \ + while (True) { \ + if (s->bsLive >= nnn) { \ + UInt32 v; \ + v = (s->bsBuff >> \ + (s->bsLive-nnn)) & ((1 << nnn)-1); \ + s->bsLive -= nnn; \ + vvv = v; \ + break; \ + } \ + if (s->strm->avail_in == 0) RETURN(BZ_OK); \ + s->bsBuff \ + = (s->bsBuff << 8) | \ + ((UInt32) \ + (*((UChar*)(s->strm->next_in)))); \ + s->bsLive += 8; \ + s->strm->next_in++; \ + s->strm->avail_in--; \ + s->strm->total_in_lo32++; \ + if (s->strm->total_in_lo32 == 0) \ + s->strm->total_in_hi32++; \ + } + +#define GET_UCHAR(lll,uuu) \ + GET_BITS(lll,uuu,8) + +#define GET_BIT(lll,uuu) \ + GET_BITS(lll,uuu,1) + +/*---------------------------------------------------*/ +#define GET_MTF_VAL(label1,label2,lval) \ +{ \ + if (groupPos == 0) { \ + groupNo++; \ + if (groupNo >= nSelectors) \ + RETURN(BZ_DATA_ERROR); \ + groupPos = BZ_G_SIZE; \ + gSel = s->selector[groupNo]; \ + gMinlen = s->minLens[gSel]; \ + gLimit = &(s->limit[gSel][0]); \ + gPerm = &(s->perm[gSel][0]); \ + gBase = &(s->base[gSel][0]); \ + } \ + groupPos--; \ + zn = gMinlen; \ + GET_BITS(label1, zvec, zn); \ + while (1) { \ + if (zn > 20 /* the longest code */) \ + RETURN(BZ_DATA_ERROR); \ + if (zvec <= gLimit[zn]) break; \ + zn++; \ + GET_BIT(label2, zj); \ + zvec = (zvec << 1) | zj; \ + }; \ + if (zvec - gBase[zn] < 0 \ + || zvec - gBase[zn] >= BZ_MAX_ALPHA_SIZE) \ + RETURN(BZ_DATA_ERROR); \ + lval = gPerm[zvec - gBase[zn]]; \ +} + + +/*---------------------------------------------------*/ +Int32 BZ2_decompress ( DState* s ) +{ + UChar uc; + Int32 retVal; + Int32 minLen, maxLen; + bz_stream* strm = s->strm; + + /* stuff that needs to be saved/restored */ + Int32 i; + Int32 j; + Int32 t; + Int32 alphaSize; + Int32 nGroups; + Int32 nSelectors; + Int32 EOB; + Int32 groupNo; + Int32 groupPos; + Int32 nextSym; + Int32 nblockMAX; + Int32 nblock; + Int32 es; + Int32 N; + Int32 curr; + Int32 zt; + Int32 zn; + Int32 zvec; + Int32 zj; + Int32 gSel; + Int32 gMinlen; + Int32* gLimit; + Int32* gBase; + Int32* gPerm; + + if (s->state == BZ_X_MAGIC_1) { + /*initialise the save area*/ + s->save_i = 0; + s->save_j = 0; + s->save_t = 0; + s->save_alphaSize = 0; + s->save_nGroups = 0; + s->save_nSelectors = 0; + s->save_EOB = 0; + s->save_groupNo = 0; + s->save_groupPos = 0; + s->save_nextSym = 0; + s->save_nblockMAX = 0; + s->save_nblock = 0; + s->save_es = 0; + s->save_N = 0; + s->save_curr = 0; + s->save_zt = 0; + s->save_zn = 0; + s->save_zvec = 0; + s->save_zj = 0; + s->save_gSel = 0; + s->save_gMinlen = 0; + s->save_gLimit = NULL; + s->save_gBase = NULL; + s->save_gPerm = NULL; + } + + /*restore from the save area*/ + i = s->save_i; + j = s->save_j; + t = s->save_t; + alphaSize = s->save_alphaSize; + nGroups = s->save_nGroups; + nSelectors = s->save_nSelectors; + EOB = s->save_EOB; + groupNo = s->save_groupNo; + groupPos = s->save_groupPos; + nextSym = s->save_nextSym; + nblockMAX = s->save_nblockMAX; + nblock = s->save_nblock; + es = s->save_es; + N = s->save_N; + curr = s->save_curr; + zt = s->save_zt; + zn = s->save_zn; + zvec = s->save_zvec; + zj = s->save_zj; + gSel = s->save_gSel; + gMinlen = s->save_gMinlen; + gLimit = s->save_gLimit; + gBase = s->save_gBase; + gPerm = s->save_gPerm; + + retVal = BZ_OK; + + switch (s->state) { + + GET_UCHAR(BZ_X_MAGIC_1, uc); + if (uc != BZ_HDR_B) RETURN(BZ_DATA_ERROR_MAGIC); + + GET_UCHAR(BZ_X_MAGIC_2, uc); + if (uc != BZ_HDR_Z) RETURN(BZ_DATA_ERROR_MAGIC); + + GET_UCHAR(BZ_X_MAGIC_3, uc) + if (uc != BZ_HDR_h) RETURN(BZ_DATA_ERROR_MAGIC); + + GET_BITS(BZ_X_MAGIC_4, s->blockSize100k, 8) + if (s->blockSize100k < (BZ_HDR_0 + 1) || + s->blockSize100k > (BZ_HDR_0 + 9)) RETURN(BZ_DATA_ERROR_MAGIC); + s->blockSize100k -= BZ_HDR_0; + + if (s->smallDecompress) { + s->ll16 = BZALLOC( s->blockSize100k * 100000 * sizeof(UInt16) ); + s->ll4 = BZALLOC( + ((1 + s->blockSize100k * 100000) >> 1) * sizeof(UChar) + ); + if (s->ll16 == NULL || s->ll4 == NULL) RETURN(BZ_MEM_ERROR); + } else { + s->tt = BZALLOC( s->blockSize100k * 100000 * sizeof(Int32) ); + if (s->tt == NULL) RETURN(BZ_MEM_ERROR); + } + + GET_UCHAR(BZ_X_BLKHDR_1, uc); + + if (uc == 0x17) goto endhdr_2; + if (uc != 0x31) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_BLKHDR_2, uc); + if (uc != 0x41) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_BLKHDR_3, uc); + if (uc != 0x59) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_BLKHDR_4, uc); + if (uc != 0x26) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_BLKHDR_5, uc); + if (uc != 0x53) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_BLKHDR_6, uc); + if (uc != 0x59) RETURN(BZ_DATA_ERROR); + + s->currBlockNo++; + if (s->verbosity >= 2) + VPrintf1 ( "\n [%d: huff+mtf ", s->currBlockNo ); + + s->storedBlockCRC = 0; + GET_UCHAR(BZ_X_BCRC_1, uc); + s->storedBlockCRC = (s->storedBlockCRC << 8) | ((UInt32)uc); + GET_UCHAR(BZ_X_BCRC_2, uc); + s->storedBlockCRC = (s->storedBlockCRC << 8) | ((UInt32)uc); + GET_UCHAR(BZ_X_BCRC_3, uc); + s->storedBlockCRC = (s->storedBlockCRC << 8) | ((UInt32)uc); + GET_UCHAR(BZ_X_BCRC_4, uc); + s->storedBlockCRC = (s->storedBlockCRC << 8) | ((UInt32)uc); + + GET_BITS(BZ_X_RANDBIT, s->blockRandomised, 1); + + s->origPtr = 0; + GET_UCHAR(BZ_X_ORIGPTR_1, uc); + s->origPtr = (s->origPtr << 8) | ((Int32)uc); + GET_UCHAR(BZ_X_ORIGPTR_2, uc); + s->origPtr = (s->origPtr << 8) | ((Int32)uc); + GET_UCHAR(BZ_X_ORIGPTR_3, uc); + s->origPtr = (s->origPtr << 8) | ((Int32)uc); + + if (s->origPtr < 0) + RETURN(BZ_DATA_ERROR); + if (s->origPtr > 10 + 100000*s->blockSize100k) + RETURN(BZ_DATA_ERROR); + + /*--- Receive the mapping table ---*/ + for (i = 0; i < 16; i++) { + GET_BIT(BZ_X_MAPPING_1, uc); + if (uc == 1) + s->inUse16[i] = True; else + s->inUse16[i] = False; + } + + for (i = 0; i < 256; i++) s->inUse[i] = False; + + for (i = 0; i < 16; i++) + if (s->inUse16[i]) + for (j = 0; j < 16; j++) { + GET_BIT(BZ_X_MAPPING_2, uc); + if (uc == 1) s->inUse[i * 16 + j] = True; + } + makeMaps_d ( s ); + if (s->nInUse == 0) RETURN(BZ_DATA_ERROR); + alphaSize = s->nInUse+2; + + /*--- Now the selectors ---*/ + GET_BITS(BZ_X_SELECTOR_1, nGroups, 3); + if (nGroups < 2 || nGroups > 6) RETURN(BZ_DATA_ERROR); + GET_BITS(BZ_X_SELECTOR_2, nSelectors, 15); + if (nSelectors < 1) RETURN(BZ_DATA_ERROR); + for (i = 0; i < nSelectors; i++) { + j = 0; + while (True) { + GET_BIT(BZ_X_SELECTOR_3, uc); + if (uc == 0) break; + j++; + if (j >= nGroups) RETURN(BZ_DATA_ERROR); + } + s->selectorMtf[i] = j; + } + + /*--- Undo the MTF values for the selectors. ---*/ + { + UChar pos[BZ_N_GROUPS], tmp, v; + for (v = 0; v < nGroups; v++) pos[v] = v; + + for (i = 0; i < nSelectors; i++) { + v = s->selectorMtf[i]; + tmp = pos[v]; + while (v > 0) { pos[v] = pos[v-1]; v--; } + pos[0] = tmp; + s->selector[i] = tmp; + } + } + + /*--- Now the coding tables ---*/ + for (t = 0; t < nGroups; t++) { + GET_BITS(BZ_X_CODING_1, curr, 5); + for (i = 0; i < alphaSize; i++) { + while (True) { + if (curr < 1 || curr > 20) RETURN(BZ_DATA_ERROR); + GET_BIT(BZ_X_CODING_2, uc); + if (uc == 0) break; + GET_BIT(BZ_X_CODING_3, uc); + if (uc == 0) curr++; else curr--; + } + s->len[t][i] = curr; + } + } + + /*--- Create the Huffman decoding tables ---*/ + for (t = 0; t < nGroups; t++) { + minLen = 32; + maxLen = 0; + for (i = 0; i < alphaSize; i++) { + if (s->len[t][i] > maxLen) maxLen = s->len[t][i]; + if (s->len[t][i] < minLen) minLen = s->len[t][i]; + } + BZ2_hbCreateDecodeTables ( + &(s->limit[t][0]), + &(s->base[t][0]), + &(s->perm[t][0]), + &(s->len[t][0]), + minLen, maxLen, alphaSize + ); + s->minLens[t] = minLen; + } + + /*--- Now the MTF values ---*/ + + EOB = s->nInUse+1; + nblockMAX = 100000 * s->blockSize100k; + groupNo = -1; + groupPos = 0; + + for (i = 0; i <= 255; i++) s->unzftab[i] = 0; + + /*-- MTF init --*/ + { + Int32 ii, jj, kk; + kk = MTFA_SIZE-1; + for (ii = 256 / MTFL_SIZE - 1; ii >= 0; ii--) { + for (jj = MTFL_SIZE-1; jj >= 0; jj--) { + s->mtfa[kk] = (UChar)(ii * MTFL_SIZE + jj); + kk--; + } + s->mtfbase[ii] = kk + 1; + } + } + /*-- end MTF init --*/ + + nblock = 0; + GET_MTF_VAL(BZ_X_MTF_1, BZ_X_MTF_2, nextSym); + + while (True) { + + if (nextSym == EOB) break; + + if (nextSym == BZ_RUNA || nextSym == BZ_RUNB) { + + es = -1; + N = 1; + do { + if (nextSym == BZ_RUNA) es = es + (0+1) * N; else + if (nextSym == BZ_RUNB) es = es + (1+1) * N; + N = N * 2; + GET_MTF_VAL(BZ_X_MTF_3, BZ_X_MTF_4, nextSym); + } + while (nextSym == BZ_RUNA || nextSym == BZ_RUNB); + + es++; + uc = s->seqToUnseq[ s->mtfa[s->mtfbase[0]] ]; + s->unzftab[uc] += es; + + if (s->smallDecompress) + while (es > 0) { + if (nblock >= nblockMAX) RETURN(BZ_DATA_ERROR); + s->ll16[nblock] = (UInt16)uc; + nblock++; + es--; + } + else + while (es > 0) { + if (nblock >= nblockMAX) RETURN(BZ_DATA_ERROR); + s->tt[nblock] = (UInt32)uc; + nblock++; + es--; + }; + + continue; + + } else { + + if (nblock >= nblockMAX) RETURN(BZ_DATA_ERROR); + + /*-- uc = MTF ( nextSym-1 ) --*/ + { + Int32 ii, jj, kk, pp, lno, off; + UInt32 nn; + nn = (UInt32)(nextSym - 1); + + if (nn < MTFL_SIZE) { + /* avoid general-case expense */ + pp = s->mtfbase[0]; + uc = s->mtfa[pp+nn]; + while (nn > 3) { + Int32 z = pp+nn; + s->mtfa[(z) ] = s->mtfa[(z)-1]; + s->mtfa[(z)-1] = s->mtfa[(z)-2]; + s->mtfa[(z)-2] = s->mtfa[(z)-3]; + s->mtfa[(z)-3] = s->mtfa[(z)-4]; + nn -= 4; + } + while (nn > 0) { + s->mtfa[(pp+nn)] = s->mtfa[(pp+nn)-1]; nn--; + }; + s->mtfa[pp] = uc; + } else { + /* general case */ + lno = nn / MTFL_SIZE; + off = nn % MTFL_SIZE; + pp = s->mtfbase[lno] + off; + uc = s->mtfa[pp]; + while (pp > s->mtfbase[lno]) { + s->mtfa[pp] = s->mtfa[pp-1]; pp--; + }; + s->mtfbase[lno]++; + while (lno > 0) { + s->mtfbase[lno]--; + s->mtfa[s->mtfbase[lno]] + = s->mtfa[s->mtfbase[lno-1] + MTFL_SIZE - 1]; + lno--; + } + s->mtfbase[0]--; + s->mtfa[s->mtfbase[0]] = uc; + if (s->mtfbase[0] == 0) { + kk = MTFA_SIZE-1; + for (ii = 256 / MTFL_SIZE-1; ii >= 0; ii--) { + for (jj = MTFL_SIZE-1; jj >= 0; jj--) { + s->mtfa[kk] = s->mtfa[s->mtfbase[ii] + jj]; + kk--; + } + s->mtfbase[ii] = kk + 1; + } + } + } + } + /*-- end uc = MTF ( nextSym-1 ) --*/ + + s->unzftab[s->seqToUnseq[uc]]++; + if (s->smallDecompress) + s->ll16[nblock] = (UInt16)(s->seqToUnseq[uc]); else + s->tt[nblock] = (UInt32)(s->seqToUnseq[uc]); + nblock++; + + GET_MTF_VAL(BZ_X_MTF_5, BZ_X_MTF_6, nextSym); + continue; + } + } + + /* Now we know what nblock is, we can do a better sanity + check on s->origPtr. + */ + if (s->origPtr < 0 || s->origPtr >= nblock) + RETURN(BZ_DATA_ERROR); + + /*-- Set up cftab to facilitate generation of T^(-1) --*/ + s->cftab[0] = 0; + for (i = 1; i <= 256; i++) s->cftab[i] = s->unzftab[i-1]; + for (i = 1; i <= 256; i++) s->cftab[i] += s->cftab[i-1]; + for (i = 0; i <= 256; i++) { + if (s->cftab[i] < 0 || s->cftab[i] > nblock) { + /* s->cftab[i] can legitimately be == nblock */ + RETURN(BZ_DATA_ERROR); + } + } + + s->state_out_len = 0; + s->state_out_ch = 0; + BZ_INITIALISE_CRC ( s->calculatedBlockCRC ); + s->state = BZ_X_OUTPUT; + if (s->verbosity >= 2) VPrintf0 ( "rt+rld" ); + + if (s->smallDecompress) { + + /*-- Make a copy of cftab, used in generation of T --*/ + for (i = 0; i <= 256; i++) s->cftabCopy[i] = s->cftab[i]; + + /*-- compute the T vector --*/ + for (i = 0; i < nblock; i++) { + uc = (UChar)(s->ll16[i]); + SET_LL(i, s->cftabCopy[uc]); + s->cftabCopy[uc]++; + } + + /*-- Compute T^(-1) by pointer reversal on T --*/ + i = s->origPtr; + j = GET_LL(i); + do { + Int32 tmp = GET_LL(j); + SET_LL(j, i); + i = j; + j = tmp; + } + while (i != s->origPtr); + + s->tPos = s->origPtr; + s->nblock_used = 0; + if (s->blockRandomised) { + BZ_RAND_INIT_MASK; + BZ_GET_SMALL(s->k0); s->nblock_used++; + BZ_RAND_UPD_MASK; s->k0 ^= BZ_RAND_MASK; + } else { + BZ_GET_SMALL(s->k0); s->nblock_used++; + } + + } else { + + /*-- compute the T^(-1) vector --*/ + for (i = 0; i < nblock; i++) { + uc = (UChar)(s->tt[i] & 0xff); + s->tt[s->cftab[uc]] |= (i << 8); + s->cftab[uc]++; + } + + s->tPos = s->tt[s->origPtr] >> 8; + s->nblock_used = 0; + if (s->blockRandomised) { + BZ_RAND_INIT_MASK; + BZ_GET_FAST(s->k0); s->nblock_used++; + BZ_RAND_UPD_MASK; s->k0 ^= BZ_RAND_MASK; + } else { + BZ_GET_FAST(s->k0); s->nblock_used++; + } + + } + + RETURN(BZ_OK); + + + + endhdr_2: + + GET_UCHAR(BZ_X_ENDHDR_2, uc); + if (uc != 0x72) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_ENDHDR_3, uc); + if (uc != 0x45) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_ENDHDR_4, uc); + if (uc != 0x38) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_ENDHDR_5, uc); + if (uc != 0x50) RETURN(BZ_DATA_ERROR); + GET_UCHAR(BZ_X_ENDHDR_6, uc); + if (uc != 0x90) RETURN(BZ_DATA_ERROR); + + s->storedCombinedCRC = 0; + GET_UCHAR(BZ_X_CCRC_1, uc); + s->storedCombinedCRC = (s->storedCombinedCRC << 8) | ((UInt32)uc); + GET_UCHAR(BZ_X_CCRC_2, uc); + s->storedCombinedCRC = (s->storedCombinedCRC << 8) | ((UInt32)uc); + GET_UCHAR(BZ_X_CCRC_3, uc); + s->storedCombinedCRC = (s->storedCombinedCRC << 8) | ((UInt32)uc); + GET_UCHAR(BZ_X_CCRC_4, uc); + s->storedCombinedCRC = (s->storedCombinedCRC << 8) | ((UInt32)uc); + + s->state = BZ_X_IDLE; + RETURN(BZ_STREAM_END); + + default: AssertH ( False, 4001 ); + } + + AssertH ( False, 4002 ); + + save_state_and_return: + + s->save_i = i; + s->save_j = j; + s->save_t = t; + s->save_alphaSize = alphaSize; + s->save_nGroups = nGroups; + s->save_nSelectors = nSelectors; + s->save_EOB = EOB; + s->save_groupNo = groupNo; + s->save_groupPos = groupPos; + s->save_nextSym = nextSym; + s->save_nblockMAX = nblockMAX; + s->save_nblock = nblock; + s->save_es = es; + s->save_N = N; + s->save_curr = curr; + s->save_zt = zt; + s->save_zn = zn; + s->save_zvec = zvec; + s->save_zj = zj; + s->save_gSel = gSel; + s->save_gMinlen = gMinlen; + s->save_gLimit = gLimit; + s->save_gBase = gBase; + s->save_gPerm = gPerm; + + return retVal; +} + + +/*-------------------------------------------------------------*/ +/*--- end decompress.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/bzip2/huffman.c b/bzip2/huffman.c new file mode 100644 index 0000000000..87e79e38af --- /dev/null +++ b/bzip2/huffman.c @@ -0,0 +1,205 @@ + +/*-------------------------------------------------------------*/ +/*--- Huffman coding low-level stuff ---*/ +/*--- huffman.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#include "bzlib_private.h" + +/*---------------------------------------------------*/ +#define WEIGHTOF(zz0) ((zz0) & 0xffffff00) +#define DEPTHOF(zz1) ((zz1) & 0x000000ff) +#define MYMAX(zz2,zz3) ((zz2) > (zz3) ? (zz2) : (zz3)) + +#define ADDWEIGHTS(zw1,zw2) \ + (WEIGHTOF(zw1)+WEIGHTOF(zw2)) | \ + (1 + MYMAX(DEPTHOF(zw1),DEPTHOF(zw2))) + +#define UPHEAP(z) \ +{ \ + Int32 zz, tmp; \ + zz = z; tmp = heap[zz]; \ + while (weight[tmp] < weight[heap[zz >> 1]]) { \ + heap[zz] = heap[zz >> 1]; \ + zz >>= 1; \ + } \ + heap[zz] = tmp; \ +} + +#define DOWNHEAP(z) \ +{ \ + Int32 zz, yy, tmp; \ + zz = z; tmp = heap[zz]; \ + while (True) { \ + yy = zz << 1; \ + if (yy > nHeap) break; \ + if (yy < nHeap && \ + weight[heap[yy+1]] < weight[heap[yy]]) \ + yy++; \ + if (weight[tmp] < weight[heap[yy]]) break; \ + heap[zz] = heap[yy]; \ + zz = yy; \ + } \ + heap[zz] = tmp; \ +} + + +/*---------------------------------------------------*/ +void BZ2_hbMakeCodeLengths ( UChar *len, + Int32 *freq, + Int32 alphaSize, + Int32 maxLen ) +{ + /*-- + Nodes and heap entries run from 1. Entry 0 + for both the heap and nodes is a sentinel. + --*/ + Int32 nNodes, nHeap, n1, n2, i, j, k; + Bool tooLong; + + Int32 heap [ BZ_MAX_ALPHA_SIZE + 2 ]; + Int32 weight [ BZ_MAX_ALPHA_SIZE * 2 ]; + Int32 parent [ BZ_MAX_ALPHA_SIZE * 2 ]; + + for (i = 0; i < alphaSize; i++) + weight[i+1] = (freq[i] == 0 ? 1 : freq[i]) << 8; + + while (True) { + + nNodes = alphaSize; + nHeap = 0; + + heap[0] = 0; + weight[0] = 0; + parent[0] = -2; + + for (i = 1; i <= alphaSize; i++) { + parent[i] = -1; + nHeap++; + heap[nHeap] = i; + UPHEAP(nHeap); + } + + AssertH( nHeap < (BZ_MAX_ALPHA_SIZE+2), 2001 ); + + while (nHeap > 1) { + n1 = heap[1]; heap[1] = heap[nHeap]; nHeap--; DOWNHEAP(1); + n2 = heap[1]; heap[1] = heap[nHeap]; nHeap--; DOWNHEAP(1); + nNodes++; + parent[n1] = parent[n2] = nNodes; + weight[nNodes] = ADDWEIGHTS(weight[n1], weight[n2]); + parent[nNodes] = -1; + nHeap++; + heap[nHeap] = nNodes; + UPHEAP(nHeap); + } + + AssertH( nNodes < (BZ_MAX_ALPHA_SIZE * 2), 2002 ); + + tooLong = False; + for (i = 1; i <= alphaSize; i++) { + j = 0; + k = i; + while (parent[k] >= 0) { k = parent[k]; j++; } + len[i-1] = j; + if (j > maxLen) tooLong = True; + } + + if (! tooLong) break; + + /* 17 Oct 04: keep-going condition for the following loop used + to be 'i < alphaSize', which missed the last element, + theoretically leading to the possibility of the compressor + looping. However, this count-scaling step is only needed if + one of the generated Huffman code words is longer than + maxLen, which up to and including version 1.0.2 was 20 bits, + which is extremely unlikely. In version 1.0.3 maxLen was + changed to 17 bits, which has minimal effect on compression + ratio, but does mean this scaling step is used from time to + time, enough to verify that it works. + + This means that bzip2-1.0.3 and later will only produce + Huffman codes with a maximum length of 17 bits. However, in + order to preserve backwards compatibility with bitstreams + produced by versions pre-1.0.3, the decompressor must still + handle lengths of up to 20. */ + + for (i = 1; i <= alphaSize; i++) { + j = weight[i] >> 8; + j = 1 + (j / 2); + weight[i] = j << 8; + } + } +} + + +/*---------------------------------------------------*/ +void BZ2_hbAssignCodes ( Int32 *code, + UChar *length, + Int32 minLen, + Int32 maxLen, + Int32 alphaSize ) +{ + Int32 n, vec, i; + + vec = 0; + for (n = minLen; n <= maxLen; n++) { + for (i = 0; i < alphaSize; i++) + if (length[i] == n) { code[i] = vec; vec++; }; + vec <<= 1; + } +} + + +/*---------------------------------------------------*/ +void BZ2_hbCreateDecodeTables ( Int32 *limit, + Int32 *base, + Int32 *perm, + UChar *length, + Int32 minLen, + Int32 maxLen, + Int32 alphaSize ) +{ + Int32 pp, i, j, vec; + + pp = 0; + for (i = minLen; i <= maxLen; i++) + for (j = 0; j < alphaSize; j++) + if (length[j] == i) { perm[pp] = j; pp++; }; + + for (i = 0; i < BZ_MAX_CODE_LEN; i++) base[i] = 0; + for (i = 0; i < alphaSize; i++) base[length[i]+1]++; + + for (i = 1; i < BZ_MAX_CODE_LEN; i++) base[i] += base[i-1]; + + for (i = 0; i < BZ_MAX_CODE_LEN; i++) limit[i] = 0; + vec = 0; + + for (i = minLen; i <= maxLen; i++) { + vec += (base[i+1] - base[i]); + limit[i] = vec-1; + vec <<= 1; + } + for (i = minLen + 1; i <= maxLen; i++) + base[i] = ((limit[i-1] + 1) << 1) - base[i]; +} + + +/*-------------------------------------------------------------*/ +/*--- end huffman.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/bzip2/randtable.c b/bzip2/randtable.c new file mode 100644 index 0000000000..068b76367b --- /dev/null +++ b/bzip2/randtable.c @@ -0,0 +1,84 @@ + +/*-------------------------------------------------------------*/ +/*--- Table for randomising repetitive blocks ---*/ +/*--- randtable.c ---*/ +/*-------------------------------------------------------------*/ + +/* ------------------------------------------------------------------ + This file is part of bzip2/libbzip2, a program and library for + lossless, block-sorting data compression. + + bzip2/libbzip2 version 1.0.5 of 10 December 2007 + Copyright (C) 1996-2007 Julian Seward + + Please read the WARNING, DISCLAIMER and PATENTS sections in the + README file. + + This program is released under the terms of the license contained + in the file LICENSE. + ------------------------------------------------------------------ */ + + +#include "bzlib_private.h" + + +/*---------------------------------------------*/ +Int32 BZ2_rNums[512] = { + 619, 720, 127, 481, 931, 816, 813, 233, 566, 247, + 985, 724, 205, 454, 863, 491, 741, 242, 949, 214, + 733, 859, 335, 708, 621, 574, 73, 654, 730, 472, + 419, 436, 278, 496, 867, 210, 399, 680, 480, 51, + 878, 465, 811, 169, 869, 675, 611, 697, 867, 561, + 862, 687, 507, 283, 482, 129, 807, 591, 733, 623, + 150, 238, 59, 379, 684, 877, 625, 169, 643, 105, + 170, 607, 520, 932, 727, 476, 693, 425, 174, 647, + 73, 122, 335, 530, 442, 853, 695, 249, 445, 515, + 909, 545, 703, 919, 874, 474, 882, 500, 594, 612, + 641, 801, 220, 162, 819, 984, 589, 513, 495, 799, + 161, 604, 958, 533, 221, 400, 386, 867, 600, 782, + 382, 596, 414, 171, 516, 375, 682, 485, 911, 276, + 98, 553, 163, 354, 666, 933, 424, 341, 533, 870, + 227, 730, 475, 186, 263, 647, 537, 686, 600, 224, + 469, 68, 770, 919, 190, 373, 294, 822, 808, 206, + 184, 943, 795, 384, 383, 461, 404, 758, 839, 887, + 715, 67, 618, 276, 204, 918, 873, 777, 604, 560, + 951, 160, 578, 722, 79, 804, 96, 409, 713, 940, + 652, 934, 970, 447, 318, 353, 859, 672, 112, 785, + 645, 863, 803, 350, 139, 93, 354, 99, 820, 908, + 609, 772, 154, 274, 580, 184, 79, 626, 630, 742, + 653, 282, 762, 623, 680, 81, 927, 626, 789, 125, + 411, 521, 938, 300, 821, 78, 343, 175, 128, 250, + 170, 774, 972, 275, 999, 639, 495, 78, 352, 126, + 857, 956, 358, 619, 580, 124, 737, 594, 701, 612, + 669, 112, 134, 694, 363, 992, 809, 743, 168, 974, + 944, 375, 748, 52, 600, 747, 642, 182, 862, 81, + 344, 805, 988, 739, 511, 655, 814, 334, 249, 515, + 897, 955, 664, 981, 649, 113, 974, 459, 893, 228, + 433, 837, 553, 268, 926, 240, 102, 654, 459, 51, + 686, 754, 806, 760, 493, 403, 415, 394, 687, 700, + 946, 670, 656, 610, 738, 392, 760, 799, 887, 653, + 978, 321, 576, 617, 626, 502, 894, 679, 243, 440, + 680, 879, 194, 572, 640, 724, 926, 56, 204, 700, + 707, 151, 457, 449, 797, 195, 791, 558, 945, 679, + 297, 59, 87, 824, 713, 663, 412, 693, 342, 606, + 134, 108, 571, 364, 631, 212, 174, 643, 304, 329, + 343, 97, 430, 751, 497, 314, 983, 374, 822, 928, + 140, 206, 73, 263, 980, 736, 876, 478, 430, 305, + 170, 514, 364, 692, 829, 82, 855, 953, 676, 246, + 369, 970, 294, 750, 807, 827, 150, 790, 288, 923, + 804, 378, 215, 828, 592, 281, 565, 555, 710, 82, + 896, 831, 547, 261, 524, 462, 293, 465, 502, 56, + 661, 821, 976, 991, 658, 869, 905, 758, 745, 193, + 768, 550, 608, 933, 378, 286, 215, 979, 792, 961, + 61, 688, 793, 644, 986, 403, 106, 366, 905, 644, + 372, 567, 466, 434, 645, 210, 389, 550, 919, 135, + 780, 773, 635, 389, 707, 100, 626, 958, 165, 504, + 920, 176, 193, 713, 857, 265, 203, 50, 668, 108, + 645, 990, 626, 197, 510, 357, 358, 850, 858, 364, + 936, 638 +}; + + +/*-------------------------------------------------------------*/ +/*--- end randtable.c ---*/ +/*-------------------------------------------------------------*/ diff --git a/docs/rh-log.txt b/docs/rh-log.txt index dc0517fc8e..7340d557e0 100644 --- a/docs/rh-log.txt +++ b/docs/rh-log.txt @@ -1,7 +1,20 @@ -March 7, 2009 (Changes by Graf Zahl) +March 10, 2009 +- Added support for zip/pk3 files with LZMA and bzip2 compression to ZDoom. +- Added more output to zipdir and a -q option to turn it off. +- Added -u option to zipdir to only recompress those files in a zip that have + changed. + +March 7, 2009 +- Added -d and -f options to zipdir. -d forces deflate compression, and -f + forces a write of the zip, even if it's newer than all the files it contains. + +March 7, 2009 (Changes by Graf Zahl) - Adjusted some gravity related thresholds for the fix from Feb 28. Also removed some unnecessary floating point math from this code. +March 6, 2009 +- Added support for bzip2 and LZMA compression to zipdir. + March 6, 2009 (Changes by Graf Zahl) - Added Hirogen2's patch for unlimited ammo CVAR. diff --git a/lzma/C/7zVersion.h b/lzma/C/7zVersion.h new file mode 100644 index 0000000000..5a6bcadac6 --- /dev/null +++ b/lzma/C/7zVersion.h @@ -0,0 +1,7 @@ +#define MY_VER_MAJOR 4 +#define MY_VER_MINOR 65 +#define MY_VER_BUILD 0 +#define MY_VERSION "4.65" +#define MY_DATE "2009-02-03" +#define MY_COPYRIGHT ": Igor Pavlov : Public domain" +#define MY_VERSION_COPYRIGHT_DATE MY_VERSION " " MY_COPYRIGHT " : " MY_DATE diff --git a/lzma/C/Alloc.c b/lzma/C/Alloc.c new file mode 100644 index 0000000000..bb24a772b3 --- /dev/null +++ b/lzma/C/Alloc.c @@ -0,0 +1,127 @@ +/* Alloc.c -- Memory allocation functions +2008-09-24 +Igor Pavlov +Public domain */ + +#ifdef _WIN32 +#include +#endif +#include + +#include "Alloc.h" + +/* #define _SZ_ALLOC_DEBUG */ + +/* use _SZ_ALLOC_DEBUG to debug alloc/free operations */ +#ifdef _SZ_ALLOC_DEBUG +#include +int g_allocCount = 0; +int g_allocCountMid = 0; +int g_allocCountBig = 0; +#endif + +void *MyAlloc(size_t size) +{ + if (size == 0) + return 0; + #ifdef _SZ_ALLOC_DEBUG + { + void *p = malloc(size); + fprintf(stderr, "\nAlloc %10d bytes, count = %10d, addr = %8X", size, g_allocCount++, (unsigned)p); + return p; + } + #else + return malloc(size); + #endif +} + +void MyFree(void *address) +{ + #ifdef _SZ_ALLOC_DEBUG + if (address != 0) + fprintf(stderr, "\nFree; count = %10d, addr = %8X", --g_allocCount, (unsigned)address); + #endif + free(address); +} + +#ifdef _WIN32 + +void *MidAlloc(size_t size) +{ + if (size == 0) + return 0; + #ifdef _SZ_ALLOC_DEBUG + fprintf(stderr, "\nAlloc_Mid %10d bytes; count = %10d", size, g_allocCountMid++); + #endif + return VirtualAlloc(0, size, MEM_COMMIT, PAGE_READWRITE); +} + +void MidFree(void *address) +{ + #ifdef _SZ_ALLOC_DEBUG + if (address != 0) + fprintf(stderr, "\nFree_Mid; count = %10d", --g_allocCountMid); + #endif + if (address == 0) + return; + VirtualFree(address, 0, MEM_RELEASE); +} + +#ifndef MEM_LARGE_PAGES +#undef _7ZIP_LARGE_PAGES +#endif + +#ifdef _7ZIP_LARGE_PAGES +SIZE_T g_LargePageSize = 0; +typedef SIZE_T (WINAPI *GetLargePageMinimumP)(); +#endif + +void SetLargePageSize() +{ + #ifdef _7ZIP_LARGE_PAGES + SIZE_T size = 0; + GetLargePageMinimumP largePageMinimum = (GetLargePageMinimumP) + GetProcAddress(GetModuleHandle(TEXT("kernel32.dll")), "GetLargePageMinimum"); + if (largePageMinimum == 0) + return; + size = largePageMinimum(); + if (size == 0 || (size & (size - 1)) != 0) + return; + g_LargePageSize = size; + #endif +} + + +void *BigAlloc(size_t size) +{ + if (size == 0) + return 0; + #ifdef _SZ_ALLOC_DEBUG + fprintf(stderr, "\nAlloc_Big %10d bytes; count = %10d", size, g_allocCountBig++); + #endif + + #ifdef _7ZIP_LARGE_PAGES + if (g_LargePageSize != 0 && g_LargePageSize <= (1 << 30) && size >= (1 << 18)) + { + void *res = VirtualAlloc(0, (size + g_LargePageSize - 1) & (~(g_LargePageSize - 1)), + MEM_COMMIT | MEM_LARGE_PAGES, PAGE_READWRITE); + if (res != 0) + return res; + } + #endif + return VirtualAlloc(0, size, MEM_COMMIT, PAGE_READWRITE); +} + +void BigFree(void *address) +{ + #ifdef _SZ_ALLOC_DEBUG + if (address != 0) + fprintf(stderr, "\nFree_Big; count = %10d", --g_allocCountBig); + #endif + + if (address == 0) + return; + VirtualFree(address, 0, MEM_RELEASE); +} + +#endif diff --git a/lzma/C/Alloc.h b/lzma/C/Alloc.h new file mode 100644 index 0000000000..a396c6b9e1 --- /dev/null +++ b/lzma/C/Alloc.h @@ -0,0 +1,32 @@ +/* Alloc.h -- Memory allocation functions +2008-03-13 +Igor Pavlov +Public domain */ + +#ifndef __COMMON_ALLOC_H +#define __COMMON_ALLOC_H + +#include + +void *MyAlloc(size_t size); +void MyFree(void *address); + +#ifdef _WIN32 + +void SetLargePageSize(); + +void *MidAlloc(size_t size); +void MidFree(void *address); +void *BigAlloc(size_t size); +void BigFree(void *address); + +#else + +#define MidAlloc(size) MyAlloc(size) +#define MidFree(address) MyFree(address) +#define BigAlloc(size) MyAlloc(size) +#define BigFree(address) MyFree(address) + +#endif + +#endif diff --git a/lzma/C/LzFind.c b/lzma/C/LzFind.c new file mode 100644 index 0000000000..e0ebe62356 --- /dev/null +++ b/lzma/C/LzFind.c @@ -0,0 +1,751 @@ +/* LzFind.c -- Match finder for LZ algorithms +2008-10-04 : Igor Pavlov : Public domain */ + +#include + +#include "LzFind.h" +#include "LzHash.h" + +#define kEmptyHashValue 0 +#define kMaxValForNormalize ((UInt32)0xFFFFFFFF) +#define kNormalizeStepMin (1 << 10) /* it must be power of 2 */ +#define kNormalizeMask (~(kNormalizeStepMin - 1)) +#define kMaxHistorySize ((UInt32)3 << 30) + +#define kStartMaxLen 3 + +static void LzInWindow_Free(CMatchFinder *p, ISzAlloc *alloc) +{ + if (!p->directInput) + { + alloc->Free(alloc, p->bufferBase); + p->bufferBase = 0; + } +} + +/* keepSizeBefore + keepSizeAfter + keepSizeReserv must be < 4G) */ + +static int LzInWindow_Create(CMatchFinder *p, UInt32 keepSizeReserv, ISzAlloc *alloc) +{ + UInt32 blockSize = p->keepSizeBefore + p->keepSizeAfter + keepSizeReserv; + if (p->directInput) + { + p->blockSize = blockSize; + return 1; + } + if (p->bufferBase == 0 || p->blockSize != blockSize) + { + LzInWindow_Free(p, alloc); + p->blockSize = blockSize; + p->bufferBase = (Byte *)alloc->Alloc(alloc, (size_t)blockSize); + } + return (p->bufferBase != 0); +} + +Byte *MatchFinder_GetPointerToCurrentPos(CMatchFinder *p) { return p->buffer; } +Byte MatchFinder_GetIndexByte(CMatchFinder *p, Int32 index) { return p->buffer[index]; } + +UInt32 MatchFinder_GetNumAvailableBytes(CMatchFinder *p) { return p->streamPos - p->pos; } + +void MatchFinder_ReduceOffsets(CMatchFinder *p, UInt32 subValue) +{ + p->posLimit -= subValue; + p->pos -= subValue; + p->streamPos -= subValue; +} + +static void MatchFinder_ReadBlock(CMatchFinder *p) +{ + if (p->streamEndWasReached || p->result != SZ_OK) + return; + for (;;) + { + Byte *dest = p->buffer + (p->streamPos - p->pos); + size_t size = (p->bufferBase + p->blockSize - dest); + if (size == 0) + return; + p->result = p->stream->Read(p->stream, dest, &size); + if (p->result != SZ_OK) + return; + if (size == 0) + { + p->streamEndWasReached = 1; + return; + } + p->streamPos += (UInt32)size; + if (p->streamPos - p->pos > p->keepSizeAfter) + return; + } +} + +void MatchFinder_MoveBlock(CMatchFinder *p) +{ + memmove(p->bufferBase, + p->buffer - p->keepSizeBefore, + (size_t)(p->streamPos - p->pos + p->keepSizeBefore)); + p->buffer = p->bufferBase + p->keepSizeBefore; +} + +int MatchFinder_NeedMove(CMatchFinder *p) +{ + /* if (p->streamEndWasReached) return 0; */ + return ((size_t)(p->bufferBase + p->blockSize - p->buffer) <= p->keepSizeAfter); +} + +void MatchFinder_ReadIfRequired(CMatchFinder *p) +{ + if (p->streamEndWasReached) + return; + if (p->keepSizeAfter >= p->streamPos - p->pos) + MatchFinder_ReadBlock(p); +} + +static void MatchFinder_CheckAndMoveAndRead(CMatchFinder *p) +{ + if (MatchFinder_NeedMove(p)) + MatchFinder_MoveBlock(p); + MatchFinder_ReadBlock(p); +} + +static void MatchFinder_SetDefaultSettings(CMatchFinder *p) +{ + p->cutValue = 32; + p->btMode = 1; + p->numHashBytes = 4; + /* p->skipModeBits = 0; */ + p->directInput = 0; + p->bigHash = 0; +} + +#define kCrcPoly 0xEDB88320 + +void MatchFinder_Construct(CMatchFinder *p) +{ + UInt32 i; + p->bufferBase = 0; + p->directInput = 0; + p->hash = 0; + MatchFinder_SetDefaultSettings(p); + + for (i = 0; i < 256; i++) + { + UInt32 r = i; + int j; + for (j = 0; j < 8; j++) + r = (r >> 1) ^ (kCrcPoly & ~((r & 1) - 1)); + p->crc[i] = r; + } +} + +static void MatchFinder_FreeThisClassMemory(CMatchFinder *p, ISzAlloc *alloc) +{ + alloc->Free(alloc, p->hash); + p->hash = 0; +} + +void MatchFinder_Free(CMatchFinder *p, ISzAlloc *alloc) +{ + MatchFinder_FreeThisClassMemory(p, alloc); + LzInWindow_Free(p, alloc); +} + +static CLzRef* AllocRefs(UInt32 num, ISzAlloc *alloc) +{ + size_t sizeInBytes = (size_t)num * sizeof(CLzRef); + if (sizeInBytes / sizeof(CLzRef) != num) + return 0; + return (CLzRef *)alloc->Alloc(alloc, sizeInBytes); +} + +int MatchFinder_Create(CMatchFinder *p, UInt32 historySize, + UInt32 keepAddBufferBefore, UInt32 matchMaxLen, UInt32 keepAddBufferAfter, + ISzAlloc *alloc) +{ + UInt32 sizeReserv; + if (historySize > kMaxHistorySize) + { + MatchFinder_Free(p, alloc); + return 0; + } + sizeReserv = historySize >> 1; + if (historySize > ((UInt32)2 << 30)) + sizeReserv = historySize >> 2; + sizeReserv += (keepAddBufferBefore + matchMaxLen + keepAddBufferAfter) / 2 + (1 << 19); + + p->keepSizeBefore = historySize + keepAddBufferBefore + 1; + p->keepSizeAfter = matchMaxLen + keepAddBufferAfter; + /* we need one additional byte, since we use MoveBlock after pos++ and before dictionary using */ + if (LzInWindow_Create(p, sizeReserv, alloc)) + { + UInt32 newCyclicBufferSize = (historySize /* >> p->skipModeBits */) + 1; + UInt32 hs; + p->matchMaxLen = matchMaxLen; + { + p->fixedHashSize = 0; + if (p->numHashBytes == 2) + hs = (1 << 16) - 1; + else + { + hs = historySize - 1; + hs |= (hs >> 1); + hs |= (hs >> 2); + hs |= (hs >> 4); + hs |= (hs >> 8); + hs >>= 1; + /* hs >>= p->skipModeBits; */ + hs |= 0xFFFF; /* don't change it! It's required for Deflate */ + if (hs > (1 << 24)) + { + if (p->numHashBytes == 3) + hs = (1 << 24) - 1; + else + hs >>= 1; + } + } + p->hashMask = hs; + hs++; + if (p->numHashBytes > 2) p->fixedHashSize += kHash2Size; + if (p->numHashBytes > 3) p->fixedHashSize += kHash3Size; + if (p->numHashBytes > 4) p->fixedHashSize += kHash4Size; + hs += p->fixedHashSize; + } + + { + UInt32 prevSize = p->hashSizeSum + p->numSons; + UInt32 newSize; + p->historySize = historySize; + p->hashSizeSum = hs; + p->cyclicBufferSize = newCyclicBufferSize; + p->numSons = (p->btMode ? newCyclicBufferSize * 2 : newCyclicBufferSize); + newSize = p->hashSizeSum + p->numSons; + if (p->hash != 0 && prevSize == newSize) + return 1; + MatchFinder_FreeThisClassMemory(p, alloc); + p->hash = AllocRefs(newSize, alloc); + if (p->hash != 0) + { + p->son = p->hash + p->hashSizeSum; + return 1; + } + } + } + MatchFinder_Free(p, alloc); + return 0; +} + +static void MatchFinder_SetLimits(CMatchFinder *p) +{ + UInt32 limit = kMaxValForNormalize - p->pos; + UInt32 limit2 = p->cyclicBufferSize - p->cyclicBufferPos; + if (limit2 < limit) + limit = limit2; + limit2 = p->streamPos - p->pos; + if (limit2 <= p->keepSizeAfter) + { + if (limit2 > 0) + limit2 = 1; + } + else + limit2 -= p->keepSizeAfter; + if (limit2 < limit) + limit = limit2; + { + UInt32 lenLimit = p->streamPos - p->pos; + if (lenLimit > p->matchMaxLen) + lenLimit = p->matchMaxLen; + p->lenLimit = lenLimit; + } + p->posLimit = p->pos + limit; +} + +void MatchFinder_Init(CMatchFinder *p) +{ + UInt32 i; + for (i = 0; i < p->hashSizeSum; i++) + p->hash[i] = kEmptyHashValue; + p->cyclicBufferPos = 0; + p->buffer = p->bufferBase; + p->pos = p->streamPos = p->cyclicBufferSize; + p->result = SZ_OK; + p->streamEndWasReached = 0; + MatchFinder_ReadBlock(p); + MatchFinder_SetLimits(p); +} + +static UInt32 MatchFinder_GetSubValue(CMatchFinder *p) +{ + return (p->pos - p->historySize - 1) & kNormalizeMask; +} + +void MatchFinder_Normalize3(UInt32 subValue, CLzRef *items, UInt32 numItems) +{ + UInt32 i; + for (i = 0; i < numItems; i++) + { + UInt32 value = items[i]; + if (value <= subValue) + value = kEmptyHashValue; + else + value -= subValue; + items[i] = value; + } +} + +static void MatchFinder_Normalize(CMatchFinder *p) +{ + UInt32 subValue = MatchFinder_GetSubValue(p); + MatchFinder_Normalize3(subValue, p->hash, p->hashSizeSum + p->numSons); + MatchFinder_ReduceOffsets(p, subValue); +} + +static void MatchFinder_CheckLimits(CMatchFinder *p) +{ + if (p->pos == kMaxValForNormalize) + MatchFinder_Normalize(p); + if (!p->streamEndWasReached && p->keepSizeAfter == p->streamPos - p->pos) + MatchFinder_CheckAndMoveAndRead(p); + if (p->cyclicBufferPos == p->cyclicBufferSize) + p->cyclicBufferPos = 0; + MatchFinder_SetLimits(p); +} + +static UInt32 * Hc_GetMatchesSpec(UInt32 lenLimit, UInt32 curMatch, UInt32 pos, const Byte *cur, CLzRef *son, + UInt32 _cyclicBufferPos, UInt32 _cyclicBufferSize, UInt32 cutValue, + UInt32 *distances, UInt32 maxLen) +{ + son[_cyclicBufferPos] = curMatch; + for (;;) + { + UInt32 delta = pos - curMatch; + if (cutValue-- == 0 || delta >= _cyclicBufferSize) + return distances; + { + const Byte *pb = cur - delta; + curMatch = son[_cyclicBufferPos - delta + ((delta > _cyclicBufferPos) ? _cyclicBufferSize : 0)]; + if (pb[maxLen] == cur[maxLen] && *pb == *cur) + { + UInt32 len = 0; + while (++len != lenLimit) + if (pb[len] != cur[len]) + break; + if (maxLen < len) + { + *distances++ = maxLen = len; + *distances++ = delta - 1; + if (len == lenLimit) + return distances; + } + } + } + } +} + +UInt32 * GetMatchesSpec1(UInt32 lenLimit, UInt32 curMatch, UInt32 pos, const Byte *cur, CLzRef *son, + UInt32 _cyclicBufferPos, UInt32 _cyclicBufferSize, UInt32 cutValue, + UInt32 *distances, UInt32 maxLen) +{ + CLzRef *ptr0 = son + (_cyclicBufferPos << 1) + 1; + CLzRef *ptr1 = son + (_cyclicBufferPos << 1); + UInt32 len0 = 0, len1 = 0; + for (;;) + { + UInt32 delta = pos - curMatch; + if (cutValue-- == 0 || delta >= _cyclicBufferSize) + { + *ptr0 = *ptr1 = kEmptyHashValue; + return distances; + } + { + CLzRef *pair = son + ((_cyclicBufferPos - delta + ((delta > _cyclicBufferPos) ? _cyclicBufferSize : 0)) << 1); + const Byte *pb = cur - delta; + UInt32 len = (len0 < len1 ? len0 : len1); + if (pb[len] == cur[len]) + { + if (++len != lenLimit && pb[len] == cur[len]) + while (++len != lenLimit) + if (pb[len] != cur[len]) + break; + if (maxLen < len) + { + *distances++ = maxLen = len; + *distances++ = delta - 1; + if (len == lenLimit) + { + *ptr1 = pair[0]; + *ptr0 = pair[1]; + return distances; + } + } + } + if (pb[len] < cur[len]) + { + *ptr1 = curMatch; + ptr1 = pair + 1; + curMatch = *ptr1; + len1 = len; + } + else + { + *ptr0 = curMatch; + ptr0 = pair; + curMatch = *ptr0; + len0 = len; + } + } + } +} + +static void SkipMatchesSpec(UInt32 lenLimit, UInt32 curMatch, UInt32 pos, const Byte *cur, CLzRef *son, + UInt32 _cyclicBufferPos, UInt32 _cyclicBufferSize, UInt32 cutValue) +{ + CLzRef *ptr0 = son + (_cyclicBufferPos << 1) + 1; + CLzRef *ptr1 = son + (_cyclicBufferPos << 1); + UInt32 len0 = 0, len1 = 0; + for (;;) + { + UInt32 delta = pos - curMatch; + if (cutValue-- == 0 || delta >= _cyclicBufferSize) + { + *ptr0 = *ptr1 = kEmptyHashValue; + return; + } + { + CLzRef *pair = son + ((_cyclicBufferPos - delta + ((delta > _cyclicBufferPos) ? _cyclicBufferSize : 0)) << 1); + const Byte *pb = cur - delta; + UInt32 len = (len0 < len1 ? len0 : len1); + if (pb[len] == cur[len]) + { + while (++len != lenLimit) + if (pb[len] != cur[len]) + break; + { + if (len == lenLimit) + { + *ptr1 = pair[0]; + *ptr0 = pair[1]; + return; + } + } + } + if (pb[len] < cur[len]) + { + *ptr1 = curMatch; + ptr1 = pair + 1; + curMatch = *ptr1; + len1 = len; + } + else + { + *ptr0 = curMatch; + ptr0 = pair; + curMatch = *ptr0; + len0 = len; + } + } + } +} + +#define MOVE_POS \ + ++p->cyclicBufferPos; \ + p->buffer++; \ + if (++p->pos == p->posLimit) MatchFinder_CheckLimits(p); + +#define MOVE_POS_RET MOVE_POS return offset; + +static void MatchFinder_MovePos(CMatchFinder *p) { MOVE_POS; } + +#define GET_MATCHES_HEADER2(minLen, ret_op) \ + UInt32 lenLimit; UInt32 hashValue; const Byte *cur; UInt32 curMatch; \ + lenLimit = p->lenLimit; { if (lenLimit < minLen) { MatchFinder_MovePos(p); ret_op; }} \ + cur = p->buffer; + +#define GET_MATCHES_HEADER(minLen) GET_MATCHES_HEADER2(minLen, return 0) +#define SKIP_HEADER(minLen) GET_MATCHES_HEADER2(minLen, continue) + +#define MF_PARAMS(p) p->pos, p->buffer, p->son, p->cyclicBufferPos, p->cyclicBufferSize, p->cutValue + +#define GET_MATCHES_FOOTER(offset, maxLen) \ + offset = (UInt32)(GetMatchesSpec1(lenLimit, curMatch, MF_PARAMS(p), \ + distances + offset, maxLen) - distances); MOVE_POS_RET; + +#define SKIP_FOOTER \ + SkipMatchesSpec(lenLimit, curMatch, MF_PARAMS(p)); MOVE_POS; + +static UInt32 Bt2_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) +{ + UInt32 offset; + GET_MATCHES_HEADER(2) + HASH2_CALC; + curMatch = p->hash[hashValue]; + p->hash[hashValue] = p->pos; + offset = 0; + GET_MATCHES_FOOTER(offset, 1) +} + +UInt32 Bt3Zip_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) +{ + UInt32 offset; + GET_MATCHES_HEADER(3) + HASH_ZIP_CALC; + curMatch = p->hash[hashValue]; + p->hash[hashValue] = p->pos; + offset = 0; + GET_MATCHES_FOOTER(offset, 2) +} + +static UInt32 Bt3_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) +{ + UInt32 hash2Value, delta2, maxLen, offset; + GET_MATCHES_HEADER(3) + + HASH3_CALC; + + delta2 = p->pos - p->hash[hash2Value]; + curMatch = p->hash[kFix3HashSize + hashValue]; + + p->hash[hash2Value] = + p->hash[kFix3HashSize + hashValue] = p->pos; + + + maxLen = 2; + offset = 0; + if (delta2 < p->cyclicBufferSize && *(cur - delta2) == *cur) + { + for (; maxLen != lenLimit; maxLen++) + if (cur[(ptrdiff_t)maxLen - delta2] != cur[maxLen]) + break; + distances[0] = maxLen; + distances[1] = delta2 - 1; + offset = 2; + if (maxLen == lenLimit) + { + SkipMatchesSpec(lenLimit, curMatch, MF_PARAMS(p)); + MOVE_POS_RET; + } + } + GET_MATCHES_FOOTER(offset, maxLen) +} + +static UInt32 Bt4_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) +{ + UInt32 hash2Value, hash3Value, delta2, delta3, maxLen, offset; + GET_MATCHES_HEADER(4) + + HASH4_CALC; + + delta2 = p->pos - p->hash[ hash2Value]; + delta3 = p->pos - p->hash[kFix3HashSize + hash3Value]; + curMatch = p->hash[kFix4HashSize + hashValue]; + + p->hash[ hash2Value] = + p->hash[kFix3HashSize + hash3Value] = + p->hash[kFix4HashSize + hashValue] = p->pos; + + maxLen = 1; + offset = 0; + if (delta2 < p->cyclicBufferSize && *(cur - delta2) == *cur) + { + distances[0] = maxLen = 2; + distances[1] = delta2 - 1; + offset = 2; + } + if (delta2 != delta3 && delta3 < p->cyclicBufferSize && *(cur - delta3) == *cur) + { + maxLen = 3; + distances[offset + 1] = delta3 - 1; + offset += 2; + delta2 = delta3; + } + if (offset != 0) + { + for (; maxLen != lenLimit; maxLen++) + if (cur[(ptrdiff_t)maxLen - delta2] != cur[maxLen]) + break; + distances[offset - 2] = maxLen; + if (maxLen == lenLimit) + { + SkipMatchesSpec(lenLimit, curMatch, MF_PARAMS(p)); + MOVE_POS_RET; + } + } + if (maxLen < 3) + maxLen = 3; + GET_MATCHES_FOOTER(offset, maxLen) +} + +static UInt32 Hc4_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) +{ + UInt32 hash2Value, hash3Value, delta2, delta3, maxLen, offset; + GET_MATCHES_HEADER(4) + + HASH4_CALC; + + delta2 = p->pos - p->hash[ hash2Value]; + delta3 = p->pos - p->hash[kFix3HashSize + hash3Value]; + curMatch = p->hash[kFix4HashSize + hashValue]; + + p->hash[ hash2Value] = + p->hash[kFix3HashSize + hash3Value] = + p->hash[kFix4HashSize + hashValue] = p->pos; + + maxLen = 1; + offset = 0; + if (delta2 < p->cyclicBufferSize && *(cur - delta2) == *cur) + { + distances[0] = maxLen = 2; + distances[1] = delta2 - 1; + offset = 2; + } + if (delta2 != delta3 && delta3 < p->cyclicBufferSize && *(cur - delta3) == *cur) + { + maxLen = 3; + distances[offset + 1] = delta3 - 1; + offset += 2; + delta2 = delta3; + } + if (offset != 0) + { + for (; maxLen != lenLimit; maxLen++) + if (cur[(ptrdiff_t)maxLen - delta2] != cur[maxLen]) + break; + distances[offset - 2] = maxLen; + if (maxLen == lenLimit) + { + p->son[p->cyclicBufferPos] = curMatch; + MOVE_POS_RET; + } + } + if (maxLen < 3) + maxLen = 3; + offset = (UInt32)(Hc_GetMatchesSpec(lenLimit, curMatch, MF_PARAMS(p), + distances + offset, maxLen) - (distances)); + MOVE_POS_RET +} + +UInt32 Hc3Zip_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) +{ + UInt32 offset; + GET_MATCHES_HEADER(3) + HASH_ZIP_CALC; + curMatch = p->hash[hashValue]; + p->hash[hashValue] = p->pos; + offset = (UInt32)(Hc_GetMatchesSpec(lenLimit, curMatch, MF_PARAMS(p), + distances, 2) - (distances)); + MOVE_POS_RET +} + +static void Bt2_MatchFinder_Skip(CMatchFinder *p, UInt32 num) +{ + do + { + SKIP_HEADER(2) + HASH2_CALC; + curMatch = p->hash[hashValue]; + p->hash[hashValue] = p->pos; + SKIP_FOOTER + } + while (--num != 0); +} + +void Bt3Zip_MatchFinder_Skip(CMatchFinder *p, UInt32 num) +{ + do + { + SKIP_HEADER(3) + HASH_ZIP_CALC; + curMatch = p->hash[hashValue]; + p->hash[hashValue] = p->pos; + SKIP_FOOTER + } + while (--num != 0); +} + +static void Bt3_MatchFinder_Skip(CMatchFinder *p, UInt32 num) +{ + do + { + UInt32 hash2Value; + SKIP_HEADER(3) + HASH3_CALC; + curMatch = p->hash[kFix3HashSize + hashValue]; + p->hash[hash2Value] = + p->hash[kFix3HashSize + hashValue] = p->pos; + SKIP_FOOTER + } + while (--num != 0); +} + +static void Bt4_MatchFinder_Skip(CMatchFinder *p, UInt32 num) +{ + do + { + UInt32 hash2Value, hash3Value; + SKIP_HEADER(4) + HASH4_CALC; + curMatch = p->hash[kFix4HashSize + hashValue]; + p->hash[ hash2Value] = + p->hash[kFix3HashSize + hash3Value] = p->pos; + p->hash[kFix4HashSize + hashValue] = p->pos; + SKIP_FOOTER + } + while (--num != 0); +} + +static void Hc4_MatchFinder_Skip(CMatchFinder *p, UInt32 num) +{ + do + { + UInt32 hash2Value, hash3Value; + SKIP_HEADER(4) + HASH4_CALC; + curMatch = p->hash[kFix4HashSize + hashValue]; + p->hash[ hash2Value] = + p->hash[kFix3HashSize + hash3Value] = + p->hash[kFix4HashSize + hashValue] = p->pos; + p->son[p->cyclicBufferPos] = curMatch; + MOVE_POS + } + while (--num != 0); +} + +void Hc3Zip_MatchFinder_Skip(CMatchFinder *p, UInt32 num) +{ + do + { + SKIP_HEADER(3) + HASH_ZIP_CALC; + curMatch = p->hash[hashValue]; + p->hash[hashValue] = p->pos; + p->son[p->cyclicBufferPos] = curMatch; + MOVE_POS + } + while (--num != 0); +} + +void MatchFinder_CreateVTable(CMatchFinder *p, IMatchFinder *vTable) +{ + vTable->Init = (Mf_Init_Func)MatchFinder_Init; + vTable->GetIndexByte = (Mf_GetIndexByte_Func)MatchFinder_GetIndexByte; + vTable->GetNumAvailableBytes = (Mf_GetNumAvailableBytes_Func)MatchFinder_GetNumAvailableBytes; + vTable->GetPointerToCurrentPos = (Mf_GetPointerToCurrentPos_Func)MatchFinder_GetPointerToCurrentPos; + if (!p->btMode) + { + vTable->GetMatches = (Mf_GetMatches_Func)Hc4_MatchFinder_GetMatches; + vTable->Skip = (Mf_Skip_Func)Hc4_MatchFinder_Skip; + } + else if (p->numHashBytes == 2) + { + vTable->GetMatches = (Mf_GetMatches_Func)Bt2_MatchFinder_GetMatches; + vTable->Skip = (Mf_Skip_Func)Bt2_MatchFinder_Skip; + } + else if (p->numHashBytes == 3) + { + vTable->GetMatches = (Mf_GetMatches_Func)Bt3_MatchFinder_GetMatches; + vTable->Skip = (Mf_Skip_Func)Bt3_MatchFinder_Skip; + } + else + { + vTable->GetMatches = (Mf_GetMatches_Func)Bt4_MatchFinder_GetMatches; + vTable->Skip = (Mf_Skip_Func)Bt4_MatchFinder_Skip; + } +} diff --git a/lzma/C/LzFind.h b/lzma/C/LzFind.h new file mode 100644 index 0000000000..423d67e0c3 --- /dev/null +++ b/lzma/C/LzFind.h @@ -0,0 +1,107 @@ +/* LzFind.h -- Match finder for LZ algorithms +2008-10-04 : Igor Pavlov : Public domain */ + +#ifndef __LZFIND_H +#define __LZFIND_H + +#include "Types.h" + +typedef UInt32 CLzRef; + +typedef struct _CMatchFinder +{ + Byte *buffer; + UInt32 pos; + UInt32 posLimit; + UInt32 streamPos; + UInt32 lenLimit; + + UInt32 cyclicBufferPos; + UInt32 cyclicBufferSize; /* it must be = (historySize + 1) */ + + UInt32 matchMaxLen; + CLzRef *hash; + CLzRef *son; + UInt32 hashMask; + UInt32 cutValue; + + Byte *bufferBase; + ISeqInStream *stream; + int streamEndWasReached; + + UInt32 blockSize; + UInt32 keepSizeBefore; + UInt32 keepSizeAfter; + + UInt32 numHashBytes; + int directInput; + int btMode; + /* int skipModeBits; */ + int bigHash; + UInt32 historySize; + UInt32 fixedHashSize; + UInt32 hashSizeSum; + UInt32 numSons; + SRes result; + UInt32 crc[256]; +} CMatchFinder; + +#define Inline_MatchFinder_GetPointerToCurrentPos(p) ((p)->buffer) +#define Inline_MatchFinder_GetIndexByte(p, index) ((p)->buffer[(Int32)(index)]) + +#define Inline_MatchFinder_GetNumAvailableBytes(p) ((p)->streamPos - (p)->pos) + +int MatchFinder_NeedMove(CMatchFinder *p); +Byte *MatchFinder_GetPointerToCurrentPos(CMatchFinder *p); +void MatchFinder_MoveBlock(CMatchFinder *p); +void MatchFinder_ReadIfRequired(CMatchFinder *p); + +void MatchFinder_Construct(CMatchFinder *p); + +/* Conditions: + historySize <= 3 GB + keepAddBufferBefore + matchMaxLen + keepAddBufferAfter < 511MB +*/ +int MatchFinder_Create(CMatchFinder *p, UInt32 historySize, + UInt32 keepAddBufferBefore, UInt32 matchMaxLen, UInt32 keepAddBufferAfter, + ISzAlloc *alloc); +void MatchFinder_Free(CMatchFinder *p, ISzAlloc *alloc); +void MatchFinder_Normalize3(UInt32 subValue, CLzRef *items, UInt32 numItems); +void MatchFinder_ReduceOffsets(CMatchFinder *p, UInt32 subValue); + +UInt32 * GetMatchesSpec1(UInt32 lenLimit, UInt32 curMatch, UInt32 pos, const Byte *buffer, CLzRef *son, + UInt32 _cyclicBufferPos, UInt32 _cyclicBufferSize, UInt32 _cutValue, + UInt32 *distances, UInt32 maxLen); + +/* +Conditions: + Mf_GetNumAvailableBytes_Func must be called before each Mf_GetMatchLen_Func. + Mf_GetPointerToCurrentPos_Func's result must be used only before any other function +*/ + +typedef void (*Mf_Init_Func)(void *object); +typedef Byte (*Mf_GetIndexByte_Func)(void *object, Int32 index); +typedef UInt32 (*Mf_GetNumAvailableBytes_Func)(void *object); +typedef const Byte * (*Mf_GetPointerToCurrentPos_Func)(void *object); +typedef UInt32 (*Mf_GetMatches_Func)(void *object, UInt32 *distances); +typedef void (*Mf_Skip_Func)(void *object, UInt32); + +typedef struct _IMatchFinder +{ + Mf_Init_Func Init; + Mf_GetIndexByte_Func GetIndexByte; + Mf_GetNumAvailableBytes_Func GetNumAvailableBytes; + Mf_GetPointerToCurrentPos_Func GetPointerToCurrentPos; + Mf_GetMatches_Func GetMatches; + Mf_Skip_Func Skip; +} IMatchFinder; + +void MatchFinder_CreateVTable(CMatchFinder *p, IMatchFinder *vTable); + +void MatchFinder_Init(CMatchFinder *p); +UInt32 Bt3Zip_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances); +UInt32 Hc3Zip_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances); +void Bt3Zip_MatchFinder_Skip(CMatchFinder *p, UInt32 num); +void Hc3Zip_MatchFinder_Skip(CMatchFinder *p, UInt32 num); + +#endif diff --git a/lzma/C/LzFindMt.c b/lzma/C/LzFindMt.c new file mode 100644 index 0000000000..a67899ab0e --- /dev/null +++ b/lzma/C/LzFindMt.c @@ -0,0 +1,793 @@ +/* LzFindMt.c -- multithreaded Match finder for LZ algorithms +2008-10-04 : Igor Pavlov : Public domain */ + +#include "LzHash.h" + +#include "LzFindMt.h" + +void MtSync_Construct(CMtSync *p) +{ + p->wasCreated = False; + p->csWasInitialized = False; + p->csWasEntered = False; + Thread_Construct(&p->thread); + Event_Construct(&p->canStart); + Event_Construct(&p->wasStarted); + Event_Construct(&p->wasStopped); + Semaphore_Construct(&p->freeSemaphore); + Semaphore_Construct(&p->filledSemaphore); +} + +void MtSync_GetNextBlock(CMtSync *p) +{ + if (p->needStart) + { + p->numProcessedBlocks = 1; + p->needStart = False; + p->stopWriting = False; + p->exit = False; + Event_Reset(&p->wasStarted); + Event_Reset(&p->wasStopped); + + Event_Set(&p->canStart); + Event_Wait(&p->wasStarted); + } + else + { + CriticalSection_Leave(&p->cs); + p->csWasEntered = False; + p->numProcessedBlocks++; + Semaphore_Release1(&p->freeSemaphore); + } + Semaphore_Wait(&p->filledSemaphore); + CriticalSection_Enter(&p->cs); + p->csWasEntered = True; +} + +/* MtSync_StopWriting must be called if Writing was started */ + +void MtSync_StopWriting(CMtSync *p) +{ + UInt32 myNumBlocks = p->numProcessedBlocks; + if (!Thread_WasCreated(&p->thread) || p->needStart) + return; + p->stopWriting = True; + if (p->csWasEntered) + { + CriticalSection_Leave(&p->cs); + p->csWasEntered = False; + } + Semaphore_Release1(&p->freeSemaphore); + + Event_Wait(&p->wasStopped); + + while (myNumBlocks++ != p->numProcessedBlocks) + { + Semaphore_Wait(&p->filledSemaphore); + Semaphore_Release1(&p->freeSemaphore); + } + p->needStart = True; +} + +void MtSync_Destruct(CMtSync *p) +{ + if (Thread_WasCreated(&p->thread)) + { + MtSync_StopWriting(p); + p->exit = True; + if (p->needStart) + Event_Set(&p->canStart); + Thread_Wait(&p->thread); + Thread_Close(&p->thread); + } + if (p->csWasInitialized) + { + CriticalSection_Delete(&p->cs); + p->csWasInitialized = False; + } + + Event_Close(&p->canStart); + Event_Close(&p->wasStarted); + Event_Close(&p->wasStopped); + Semaphore_Close(&p->freeSemaphore); + Semaphore_Close(&p->filledSemaphore); + + p->wasCreated = False; +} + +#define RINOK_THREAD(x) { if ((x) != 0) return SZ_ERROR_THREAD; } + +static SRes MtSync_Create2(CMtSync *p, unsigned (THREAD_FUNC_CALL_TYPE *startAddress)(void *), void *obj, UInt32 numBlocks) +{ + if (p->wasCreated) + return SZ_OK; + + RINOK_THREAD(CriticalSection_Init(&p->cs)); + p->csWasInitialized = True; + + RINOK_THREAD(AutoResetEvent_CreateNotSignaled(&p->canStart)); + RINOK_THREAD(AutoResetEvent_CreateNotSignaled(&p->wasStarted)); + RINOK_THREAD(AutoResetEvent_CreateNotSignaled(&p->wasStopped)); + + RINOK_THREAD(Semaphore_Create(&p->freeSemaphore, numBlocks, numBlocks)); + RINOK_THREAD(Semaphore_Create(&p->filledSemaphore, 0, numBlocks)); + + p->needStart = True; + + RINOK_THREAD(Thread_Create(&p->thread, startAddress, obj)); + p->wasCreated = True; + return SZ_OK; +} + +static SRes MtSync_Create(CMtSync *p, unsigned (THREAD_FUNC_CALL_TYPE *startAddress)(void *), void *obj, UInt32 numBlocks) +{ + SRes res = MtSync_Create2(p, startAddress, obj, numBlocks); + if (res != SZ_OK) + MtSync_Destruct(p); + return res; +} + +void MtSync_Init(CMtSync *p) { p->needStart = True; } + +#define kMtMaxValForNormalize 0xFFFFFFFF + +#define DEF_GetHeads2(name, v, action) \ +static void GetHeads ## name(const Byte *p, UInt32 pos, \ +UInt32 *hash, UInt32 hashMask, UInt32 *heads, UInt32 numHeads, const UInt32 *crc) \ +{ action; for (; numHeads != 0; numHeads--) { \ +const UInt32 value = (v); p++; *heads++ = pos - hash[value]; hash[value] = pos++; } } + +#define DEF_GetHeads(name, v) DEF_GetHeads2(name, v, ;) + +DEF_GetHeads2(2, (p[0] | ((UInt32)p[1] << 8)), hashMask = hashMask; crc = crc; ) +DEF_GetHeads(3, (crc[p[0]] ^ p[1] ^ ((UInt32)p[2] << 8)) & hashMask) +DEF_GetHeads(4, (crc[p[0]] ^ p[1] ^ ((UInt32)p[2] << 8) ^ (crc[p[3]] << 5)) & hashMask) +DEF_GetHeads(4b, (crc[p[0]] ^ p[1] ^ ((UInt32)p[2] << 8) ^ ((UInt32)p[3] << 16)) & hashMask) +DEF_GetHeads(5, (crc[p[0]] ^ p[1] ^ ((UInt32)p[2] << 8) ^ (crc[p[3]] << 5) ^ (crc[p[4]] << 3)) & hashMask) + +void HashThreadFunc(CMatchFinderMt *mt) +{ + CMtSync *p = &mt->hashSync; + for (;;) + { + UInt32 numProcessedBlocks = 0; + Event_Wait(&p->canStart); + Event_Set(&p->wasStarted); + for (;;) + { + if (p->exit) + return; + if (p->stopWriting) + { + p->numProcessedBlocks = numProcessedBlocks; + Event_Set(&p->wasStopped); + break; + } + + { + CMatchFinder *mf = mt->MatchFinder; + if (MatchFinder_NeedMove(mf)) + { + CriticalSection_Enter(&mt->btSync.cs); + CriticalSection_Enter(&mt->hashSync.cs); + { + const Byte *beforePtr = MatchFinder_GetPointerToCurrentPos(mf); + const Byte *afterPtr; + MatchFinder_MoveBlock(mf); + afterPtr = MatchFinder_GetPointerToCurrentPos(mf); + mt->pointerToCurPos -= beforePtr - afterPtr; + mt->buffer -= beforePtr - afterPtr; + } + CriticalSection_Leave(&mt->btSync.cs); + CriticalSection_Leave(&mt->hashSync.cs); + continue; + } + + Semaphore_Wait(&p->freeSemaphore); + + MatchFinder_ReadIfRequired(mf); + if (mf->pos > (kMtMaxValForNormalize - kMtHashBlockSize)) + { + UInt32 subValue = (mf->pos - mf->historySize - 1); + MatchFinder_ReduceOffsets(mf, subValue); + MatchFinder_Normalize3(subValue, mf->hash + mf->fixedHashSize, mf->hashMask + 1); + } + { + UInt32 *heads = mt->hashBuf + ((numProcessedBlocks++) & kMtHashNumBlocksMask) * kMtHashBlockSize; + UInt32 num = mf->streamPos - mf->pos; + heads[0] = 2; + heads[1] = num; + if (num >= mf->numHashBytes) + { + num = num - mf->numHashBytes + 1; + if (num > kMtHashBlockSize - 2) + num = kMtHashBlockSize - 2; + mt->GetHeadsFunc(mf->buffer, mf->pos, mf->hash + mf->fixedHashSize, mf->hashMask, heads + 2, num, mf->crc); + heads[0] += num; + } + mf->pos += num; + mf->buffer += num; + } + } + + Semaphore_Release1(&p->filledSemaphore); + } + } +} + +void MatchFinderMt_GetNextBlock_Hash(CMatchFinderMt *p) +{ + MtSync_GetNextBlock(&p->hashSync); + p->hashBufPosLimit = p->hashBufPos = ((p->hashSync.numProcessedBlocks - 1) & kMtHashNumBlocksMask) * kMtHashBlockSize; + p->hashBufPosLimit += p->hashBuf[p->hashBufPos++]; + p->hashNumAvail = p->hashBuf[p->hashBufPos++]; +} + +#define kEmptyHashValue 0 + +/* #define MFMT_GM_INLINE */ + +#ifdef MFMT_GM_INLINE + +#define NO_INLINE MY_FAST_CALL + +Int32 NO_INLINE GetMatchesSpecN(UInt32 lenLimit, UInt32 pos, const Byte *cur, CLzRef *son, + UInt32 _cyclicBufferPos, UInt32 _cyclicBufferSize, UInt32 _cutValue, + UInt32 *_distances, UInt32 _maxLen, const UInt32 *hash, Int32 limit, UInt32 size, UInt32 *posRes) +{ + do + { + UInt32 *distances = _distances + 1; + UInt32 curMatch = pos - *hash++; + + CLzRef *ptr0 = son + (_cyclicBufferPos << 1) + 1; + CLzRef *ptr1 = son + (_cyclicBufferPos << 1); + UInt32 len0 = 0, len1 = 0; + UInt32 cutValue = _cutValue; + UInt32 maxLen = _maxLen; + for (;;) + { + UInt32 delta = pos - curMatch; + if (cutValue-- == 0 || delta >= _cyclicBufferSize) + { + *ptr0 = *ptr1 = kEmptyHashValue; + break; + } + { + CLzRef *pair = son + ((_cyclicBufferPos - delta + ((delta > _cyclicBufferPos) ? _cyclicBufferSize : 0)) << 1); + const Byte *pb = cur - delta; + UInt32 len = (len0 < len1 ? len0 : len1); + if (pb[len] == cur[len]) + { + if (++len != lenLimit && pb[len] == cur[len]) + while (++len != lenLimit) + if (pb[len] != cur[len]) + break; + if (maxLen < len) + { + *distances++ = maxLen = len; + *distances++ = delta - 1; + if (len == lenLimit) + { + *ptr1 = pair[0]; + *ptr0 = pair[1]; + break; + } + } + } + if (pb[len] < cur[len]) + { + *ptr1 = curMatch; + ptr1 = pair + 1; + curMatch = *ptr1; + len1 = len; + } + else + { + *ptr0 = curMatch; + ptr0 = pair; + curMatch = *ptr0; + len0 = len; + } + } + } + pos++; + _cyclicBufferPos++; + cur++; + { + UInt32 num = (UInt32)(distances - _distances); + *_distances = num - 1; + _distances += num; + limit -= num; + } + } + while (limit > 0 && --size != 0); + *posRes = pos; + return limit; +} + +#endif + +void BtGetMatches(CMatchFinderMt *p, UInt32 *distances) +{ + UInt32 numProcessed = 0; + UInt32 curPos = 2; + UInt32 limit = kMtBtBlockSize - (p->matchMaxLen * 2); + distances[1] = p->hashNumAvail; + while (curPos < limit) + { + if (p->hashBufPos == p->hashBufPosLimit) + { + MatchFinderMt_GetNextBlock_Hash(p); + distances[1] = numProcessed + p->hashNumAvail; + if (p->hashNumAvail >= p->numHashBytes) + continue; + for (; p->hashNumAvail != 0; p->hashNumAvail--) + distances[curPos++] = 0; + break; + } + { + UInt32 size = p->hashBufPosLimit - p->hashBufPos; + UInt32 lenLimit = p->matchMaxLen; + UInt32 pos = p->pos; + UInt32 cyclicBufferPos = p->cyclicBufferPos; + if (lenLimit >= p->hashNumAvail) + lenLimit = p->hashNumAvail; + { + UInt32 size2 = p->hashNumAvail - lenLimit + 1; + if (size2 < size) + size = size2; + size2 = p->cyclicBufferSize - cyclicBufferPos; + if (size2 < size) + size = size2; + } + #ifndef MFMT_GM_INLINE + while (curPos < limit && size-- != 0) + { + UInt32 *startDistances = distances + curPos; + UInt32 num = (UInt32)(GetMatchesSpec1(lenLimit, pos - p->hashBuf[p->hashBufPos++], + pos, p->buffer, p->son, cyclicBufferPos, p->cyclicBufferSize, p->cutValue, + startDistances + 1, p->numHashBytes - 1) - startDistances); + *startDistances = num - 1; + curPos += num; + cyclicBufferPos++; + pos++; + p->buffer++; + } + #else + { + UInt32 posRes; + curPos = limit - GetMatchesSpecN(lenLimit, pos, p->buffer, p->son, cyclicBufferPos, p->cyclicBufferSize, p->cutValue, + distances + curPos, p->numHashBytes - 1, p->hashBuf + p->hashBufPos, (Int32)(limit - curPos) , size, &posRes); + p->hashBufPos += posRes - pos; + cyclicBufferPos += posRes - pos; + p->buffer += posRes - pos; + pos = posRes; + } + #endif + + numProcessed += pos - p->pos; + p->hashNumAvail -= pos - p->pos; + p->pos = pos; + if (cyclicBufferPos == p->cyclicBufferSize) + cyclicBufferPos = 0; + p->cyclicBufferPos = cyclicBufferPos; + } + } + distances[0] = curPos; +} + +void BtFillBlock(CMatchFinderMt *p, UInt32 globalBlockIndex) +{ + CMtSync *sync = &p->hashSync; + if (!sync->needStart) + { + CriticalSection_Enter(&sync->cs); + sync->csWasEntered = True; + } + + BtGetMatches(p, p->btBuf + (globalBlockIndex & kMtBtNumBlocksMask) * kMtBtBlockSize); + + if (p->pos > kMtMaxValForNormalize - kMtBtBlockSize) + { + UInt32 subValue = p->pos - p->cyclicBufferSize; + MatchFinder_Normalize3(subValue, p->son, p->cyclicBufferSize * 2); + p->pos -= subValue; + } + + if (!sync->needStart) + { + CriticalSection_Leave(&sync->cs); + sync->csWasEntered = False; + } +} + +void BtThreadFunc(CMatchFinderMt *mt) +{ + CMtSync *p = &mt->btSync; + for (;;) + { + UInt32 blockIndex = 0; + Event_Wait(&p->canStart); + Event_Set(&p->wasStarted); + for (;;) + { + if (p->exit) + return; + if (p->stopWriting) + { + p->numProcessedBlocks = blockIndex; + MtSync_StopWriting(&mt->hashSync); + Event_Set(&p->wasStopped); + break; + } + Semaphore_Wait(&p->freeSemaphore); + BtFillBlock(mt, blockIndex++); + Semaphore_Release1(&p->filledSemaphore); + } + } +} + +void MatchFinderMt_Construct(CMatchFinderMt *p) +{ + p->hashBuf = 0; + MtSync_Construct(&p->hashSync); + MtSync_Construct(&p->btSync); +} + +void MatchFinderMt_FreeMem(CMatchFinderMt *p, ISzAlloc *alloc) +{ + alloc->Free(alloc, p->hashBuf); + p->hashBuf = 0; +} + +void MatchFinderMt_Destruct(CMatchFinderMt *p, ISzAlloc *alloc) +{ + MtSync_Destruct(&p->hashSync); + MtSync_Destruct(&p->btSync); + MatchFinderMt_FreeMem(p, alloc); +} + +#define kHashBufferSize (kMtHashBlockSize * kMtHashNumBlocks) +#define kBtBufferSize (kMtBtBlockSize * kMtBtNumBlocks) + +static unsigned THREAD_FUNC_CALL_TYPE HashThreadFunc2(void *p) { HashThreadFunc((CMatchFinderMt *)p); return 0; } +static unsigned THREAD_FUNC_CALL_TYPE BtThreadFunc2(void *p) +{ + Byte allocaDummy[0x180]; + int i = 0; + for (i = 0; i < 16; i++) + allocaDummy[i] = (Byte)i; + BtThreadFunc((CMatchFinderMt *)p); + return 0; +} + +SRes MatchFinderMt_Create(CMatchFinderMt *p, UInt32 historySize, UInt32 keepAddBufferBefore, + UInt32 matchMaxLen, UInt32 keepAddBufferAfter, ISzAlloc *alloc) +{ + CMatchFinder *mf = p->MatchFinder; + p->historySize = historySize; + if (kMtBtBlockSize <= matchMaxLen * 4) + return SZ_ERROR_PARAM; + if (p->hashBuf == 0) + { + p->hashBuf = (UInt32 *)alloc->Alloc(alloc, (kHashBufferSize + kBtBufferSize) * sizeof(UInt32)); + if (p->hashBuf == 0) + return SZ_ERROR_MEM; + p->btBuf = p->hashBuf + kHashBufferSize; + } + keepAddBufferBefore += (kHashBufferSize + kBtBufferSize); + keepAddBufferAfter += kMtHashBlockSize; + if (!MatchFinder_Create(mf, historySize, keepAddBufferBefore, matchMaxLen, keepAddBufferAfter, alloc)) + return SZ_ERROR_MEM; + + RINOK(MtSync_Create(&p->hashSync, HashThreadFunc2, p, kMtHashNumBlocks)); + RINOK(MtSync_Create(&p->btSync, BtThreadFunc2, p, kMtBtNumBlocks)); + return SZ_OK; +} + +/* Call it after ReleaseStream / SetStream */ +void MatchFinderMt_Init(CMatchFinderMt *p) +{ + CMatchFinder *mf = p->MatchFinder; + p->btBufPos = p->btBufPosLimit = 0; + p->hashBufPos = p->hashBufPosLimit = 0; + MatchFinder_Init(mf); + p->pointerToCurPos = MatchFinder_GetPointerToCurrentPos(mf); + p->btNumAvailBytes = 0; + p->lzPos = p->historySize + 1; + + p->hash = mf->hash; + p->fixedHashSize = mf->fixedHashSize; + p->crc = mf->crc; + + p->son = mf->son; + p->matchMaxLen = mf->matchMaxLen; + p->numHashBytes = mf->numHashBytes; + p->pos = mf->pos; + p->buffer = mf->buffer; + p->cyclicBufferPos = mf->cyclicBufferPos; + p->cyclicBufferSize = mf->cyclicBufferSize; + p->cutValue = mf->cutValue; +} + +/* ReleaseStream is required to finish multithreading */ +void MatchFinderMt_ReleaseStream(CMatchFinderMt *p) +{ + MtSync_StopWriting(&p->btSync); + /* p->MatchFinder->ReleaseStream(); */ +} + +void MatchFinderMt_Normalize(CMatchFinderMt *p) +{ + MatchFinder_Normalize3(p->lzPos - p->historySize - 1, p->hash, p->fixedHashSize); + p->lzPos = p->historySize + 1; +} + +void MatchFinderMt_GetNextBlock_Bt(CMatchFinderMt *p) +{ + UInt32 blockIndex; + MtSync_GetNextBlock(&p->btSync); + blockIndex = ((p->btSync.numProcessedBlocks - 1) & kMtBtNumBlocksMask); + p->btBufPosLimit = p->btBufPos = blockIndex * kMtBtBlockSize; + p->btBufPosLimit += p->btBuf[p->btBufPos++]; + p->btNumAvailBytes = p->btBuf[p->btBufPos++]; + if (p->lzPos >= kMtMaxValForNormalize - kMtBtBlockSize) + MatchFinderMt_Normalize(p); +} + +const Byte * MatchFinderMt_GetPointerToCurrentPos(CMatchFinderMt *p) +{ + return p->pointerToCurPos; +} + +#define GET_NEXT_BLOCK_IF_REQUIRED if (p->btBufPos == p->btBufPosLimit) MatchFinderMt_GetNextBlock_Bt(p); + +UInt32 MatchFinderMt_GetNumAvailableBytes(CMatchFinderMt *p) +{ + GET_NEXT_BLOCK_IF_REQUIRED; + return p->btNumAvailBytes; +} + +Byte MatchFinderMt_GetIndexByte(CMatchFinderMt *p, Int32 index) +{ + return p->pointerToCurPos[index]; +} + +UInt32 * MixMatches2(CMatchFinderMt *p, UInt32 matchMinPos, UInt32 *distances) +{ + UInt32 hash2Value, curMatch2; + UInt32 *hash = p->hash; + const Byte *cur = p->pointerToCurPos; + UInt32 lzPos = p->lzPos; + MT_HASH2_CALC + + curMatch2 = hash[hash2Value]; + hash[hash2Value] = lzPos; + + if (curMatch2 >= matchMinPos) + if (cur[(ptrdiff_t)curMatch2 - lzPos] == cur[0]) + { + *distances++ = 2; + *distances++ = lzPos - curMatch2 - 1; + } + return distances; +} + +UInt32 * MixMatches3(CMatchFinderMt *p, UInt32 matchMinPos, UInt32 *distances) +{ + UInt32 hash2Value, hash3Value, curMatch2, curMatch3; + UInt32 *hash = p->hash; + const Byte *cur = p->pointerToCurPos; + UInt32 lzPos = p->lzPos; + MT_HASH3_CALC + + curMatch2 = hash[ hash2Value]; + curMatch3 = hash[kFix3HashSize + hash3Value]; + + hash[ hash2Value] = + hash[kFix3HashSize + hash3Value] = + lzPos; + + if (curMatch2 >= matchMinPos && cur[(ptrdiff_t)curMatch2 - lzPos] == cur[0]) + { + distances[1] = lzPos - curMatch2 - 1; + if (cur[(ptrdiff_t)curMatch2 - lzPos + 2] == cur[2]) + { + distances[0] = 3; + return distances + 2; + } + distances[0] = 2; + distances += 2; + } + if (curMatch3 >= matchMinPos && cur[(ptrdiff_t)curMatch3 - lzPos] == cur[0]) + { + *distances++ = 3; + *distances++ = lzPos - curMatch3 - 1; + } + return distances; +} + +/* +UInt32 *MixMatches4(CMatchFinderMt *p, UInt32 matchMinPos, UInt32 *distances) +{ + UInt32 hash2Value, hash3Value, hash4Value, curMatch2, curMatch3, curMatch4; + UInt32 *hash = p->hash; + const Byte *cur = p->pointerToCurPos; + UInt32 lzPos = p->lzPos; + MT_HASH4_CALC + + curMatch2 = hash[ hash2Value]; + curMatch3 = hash[kFix3HashSize + hash3Value]; + curMatch4 = hash[kFix4HashSize + hash4Value]; + + hash[ hash2Value] = + hash[kFix3HashSize + hash3Value] = + hash[kFix4HashSize + hash4Value] = + lzPos; + + if (curMatch2 >= matchMinPos && cur[(ptrdiff_t)curMatch2 - lzPos] == cur[0]) + { + distances[1] = lzPos - curMatch2 - 1; + if (cur[(ptrdiff_t)curMatch2 - lzPos + 2] == cur[2]) + { + distances[0] = (cur[(ptrdiff_t)curMatch2 - lzPos + 3] == cur[3]) ? 4 : 3; + return distances + 2; + } + distances[0] = 2; + distances += 2; + } + if (curMatch3 >= matchMinPos && cur[(ptrdiff_t)curMatch3 - lzPos] == cur[0]) + { + distances[1] = lzPos - curMatch3 - 1; + if (cur[(ptrdiff_t)curMatch3 - lzPos + 3] == cur[3]) + { + distances[0] = 4; + return distances + 2; + } + distances[0] = 3; + distances += 2; + } + + if (curMatch4 >= matchMinPos) + if ( + cur[(ptrdiff_t)curMatch4 - lzPos] == cur[0] && + cur[(ptrdiff_t)curMatch4 - lzPos + 3] == cur[3] + ) + { + *distances++ = 4; + *distances++ = lzPos - curMatch4 - 1; + } + return distances; +} +*/ + +#define INCREASE_LZ_POS p->lzPos++; p->pointerToCurPos++; + +UInt32 MatchFinderMt2_GetMatches(CMatchFinderMt *p, UInt32 *distances) +{ + const UInt32 *btBuf = p->btBuf + p->btBufPos; + UInt32 len = *btBuf++; + p->btBufPos += 1 + len; + p->btNumAvailBytes--; + { + UInt32 i; + for (i = 0; i < len; i += 2) + { + *distances++ = *btBuf++; + *distances++ = *btBuf++; + } + } + INCREASE_LZ_POS + return len; +} + +UInt32 MatchFinderMt_GetMatches(CMatchFinderMt *p, UInt32 *distances) +{ + const UInt32 *btBuf = p->btBuf + p->btBufPos; + UInt32 len = *btBuf++; + p->btBufPos += 1 + len; + + if (len == 0) + { + if (p->btNumAvailBytes-- >= 4) + len = (UInt32)(p->MixMatchesFunc(p, p->lzPos - p->historySize, distances) - (distances)); + } + else + { + /* Condition: there are matches in btBuf with length < p->numHashBytes */ + UInt32 *distances2; + p->btNumAvailBytes--; + distances2 = p->MixMatchesFunc(p, p->lzPos - btBuf[1], distances); + do + { + *distances2++ = *btBuf++; + *distances2++ = *btBuf++; + } + while ((len -= 2) != 0); + len = (UInt32)(distances2 - (distances)); + } + INCREASE_LZ_POS + return len; +} + +#define SKIP_HEADER2 do { GET_NEXT_BLOCK_IF_REQUIRED +#define SKIP_HEADER(n) SKIP_HEADER2 if (p->btNumAvailBytes-- >= (n)) { const Byte *cur = p->pointerToCurPos; UInt32 *hash = p->hash; +#define SKIP_FOOTER } INCREASE_LZ_POS p->btBufPos += p->btBuf[p->btBufPos] + 1; } while (--num != 0); + +void MatchFinderMt0_Skip(CMatchFinderMt *p, UInt32 num) +{ + SKIP_HEADER2 { p->btNumAvailBytes--; + SKIP_FOOTER +} + +void MatchFinderMt2_Skip(CMatchFinderMt *p, UInt32 num) +{ + SKIP_HEADER(2) + UInt32 hash2Value; + MT_HASH2_CALC + hash[hash2Value] = p->lzPos; + SKIP_FOOTER +} + +void MatchFinderMt3_Skip(CMatchFinderMt *p, UInt32 num) +{ + SKIP_HEADER(3) + UInt32 hash2Value, hash3Value; + MT_HASH3_CALC + hash[kFix3HashSize + hash3Value] = + hash[ hash2Value] = + p->lzPos; + SKIP_FOOTER +} + +/* +void MatchFinderMt4_Skip(CMatchFinderMt *p, UInt32 num) +{ + SKIP_HEADER(4) + UInt32 hash2Value, hash3Value, hash4Value; + MT_HASH4_CALC + hash[kFix4HashSize + hash4Value] = + hash[kFix3HashSize + hash3Value] = + hash[ hash2Value] = + p->lzPos; + SKIP_FOOTER +} +*/ + +void MatchFinderMt_CreateVTable(CMatchFinderMt *p, IMatchFinder *vTable) +{ + vTable->Init = (Mf_Init_Func)MatchFinderMt_Init; + vTable->GetIndexByte = (Mf_GetIndexByte_Func)MatchFinderMt_GetIndexByte; + vTable->GetNumAvailableBytes = (Mf_GetNumAvailableBytes_Func)MatchFinderMt_GetNumAvailableBytes; + vTable->GetPointerToCurrentPos = (Mf_GetPointerToCurrentPos_Func)MatchFinderMt_GetPointerToCurrentPos; + vTable->GetMatches = (Mf_GetMatches_Func)MatchFinderMt_GetMatches; + switch(p->MatchFinder->numHashBytes) + { + case 2: + p->GetHeadsFunc = GetHeads2; + p->MixMatchesFunc = (Mf_Mix_Matches)0; + vTable->Skip = (Mf_Skip_Func)MatchFinderMt0_Skip; + vTable->GetMatches = (Mf_GetMatches_Func)MatchFinderMt2_GetMatches; + break; + case 3: + p->GetHeadsFunc = GetHeads3; + p->MixMatchesFunc = (Mf_Mix_Matches)MixMatches2; + vTable->Skip = (Mf_Skip_Func)MatchFinderMt2_Skip; + break; + default: + /* case 4: */ + p->GetHeadsFunc = p->MatchFinder->bigHash ? GetHeads4b : GetHeads4; + /* p->GetHeadsFunc = GetHeads4; */ + p->MixMatchesFunc = (Mf_Mix_Matches)MixMatches3; + vTable->Skip = (Mf_Skip_Func)MatchFinderMt3_Skip; + break; + /* + default: + p->GetHeadsFunc = GetHeads5; + p->MixMatchesFunc = (Mf_Mix_Matches)MixMatches4; + vTable->Skip = (Mf_Skip_Func)MatchFinderMt4_Skip; + break; + */ + } +} diff --git a/lzma/C/LzFindMt.h b/lzma/C/LzFindMt.h new file mode 100644 index 0000000000..b7ead2d472 --- /dev/null +++ b/lzma/C/LzFindMt.h @@ -0,0 +1,97 @@ +/* LzFindMt.h -- multithreaded Match finder for LZ algorithms +2008-10-04 : Igor Pavlov : Public domain */ + +#ifndef __LZFINDMT_H +#define __LZFINDMT_H + +#include "Threads.h" +#include "LzFind.h" + +#define kMtHashBlockSize (1 << 13) +#define kMtHashNumBlocks (1 << 3) +#define kMtHashNumBlocksMask (kMtHashNumBlocks - 1) + +#define kMtBtBlockSize (1 << 14) +#define kMtBtNumBlocks (1 << 6) +#define kMtBtNumBlocksMask (kMtBtNumBlocks - 1) + +typedef struct _CMtSync +{ + Bool wasCreated; + Bool needStart; + Bool exit; + Bool stopWriting; + + CThread thread; + CAutoResetEvent canStart; + CAutoResetEvent wasStarted; + CAutoResetEvent wasStopped; + CSemaphore freeSemaphore; + CSemaphore filledSemaphore; + Bool csWasInitialized; + Bool csWasEntered; + CCriticalSection cs; + UInt32 numProcessedBlocks; +} CMtSync; + +typedef UInt32 * (*Mf_Mix_Matches)(void *p, UInt32 matchMinPos, UInt32 *distances); + +/* kMtCacheLineDummy must be >= size_of_CPU_cache_line */ +#define kMtCacheLineDummy 128 + +typedef void (*Mf_GetHeads)(const Byte *buffer, UInt32 pos, + UInt32 *hash, UInt32 hashMask, UInt32 *heads, UInt32 numHeads, const UInt32 *crc); + +typedef struct _CMatchFinderMt +{ + /* LZ */ + const Byte *pointerToCurPos; + UInt32 *btBuf; + UInt32 btBufPos; + UInt32 btBufPosLimit; + UInt32 lzPos; + UInt32 btNumAvailBytes; + + UInt32 *hash; + UInt32 fixedHashSize; + UInt32 historySize; + const UInt32 *crc; + + Mf_Mix_Matches MixMatchesFunc; + + /* LZ + BT */ + CMtSync btSync; + Byte btDummy[kMtCacheLineDummy]; + + /* BT */ + UInt32 *hashBuf; + UInt32 hashBufPos; + UInt32 hashBufPosLimit; + UInt32 hashNumAvail; + + CLzRef *son; + UInt32 matchMaxLen; + UInt32 numHashBytes; + UInt32 pos; + Byte *buffer; + UInt32 cyclicBufferPos; + UInt32 cyclicBufferSize; /* it must be historySize + 1 */ + UInt32 cutValue; + + /* BT + Hash */ + CMtSync hashSync; + /* Byte hashDummy[kMtCacheLineDummy]; */ + + /* Hash */ + Mf_GetHeads GetHeadsFunc; + CMatchFinder *MatchFinder; +} CMatchFinderMt; + +void MatchFinderMt_Construct(CMatchFinderMt *p); +void MatchFinderMt_Destruct(CMatchFinderMt *p, ISzAlloc *alloc); +SRes MatchFinderMt_Create(CMatchFinderMt *p, UInt32 historySize, UInt32 keepAddBufferBefore, + UInt32 matchMaxLen, UInt32 keepAddBufferAfter, ISzAlloc *alloc); +void MatchFinderMt_CreateVTable(CMatchFinderMt *p, IMatchFinder *vTable); +void MatchFinderMt_ReleaseStream(CMatchFinderMt *p); + +#endif diff --git a/lzma/C/LzHash.h b/lzma/C/LzHash.h new file mode 100644 index 0000000000..c923417501 --- /dev/null +++ b/lzma/C/LzHash.h @@ -0,0 +1,54 @@ +/* LzHash.h -- HASH functions for LZ algorithms +2008-10-04 : Igor Pavlov : Public domain */ + +#ifndef __LZHASH_H +#define __LZHASH_H + +#define kHash2Size (1 << 10) +#define kHash3Size (1 << 16) +#define kHash4Size (1 << 20) + +#define kFix3HashSize (kHash2Size) +#define kFix4HashSize (kHash2Size + kHash3Size) +#define kFix5HashSize (kHash2Size + kHash3Size + kHash4Size) + +#define HASH2_CALC hashValue = cur[0] | ((UInt32)cur[1] << 8); + +#define HASH3_CALC { \ + UInt32 temp = p->crc[cur[0]] ^ cur[1]; \ + hash2Value = temp & (kHash2Size - 1); \ + hashValue = (temp ^ ((UInt32)cur[2] << 8)) & p->hashMask; } + +#define HASH4_CALC { \ + UInt32 temp = p->crc[cur[0]] ^ cur[1]; \ + hash2Value = temp & (kHash2Size - 1); \ + hash3Value = (temp ^ ((UInt32)cur[2] << 8)) & (kHash3Size - 1); \ + hashValue = (temp ^ ((UInt32)cur[2] << 8) ^ (p->crc[cur[3]] << 5)) & p->hashMask; } + +#define HASH5_CALC { \ + UInt32 temp = p->crc[cur[0]] ^ cur[1]; \ + hash2Value = temp & (kHash2Size - 1); \ + hash3Value = (temp ^ ((UInt32)cur[2] << 8)) & (kHash3Size - 1); \ + hash4Value = (temp ^ ((UInt32)cur[2] << 8) ^ (p->crc[cur[3]] << 5)); \ + hashValue = (hash4Value ^ (p->crc[cur[4]] << 3)) & p->hashMask; \ + hash4Value &= (kHash4Size - 1); } + +/* #define HASH_ZIP_CALC hashValue = ((cur[0] | ((UInt32)cur[1] << 8)) ^ p->crc[cur[2]]) & 0xFFFF; */ +#define HASH_ZIP_CALC hashValue = ((cur[2] | ((UInt32)cur[0] << 8)) ^ p->crc[cur[1]]) & 0xFFFF; + + +#define MT_HASH2_CALC \ + hash2Value = (p->crc[cur[0]] ^ cur[1]) & (kHash2Size - 1); + +#define MT_HASH3_CALC { \ + UInt32 temp = p->crc[cur[0]] ^ cur[1]; \ + hash2Value = temp & (kHash2Size - 1); \ + hash3Value = (temp ^ ((UInt32)cur[2] << 8)) & (kHash3Size - 1); } + +#define MT_HASH4_CALC { \ + UInt32 temp = p->crc[cur[0]] ^ cur[1]; \ + hash2Value = temp & (kHash2Size - 1); \ + hash3Value = (temp ^ ((UInt32)cur[2] << 8)) & (kHash3Size - 1); \ + hash4Value = (temp ^ ((UInt32)cur[2] << 8) ^ (p->crc[cur[3]] << 5)) & (kHash4Size - 1); } + +#endif diff --git a/lzma/C/LzmaDec.c b/lzma/C/LzmaDec.c new file mode 100644 index 0000000000..eb5250174b --- /dev/null +++ b/lzma/C/LzmaDec.c @@ -0,0 +1,1007 @@ +/* LzmaDec.c -- LZMA Decoder +2008-11-06 : Igor Pavlov : Public domain */ + +#include "LzmaDec.h" + +#include + +#define kNumTopBits 24 +#define kTopValue ((UInt32)1 << kNumTopBits) + +#define kNumBitModelTotalBits 11 +#define kBitModelTotal (1 << kNumBitModelTotalBits) +#define kNumMoveBits 5 + +#define RC_INIT_SIZE 5 + +#define NORMALIZE if (range < kTopValue) { range <<= 8; code = (code << 8) | (*buf++); } + +#define IF_BIT_0(p) ttt = *(p); NORMALIZE; bound = (range >> kNumBitModelTotalBits) * ttt; if (code < bound) +#define UPDATE_0(p) range = bound; *(p) = (CLzmaProb)(ttt + ((kBitModelTotal - ttt) >> kNumMoveBits)); +#define UPDATE_1(p) range -= bound; code -= bound; *(p) = (CLzmaProb)(ttt - (ttt >> kNumMoveBits)); +#define GET_BIT2(p, i, A0, A1) IF_BIT_0(p) \ + { UPDATE_0(p); i = (i + i); A0; } else \ + { UPDATE_1(p); i = (i + i) + 1; A1; } +#define GET_BIT(p, i) GET_BIT2(p, i, ; , ;) + +#define TREE_GET_BIT(probs, i) { GET_BIT((probs + i), i); } +#define TREE_DECODE(probs, limit, i) \ + { i = 1; do { TREE_GET_BIT(probs, i); } while (i < limit); i -= limit; } + +/* #define _LZMA_SIZE_OPT */ + +#ifdef _LZMA_SIZE_OPT +#define TREE_6_DECODE(probs, i) TREE_DECODE(probs, (1 << 6), i) +#else +#define TREE_6_DECODE(probs, i) \ + { i = 1; \ + TREE_GET_BIT(probs, i); \ + TREE_GET_BIT(probs, i); \ + TREE_GET_BIT(probs, i); \ + TREE_GET_BIT(probs, i); \ + TREE_GET_BIT(probs, i); \ + TREE_GET_BIT(probs, i); \ + i -= 0x40; } +#endif + +#define NORMALIZE_CHECK if (range < kTopValue) { if (buf >= bufLimit) return DUMMY_ERROR; range <<= 8; code = (code << 8) | (*buf++); } + +#define IF_BIT_0_CHECK(p) ttt = *(p); NORMALIZE_CHECK; bound = (range >> kNumBitModelTotalBits) * ttt; if (code < bound) +#define UPDATE_0_CHECK range = bound; +#define UPDATE_1_CHECK range -= bound; code -= bound; +#define GET_BIT2_CHECK(p, i, A0, A1) IF_BIT_0_CHECK(p) \ + { UPDATE_0_CHECK; i = (i + i); A0; } else \ + { UPDATE_1_CHECK; i = (i + i) + 1; A1; } +#define GET_BIT_CHECK(p, i) GET_BIT2_CHECK(p, i, ; , ;) +#define TREE_DECODE_CHECK(probs, limit, i) \ + { i = 1; do { GET_BIT_CHECK(probs + i, i) } while (i < limit); i -= limit; } + + +#define kNumPosBitsMax 4 +#define kNumPosStatesMax (1 << kNumPosBitsMax) + +#define kLenNumLowBits 3 +#define kLenNumLowSymbols (1 << kLenNumLowBits) +#define kLenNumMidBits 3 +#define kLenNumMidSymbols (1 << kLenNumMidBits) +#define kLenNumHighBits 8 +#define kLenNumHighSymbols (1 << kLenNumHighBits) + +#define LenChoice 0 +#define LenChoice2 (LenChoice + 1) +#define LenLow (LenChoice2 + 1) +#define LenMid (LenLow + (kNumPosStatesMax << kLenNumLowBits)) +#define LenHigh (LenMid + (kNumPosStatesMax << kLenNumMidBits)) +#define kNumLenProbs (LenHigh + kLenNumHighSymbols) + + +#define kNumStates 12 +#define kNumLitStates 7 + +#define kStartPosModelIndex 4 +#define kEndPosModelIndex 14 +#define kNumFullDistances (1 << (kEndPosModelIndex >> 1)) + +#define kNumPosSlotBits 6 +#define kNumLenToPosStates 4 + +#define kNumAlignBits 4 +#define kAlignTableSize (1 << kNumAlignBits) + +#define kMatchMinLen 2 +#define kMatchSpecLenStart (kMatchMinLen + kLenNumLowSymbols + kLenNumMidSymbols + kLenNumHighSymbols) + +#define IsMatch 0 +#define IsRep (IsMatch + (kNumStates << kNumPosBitsMax)) +#define IsRepG0 (IsRep + kNumStates) +#define IsRepG1 (IsRepG0 + kNumStates) +#define IsRepG2 (IsRepG1 + kNumStates) +#define IsRep0Long (IsRepG2 + kNumStates) +#define PosSlot (IsRep0Long + (kNumStates << kNumPosBitsMax)) +#define SpecPos (PosSlot + (kNumLenToPosStates << kNumPosSlotBits)) +#define Align (SpecPos + kNumFullDistances - kEndPosModelIndex) +#define LenCoder (Align + kAlignTableSize) +#define RepLenCoder (LenCoder + kNumLenProbs) +#define Literal (RepLenCoder + kNumLenProbs) + +#define LZMA_BASE_SIZE 1846 +#define LZMA_LIT_SIZE 768 + +#define LzmaProps_GetNumProbs(p) ((UInt32)LZMA_BASE_SIZE + (LZMA_LIT_SIZE << ((p)->lc + (p)->lp))) + +#if Literal != LZMA_BASE_SIZE +StopCompilingDueBUG +#endif + +static const Byte kLiteralNextStates[kNumStates * 2] = +{ + 0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 4, 5, + 7, 7, 7, 7, 7, 7, 7, 10, 10, 10, 10, 10 +}; + +#define LZMA_DIC_MIN (1 << 12) + +/* First LZMA-symbol is always decoded. +And it decodes new LZMA-symbols while (buf < bufLimit), but "buf" is without last normalization +Out: + Result: + SZ_OK - OK + SZ_ERROR_DATA - Error + p->remainLen: + < kMatchSpecLenStart : normal remain + = kMatchSpecLenStart : finished + = kMatchSpecLenStart + 1 : Flush marker + = kMatchSpecLenStart + 2 : State Init Marker +*/ + +static int MY_FAST_CALL LzmaDec_DecodeReal(CLzmaDec *p, SizeT limit, const Byte *bufLimit) +{ + CLzmaProb *probs = p->probs; + + unsigned state = p->state; + UInt32 rep0 = p->reps[0], rep1 = p->reps[1], rep2 = p->reps[2], rep3 = p->reps[3]; + unsigned pbMask = ((unsigned)1 << (p->prop.pb)) - 1; + unsigned lpMask = ((unsigned)1 << (p->prop.lp)) - 1; + unsigned lc = p->prop.lc; + + Byte *dic = p->dic; + SizeT dicBufSize = p->dicBufSize; + SizeT dicPos = p->dicPos; + + UInt32 processedPos = p->processedPos; + UInt32 checkDicSize = p->checkDicSize; + unsigned len = 0; + + const Byte *buf = p->buf; + UInt32 range = p->range; + UInt32 code = p->code; + + do + { + CLzmaProb *prob; + UInt32 bound; + unsigned ttt; + unsigned posState = processedPos & pbMask; + + prob = probs + IsMatch + (state << kNumPosBitsMax) + posState; + IF_BIT_0(prob) + { + unsigned symbol; + UPDATE_0(prob); + prob = probs + Literal; + if (checkDicSize != 0 || processedPos != 0) + prob += (LZMA_LIT_SIZE * (((processedPos & lpMask) << lc) + + (dic[(dicPos == 0 ? dicBufSize : dicPos) - 1] >> (8 - lc)))); + + if (state < kNumLitStates) + { + symbol = 1; + do { GET_BIT(prob + symbol, symbol) } while (symbol < 0x100); + } + else + { + unsigned matchByte = p->dic[(dicPos - rep0) + ((dicPos < rep0) ? dicBufSize : 0)]; + unsigned offs = 0x100; + symbol = 1; + do + { + unsigned bit; + CLzmaProb *probLit; + matchByte <<= 1; + bit = (matchByte & offs); + probLit = prob + offs + bit + symbol; + GET_BIT2(probLit, symbol, offs &= ~bit, offs &= bit) + } + while (symbol < 0x100); + } + dic[dicPos++] = (Byte)symbol; + processedPos++; + + state = kLiteralNextStates[state]; + /* if (state < 4) state = 0; else if (state < 10) state -= 3; else state -= 6; */ + continue; + } + else + { + UPDATE_1(prob); + prob = probs + IsRep + state; + IF_BIT_0(prob) + { + UPDATE_0(prob); + state += kNumStates; + prob = probs + LenCoder; + } + else + { + UPDATE_1(prob); + if (checkDicSize == 0 && processedPos == 0) + return SZ_ERROR_DATA; + prob = probs + IsRepG0 + state; + IF_BIT_0(prob) + { + UPDATE_0(prob); + prob = probs + IsRep0Long + (state << kNumPosBitsMax) + posState; + IF_BIT_0(prob) + { + UPDATE_0(prob); + dic[dicPos] = dic[(dicPos - rep0) + ((dicPos < rep0) ? dicBufSize : 0)]; + dicPos++; + processedPos++; + state = state < kNumLitStates ? 9 : 11; + continue; + } + UPDATE_1(prob); + } + else + { + UInt32 distance; + UPDATE_1(prob); + prob = probs + IsRepG1 + state; + IF_BIT_0(prob) + { + UPDATE_0(prob); + distance = rep1; + } + else + { + UPDATE_1(prob); + prob = probs + IsRepG2 + state; + IF_BIT_0(prob) + { + UPDATE_0(prob); + distance = rep2; + } + else + { + UPDATE_1(prob); + distance = rep3; + rep3 = rep2; + } + rep2 = rep1; + } + rep1 = rep0; + rep0 = distance; + } + state = state < kNumLitStates ? 8 : 11; + prob = probs + RepLenCoder; + } + { + unsigned limit, offset; + CLzmaProb *probLen = prob + LenChoice; + IF_BIT_0(probLen) + { + UPDATE_0(probLen); + probLen = prob + LenLow + (posState << kLenNumLowBits); + offset = 0; + limit = (1 << kLenNumLowBits); + } + else + { + UPDATE_1(probLen); + probLen = prob + LenChoice2; + IF_BIT_0(probLen) + { + UPDATE_0(probLen); + probLen = prob + LenMid + (posState << kLenNumMidBits); + offset = kLenNumLowSymbols; + limit = (1 << kLenNumMidBits); + } + else + { + UPDATE_1(probLen); + probLen = prob + LenHigh; + offset = kLenNumLowSymbols + kLenNumMidSymbols; + limit = (1 << kLenNumHighBits); + } + } + TREE_DECODE(probLen, limit, len); + len += offset; + } + + if (state >= kNumStates) + { + UInt32 distance; + prob = probs + PosSlot + + ((len < kNumLenToPosStates ? len : kNumLenToPosStates - 1) << kNumPosSlotBits); + TREE_6_DECODE(prob, distance); + if (distance >= kStartPosModelIndex) + { + unsigned posSlot = (unsigned)distance; + int numDirectBits = (int)(((distance >> 1) - 1)); + distance = (2 | (distance & 1)); + if (posSlot < kEndPosModelIndex) + { + distance <<= numDirectBits; + prob = probs + SpecPos + distance - posSlot - 1; + { + UInt32 mask = 1; + unsigned i = 1; + do + { + GET_BIT2(prob + i, i, ; , distance |= mask); + mask <<= 1; + } + while (--numDirectBits != 0); + } + } + else + { + numDirectBits -= kNumAlignBits; + do + { + NORMALIZE + range >>= 1; + + { + UInt32 t; + code -= range; + t = (0 - ((UInt32)code >> 31)); /* (UInt32)((Int32)code >> 31) */ + distance = (distance << 1) + (t + 1); + code += range & t; + } + /* + distance <<= 1; + if (code >= range) + { + code -= range; + distance |= 1; + } + */ + } + while (--numDirectBits != 0); + prob = probs + Align; + distance <<= kNumAlignBits; + { + unsigned i = 1; + GET_BIT2(prob + i, i, ; , distance |= 1); + GET_BIT2(prob + i, i, ; , distance |= 2); + GET_BIT2(prob + i, i, ; , distance |= 4); + GET_BIT2(prob + i, i, ; , distance |= 8); + } + if (distance == (UInt32)0xFFFFFFFF) + { + len += kMatchSpecLenStart; + state -= kNumStates; + break; + } + } + } + rep3 = rep2; + rep2 = rep1; + rep1 = rep0; + rep0 = distance + 1; + if (checkDicSize == 0) + { + if (distance >= processedPos) + return SZ_ERROR_DATA; + } + else if (distance >= checkDicSize) + return SZ_ERROR_DATA; + state = (state < kNumStates + kNumLitStates) ? kNumLitStates : kNumLitStates + 3; + /* state = kLiteralNextStates[state]; */ + } + + len += kMatchMinLen; + + if (limit == dicPos) + return SZ_ERROR_DATA; + { + SizeT rem = limit - dicPos; + unsigned curLen = ((rem < len) ? (unsigned)rem : len); + SizeT pos = (dicPos - rep0) + ((dicPos < rep0) ? dicBufSize : 0); + + processedPos += curLen; + + len -= curLen; + if (pos + curLen <= dicBufSize) + { + Byte *dest = dic + dicPos; + ptrdiff_t src = (ptrdiff_t)pos - (ptrdiff_t)dicPos; + const Byte *lim = dest + curLen; + dicPos += curLen; + do + *(dest) = (Byte)*(dest + src); + while (++dest != lim); + } + else + { + do + { + dic[dicPos++] = dic[pos]; + if (++pos == dicBufSize) + pos = 0; + } + while (--curLen != 0); + } + } + } + } + while (dicPos < limit && buf < bufLimit); + NORMALIZE; + p->buf = buf; + p->range = range; + p->code = code; + p->remainLen = len; + p->dicPos = dicPos; + p->processedPos = processedPos; + p->reps[0] = rep0; + p->reps[1] = rep1; + p->reps[2] = rep2; + p->reps[3] = rep3; + p->state = state; + + return SZ_OK; +} + +static void MY_FAST_CALL LzmaDec_WriteRem(CLzmaDec *p, SizeT limit) +{ + if (p->remainLen != 0 && p->remainLen < kMatchSpecLenStart) + { + Byte *dic = p->dic; + SizeT dicPos = p->dicPos; + SizeT dicBufSize = p->dicBufSize; + unsigned len = p->remainLen; + UInt32 rep0 = p->reps[0]; + if (limit - dicPos < len) + len = (unsigned)(limit - dicPos); + + if (p->checkDicSize == 0 && p->prop.dicSize - p->processedPos <= len) + p->checkDicSize = p->prop.dicSize; + + p->processedPos += len; + p->remainLen -= len; + while (len-- != 0) + { + dic[dicPos] = dic[(dicPos - rep0) + ((dicPos < rep0) ? dicBufSize : 0)]; + dicPos++; + } + p->dicPos = dicPos; + } +} + +static int MY_FAST_CALL LzmaDec_DecodeReal2(CLzmaDec *p, SizeT limit, const Byte *bufLimit) +{ + do + { + SizeT limit2 = limit; + if (p->checkDicSize == 0) + { + UInt32 rem = p->prop.dicSize - p->processedPos; + if (limit - p->dicPos > rem) + limit2 = p->dicPos + rem; + } + RINOK(LzmaDec_DecodeReal(p, limit2, bufLimit)); + if (p->processedPos >= p->prop.dicSize) + p->checkDicSize = p->prop.dicSize; + LzmaDec_WriteRem(p, limit); + } + while (p->dicPos < limit && p->buf < bufLimit && p->remainLen < kMatchSpecLenStart); + + if (p->remainLen > kMatchSpecLenStart) + { + p->remainLen = kMatchSpecLenStart; + } + return 0; +} + +typedef enum +{ + DUMMY_ERROR, /* unexpected end of input stream */ + DUMMY_LIT, + DUMMY_MATCH, + DUMMY_REP +} ELzmaDummy; + +static ELzmaDummy LzmaDec_TryDummy(const CLzmaDec *p, const Byte *buf, SizeT inSize) +{ + UInt32 range = p->range; + UInt32 code = p->code; + const Byte *bufLimit = buf + inSize; + CLzmaProb *probs = p->probs; + unsigned state = p->state; + ELzmaDummy res; + + { + CLzmaProb *prob; + UInt32 bound; + unsigned ttt; + unsigned posState = (p->processedPos) & ((1 << p->prop.pb) - 1); + + prob = probs + IsMatch + (state << kNumPosBitsMax) + posState; + IF_BIT_0_CHECK(prob) + { + UPDATE_0_CHECK + + /* if (bufLimit - buf >= 7) return DUMMY_LIT; */ + + prob = probs + Literal; + if (p->checkDicSize != 0 || p->processedPos != 0) + prob += (LZMA_LIT_SIZE * + ((((p->processedPos) & ((1 << (p->prop.lp)) - 1)) << p->prop.lc) + + (p->dic[(p->dicPos == 0 ? p->dicBufSize : p->dicPos) - 1] >> (8 - p->prop.lc)))); + + if (state < kNumLitStates) + { + unsigned symbol = 1; + do { GET_BIT_CHECK(prob + symbol, symbol) } while (symbol < 0x100); + } + else + { + unsigned matchByte = p->dic[p->dicPos - p->reps[0] + + ((p->dicPos < p->reps[0]) ? p->dicBufSize : 0)]; + unsigned offs = 0x100; + unsigned symbol = 1; + do + { + unsigned bit; + CLzmaProb *probLit; + matchByte <<= 1; + bit = (matchByte & offs); + probLit = prob + offs + bit + symbol; + GET_BIT2_CHECK(probLit, symbol, offs &= ~bit, offs &= bit) + } + while (symbol < 0x100); + } + res = DUMMY_LIT; + } + else + { + unsigned len; + UPDATE_1_CHECK; + + prob = probs + IsRep + state; + IF_BIT_0_CHECK(prob) + { + UPDATE_0_CHECK; + state = 0; + prob = probs + LenCoder; + res = DUMMY_MATCH; + } + else + { + UPDATE_1_CHECK; + res = DUMMY_REP; + prob = probs + IsRepG0 + state; + IF_BIT_0_CHECK(prob) + { + UPDATE_0_CHECK; + prob = probs + IsRep0Long + (state << kNumPosBitsMax) + posState; + IF_BIT_0_CHECK(prob) + { + UPDATE_0_CHECK; + NORMALIZE_CHECK; + return DUMMY_REP; + } + else + { + UPDATE_1_CHECK; + } + } + else + { + UPDATE_1_CHECK; + prob = probs + IsRepG1 + state; + IF_BIT_0_CHECK(prob) + { + UPDATE_0_CHECK; + } + else + { + UPDATE_1_CHECK; + prob = probs + IsRepG2 + state; + IF_BIT_0_CHECK(prob) + { + UPDATE_0_CHECK; + } + else + { + UPDATE_1_CHECK; + } + } + } + state = kNumStates; + prob = probs + RepLenCoder; + } + { + unsigned limit, offset; + CLzmaProb *probLen = prob + LenChoice; + IF_BIT_0_CHECK(probLen) + { + UPDATE_0_CHECK; + probLen = prob + LenLow + (posState << kLenNumLowBits); + offset = 0; + limit = 1 << kLenNumLowBits; + } + else + { + UPDATE_1_CHECK; + probLen = prob + LenChoice2; + IF_BIT_0_CHECK(probLen) + { + UPDATE_0_CHECK; + probLen = prob + LenMid + (posState << kLenNumMidBits); + offset = kLenNumLowSymbols; + limit = 1 << kLenNumMidBits; + } + else + { + UPDATE_1_CHECK; + probLen = prob + LenHigh; + offset = kLenNumLowSymbols + kLenNumMidSymbols; + limit = 1 << kLenNumHighBits; + } + } + TREE_DECODE_CHECK(probLen, limit, len); + len += offset; + } + + if (state < 4) + { + unsigned posSlot; + prob = probs + PosSlot + + ((len < kNumLenToPosStates ? len : kNumLenToPosStates - 1) << + kNumPosSlotBits); + TREE_DECODE_CHECK(prob, 1 << kNumPosSlotBits, posSlot); + if (posSlot >= kStartPosModelIndex) + { + int numDirectBits = ((posSlot >> 1) - 1); + + /* if (bufLimit - buf >= 8) return DUMMY_MATCH; */ + + if (posSlot < kEndPosModelIndex) + { + prob = probs + SpecPos + ((2 | (posSlot & 1)) << numDirectBits) - posSlot - 1; + } + else + { + numDirectBits -= kNumAlignBits; + do + { + NORMALIZE_CHECK + range >>= 1; + code -= range & (((code - range) >> 31) - 1); + /* if (code >= range) code -= range; */ + } + while (--numDirectBits != 0); + prob = probs + Align; + numDirectBits = kNumAlignBits; + } + { + unsigned i = 1; + do + { + GET_BIT_CHECK(prob + i, i); + } + while (--numDirectBits != 0); + } + } + } + } + } + NORMALIZE_CHECK; + return res; +} + + +static void LzmaDec_InitRc(CLzmaDec *p, const Byte *data) +{ + p->code = ((UInt32)data[1] << 24) | ((UInt32)data[2] << 16) | ((UInt32)data[3] << 8) | ((UInt32)data[4]); + p->range = 0xFFFFFFFF; + p->needFlush = 0; +} + +void LzmaDec_InitDicAndState(CLzmaDec *p, Bool initDic, Bool initState) +{ + p->needFlush = 1; + p->remainLen = 0; + p->tempBufSize = 0; + + if (initDic) + { + p->processedPos = 0; + p->checkDicSize = 0; + p->needInitState = 1; + } + if (initState) + p->needInitState = 1; +} + +void LzmaDec_Init(CLzmaDec *p) +{ + p->dicPos = 0; + LzmaDec_InitDicAndState(p, True, True); +} + +static void LzmaDec_InitStateReal(CLzmaDec *p) +{ + UInt32 numProbs = Literal + ((UInt32)LZMA_LIT_SIZE << (p->prop.lc + p->prop.lp)); + UInt32 i; + CLzmaProb *probs = p->probs; + for (i = 0; i < numProbs; i++) + probs[i] = kBitModelTotal >> 1; + p->reps[0] = p->reps[1] = p->reps[2] = p->reps[3] = 1; + p->state = 0; + p->needInitState = 0; +} + +SRes LzmaDec_DecodeToDic(CLzmaDec *p, SizeT dicLimit, const Byte *src, SizeT *srcLen, + ELzmaFinishMode finishMode, ELzmaStatus *status) +{ + SizeT inSize = *srcLen; + (*srcLen) = 0; + LzmaDec_WriteRem(p, dicLimit); + + *status = LZMA_STATUS_NOT_SPECIFIED; + + while (p->remainLen != kMatchSpecLenStart) + { + int checkEndMarkNow; + + if (p->needFlush != 0) + { + for (; inSize > 0 && p->tempBufSize < RC_INIT_SIZE; (*srcLen)++, inSize--) + p->tempBuf[p->tempBufSize++] = *src++; + if (p->tempBufSize < RC_INIT_SIZE) + { + *status = LZMA_STATUS_NEEDS_MORE_INPUT; + return SZ_OK; + } + if (p->tempBuf[0] != 0) + return SZ_ERROR_DATA; + + LzmaDec_InitRc(p, p->tempBuf); + p->tempBufSize = 0; + } + + checkEndMarkNow = 0; + if (p->dicPos >= dicLimit) + { + if (p->remainLen == 0 && p->code == 0) + { + *status = LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK; + return SZ_OK; + } + if (finishMode == LZMA_FINISH_ANY) + { + *status = LZMA_STATUS_NOT_FINISHED; + return SZ_OK; + } + if (p->remainLen != 0) + { + *status = LZMA_STATUS_NOT_FINISHED; + return SZ_ERROR_DATA; + } + checkEndMarkNow = 1; + } + + if (p->needInitState) + LzmaDec_InitStateReal(p); + + if (p->tempBufSize == 0) + { + SizeT processed; + const Byte *bufLimit; + if (inSize < LZMA_REQUIRED_INPUT_MAX || checkEndMarkNow) + { + int dummyRes = LzmaDec_TryDummy(p, src, inSize); + if (dummyRes == DUMMY_ERROR) + { + memcpy(p->tempBuf, src, inSize); + p->tempBufSize = (unsigned)inSize; + (*srcLen) += inSize; + *status = LZMA_STATUS_NEEDS_MORE_INPUT; + return SZ_OK; + } + if (checkEndMarkNow && dummyRes != DUMMY_MATCH) + { + *status = LZMA_STATUS_NOT_FINISHED; + return SZ_ERROR_DATA; + } + bufLimit = src; + } + else + bufLimit = src + inSize - LZMA_REQUIRED_INPUT_MAX; + p->buf = src; + if (LzmaDec_DecodeReal2(p, dicLimit, bufLimit) != 0) + return SZ_ERROR_DATA; + processed = (SizeT)(p->buf - src); + (*srcLen) += processed; + src += processed; + inSize -= processed; + } + else + { + unsigned rem = p->tempBufSize, lookAhead = 0; + while (rem < LZMA_REQUIRED_INPUT_MAX && lookAhead < inSize) + p->tempBuf[rem++] = src[lookAhead++]; + p->tempBufSize = rem; + if (rem < LZMA_REQUIRED_INPUT_MAX || checkEndMarkNow) + { + int dummyRes = LzmaDec_TryDummy(p, p->tempBuf, rem); + if (dummyRes == DUMMY_ERROR) + { + (*srcLen) += lookAhead; + *status = LZMA_STATUS_NEEDS_MORE_INPUT; + return SZ_OK; + } + if (checkEndMarkNow && dummyRes != DUMMY_MATCH) + { + *status = LZMA_STATUS_NOT_FINISHED; + return SZ_ERROR_DATA; + } + } + p->buf = p->tempBuf; + if (LzmaDec_DecodeReal2(p, dicLimit, p->buf) != 0) + return SZ_ERROR_DATA; + lookAhead -= (rem - (unsigned)(p->buf - p->tempBuf)); + (*srcLen) += lookAhead; + src += lookAhead; + inSize -= lookAhead; + p->tempBufSize = 0; + } + } + if (p->code == 0) + *status = LZMA_STATUS_FINISHED_WITH_MARK; + return (p->code == 0) ? SZ_OK : SZ_ERROR_DATA; +} + +SRes LzmaDec_DecodeToBuf(CLzmaDec *p, Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode, ELzmaStatus *status) +{ + SizeT outSize = *destLen; + SizeT inSize = *srcLen; + *srcLen = *destLen = 0; + for (;;) + { + SizeT inSizeCur = inSize, outSizeCur, dicPos; + ELzmaFinishMode curFinishMode; + SRes res; + if (p->dicPos == p->dicBufSize) + p->dicPos = 0; + dicPos = p->dicPos; + if (outSize > p->dicBufSize - dicPos) + { + outSizeCur = p->dicBufSize; + curFinishMode = LZMA_FINISH_ANY; + } + else + { + outSizeCur = dicPos + outSize; + curFinishMode = finishMode; + } + + res = LzmaDec_DecodeToDic(p, outSizeCur, src, &inSizeCur, curFinishMode, status); + src += inSizeCur; + inSize -= inSizeCur; + *srcLen += inSizeCur; + outSizeCur = p->dicPos - dicPos; + memcpy(dest, p->dic + dicPos, outSizeCur); + dest += outSizeCur; + outSize -= outSizeCur; + *destLen += outSizeCur; + if (res != 0) + return res; + if (outSizeCur == 0 || outSize == 0) + return SZ_OK; + } +} + +void LzmaDec_FreeProbs(CLzmaDec *p, ISzAlloc *alloc) +{ + alloc->Free(alloc, p->probs); + p->probs = 0; +} + +static void LzmaDec_FreeDict(CLzmaDec *p, ISzAlloc *alloc) +{ + alloc->Free(alloc, p->dic); + p->dic = 0; +} + +void LzmaDec_Free(CLzmaDec *p, ISzAlloc *alloc) +{ + LzmaDec_FreeProbs(p, alloc); + LzmaDec_FreeDict(p, alloc); +} + +SRes LzmaProps_Decode(CLzmaProps *p, const Byte *data, unsigned size) +{ + UInt32 dicSize; + Byte d; + + if (size < LZMA_PROPS_SIZE) + return SZ_ERROR_UNSUPPORTED; + else + dicSize = data[1] | ((UInt32)data[2] << 8) | ((UInt32)data[3] << 16) | ((UInt32)data[4] << 24); + + if (dicSize < LZMA_DIC_MIN) + dicSize = LZMA_DIC_MIN; + p->dicSize = dicSize; + + d = data[0]; + if (d >= (9 * 5 * 5)) + return SZ_ERROR_UNSUPPORTED; + + p->lc = d % 9; + d /= 9; + p->pb = d / 5; + p->lp = d % 5; + + return SZ_OK; +} + +static SRes LzmaDec_AllocateProbs2(CLzmaDec *p, const CLzmaProps *propNew, ISzAlloc *alloc) +{ + UInt32 numProbs = LzmaProps_GetNumProbs(propNew); + if (p->probs == 0 || numProbs != p->numProbs) + { + LzmaDec_FreeProbs(p, alloc); + p->probs = (CLzmaProb *)alloc->Alloc(alloc, numProbs * sizeof(CLzmaProb)); + p->numProbs = numProbs; + if (p->probs == 0) + return SZ_ERROR_MEM; + } + return SZ_OK; +} + +SRes LzmaDec_AllocateProbs(CLzmaDec *p, const Byte *props, unsigned propsSize, ISzAlloc *alloc) +{ + CLzmaProps propNew; + RINOK(LzmaProps_Decode(&propNew, props, propsSize)); + RINOK(LzmaDec_AllocateProbs2(p, &propNew, alloc)); + p->prop = propNew; + return SZ_OK; +} + +SRes LzmaDec_Allocate(CLzmaDec *p, const Byte *props, unsigned propsSize, ISzAlloc *alloc) +{ + CLzmaProps propNew; + SizeT dicBufSize; + RINOK(LzmaProps_Decode(&propNew, props, propsSize)); + RINOK(LzmaDec_AllocateProbs2(p, &propNew, alloc)); + dicBufSize = propNew.dicSize; + if (p->dic == 0 || dicBufSize != p->dicBufSize) + { + LzmaDec_FreeDict(p, alloc); + p->dic = (Byte *)alloc->Alloc(alloc, dicBufSize); + if (p->dic == 0) + { + LzmaDec_FreeProbs(p, alloc); + return SZ_ERROR_MEM; + } + } + p->dicBufSize = dicBufSize; + p->prop = propNew; + return SZ_OK; +} + +SRes LzmaDecode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, + const Byte *propData, unsigned propSize, ELzmaFinishMode finishMode, + ELzmaStatus *status, ISzAlloc *alloc) +{ + CLzmaDec p; + SRes res; + SizeT inSize = *srcLen; + SizeT outSize = *destLen; + *srcLen = *destLen = 0; + if (inSize < RC_INIT_SIZE) + return SZ_ERROR_INPUT_EOF; + + LzmaDec_Construct(&p); + res = LzmaDec_AllocateProbs(&p, propData, propSize, alloc); + if (res != 0) + return res; + p.dic = dest; + p.dicBufSize = outSize; + + LzmaDec_Init(&p); + + *srcLen = inSize; + res = LzmaDec_DecodeToDic(&p, outSize, src, srcLen, finishMode, status); + + if (res == SZ_OK && *status == LZMA_STATUS_NEEDS_MORE_INPUT) + res = SZ_ERROR_INPUT_EOF; + + (*destLen) = p.dicPos; + LzmaDec_FreeProbs(&p, alloc); + return res; +} diff --git a/lzma/C/LzmaDec.h b/lzma/C/LzmaDec.h new file mode 100644 index 0000000000..9ec549c86a --- /dev/null +++ b/lzma/C/LzmaDec.h @@ -0,0 +1,237 @@ +/* LzmaDec.h -- LZMA Decoder +2008-10-04 : Igor Pavlov : Public domain */ + +#ifndef __LZMADEC_H +#define __LZMADEC_H + +#include "Types.h" + +/* #define _LZMA_PROB32 */ +/* _LZMA_PROB32 can increase the speed on some CPUs, + but memory usage for CLzmaDec::probs will be doubled in that case */ + +#ifdef _LZMA_PROB32 +#define CLzmaProb UInt32 +#else +#define CLzmaProb UInt16 +#endif + + +/* ---------- LZMA Properties ---------- */ + +#define LZMA_PROPS_SIZE 5 + +typedef struct _CLzmaProps +{ + unsigned lc, lp, pb; + UInt32 dicSize; +} CLzmaProps; + +/* LzmaProps_Decode - decodes properties +Returns: + SZ_OK + SZ_ERROR_UNSUPPORTED - Unsupported properties +*/ + +SRes LzmaProps_Decode(CLzmaProps *p, const Byte *data, unsigned size); + + +/* ---------- LZMA Decoder state ---------- */ + +/* LZMA_REQUIRED_INPUT_MAX = number of required input bytes for worst case. + Num bits = log2((2^11 / 31) ^ 22) + 26 < 134 + 26 = 160; */ + +#define LZMA_REQUIRED_INPUT_MAX 20 + +typedef struct +{ + CLzmaProps prop; + CLzmaProb *probs; + Byte *dic; + const Byte *buf; + UInt32 range, code; + SizeT dicPos; + SizeT dicBufSize; + UInt32 processedPos; + UInt32 checkDicSize; + unsigned state; + UInt32 reps[4]; + unsigned remainLen; + int needFlush; + int needInitState; + UInt32 numProbs; + unsigned tempBufSize; + Byte tempBuf[LZMA_REQUIRED_INPUT_MAX]; +} CLzmaDec; + +#define LzmaDec_Construct(p) { (p)->dic = 0; (p)->probs = 0; } + +#ifdef __cplusplus +extern "C" { +#endif +void LzmaDec_Init(CLzmaDec *p); +#ifdef __cplusplus +} +#endif + +/* There are two types of LZMA streams: + 0) Stream with end mark. That end mark adds about 6 bytes to compressed size. + 1) Stream without end mark. You must know exact uncompressed size to decompress such stream. */ + +typedef enum +{ + LZMA_FINISH_ANY, /* finish at any point */ + LZMA_FINISH_END /* block must be finished at the end */ +} ELzmaFinishMode; + +/* ELzmaFinishMode has meaning only if the decoding reaches output limit !!! + + You must use LZMA_FINISH_END, when you know that current output buffer + covers last bytes of block. In other cases you must use LZMA_FINISH_ANY. + + If LZMA decoder sees end marker before reaching output limit, it returns SZ_OK, + and output value of destLen will be less than output buffer size limit. + You can check status result also. + + You can use multiple checks to test data integrity after full decompression: + 1) Check Result and "status" variable. + 2) Check that output(destLen) = uncompressedSize, if you know real uncompressedSize. + 3) Check that output(srcLen) = compressedSize, if you know real compressedSize. + You must use correct finish mode in that case. */ + +typedef enum +{ + LZMA_STATUS_NOT_SPECIFIED, /* use main error code instead */ + LZMA_STATUS_FINISHED_WITH_MARK, /* stream was finished with end mark. */ + LZMA_STATUS_NOT_FINISHED, /* stream was not finished */ + LZMA_STATUS_NEEDS_MORE_INPUT, /* you must provide more input bytes */ + LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK /* there is probability that stream was finished without end mark */ +} ELzmaStatus; + +/* ELzmaStatus is used only as output value for function call */ + + +/* ---------- Interfaces ---------- */ + +/* There are 3 levels of interfaces: + 1) Dictionary Interface + 2) Buffer Interface + 3) One Call Interface + You can select any of these interfaces, but don't mix functions from different + groups for same object. */ + + +/* There are two variants to allocate state for Dictionary Interface: + 1) LzmaDec_Allocate / LzmaDec_Free + 2) LzmaDec_AllocateProbs / LzmaDec_FreeProbs + You can use variant 2, if you set dictionary buffer manually. + For Buffer Interface you must always use variant 1. + +LzmaDec_Allocate* can return: + SZ_OK + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_UNSUPPORTED - Unsupported properties +*/ + +#ifdef __cplusplus +extern "C" { +#endif + +SRes LzmaDec_AllocateProbs(CLzmaDec *p, const Byte *props, unsigned propsSize, ISzAlloc *alloc); +void LzmaDec_FreeProbs(CLzmaDec *p, ISzAlloc *alloc); + +SRes LzmaDec_Allocate(CLzmaDec *state, const Byte *prop, unsigned propsSize, ISzAlloc *alloc); +void LzmaDec_Free(CLzmaDec *state, ISzAlloc *alloc); + +/* ---------- Dictionary Interface ---------- */ + +/* You can use it, if you want to eliminate the overhead for data copying from + dictionary to some other external buffer. + You must work with CLzmaDec variables directly in this interface. + + STEPS: + LzmaDec_Constr() + LzmaDec_Allocate() + for (each new stream) + { + LzmaDec_Init() + while (it needs more decompression) + { + LzmaDec_DecodeToDic() + use data from CLzmaDec::dic and update CLzmaDec::dicPos + } + } + LzmaDec_Free() +*/ + +/* LzmaDec_DecodeToDic + + The decoding to internal dictionary buffer (CLzmaDec::dic). + You must manually update CLzmaDec::dicPos, if it reaches CLzmaDec::dicBufSize !!! + +finishMode: + It has meaning only if the decoding reaches output limit (dicLimit). + LZMA_FINISH_ANY - Decode just dicLimit bytes. + LZMA_FINISH_END - Stream must be finished after dicLimit. + +Returns: + SZ_OK + status: + LZMA_STATUS_FINISHED_WITH_MARK + LZMA_STATUS_NOT_FINISHED + LZMA_STATUS_NEEDS_MORE_INPUT + LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK + SZ_ERROR_DATA - Data error +*/ + +SRes LzmaDec_DecodeToDic(CLzmaDec *p, SizeT dicLimit, + const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode, ELzmaStatus *status); + + +/* ---------- Buffer Interface ---------- */ + +/* It's zlib-like interface. + See LzmaDec_DecodeToDic description for information about STEPS and return results, + but you must use LzmaDec_DecodeToBuf instead of LzmaDec_DecodeToDic and you don't need + to work with CLzmaDec variables manually. + +finishMode: + It has meaning only if the decoding reaches output limit (*destLen). + LZMA_FINISH_ANY - Decode just destLen bytes. + LZMA_FINISH_END - Stream must be finished after (*destLen). +*/ + +SRes LzmaDec_DecodeToBuf(CLzmaDec *p, Byte *dest, SizeT *destLen, + const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode, ELzmaStatus *status); + + +/* ---------- One Call Interface ---------- */ + +/* LzmaDecode + +finishMode: + It has meaning only if the decoding reaches output limit (*destLen). + LZMA_FINISH_ANY - Decode just destLen bytes. + LZMA_FINISH_END - Stream must be finished after (*destLen). + +Returns: + SZ_OK + status: + LZMA_STATUS_FINISHED_WITH_MARK + LZMA_STATUS_NOT_FINISHED + LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK + SZ_ERROR_DATA - Data error + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_UNSUPPORTED - Unsupported properties + SZ_ERROR_INPUT_EOF - It needs more bytes in input buffer (src). +*/ + +SRes LzmaDecode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, + const Byte *propData, unsigned propSize, ELzmaFinishMode finishMode, + ELzmaStatus *status, ISzAlloc *alloc); + +#ifdef __cplusplus +} +#endif + +#endif diff --git a/lzma/C/LzmaEnc.c b/lzma/C/LzmaEnc.c new file mode 100644 index 0000000000..529fd98f87 --- /dev/null +++ b/lzma/C/LzmaEnc.c @@ -0,0 +1,2281 @@ +/* LzmaEnc.c -- LZMA Encoder +2009-02-02 : Igor Pavlov : Public domain */ + +#include + +/* #define SHOW_STAT */ +/* #define SHOW_STAT2 */ + +#if defined(SHOW_STAT) || defined(SHOW_STAT2) +#include +#endif + +#include "LzmaEnc.h" + +#include "LzFind.h" +#ifdef COMPRESS_MF_MT +#include "LzFindMt.h" +#endif + +#ifdef SHOW_STAT +static int ttt = 0; +#endif + +#define kBlockSizeMax ((1 << LZMA_NUM_BLOCK_SIZE_BITS) - 1) + +#define kBlockSize (9 << 10) +#define kUnpackBlockSize (1 << 18) +#define kMatchArraySize (1 << 21) +#define kMatchRecordMaxSize ((LZMA_MATCH_LEN_MAX * 2 + 3) * LZMA_MATCH_LEN_MAX) + +#define kNumMaxDirectBits (31) + +#define kNumTopBits 24 +#define kTopValue ((UInt32)1 << kNumTopBits) + +#define kNumBitModelTotalBits 11 +#define kBitModelTotal (1 << kNumBitModelTotalBits) +#define kNumMoveBits 5 +#define kProbInitValue (kBitModelTotal >> 1) + +#define kNumMoveReducingBits 4 +#define kNumBitPriceShiftBits 4 +#define kBitPrice (1 << kNumBitPriceShiftBits) + +void LzmaEncProps_Init(CLzmaEncProps *p) +{ + p->level = 5; + p->dictSize = p->mc = 0; + p->lc = p->lp = p->pb = p->algo = p->fb = p->btMode = p->numHashBytes = p->numThreads = -1; + p->writeEndMark = 0; +} + +void LzmaEncProps_Normalize(CLzmaEncProps *p) +{ + int level = p->level; + if (level < 0) level = 5; + p->level = level; + if (p->dictSize == 0) p->dictSize = (level <= 5 ? (1 << (level * 2 + 14)) : (level == 6 ? (1 << 25) : (1 << 26))); + if (p->lc < 0) p->lc = 3; + if (p->lp < 0) p->lp = 0; + if (p->pb < 0) p->pb = 2; + if (p->algo < 0) p->algo = (level < 5 ? 0 : 1); + if (p->fb < 0) p->fb = (level < 7 ? 32 : 64); + if (p->btMode < 0) p->btMode = (p->algo == 0 ? 0 : 1); + if (p->numHashBytes < 0) p->numHashBytes = 4; + if (p->mc == 0) p->mc = (16 + (p->fb >> 1)) >> (p->btMode ? 0 : 1); + if (p->numThreads < 0) + p->numThreads = + #ifdef COMPRESS_MF_MT + ((p->btMode && p->algo) ? 2 : 1); + #else + 1; + #endif +} + +UInt32 LzmaEncProps_GetDictSize(const CLzmaEncProps *props2) +{ + CLzmaEncProps props = *props2; + LzmaEncProps_Normalize(&props); + return props.dictSize; +} + +/* #define LZMA_LOG_BSR */ +/* Define it for Intel's CPU */ + + +#ifdef LZMA_LOG_BSR + +#define kDicLogSizeMaxCompress 30 + +#define BSR2_RET(pos, res) { unsigned long i; _BitScanReverse(&i, (pos)); res = (i + i) + ((pos >> (i - 1)) & 1); } + +UInt32 GetPosSlot1(UInt32 pos) +{ + UInt32 res; + BSR2_RET(pos, res); + return res; +} +#define GetPosSlot2(pos, res) { BSR2_RET(pos, res); } +#define GetPosSlot(pos, res) { if (pos < 2) res = pos; else BSR2_RET(pos, res); } + +#else + +#define kNumLogBits (9 + (int)sizeof(size_t) / 2) +#define kDicLogSizeMaxCompress ((kNumLogBits - 1) * 2 + 7) + +void LzmaEnc_FastPosInit(Byte *g_FastPos) +{ + int c = 2, slotFast; + g_FastPos[0] = 0; + g_FastPos[1] = 1; + + for (slotFast = 2; slotFast < kNumLogBits * 2; slotFast++) + { + UInt32 k = (1 << ((slotFast >> 1) - 1)); + UInt32 j; + for (j = 0; j < k; j++, c++) + g_FastPos[c] = (Byte)slotFast; + } +} + +#define BSR2_RET(pos, res) { UInt32 i = 6 + ((kNumLogBits - 1) & \ + (0 - (((((UInt32)1 << (kNumLogBits + 6)) - 1) - pos) >> 31))); \ + res = p->g_FastPos[pos >> i] + (i * 2); } +/* +#define BSR2_RET(pos, res) { res = (pos < (1 << (kNumLogBits + 6))) ? \ + p->g_FastPos[pos >> 6] + 12 : \ + p->g_FastPos[pos >> (6 + kNumLogBits - 1)] + (6 + (kNumLogBits - 1)) * 2; } +*/ + +#define GetPosSlot1(pos) p->g_FastPos[pos] +#define GetPosSlot2(pos, res) { BSR2_RET(pos, res); } +#define GetPosSlot(pos, res) { if (pos < kNumFullDistances) res = p->g_FastPos[pos]; else BSR2_RET(pos, res); } + +#endif + + +#define LZMA_NUM_REPS 4 + +typedef unsigned CState; + +typedef struct _COptimal +{ + UInt32 price; + + CState state; + int prev1IsChar; + int prev2; + + UInt32 posPrev2; + UInt32 backPrev2; + + UInt32 posPrev; + UInt32 backPrev; + UInt32 backs[LZMA_NUM_REPS]; +} COptimal; + +#define kNumOpts (1 << 12) + +#define kNumLenToPosStates 4 +#define kNumPosSlotBits 6 +#define kDicLogSizeMin 0 +#define kDicLogSizeMax 32 +#define kDistTableSizeMax (kDicLogSizeMax * 2) + + +#define kNumAlignBits 4 +#define kAlignTableSize (1 << kNumAlignBits) +#define kAlignMask (kAlignTableSize - 1) + +#define kStartPosModelIndex 4 +#define kEndPosModelIndex 14 +#define kNumPosModels (kEndPosModelIndex - kStartPosModelIndex) + +#define kNumFullDistances (1 << (kEndPosModelIndex / 2)) + +#ifdef _LZMA_PROB32 +#define CLzmaProb UInt32 +#else +#define CLzmaProb UInt16 +#endif + +#define LZMA_PB_MAX 4 +#define LZMA_LC_MAX 8 +#define LZMA_LP_MAX 4 + +#define LZMA_NUM_PB_STATES_MAX (1 << LZMA_PB_MAX) + + +#define kLenNumLowBits 3 +#define kLenNumLowSymbols (1 << kLenNumLowBits) +#define kLenNumMidBits 3 +#define kLenNumMidSymbols (1 << kLenNumMidBits) +#define kLenNumHighBits 8 +#define kLenNumHighSymbols (1 << kLenNumHighBits) + +#define kLenNumSymbolsTotal (kLenNumLowSymbols + kLenNumMidSymbols + kLenNumHighSymbols) + +#define LZMA_MATCH_LEN_MIN 2 +#define LZMA_MATCH_LEN_MAX (LZMA_MATCH_LEN_MIN + kLenNumSymbolsTotal - 1) + +#define kNumStates 12 + +typedef struct +{ + CLzmaProb choice; + CLzmaProb choice2; + CLzmaProb low[LZMA_NUM_PB_STATES_MAX << kLenNumLowBits]; + CLzmaProb mid[LZMA_NUM_PB_STATES_MAX << kLenNumMidBits]; + CLzmaProb high[kLenNumHighSymbols]; +} CLenEnc; + +typedef struct +{ + CLenEnc p; + UInt32 prices[LZMA_NUM_PB_STATES_MAX][kLenNumSymbolsTotal]; + UInt32 tableSize; + UInt32 counters[LZMA_NUM_PB_STATES_MAX]; +} CLenPriceEnc; + +typedef struct _CRangeEnc +{ + UInt32 range; + Byte cache; + UInt64 low; + UInt64 cacheSize; + Byte *buf; + Byte *bufLim; + Byte *bufBase; + ISeqOutStream *outStream; + UInt64 processed; + SRes res; +} CRangeEnc; + +typedef struct _CSeqInStreamBuf +{ + ISeqInStream funcTable; + const Byte *data; + SizeT rem; +} CSeqInStreamBuf; + +static SRes MyRead(void *pp, void *data, size_t *size) +{ + size_t curSize = *size; + CSeqInStreamBuf *p = (CSeqInStreamBuf *)pp; + if (p->rem < curSize) + curSize = p->rem; + memcpy(data, p->data, curSize); + p->rem -= curSize; + p->data += curSize; + *size = curSize; + return SZ_OK; +} + +typedef struct +{ + CLzmaProb *litProbs; + + CLzmaProb isMatch[kNumStates][LZMA_NUM_PB_STATES_MAX]; + CLzmaProb isRep[kNumStates]; + CLzmaProb isRepG0[kNumStates]; + CLzmaProb isRepG1[kNumStates]; + CLzmaProb isRepG2[kNumStates]; + CLzmaProb isRep0Long[kNumStates][LZMA_NUM_PB_STATES_MAX]; + + CLzmaProb posSlotEncoder[kNumLenToPosStates][1 << kNumPosSlotBits]; + CLzmaProb posEncoders[kNumFullDistances - kEndPosModelIndex]; + CLzmaProb posAlignEncoder[1 << kNumAlignBits]; + + CLenPriceEnc lenEnc; + CLenPriceEnc repLenEnc; + + UInt32 reps[LZMA_NUM_REPS]; + UInt32 state; +} CSaveState; + +typedef struct _CLzmaEnc +{ + IMatchFinder matchFinder; + void *matchFinderObj; + + #ifdef COMPRESS_MF_MT + Bool mtMode; + CMatchFinderMt matchFinderMt; + #endif + + CMatchFinder matchFinderBase; + + #ifdef COMPRESS_MF_MT + Byte pad[128]; + #endif + + UInt32 optimumEndIndex; + UInt32 optimumCurrentIndex; + + UInt32 longestMatchLength; + UInt32 numPairs; + UInt32 numAvail; + COptimal opt[kNumOpts]; + + #ifndef LZMA_LOG_BSR + Byte g_FastPos[1 << kNumLogBits]; + #endif + + UInt32 ProbPrices[kBitModelTotal >> kNumMoveReducingBits]; + UInt32 matches[LZMA_MATCH_LEN_MAX * 2 + 2 + 1]; + UInt32 numFastBytes; + UInt32 additionalOffset; + UInt32 reps[LZMA_NUM_REPS]; + UInt32 state; + + UInt32 posSlotPrices[kNumLenToPosStates][kDistTableSizeMax]; + UInt32 distancesPrices[kNumLenToPosStates][kNumFullDistances]; + UInt32 alignPrices[kAlignTableSize]; + UInt32 alignPriceCount; + + UInt32 distTableSize; + + unsigned lc, lp, pb; + unsigned lpMask, pbMask; + + CLzmaProb *litProbs; + + CLzmaProb isMatch[kNumStates][LZMA_NUM_PB_STATES_MAX]; + CLzmaProb isRep[kNumStates]; + CLzmaProb isRepG0[kNumStates]; + CLzmaProb isRepG1[kNumStates]; + CLzmaProb isRepG2[kNumStates]; + CLzmaProb isRep0Long[kNumStates][LZMA_NUM_PB_STATES_MAX]; + + CLzmaProb posSlotEncoder[kNumLenToPosStates][1 << kNumPosSlotBits]; + CLzmaProb posEncoders[kNumFullDistances - kEndPosModelIndex]; + CLzmaProb posAlignEncoder[1 << kNumAlignBits]; + + CLenPriceEnc lenEnc; + CLenPriceEnc repLenEnc; + + unsigned lclp; + + Bool fastMode; + + CRangeEnc rc; + + Bool writeEndMark; + UInt64 nowPos64; + UInt32 matchPriceCount; + Bool finished; + Bool multiThread; + + SRes result; + UInt32 dictSize; + UInt32 matchFinderCycles; + + ISeqInStream *inStream; + CSeqInStreamBuf seqBufInStream; + + CSaveState saveState; +} CLzmaEnc; + +void LzmaEnc_SaveState(CLzmaEncHandle pp) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + CSaveState *dest = &p->saveState; + int i; + dest->lenEnc = p->lenEnc; + dest->repLenEnc = p->repLenEnc; + dest->state = p->state; + + for (i = 0; i < kNumStates; i++) + { + memcpy(dest->isMatch[i], p->isMatch[i], sizeof(p->isMatch[i])); + memcpy(dest->isRep0Long[i], p->isRep0Long[i], sizeof(p->isRep0Long[i])); + } + for (i = 0; i < kNumLenToPosStates; i++) + memcpy(dest->posSlotEncoder[i], p->posSlotEncoder[i], sizeof(p->posSlotEncoder[i])); + memcpy(dest->isRep, p->isRep, sizeof(p->isRep)); + memcpy(dest->isRepG0, p->isRepG0, sizeof(p->isRepG0)); + memcpy(dest->isRepG1, p->isRepG1, sizeof(p->isRepG1)); + memcpy(dest->isRepG2, p->isRepG2, sizeof(p->isRepG2)); + memcpy(dest->posEncoders, p->posEncoders, sizeof(p->posEncoders)); + memcpy(dest->posAlignEncoder, p->posAlignEncoder, sizeof(p->posAlignEncoder)); + memcpy(dest->reps, p->reps, sizeof(p->reps)); + memcpy(dest->litProbs, p->litProbs, (0x300 << p->lclp) * sizeof(CLzmaProb)); +} + +void LzmaEnc_RestoreState(CLzmaEncHandle pp) +{ + CLzmaEnc *dest = (CLzmaEnc *)pp; + const CSaveState *p = &dest->saveState; + int i; + dest->lenEnc = p->lenEnc; + dest->repLenEnc = p->repLenEnc; + dest->state = p->state; + + for (i = 0; i < kNumStates; i++) + { + memcpy(dest->isMatch[i], p->isMatch[i], sizeof(p->isMatch[i])); + memcpy(dest->isRep0Long[i], p->isRep0Long[i], sizeof(p->isRep0Long[i])); + } + for (i = 0; i < kNumLenToPosStates; i++) + memcpy(dest->posSlotEncoder[i], p->posSlotEncoder[i], sizeof(p->posSlotEncoder[i])); + memcpy(dest->isRep, p->isRep, sizeof(p->isRep)); + memcpy(dest->isRepG0, p->isRepG0, sizeof(p->isRepG0)); + memcpy(dest->isRepG1, p->isRepG1, sizeof(p->isRepG1)); + memcpy(dest->isRepG2, p->isRepG2, sizeof(p->isRepG2)); + memcpy(dest->posEncoders, p->posEncoders, sizeof(p->posEncoders)); + memcpy(dest->posAlignEncoder, p->posAlignEncoder, sizeof(p->posAlignEncoder)); + memcpy(dest->reps, p->reps, sizeof(p->reps)); + memcpy(dest->litProbs, p->litProbs, (0x300 << dest->lclp) * sizeof(CLzmaProb)); +} + +SRes LzmaEnc_SetProps(CLzmaEncHandle pp, const CLzmaEncProps *props2) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + CLzmaEncProps props = *props2; + LzmaEncProps_Normalize(&props); + + if (props.lc > LZMA_LC_MAX || props.lp > LZMA_LP_MAX || props.pb > LZMA_PB_MAX || + props.dictSize > (1 << kDicLogSizeMaxCompress) || props.dictSize > (1 << 30)) + return SZ_ERROR_PARAM; + p->dictSize = props.dictSize; + p->matchFinderCycles = props.mc; + { + unsigned fb = props.fb; + if (fb < 5) + fb = 5; + if (fb > LZMA_MATCH_LEN_MAX) + fb = LZMA_MATCH_LEN_MAX; + p->numFastBytes = fb; + } + p->lc = props.lc; + p->lp = props.lp; + p->pb = props.pb; + p->fastMode = (props.algo == 0); + p->matchFinderBase.btMode = props.btMode; + { + UInt32 numHashBytes = 4; + if (props.btMode) + { + if (props.numHashBytes < 2) + numHashBytes = 2; + else if (props.numHashBytes < 4) + numHashBytes = props.numHashBytes; + } + p->matchFinderBase.numHashBytes = numHashBytes; + } + + p->matchFinderBase.cutValue = props.mc; + + p->writeEndMark = props.writeEndMark; + + #ifdef COMPRESS_MF_MT + /* + if (newMultiThread != _multiThread) + { + ReleaseMatchFinder(); + _multiThread = newMultiThread; + } + */ + p->multiThread = (props.numThreads > 1); + #endif + + return SZ_OK; +} + +static const int kLiteralNextStates[kNumStates] = {0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 4, 5}; +static const int kMatchNextStates[kNumStates] = {7, 7, 7, 7, 7, 7, 7, 10, 10, 10, 10, 10}; +static const int kRepNextStates[kNumStates] = {8, 8, 8, 8, 8, 8, 8, 11, 11, 11, 11, 11}; +static const int kShortRepNextStates[kNumStates]= {9, 9, 9, 9, 9, 9, 9, 11, 11, 11, 11, 11}; + +#define IsCharState(s) ((s) < 7) + +#define GetLenToPosState(len) (((len) < kNumLenToPosStates + 1) ? (len) - 2 : kNumLenToPosStates - 1) + +#define kInfinityPrice (1 << 30) + +static void RangeEnc_Construct(CRangeEnc *p) +{ + p->outStream = 0; + p->bufBase = 0; +} + +#define RangeEnc_GetProcessed(p) ((p)->processed + ((p)->buf - (p)->bufBase) + (p)->cacheSize) + +#define RC_BUF_SIZE (1 << 16) +static int RangeEnc_Alloc(CRangeEnc *p, ISzAlloc *alloc) +{ + if (p->bufBase == 0) + { + p->bufBase = (Byte *)alloc->Alloc(alloc, RC_BUF_SIZE); + if (p->bufBase == 0) + return 0; + p->bufLim = p->bufBase + RC_BUF_SIZE; + } + return 1; +} + +static void RangeEnc_Free(CRangeEnc *p, ISzAlloc *alloc) +{ + alloc->Free(alloc, p->bufBase); + p->bufBase = 0; +} + +static void RangeEnc_Init(CRangeEnc *p) +{ + /* Stream.Init(); */ + p->low = 0; + p->range = 0xFFFFFFFF; + p->cacheSize = 1; + p->cache = 0; + + p->buf = p->bufBase; + + p->processed = 0; + p->res = SZ_OK; +} + +static void RangeEnc_FlushStream(CRangeEnc *p) +{ + size_t num; + if (p->res != SZ_OK) + return; + num = p->buf - p->bufBase; + if (num != p->outStream->Write(p->outStream, p->bufBase, num)) + p->res = SZ_ERROR_WRITE; + p->processed += num; + p->buf = p->bufBase; +} + +static void MY_FAST_CALL RangeEnc_ShiftLow(CRangeEnc *p) +{ + if ((UInt32)p->low < (UInt32)0xFF000000 || (int)(p->low >> 32) != 0) + { + Byte temp = p->cache; + do + { + Byte *buf = p->buf; + *buf++ = (Byte)(temp + (Byte)(p->low >> 32)); + p->buf = buf; + if (buf == p->bufLim) + RangeEnc_FlushStream(p); + temp = 0xFF; + } + while (--p->cacheSize != 0); + p->cache = (Byte)((UInt32)p->low >> 24); + } + p->cacheSize++; + p->low = (UInt32)p->low << 8; +} + +static void RangeEnc_FlushData(CRangeEnc *p) +{ + int i; + for (i = 0; i < 5; i++) + RangeEnc_ShiftLow(p); +} + +static void RangeEnc_EncodeDirectBits(CRangeEnc *p, UInt32 value, int numBits) +{ + do + { + p->range >>= 1; + p->low += p->range & (0 - ((value >> --numBits) & 1)); + if (p->range < kTopValue) + { + p->range <<= 8; + RangeEnc_ShiftLow(p); + } + } + while (numBits != 0); +} + +static void RangeEnc_EncodeBit(CRangeEnc *p, CLzmaProb *prob, UInt32 symbol) +{ + UInt32 ttt = *prob; + UInt32 newBound = (p->range >> kNumBitModelTotalBits) * ttt; + if (symbol == 0) + { + p->range = newBound; + ttt += (kBitModelTotal - ttt) >> kNumMoveBits; + } + else + { + p->low += newBound; + p->range -= newBound; + ttt -= ttt >> kNumMoveBits; + } + *prob = (CLzmaProb)ttt; + if (p->range < kTopValue) + { + p->range <<= 8; + RangeEnc_ShiftLow(p); + } +} + +static void LitEnc_Encode(CRangeEnc *p, CLzmaProb *probs, UInt32 symbol) +{ + symbol |= 0x100; + do + { + RangeEnc_EncodeBit(p, probs + (symbol >> 8), (symbol >> 7) & 1); + symbol <<= 1; + } + while (symbol < 0x10000); +} + +static void LitEnc_EncodeMatched(CRangeEnc *p, CLzmaProb *probs, UInt32 symbol, UInt32 matchByte) +{ + UInt32 offs = 0x100; + symbol |= 0x100; + do + { + matchByte <<= 1; + RangeEnc_EncodeBit(p, probs + (offs + (matchByte & offs) + (symbol >> 8)), (symbol >> 7) & 1); + symbol <<= 1; + offs &= ~(matchByte ^ symbol); + } + while (symbol < 0x10000); +} + +void LzmaEnc_InitPriceTables(UInt32 *ProbPrices) +{ + UInt32 i; + for (i = (1 << kNumMoveReducingBits) / 2; i < kBitModelTotal; i += (1 << kNumMoveReducingBits)) + { + const int kCyclesBits = kNumBitPriceShiftBits; + UInt32 w = i; + UInt32 bitCount = 0; + int j; + for (j = 0; j < kCyclesBits; j++) + { + w = w * w; + bitCount <<= 1; + while (w >= ((UInt32)1 << 16)) + { + w >>= 1; + bitCount++; + } + } + ProbPrices[i >> kNumMoveReducingBits] = ((kNumBitModelTotalBits << kCyclesBits) - 15 - bitCount); + } +} + + +#define GET_PRICE(prob, symbol) \ + p->ProbPrices[((prob) ^ (((-(int)(symbol))) & (kBitModelTotal - 1))) >> kNumMoveReducingBits]; + +#define GET_PRICEa(prob, symbol) \ + ProbPrices[((prob) ^ ((-((int)(symbol))) & (kBitModelTotal - 1))) >> kNumMoveReducingBits]; + +#define GET_PRICE_0(prob) p->ProbPrices[(prob) >> kNumMoveReducingBits] +#define GET_PRICE_1(prob) p->ProbPrices[((prob) ^ (kBitModelTotal - 1)) >> kNumMoveReducingBits] + +#define GET_PRICE_0a(prob) ProbPrices[(prob) >> kNumMoveReducingBits] +#define GET_PRICE_1a(prob) ProbPrices[((prob) ^ (kBitModelTotal - 1)) >> kNumMoveReducingBits] + +static UInt32 LitEnc_GetPrice(const CLzmaProb *probs, UInt32 symbol, UInt32 *ProbPrices) +{ + UInt32 price = 0; + symbol |= 0x100; + do + { + price += GET_PRICEa(probs[symbol >> 8], (symbol >> 7) & 1); + symbol <<= 1; + } + while (symbol < 0x10000); + return price; +} + +static UInt32 LitEnc_GetPriceMatched(const CLzmaProb *probs, UInt32 symbol, UInt32 matchByte, UInt32 *ProbPrices) +{ + UInt32 price = 0; + UInt32 offs = 0x100; + symbol |= 0x100; + do + { + matchByte <<= 1; + price += GET_PRICEa(probs[offs + (matchByte & offs) + (symbol >> 8)], (symbol >> 7) & 1); + symbol <<= 1; + offs &= ~(matchByte ^ symbol); + } + while (symbol < 0x10000); + return price; +} + + +static void RcTree_Encode(CRangeEnc *rc, CLzmaProb *probs, int numBitLevels, UInt32 symbol) +{ + UInt32 m = 1; + int i; + for (i = numBitLevels; i != 0;) + { + UInt32 bit; + i--; + bit = (symbol >> i) & 1; + RangeEnc_EncodeBit(rc, probs + m, bit); + m = (m << 1) | bit; + } +} + +static void RcTree_ReverseEncode(CRangeEnc *rc, CLzmaProb *probs, int numBitLevels, UInt32 symbol) +{ + UInt32 m = 1; + int i; + for (i = 0; i < numBitLevels; i++) + { + UInt32 bit = symbol & 1; + RangeEnc_EncodeBit(rc, probs + m, bit); + m = (m << 1) | bit; + symbol >>= 1; + } +} + +static UInt32 RcTree_GetPrice(const CLzmaProb *probs, int numBitLevels, UInt32 symbol, UInt32 *ProbPrices) +{ + UInt32 price = 0; + symbol |= (1 << numBitLevels); + while (symbol != 1) + { + price += GET_PRICEa(probs[symbol >> 1], symbol & 1); + symbol >>= 1; + } + return price; +} + +static UInt32 RcTree_ReverseGetPrice(const CLzmaProb *probs, int numBitLevels, UInt32 symbol, UInt32 *ProbPrices) +{ + UInt32 price = 0; + UInt32 m = 1; + int i; + for (i = numBitLevels; i != 0; i--) + { + UInt32 bit = symbol & 1; + symbol >>= 1; + price += GET_PRICEa(probs[m], bit); + m = (m << 1) | bit; + } + return price; +} + + +static void LenEnc_Init(CLenEnc *p) +{ + unsigned i; + p->choice = p->choice2 = kProbInitValue; + for (i = 0; i < (LZMA_NUM_PB_STATES_MAX << kLenNumLowBits); i++) + p->low[i] = kProbInitValue; + for (i = 0; i < (LZMA_NUM_PB_STATES_MAX << kLenNumMidBits); i++) + p->mid[i] = kProbInitValue; + for (i = 0; i < kLenNumHighSymbols; i++) + p->high[i] = kProbInitValue; +} + +static void LenEnc_Encode(CLenEnc *p, CRangeEnc *rc, UInt32 symbol, UInt32 posState) +{ + if (symbol < kLenNumLowSymbols) + { + RangeEnc_EncodeBit(rc, &p->choice, 0); + RcTree_Encode(rc, p->low + (posState << kLenNumLowBits), kLenNumLowBits, symbol); + } + else + { + RangeEnc_EncodeBit(rc, &p->choice, 1); + if (symbol < kLenNumLowSymbols + kLenNumMidSymbols) + { + RangeEnc_EncodeBit(rc, &p->choice2, 0); + RcTree_Encode(rc, p->mid + (posState << kLenNumMidBits), kLenNumMidBits, symbol - kLenNumLowSymbols); + } + else + { + RangeEnc_EncodeBit(rc, &p->choice2, 1); + RcTree_Encode(rc, p->high, kLenNumHighBits, symbol - kLenNumLowSymbols - kLenNumMidSymbols); + } + } +} + +static void LenEnc_SetPrices(CLenEnc *p, UInt32 posState, UInt32 numSymbols, UInt32 *prices, UInt32 *ProbPrices) +{ + UInt32 a0 = GET_PRICE_0a(p->choice); + UInt32 a1 = GET_PRICE_1a(p->choice); + UInt32 b0 = a1 + GET_PRICE_0a(p->choice2); + UInt32 b1 = a1 + GET_PRICE_1a(p->choice2); + UInt32 i = 0; + for (i = 0; i < kLenNumLowSymbols; i++) + { + if (i >= numSymbols) + return; + prices[i] = a0 + RcTree_GetPrice(p->low + (posState << kLenNumLowBits), kLenNumLowBits, i, ProbPrices); + } + for (; i < kLenNumLowSymbols + kLenNumMidSymbols; i++) + { + if (i >= numSymbols) + return; + prices[i] = b0 + RcTree_GetPrice(p->mid + (posState << kLenNumMidBits), kLenNumMidBits, i - kLenNumLowSymbols, ProbPrices); + } + for (; i < numSymbols; i++) + prices[i] = b1 + RcTree_GetPrice(p->high, kLenNumHighBits, i - kLenNumLowSymbols - kLenNumMidSymbols, ProbPrices); +} + +static void MY_FAST_CALL LenPriceEnc_UpdateTable(CLenPriceEnc *p, UInt32 posState, UInt32 *ProbPrices) +{ + LenEnc_SetPrices(&p->p, posState, p->tableSize, p->prices[posState], ProbPrices); + p->counters[posState] = p->tableSize; +} + +static void LenPriceEnc_UpdateTables(CLenPriceEnc *p, UInt32 numPosStates, UInt32 *ProbPrices) +{ + UInt32 posState; + for (posState = 0; posState < numPosStates; posState++) + LenPriceEnc_UpdateTable(p, posState, ProbPrices); +} + +static void LenEnc_Encode2(CLenPriceEnc *p, CRangeEnc *rc, UInt32 symbol, UInt32 posState, Bool updatePrice, UInt32 *ProbPrices) +{ + LenEnc_Encode(&p->p, rc, symbol, posState); + if (updatePrice) + if (--p->counters[posState] == 0) + LenPriceEnc_UpdateTable(p, posState, ProbPrices); +} + + + + +static void MovePos(CLzmaEnc *p, UInt32 num) +{ + #ifdef SHOW_STAT + ttt += num; + printf("\n MovePos %d", num); + #endif + if (num != 0) + { + p->additionalOffset += num; + p->matchFinder.Skip(p->matchFinderObj, num); + } +} + +static UInt32 ReadMatchDistances(CLzmaEnc *p, UInt32 *numDistancePairsRes) +{ + UInt32 lenRes = 0, numPairs; + p->numAvail = p->matchFinder.GetNumAvailableBytes(p->matchFinderObj); + numPairs = p->matchFinder.GetMatches(p->matchFinderObj, p->matches); + #ifdef SHOW_STAT + printf("\n i = %d numPairs = %d ", ttt, numPairs / 2); + ttt++; + { + UInt32 i; + for (i = 0; i < numPairs; i += 2) + printf("%2d %6d | ", p->matches[i], p->matches[i + 1]); + } + #endif + if (numPairs > 0) + { + lenRes = p->matches[numPairs - 2]; + if (lenRes == p->numFastBytes) + { + const Byte *pby = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1; + UInt32 distance = p->matches[numPairs - 1] + 1; + UInt32 numAvail = p->numAvail; + if (numAvail > LZMA_MATCH_LEN_MAX) + numAvail = LZMA_MATCH_LEN_MAX; + { + const Byte *pby2 = pby - distance; + for (; lenRes < numAvail && pby[lenRes] == pby2[lenRes]; lenRes++); + } + } + } + p->additionalOffset++; + *numDistancePairsRes = numPairs; + return lenRes; +} + + +#define MakeAsChar(p) (p)->backPrev = (UInt32)(-1); (p)->prev1IsChar = False; +#define MakeAsShortRep(p) (p)->backPrev = 0; (p)->prev1IsChar = False; +#define IsShortRep(p) ((p)->backPrev == 0) + +static UInt32 GetRepLen1Price(CLzmaEnc *p, UInt32 state, UInt32 posState) +{ + return + GET_PRICE_0(p->isRepG0[state]) + + GET_PRICE_0(p->isRep0Long[state][posState]); +} + +static UInt32 GetPureRepPrice(CLzmaEnc *p, UInt32 repIndex, UInt32 state, UInt32 posState) +{ + UInt32 price; + if (repIndex == 0) + { + price = GET_PRICE_0(p->isRepG0[state]); + price += GET_PRICE_1(p->isRep0Long[state][posState]); + } + else + { + price = GET_PRICE_1(p->isRepG0[state]); + if (repIndex == 1) + price += GET_PRICE_0(p->isRepG1[state]); + else + { + price += GET_PRICE_1(p->isRepG1[state]); + price += GET_PRICE(p->isRepG2[state], repIndex - 2); + } + } + return price; +} + +static UInt32 GetRepPrice(CLzmaEnc *p, UInt32 repIndex, UInt32 len, UInt32 state, UInt32 posState) +{ + return p->repLenEnc.prices[posState][len - LZMA_MATCH_LEN_MIN] + + GetPureRepPrice(p, repIndex, state, posState); +} + +static UInt32 Backward(CLzmaEnc *p, UInt32 *backRes, UInt32 cur) +{ + UInt32 posMem = p->opt[cur].posPrev; + UInt32 backMem = p->opt[cur].backPrev; + p->optimumEndIndex = cur; + do + { + if (p->opt[cur].prev1IsChar) + { + MakeAsChar(&p->opt[posMem]) + p->opt[posMem].posPrev = posMem - 1; + if (p->opt[cur].prev2) + { + p->opt[posMem - 1].prev1IsChar = False; + p->opt[posMem - 1].posPrev = p->opt[cur].posPrev2; + p->opt[posMem - 1].backPrev = p->opt[cur].backPrev2; + } + } + { + UInt32 posPrev = posMem; + UInt32 backCur = backMem; + + backMem = p->opt[posPrev].backPrev; + posMem = p->opt[posPrev].posPrev; + + p->opt[posPrev].backPrev = backCur; + p->opt[posPrev].posPrev = cur; + cur = posPrev; + } + } + while (cur != 0); + *backRes = p->opt[0].backPrev; + p->optimumCurrentIndex = p->opt[0].posPrev; + return p->optimumCurrentIndex; +} + +#define LIT_PROBS(pos, prevByte) (p->litProbs + ((((pos) & p->lpMask) << p->lc) + ((prevByte) >> (8 - p->lc))) * 0x300) + +static UInt32 GetOptimum(CLzmaEnc *p, UInt32 position, UInt32 *backRes) +{ + UInt32 numAvail, mainLen, numPairs, repMaxIndex, i, posState, lenEnd, len, cur; + UInt32 matchPrice, repMatchPrice, normalMatchPrice; + UInt32 reps[LZMA_NUM_REPS], repLens[LZMA_NUM_REPS]; + UInt32 *matches; + const Byte *data; + Byte curByte, matchByte; + if (p->optimumEndIndex != p->optimumCurrentIndex) + { + const COptimal *opt = &p->opt[p->optimumCurrentIndex]; + UInt32 lenRes = opt->posPrev - p->optimumCurrentIndex; + *backRes = opt->backPrev; + p->optimumCurrentIndex = opt->posPrev; + return lenRes; + } + p->optimumCurrentIndex = p->optimumEndIndex = 0; + + if (p->additionalOffset == 0) + mainLen = ReadMatchDistances(p, &numPairs); + else + { + mainLen = p->longestMatchLength; + numPairs = p->numPairs; + } + + numAvail = p->numAvail; + if (numAvail < 2) + { + *backRes = (UInt32)(-1); + return 1; + } + if (numAvail > LZMA_MATCH_LEN_MAX) + numAvail = LZMA_MATCH_LEN_MAX; + + data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1; + repMaxIndex = 0; + for (i = 0; i < LZMA_NUM_REPS; i++) + { + UInt32 lenTest; + const Byte *data2; + reps[i] = p->reps[i]; + data2 = data - (reps[i] + 1); + if (data[0] != data2[0] || data[1] != data2[1]) + { + repLens[i] = 0; + continue; + } + for (lenTest = 2; lenTest < numAvail && data[lenTest] == data2[lenTest]; lenTest++); + repLens[i] = lenTest; + if (lenTest > repLens[repMaxIndex]) + repMaxIndex = i; + } + if (repLens[repMaxIndex] >= p->numFastBytes) + { + UInt32 lenRes; + *backRes = repMaxIndex; + lenRes = repLens[repMaxIndex]; + MovePos(p, lenRes - 1); + return lenRes; + } + + matches = p->matches; + if (mainLen >= p->numFastBytes) + { + *backRes = matches[numPairs - 1] + LZMA_NUM_REPS; + MovePos(p, mainLen - 1); + return mainLen; + } + curByte = *data; + matchByte = *(data - (reps[0] + 1)); + + if (mainLen < 2 && curByte != matchByte && repLens[repMaxIndex] < 2) + { + *backRes = (UInt32)-1; + return 1; + } + + p->opt[0].state = (CState)p->state; + + posState = (position & p->pbMask); + + { + const CLzmaProb *probs = LIT_PROBS(position, *(data - 1)); + p->opt[1].price = GET_PRICE_0(p->isMatch[p->state][posState]) + + (!IsCharState(p->state) ? + LitEnc_GetPriceMatched(probs, curByte, matchByte, p->ProbPrices) : + LitEnc_GetPrice(probs, curByte, p->ProbPrices)); + } + + MakeAsChar(&p->opt[1]); + + matchPrice = GET_PRICE_1(p->isMatch[p->state][posState]); + repMatchPrice = matchPrice + GET_PRICE_1(p->isRep[p->state]); + + if (matchByte == curByte) + { + UInt32 shortRepPrice = repMatchPrice + GetRepLen1Price(p, p->state, posState); + if (shortRepPrice < p->opt[1].price) + { + p->opt[1].price = shortRepPrice; + MakeAsShortRep(&p->opt[1]); + } + } + lenEnd = ((mainLen >= repLens[repMaxIndex]) ? mainLen : repLens[repMaxIndex]); + + if (lenEnd < 2) + { + *backRes = p->opt[1].backPrev; + return 1; + } + + p->opt[1].posPrev = 0; + for (i = 0; i < LZMA_NUM_REPS; i++) + p->opt[0].backs[i] = reps[i]; + + len = lenEnd; + do + p->opt[len--].price = kInfinityPrice; + while (len >= 2); + + for (i = 0; i < LZMA_NUM_REPS; i++) + { + UInt32 repLen = repLens[i]; + UInt32 price; + if (repLen < 2) + continue; + price = repMatchPrice + GetPureRepPrice(p, i, p->state, posState); + do + { + UInt32 curAndLenPrice = price + p->repLenEnc.prices[posState][repLen - 2]; + COptimal *opt = &p->opt[repLen]; + if (curAndLenPrice < opt->price) + { + opt->price = curAndLenPrice; + opt->posPrev = 0; + opt->backPrev = i; + opt->prev1IsChar = False; + } + } + while (--repLen >= 2); + } + + normalMatchPrice = matchPrice + GET_PRICE_0(p->isRep[p->state]); + + len = ((repLens[0] >= 2) ? repLens[0] + 1 : 2); + if (len <= mainLen) + { + UInt32 offs = 0; + while (len > matches[offs]) + offs += 2; + for (; ; len++) + { + COptimal *opt; + UInt32 distance = matches[offs + 1]; + + UInt32 curAndLenPrice = normalMatchPrice + p->lenEnc.prices[posState][len - LZMA_MATCH_LEN_MIN]; + UInt32 lenToPosState = GetLenToPosState(len); + if (distance < kNumFullDistances) + curAndLenPrice += p->distancesPrices[lenToPosState][distance]; + else + { + UInt32 slot; + GetPosSlot2(distance, slot); + curAndLenPrice += p->alignPrices[distance & kAlignMask] + p->posSlotPrices[lenToPosState][slot]; + } + opt = &p->opt[len]; + if (curAndLenPrice < opt->price) + { + opt->price = curAndLenPrice; + opt->posPrev = 0; + opt->backPrev = distance + LZMA_NUM_REPS; + opt->prev1IsChar = False; + } + if (len == matches[offs]) + { + offs += 2; + if (offs == numPairs) + break; + } + } + } + + cur = 0; + + #ifdef SHOW_STAT2 + if (position >= 0) + { + unsigned i; + printf("\n pos = %4X", position); + for (i = cur; i <= lenEnd; i++) + printf("\nprice[%4X] = %d", position - cur + i, p->opt[i].price); + } + #endif + + for (;;) + { + UInt32 numAvailFull, newLen, numPairs, posPrev, state, posState, startLen; + UInt32 curPrice, curAnd1Price, matchPrice, repMatchPrice; + Bool nextIsChar; + Byte curByte, matchByte; + const Byte *data; + COptimal *curOpt; + COptimal *nextOpt; + + cur++; + if (cur == lenEnd) + return Backward(p, backRes, cur); + + newLen = ReadMatchDistances(p, &numPairs); + if (newLen >= p->numFastBytes) + { + p->numPairs = numPairs; + p->longestMatchLength = newLen; + return Backward(p, backRes, cur); + } + position++; + curOpt = &p->opt[cur]; + posPrev = curOpt->posPrev; + if (curOpt->prev1IsChar) + { + posPrev--; + if (curOpt->prev2) + { + state = p->opt[curOpt->posPrev2].state; + if (curOpt->backPrev2 < LZMA_NUM_REPS) + state = kRepNextStates[state]; + else + state = kMatchNextStates[state]; + } + else + state = p->opt[posPrev].state; + state = kLiteralNextStates[state]; + } + else + state = p->opt[posPrev].state; + if (posPrev == cur - 1) + { + if (IsShortRep(curOpt)) + state = kShortRepNextStates[state]; + else + state = kLiteralNextStates[state]; + } + else + { + UInt32 pos; + const COptimal *prevOpt; + if (curOpt->prev1IsChar && curOpt->prev2) + { + posPrev = curOpt->posPrev2; + pos = curOpt->backPrev2; + state = kRepNextStates[state]; + } + else + { + pos = curOpt->backPrev; + if (pos < LZMA_NUM_REPS) + state = kRepNextStates[state]; + else + state = kMatchNextStates[state]; + } + prevOpt = &p->opt[posPrev]; + if (pos < LZMA_NUM_REPS) + { + UInt32 i; + reps[0] = prevOpt->backs[pos]; + for (i = 1; i <= pos; i++) + reps[i] = prevOpt->backs[i - 1]; + for (; i < LZMA_NUM_REPS; i++) + reps[i] = prevOpt->backs[i]; + } + else + { + UInt32 i; + reps[0] = (pos - LZMA_NUM_REPS); + for (i = 1; i < LZMA_NUM_REPS; i++) + reps[i] = prevOpt->backs[i - 1]; + } + } + curOpt->state = (CState)state; + + curOpt->backs[0] = reps[0]; + curOpt->backs[1] = reps[1]; + curOpt->backs[2] = reps[2]; + curOpt->backs[3] = reps[3]; + + curPrice = curOpt->price; + nextIsChar = False; + data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1; + curByte = *data; + matchByte = *(data - (reps[0] + 1)); + + posState = (position & p->pbMask); + + curAnd1Price = curPrice + GET_PRICE_0(p->isMatch[state][posState]); + { + const CLzmaProb *probs = LIT_PROBS(position, *(data - 1)); + curAnd1Price += + (!IsCharState(state) ? + LitEnc_GetPriceMatched(probs, curByte, matchByte, p->ProbPrices) : + LitEnc_GetPrice(probs, curByte, p->ProbPrices)); + } + + nextOpt = &p->opt[cur + 1]; + + if (curAnd1Price < nextOpt->price) + { + nextOpt->price = curAnd1Price; + nextOpt->posPrev = cur; + MakeAsChar(nextOpt); + nextIsChar = True; + } + + matchPrice = curPrice + GET_PRICE_1(p->isMatch[state][posState]); + repMatchPrice = matchPrice + GET_PRICE_1(p->isRep[state]); + + if (matchByte == curByte && !(nextOpt->posPrev < cur && nextOpt->backPrev == 0)) + { + UInt32 shortRepPrice = repMatchPrice + GetRepLen1Price(p, state, posState); + if (shortRepPrice <= nextOpt->price) + { + nextOpt->price = shortRepPrice; + nextOpt->posPrev = cur; + MakeAsShortRep(nextOpt); + nextIsChar = True; + } + } + numAvailFull = p->numAvail; + { + UInt32 temp = kNumOpts - 1 - cur; + if (temp < numAvailFull) + numAvailFull = temp; + } + + if (numAvailFull < 2) + continue; + numAvail = (numAvailFull <= p->numFastBytes ? numAvailFull : p->numFastBytes); + + if (!nextIsChar && matchByte != curByte) /* speed optimization */ + { + /* try Literal + rep0 */ + UInt32 temp; + UInt32 lenTest2; + const Byte *data2 = data - (reps[0] + 1); + UInt32 limit = p->numFastBytes + 1; + if (limit > numAvailFull) + limit = numAvailFull; + + for (temp = 1; temp < limit && data[temp] == data2[temp]; temp++); + lenTest2 = temp - 1; + if (lenTest2 >= 2) + { + UInt32 state2 = kLiteralNextStates[state]; + UInt32 posStateNext = (position + 1) & p->pbMask; + UInt32 nextRepMatchPrice = curAnd1Price + + GET_PRICE_1(p->isMatch[state2][posStateNext]) + + GET_PRICE_1(p->isRep[state2]); + /* for (; lenTest2 >= 2; lenTest2--) */ + { + UInt32 curAndLenPrice; + COptimal *opt; + UInt32 offset = cur + 1 + lenTest2; + while (lenEnd < offset) + p->opt[++lenEnd].price = kInfinityPrice; + curAndLenPrice = nextRepMatchPrice + GetRepPrice(p, 0, lenTest2, state2, posStateNext); + opt = &p->opt[offset]; + if (curAndLenPrice < opt->price) + { + opt->price = curAndLenPrice; + opt->posPrev = cur + 1; + opt->backPrev = 0; + opt->prev1IsChar = True; + opt->prev2 = False; + } + } + } + } + + startLen = 2; /* speed optimization */ + { + UInt32 repIndex; + for (repIndex = 0; repIndex < LZMA_NUM_REPS; repIndex++) + { + UInt32 lenTest; + UInt32 lenTestTemp; + UInt32 price; + const Byte *data2 = data - (reps[repIndex] + 1); + if (data[0] != data2[0] || data[1] != data2[1]) + continue; + for (lenTest = 2; lenTest < numAvail && data[lenTest] == data2[lenTest]; lenTest++); + while (lenEnd < cur + lenTest) + p->opt[++lenEnd].price = kInfinityPrice; + lenTestTemp = lenTest; + price = repMatchPrice + GetPureRepPrice(p, repIndex, state, posState); + do + { + UInt32 curAndLenPrice = price + p->repLenEnc.prices[posState][lenTest - 2]; + COptimal *opt = &p->opt[cur + lenTest]; + if (curAndLenPrice < opt->price) + { + opt->price = curAndLenPrice; + opt->posPrev = cur; + opt->backPrev = repIndex; + opt->prev1IsChar = False; + } + } + while (--lenTest >= 2); + lenTest = lenTestTemp; + + if (repIndex == 0) + startLen = lenTest + 1; + + /* if (_maxMode) */ + { + UInt32 lenTest2 = lenTest + 1; + UInt32 limit = lenTest2 + p->numFastBytes; + UInt32 nextRepMatchPrice; + if (limit > numAvailFull) + limit = numAvailFull; + for (; lenTest2 < limit && data[lenTest2] == data2[lenTest2]; lenTest2++); + lenTest2 -= lenTest + 1; + if (lenTest2 >= 2) + { + UInt32 state2 = kRepNextStates[state]; + UInt32 posStateNext = (position + lenTest) & p->pbMask; + UInt32 curAndLenCharPrice = + price + p->repLenEnc.prices[posState][lenTest - 2] + + GET_PRICE_0(p->isMatch[state2][posStateNext]) + + LitEnc_GetPriceMatched(LIT_PROBS(position + lenTest, data[lenTest - 1]), + data[lenTest], data2[lenTest], p->ProbPrices); + state2 = kLiteralNextStates[state2]; + posStateNext = (position + lenTest + 1) & p->pbMask; + nextRepMatchPrice = curAndLenCharPrice + + GET_PRICE_1(p->isMatch[state2][posStateNext]) + + GET_PRICE_1(p->isRep[state2]); + + /* for (; lenTest2 >= 2; lenTest2--) */ + { + UInt32 curAndLenPrice; + COptimal *opt; + UInt32 offset = cur + lenTest + 1 + lenTest2; + while (lenEnd < offset) + p->opt[++lenEnd].price = kInfinityPrice; + curAndLenPrice = nextRepMatchPrice + GetRepPrice(p, 0, lenTest2, state2, posStateNext); + opt = &p->opt[offset]; + if (curAndLenPrice < opt->price) + { + opt->price = curAndLenPrice; + opt->posPrev = cur + lenTest + 1; + opt->backPrev = 0; + opt->prev1IsChar = True; + opt->prev2 = True; + opt->posPrev2 = cur; + opt->backPrev2 = repIndex; + } + } + } + } + } + } + /* for (UInt32 lenTest = 2; lenTest <= newLen; lenTest++) */ + if (newLen > numAvail) + { + newLen = numAvail; + for (numPairs = 0; newLen > matches[numPairs]; numPairs += 2); + matches[numPairs] = newLen; + numPairs += 2; + } + if (newLen >= startLen) + { + UInt32 normalMatchPrice = matchPrice + GET_PRICE_0(p->isRep[state]); + UInt32 offs, curBack, posSlot; + UInt32 lenTest; + while (lenEnd < cur + newLen) + p->opt[++lenEnd].price = kInfinityPrice; + + offs = 0; + while (startLen > matches[offs]) + offs += 2; + curBack = matches[offs + 1]; + GetPosSlot2(curBack, posSlot); + for (lenTest = /*2*/ startLen; ; lenTest++) + { + UInt32 curAndLenPrice = normalMatchPrice + p->lenEnc.prices[posState][lenTest - LZMA_MATCH_LEN_MIN]; + UInt32 lenToPosState = GetLenToPosState(lenTest); + COptimal *opt; + if (curBack < kNumFullDistances) + curAndLenPrice += p->distancesPrices[lenToPosState][curBack]; + else + curAndLenPrice += p->posSlotPrices[lenToPosState][posSlot] + p->alignPrices[curBack & kAlignMask]; + + opt = &p->opt[cur + lenTest]; + if (curAndLenPrice < opt->price) + { + opt->price = curAndLenPrice; + opt->posPrev = cur; + opt->backPrev = curBack + LZMA_NUM_REPS; + opt->prev1IsChar = False; + } + + if (/*_maxMode && */lenTest == matches[offs]) + { + /* Try Match + Literal + Rep0 */ + const Byte *data2 = data - (curBack + 1); + UInt32 lenTest2 = lenTest + 1; + UInt32 limit = lenTest2 + p->numFastBytes; + UInt32 nextRepMatchPrice; + if (limit > numAvailFull) + limit = numAvailFull; + for (; lenTest2 < limit && data[lenTest2] == data2[lenTest2]; lenTest2++); + lenTest2 -= lenTest + 1; + if (lenTest2 >= 2) + { + UInt32 state2 = kMatchNextStates[state]; + UInt32 posStateNext = (position + lenTest) & p->pbMask; + UInt32 curAndLenCharPrice = curAndLenPrice + + GET_PRICE_0(p->isMatch[state2][posStateNext]) + + LitEnc_GetPriceMatched(LIT_PROBS(position + lenTest, data[lenTest - 1]), + data[lenTest], data2[lenTest], p->ProbPrices); + state2 = kLiteralNextStates[state2]; + posStateNext = (posStateNext + 1) & p->pbMask; + nextRepMatchPrice = curAndLenCharPrice + + GET_PRICE_1(p->isMatch[state2][posStateNext]) + + GET_PRICE_1(p->isRep[state2]); + + /* for (; lenTest2 >= 2; lenTest2--) */ + { + UInt32 offset = cur + lenTest + 1 + lenTest2; + UInt32 curAndLenPrice; + COptimal *opt; + while (lenEnd < offset) + p->opt[++lenEnd].price = kInfinityPrice; + curAndLenPrice = nextRepMatchPrice + GetRepPrice(p, 0, lenTest2, state2, posStateNext); + opt = &p->opt[offset]; + if (curAndLenPrice < opt->price) + { + opt->price = curAndLenPrice; + opt->posPrev = cur + lenTest + 1; + opt->backPrev = 0; + opt->prev1IsChar = True; + opt->prev2 = True; + opt->posPrev2 = cur; + opt->backPrev2 = curBack + LZMA_NUM_REPS; + } + } + } + offs += 2; + if (offs == numPairs) + break; + curBack = matches[offs + 1]; + if (curBack >= kNumFullDistances) + GetPosSlot2(curBack, posSlot); + } + } + } + } +} + +#define ChangePair(smallDist, bigDist) (((bigDist) >> 7) > (smallDist)) + +static UInt32 GetOptimumFast(CLzmaEnc *p, UInt32 *backRes) +{ + UInt32 numAvail, mainLen, mainDist, numPairs, repIndex, repLen, i; + const Byte *data; + const UInt32 *matches; + + if (p->additionalOffset == 0) + mainLen = ReadMatchDistances(p, &numPairs); + else + { + mainLen = p->longestMatchLength; + numPairs = p->numPairs; + } + + numAvail = p->numAvail; + *backRes = (UInt32)-1; + if (numAvail < 2) + return 1; + if (numAvail > LZMA_MATCH_LEN_MAX) + numAvail = LZMA_MATCH_LEN_MAX; + data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1; + + repLen = repIndex = 0; + for (i = 0; i < LZMA_NUM_REPS; i++) + { + UInt32 len; + const Byte *data2 = data - (p->reps[i] + 1); + if (data[0] != data2[0] || data[1] != data2[1]) + continue; + for (len = 2; len < numAvail && data[len] == data2[len]; len++); + if (len >= p->numFastBytes) + { + *backRes = i; + MovePos(p, len - 1); + return len; + } + if (len > repLen) + { + repIndex = i; + repLen = len; + } + } + + matches = p->matches; + if (mainLen >= p->numFastBytes) + { + *backRes = matches[numPairs - 1] + LZMA_NUM_REPS; + MovePos(p, mainLen - 1); + return mainLen; + } + + mainDist = 0; /* for GCC */ + if (mainLen >= 2) + { + mainDist = matches[numPairs - 1]; + while (numPairs > 2 && mainLen == matches[numPairs - 4] + 1) + { + if (!ChangePair(matches[numPairs - 3], mainDist)) + break; + numPairs -= 2; + mainLen = matches[numPairs - 2]; + mainDist = matches[numPairs - 1]; + } + if (mainLen == 2 && mainDist >= 0x80) + mainLen = 1; + } + + if (repLen >= 2 && ( + (repLen + 1 >= mainLen) || + (repLen + 2 >= mainLen && mainDist >= (1 << 9)) || + (repLen + 3 >= mainLen && mainDist >= (1 << 15)))) + { + *backRes = repIndex; + MovePos(p, repLen - 1); + return repLen; + } + + if (mainLen < 2 || numAvail <= 2) + return 1; + + p->longestMatchLength = ReadMatchDistances(p, &p->numPairs); + if (p->longestMatchLength >= 2) + { + UInt32 newDistance = matches[p->numPairs - 1]; + if ((p->longestMatchLength >= mainLen && newDistance < mainDist) || + (p->longestMatchLength == mainLen + 1 && !ChangePair(mainDist, newDistance)) || + (p->longestMatchLength > mainLen + 1) || + (p->longestMatchLength + 1 >= mainLen && mainLen >= 3 && ChangePair(newDistance, mainDist))) + return 1; + } + + data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1; + for (i = 0; i < LZMA_NUM_REPS; i++) + { + UInt32 len, limit; + const Byte *data2 = data - (p->reps[i] + 1); + if (data[0] != data2[0] || data[1] != data2[1]) + continue; + limit = mainLen - 1; + for (len = 2; len < limit && data[len] == data2[len]; len++); + if (len >= limit) + return 1; + } + *backRes = mainDist + LZMA_NUM_REPS; + MovePos(p, mainLen - 2); + return mainLen; +} + +static void WriteEndMarker(CLzmaEnc *p, UInt32 posState) +{ + UInt32 len; + RangeEnc_EncodeBit(&p->rc, &p->isMatch[p->state][posState], 1); + RangeEnc_EncodeBit(&p->rc, &p->isRep[p->state], 0); + p->state = kMatchNextStates[p->state]; + len = LZMA_MATCH_LEN_MIN; + LenEnc_Encode2(&p->lenEnc, &p->rc, len - LZMA_MATCH_LEN_MIN, posState, !p->fastMode, p->ProbPrices); + RcTree_Encode(&p->rc, p->posSlotEncoder[GetLenToPosState(len)], kNumPosSlotBits, (1 << kNumPosSlotBits) - 1); + RangeEnc_EncodeDirectBits(&p->rc, (((UInt32)1 << 30) - 1) >> kNumAlignBits, 30 - kNumAlignBits); + RcTree_ReverseEncode(&p->rc, p->posAlignEncoder, kNumAlignBits, kAlignMask); +} + +static SRes CheckErrors(CLzmaEnc *p) +{ + if (p->result != SZ_OK) + return p->result; + if (p->rc.res != SZ_OK) + p->result = SZ_ERROR_WRITE; + if (p->matchFinderBase.result != SZ_OK) + p->result = SZ_ERROR_READ; + if (p->result != SZ_OK) + p->finished = True; + return p->result; +} + +static SRes Flush(CLzmaEnc *p, UInt32 nowPos) +{ + /* ReleaseMFStream(); */ + p->finished = True; + if (p->writeEndMark) + WriteEndMarker(p, nowPos & p->pbMask); + RangeEnc_FlushData(&p->rc); + RangeEnc_FlushStream(&p->rc); + return CheckErrors(p); +} + +static void FillAlignPrices(CLzmaEnc *p) +{ + UInt32 i; + for (i = 0; i < kAlignTableSize; i++) + p->alignPrices[i] = RcTree_ReverseGetPrice(p->posAlignEncoder, kNumAlignBits, i, p->ProbPrices); + p->alignPriceCount = 0; +} + +static void FillDistancesPrices(CLzmaEnc *p) +{ + UInt32 tempPrices[kNumFullDistances]; + UInt32 i, lenToPosState; + for (i = kStartPosModelIndex; i < kNumFullDistances; i++) + { + UInt32 posSlot = GetPosSlot1(i); + UInt32 footerBits = ((posSlot >> 1) - 1); + UInt32 base = ((2 | (posSlot & 1)) << footerBits); + tempPrices[i] = RcTree_ReverseGetPrice(p->posEncoders + base - posSlot - 1, footerBits, i - base, p->ProbPrices); + } + + for (lenToPosState = 0; lenToPosState < kNumLenToPosStates; lenToPosState++) + { + UInt32 posSlot; + const CLzmaProb *encoder = p->posSlotEncoder[lenToPosState]; + UInt32 *posSlotPrices = p->posSlotPrices[lenToPosState]; + for (posSlot = 0; posSlot < p->distTableSize; posSlot++) + posSlotPrices[posSlot] = RcTree_GetPrice(encoder, kNumPosSlotBits, posSlot, p->ProbPrices); + for (posSlot = kEndPosModelIndex; posSlot < p->distTableSize; posSlot++) + posSlotPrices[posSlot] += ((((posSlot >> 1) - 1) - kNumAlignBits) << kNumBitPriceShiftBits); + + { + UInt32 *distancesPrices = p->distancesPrices[lenToPosState]; + UInt32 i; + for (i = 0; i < kStartPosModelIndex; i++) + distancesPrices[i] = posSlotPrices[i]; + for (; i < kNumFullDistances; i++) + distancesPrices[i] = posSlotPrices[GetPosSlot1(i)] + tempPrices[i]; + } + } + p->matchPriceCount = 0; +} + +void LzmaEnc_Construct(CLzmaEnc *p) +{ + RangeEnc_Construct(&p->rc); + MatchFinder_Construct(&p->matchFinderBase); + #ifdef COMPRESS_MF_MT + MatchFinderMt_Construct(&p->matchFinderMt); + p->matchFinderMt.MatchFinder = &p->matchFinderBase; + #endif + + { + CLzmaEncProps props; + LzmaEncProps_Init(&props); + LzmaEnc_SetProps(p, &props); + } + + #ifndef LZMA_LOG_BSR + LzmaEnc_FastPosInit(p->g_FastPos); + #endif + + LzmaEnc_InitPriceTables(p->ProbPrices); + p->litProbs = 0; + p->saveState.litProbs = 0; +} + +CLzmaEncHandle LzmaEnc_Create(ISzAlloc *alloc) +{ + void *p; + p = alloc->Alloc(alloc, sizeof(CLzmaEnc)); + if (p != 0) + LzmaEnc_Construct((CLzmaEnc *)p); + return p; +} + +void LzmaEnc_FreeLits(CLzmaEnc *p, ISzAlloc *alloc) +{ + alloc->Free(alloc, p->litProbs); + alloc->Free(alloc, p->saveState.litProbs); + p->litProbs = 0; + p->saveState.litProbs = 0; +} + +void LzmaEnc_Destruct(CLzmaEnc *p, ISzAlloc *alloc, ISzAlloc *allocBig) +{ + #ifdef COMPRESS_MF_MT + MatchFinderMt_Destruct(&p->matchFinderMt, allocBig); + #endif + MatchFinder_Free(&p->matchFinderBase, allocBig); + LzmaEnc_FreeLits(p, alloc); + RangeEnc_Free(&p->rc, alloc); +} + +void LzmaEnc_Destroy(CLzmaEncHandle p, ISzAlloc *alloc, ISzAlloc *allocBig) +{ + LzmaEnc_Destruct((CLzmaEnc *)p, alloc, allocBig); + alloc->Free(alloc, p); +} + +static SRes LzmaEnc_CodeOneBlock(CLzmaEnc *p, Bool useLimits, UInt32 maxPackSize, UInt32 maxUnpackSize) +{ + UInt32 nowPos32, startPos32; + if (p->inStream != 0) + { + p->matchFinderBase.stream = p->inStream; + p->matchFinder.Init(p->matchFinderObj); + p->inStream = 0; + } + + if (p->finished) + return p->result; + RINOK(CheckErrors(p)); + + nowPos32 = (UInt32)p->nowPos64; + startPos32 = nowPos32; + + if (p->nowPos64 == 0) + { + UInt32 numPairs; + Byte curByte; + if (p->matchFinder.GetNumAvailableBytes(p->matchFinderObj) == 0) + return Flush(p, nowPos32); + ReadMatchDistances(p, &numPairs); + RangeEnc_EncodeBit(&p->rc, &p->isMatch[p->state][0], 0); + p->state = kLiteralNextStates[p->state]; + curByte = p->matchFinder.GetIndexByte(p->matchFinderObj, 0 - p->additionalOffset); + LitEnc_Encode(&p->rc, p->litProbs, curByte); + p->additionalOffset--; + nowPos32++; + } + + if (p->matchFinder.GetNumAvailableBytes(p->matchFinderObj) != 0) + for (;;) + { + UInt32 pos, len, posState; + + if (p->fastMode) + len = GetOptimumFast(p, &pos); + else + len = GetOptimum(p, nowPos32, &pos); + + #ifdef SHOW_STAT2 + printf("\n pos = %4X, len = %d pos = %d", nowPos32, len, pos); + #endif + + posState = nowPos32 & p->pbMask; + if (len == 1 && pos == (UInt32)-1) + { + Byte curByte; + CLzmaProb *probs; + const Byte *data; + + RangeEnc_EncodeBit(&p->rc, &p->isMatch[p->state][posState], 0); + data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - p->additionalOffset; + curByte = *data; + probs = LIT_PROBS(nowPos32, *(data - 1)); + if (IsCharState(p->state)) + LitEnc_Encode(&p->rc, probs, curByte); + else + LitEnc_EncodeMatched(&p->rc, probs, curByte, *(data - p->reps[0] - 1)); + p->state = kLiteralNextStates[p->state]; + } + else + { + RangeEnc_EncodeBit(&p->rc, &p->isMatch[p->state][posState], 1); + if (pos < LZMA_NUM_REPS) + { + RangeEnc_EncodeBit(&p->rc, &p->isRep[p->state], 1); + if (pos == 0) + { + RangeEnc_EncodeBit(&p->rc, &p->isRepG0[p->state], 0); + RangeEnc_EncodeBit(&p->rc, &p->isRep0Long[p->state][posState], ((len == 1) ? 0 : 1)); + } + else + { + UInt32 distance = p->reps[pos]; + RangeEnc_EncodeBit(&p->rc, &p->isRepG0[p->state], 1); + if (pos == 1) + RangeEnc_EncodeBit(&p->rc, &p->isRepG1[p->state], 0); + else + { + RangeEnc_EncodeBit(&p->rc, &p->isRepG1[p->state], 1); + RangeEnc_EncodeBit(&p->rc, &p->isRepG2[p->state], pos - 2); + if (pos == 3) + p->reps[3] = p->reps[2]; + p->reps[2] = p->reps[1]; + } + p->reps[1] = p->reps[0]; + p->reps[0] = distance; + } + if (len == 1) + p->state = kShortRepNextStates[p->state]; + else + { + LenEnc_Encode2(&p->repLenEnc, &p->rc, len - LZMA_MATCH_LEN_MIN, posState, !p->fastMode, p->ProbPrices); + p->state = kRepNextStates[p->state]; + } + } + else + { + UInt32 posSlot; + RangeEnc_EncodeBit(&p->rc, &p->isRep[p->state], 0); + p->state = kMatchNextStates[p->state]; + LenEnc_Encode2(&p->lenEnc, &p->rc, len - LZMA_MATCH_LEN_MIN, posState, !p->fastMode, p->ProbPrices); + pos -= LZMA_NUM_REPS; + GetPosSlot(pos, posSlot); + RcTree_Encode(&p->rc, p->posSlotEncoder[GetLenToPosState(len)], kNumPosSlotBits, posSlot); + + if (posSlot >= kStartPosModelIndex) + { + UInt32 footerBits = ((posSlot >> 1) - 1); + UInt32 base = ((2 | (posSlot & 1)) << footerBits); + UInt32 posReduced = pos - base; + + if (posSlot < kEndPosModelIndex) + RcTree_ReverseEncode(&p->rc, p->posEncoders + base - posSlot - 1, footerBits, posReduced); + else + { + RangeEnc_EncodeDirectBits(&p->rc, posReduced >> kNumAlignBits, footerBits - kNumAlignBits); + RcTree_ReverseEncode(&p->rc, p->posAlignEncoder, kNumAlignBits, posReduced & kAlignMask); + p->alignPriceCount++; + } + } + p->reps[3] = p->reps[2]; + p->reps[2] = p->reps[1]; + p->reps[1] = p->reps[0]; + p->reps[0] = pos; + p->matchPriceCount++; + } + } + p->additionalOffset -= len; + nowPos32 += len; + if (p->additionalOffset == 0) + { + UInt32 processed; + if (!p->fastMode) + { + if (p->matchPriceCount >= (1 << 7)) + FillDistancesPrices(p); + if (p->alignPriceCount >= kAlignTableSize) + FillAlignPrices(p); + } + if (p->matchFinder.GetNumAvailableBytes(p->matchFinderObj) == 0) + break; + processed = nowPos32 - startPos32; + if (useLimits) + { + if (processed + kNumOpts + 300 >= maxUnpackSize || + RangeEnc_GetProcessed(&p->rc) + kNumOpts * 2 >= maxPackSize) + break; + } + else if (processed >= (1 << 15)) + { + p->nowPos64 += nowPos32 - startPos32; + return CheckErrors(p); + } + } + } + p->nowPos64 += nowPos32 - startPos32; + return Flush(p, nowPos32); +} + +#define kBigHashDicLimit ((UInt32)1 << 24) + +static SRes LzmaEnc_Alloc(CLzmaEnc *p, UInt32 keepWindowSize, ISzAlloc *alloc, ISzAlloc *allocBig) +{ + UInt32 beforeSize = kNumOpts; + Bool btMode; + if (!RangeEnc_Alloc(&p->rc, alloc)) + return SZ_ERROR_MEM; + btMode = (p->matchFinderBase.btMode != 0); + #ifdef COMPRESS_MF_MT + p->mtMode = (p->multiThread && !p->fastMode && btMode); + #endif + + { + unsigned lclp = p->lc + p->lp; + if (p->litProbs == 0 || p->saveState.litProbs == 0 || p->lclp != lclp) + { + LzmaEnc_FreeLits(p, alloc); + p->litProbs = (CLzmaProb *)alloc->Alloc(alloc, (0x300 << lclp) * sizeof(CLzmaProb)); + p->saveState.litProbs = (CLzmaProb *)alloc->Alloc(alloc, (0x300 << lclp) * sizeof(CLzmaProb)); + if (p->litProbs == 0 || p->saveState.litProbs == 0) + { + LzmaEnc_FreeLits(p, alloc); + return SZ_ERROR_MEM; + } + p->lclp = lclp; + } + } + + p->matchFinderBase.bigHash = (p->dictSize > kBigHashDicLimit); + + if (beforeSize + p->dictSize < keepWindowSize) + beforeSize = keepWindowSize - p->dictSize; + + #ifdef COMPRESS_MF_MT + if (p->mtMode) + { + RINOK(MatchFinderMt_Create(&p->matchFinderMt, p->dictSize, beforeSize, p->numFastBytes, LZMA_MATCH_LEN_MAX, allocBig)); + p->matchFinderObj = &p->matchFinderMt; + MatchFinderMt_CreateVTable(&p->matchFinderMt, &p->matchFinder); + } + else + #endif + { + if (!MatchFinder_Create(&p->matchFinderBase, p->dictSize, beforeSize, p->numFastBytes, LZMA_MATCH_LEN_MAX, allocBig)) + return SZ_ERROR_MEM; + p->matchFinderObj = &p->matchFinderBase; + MatchFinder_CreateVTable(&p->matchFinderBase, &p->matchFinder); + } + return SZ_OK; +} + +void LzmaEnc_Init(CLzmaEnc *p) +{ + UInt32 i; + p->state = 0; + for (i = 0 ; i < LZMA_NUM_REPS; i++) + p->reps[i] = 0; + + RangeEnc_Init(&p->rc); + + + for (i = 0; i < kNumStates; i++) + { + UInt32 j; + for (j = 0; j < LZMA_NUM_PB_STATES_MAX; j++) + { + p->isMatch[i][j] = kProbInitValue; + p->isRep0Long[i][j] = kProbInitValue; + } + p->isRep[i] = kProbInitValue; + p->isRepG0[i] = kProbInitValue; + p->isRepG1[i] = kProbInitValue; + p->isRepG2[i] = kProbInitValue; + } + + { + UInt32 num = 0x300 << (p->lp + p->lc); + for (i = 0; i < num; i++) + p->litProbs[i] = kProbInitValue; + } + + { + for (i = 0; i < kNumLenToPosStates; i++) + { + CLzmaProb *probs = p->posSlotEncoder[i]; + UInt32 j; + for (j = 0; j < (1 << kNumPosSlotBits); j++) + probs[j] = kProbInitValue; + } + } + { + for (i = 0; i < kNumFullDistances - kEndPosModelIndex; i++) + p->posEncoders[i] = kProbInitValue; + } + + LenEnc_Init(&p->lenEnc.p); + LenEnc_Init(&p->repLenEnc.p); + + for (i = 0; i < (1 << kNumAlignBits); i++) + p->posAlignEncoder[i] = kProbInitValue; + + p->optimumEndIndex = 0; + p->optimumCurrentIndex = 0; + p->additionalOffset = 0; + + p->pbMask = (1 << p->pb) - 1; + p->lpMask = (1 << p->lp) - 1; +} + +void LzmaEnc_InitPrices(CLzmaEnc *p) +{ + if (!p->fastMode) + { + FillDistancesPrices(p); + FillAlignPrices(p); + } + + p->lenEnc.tableSize = + p->repLenEnc.tableSize = + p->numFastBytes + 1 - LZMA_MATCH_LEN_MIN; + LenPriceEnc_UpdateTables(&p->lenEnc, 1 << p->pb, p->ProbPrices); + LenPriceEnc_UpdateTables(&p->repLenEnc, 1 << p->pb, p->ProbPrices); +} + +static SRes LzmaEnc_AllocAndInit(CLzmaEnc *p, UInt32 keepWindowSize, ISzAlloc *alloc, ISzAlloc *allocBig) +{ + UInt32 i; + for (i = 0; i < (UInt32)kDicLogSizeMaxCompress; i++) + if (p->dictSize <= ((UInt32)1 << i)) + break; + p->distTableSize = i * 2; + + p->finished = False; + p->result = SZ_OK; + RINOK(LzmaEnc_Alloc(p, keepWindowSize, alloc, allocBig)); + LzmaEnc_Init(p); + LzmaEnc_InitPrices(p); + p->nowPos64 = 0; + return SZ_OK; +} + +static SRes LzmaEnc_Prepare(CLzmaEncHandle pp, ISeqInStream *inStream, ISeqOutStream *outStream, + ISzAlloc *alloc, ISzAlloc *allocBig) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + p->inStream = inStream; + p->rc.outStream = outStream; + return LzmaEnc_AllocAndInit(p, 0, alloc, allocBig); +} + +SRes LzmaEnc_PrepareForLzma2(CLzmaEncHandle pp, + ISeqInStream *inStream, UInt32 keepWindowSize, + ISzAlloc *alloc, ISzAlloc *allocBig) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + p->inStream = inStream; + return LzmaEnc_AllocAndInit(p, keepWindowSize, alloc, allocBig); +} + +static void LzmaEnc_SetInputBuf(CLzmaEnc *p, const Byte *src, SizeT srcLen) +{ + p->seqBufInStream.funcTable.Read = MyRead; + p->seqBufInStream.data = src; + p->seqBufInStream.rem = srcLen; +} + +SRes LzmaEnc_MemPrepare(CLzmaEncHandle pp, const Byte *src, SizeT srcLen, + UInt32 keepWindowSize, ISzAlloc *alloc, ISzAlloc *allocBig) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + LzmaEnc_SetInputBuf(p, src, srcLen); + p->inStream = &p->seqBufInStream.funcTable; + return LzmaEnc_AllocAndInit(p, keepWindowSize, alloc, allocBig); +} + +void LzmaEnc_Finish(CLzmaEncHandle pp) +{ + #ifdef COMPRESS_MF_MT + CLzmaEnc *p = (CLzmaEnc *)pp; + if (p->mtMode) + MatchFinderMt_ReleaseStream(&p->matchFinderMt); + #else + pp = pp; + #endif +} + +typedef struct _CSeqOutStreamBuf +{ + ISeqOutStream funcTable; + Byte *data; + SizeT rem; + Bool overflow; +} CSeqOutStreamBuf; + +static size_t MyWrite(void *pp, const void *data, size_t size) +{ + CSeqOutStreamBuf *p = (CSeqOutStreamBuf *)pp; + if (p->rem < size) + { + size = p->rem; + p->overflow = True; + } + memcpy(p->data, data, size); + p->rem -= size; + p->data += size; + return size; +} + + +UInt32 LzmaEnc_GetNumAvailableBytes(CLzmaEncHandle pp) +{ + const CLzmaEnc *p = (CLzmaEnc *)pp; + return p->matchFinder.GetNumAvailableBytes(p->matchFinderObj); +} + +const Byte *LzmaEnc_GetCurBuf(CLzmaEncHandle pp) +{ + const CLzmaEnc *p = (CLzmaEnc *)pp; + return p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - p->additionalOffset; +} + +SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle pp, Bool reInit, + Byte *dest, size_t *destLen, UInt32 desiredPackSize, UInt32 *unpackSize) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + UInt64 nowPos64; + SRes res; + CSeqOutStreamBuf outStream; + + outStream.funcTable.Write = MyWrite; + outStream.data = dest; + outStream.rem = *destLen; + outStream.overflow = False; + + p->writeEndMark = False; + p->finished = False; + p->result = SZ_OK; + + if (reInit) + LzmaEnc_Init(p); + LzmaEnc_InitPrices(p); + nowPos64 = p->nowPos64; + RangeEnc_Init(&p->rc); + p->rc.outStream = &outStream.funcTable; + + res = LzmaEnc_CodeOneBlock(p, True, desiredPackSize, *unpackSize); + + *unpackSize = (UInt32)(p->nowPos64 - nowPos64); + *destLen -= outStream.rem; + if (outStream.overflow) + return SZ_ERROR_OUTPUT_EOF; + + return res; +} + +SRes LzmaEnc_Encode(CLzmaEncHandle pp, ISeqOutStream *outStream, ISeqInStream *inStream, ICompressProgress *progress, + ISzAlloc *alloc, ISzAlloc *allocBig) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + SRes res = SZ_OK; + + #ifdef COMPRESS_MF_MT + Byte allocaDummy[0x300]; + int i = 0; + for (i = 0; i < 16; i++) + allocaDummy[i] = (Byte)i; + #endif + + RINOK(LzmaEnc_Prepare(pp, inStream, outStream, alloc, allocBig)); + + for (;;) + { + res = LzmaEnc_CodeOneBlock(p, False, 0, 0); + if (res != SZ_OK || p->finished != 0) + break; + if (progress != 0) + { + res = progress->Progress(progress, p->nowPos64, RangeEnc_GetProcessed(&p->rc)); + if (res != SZ_OK) + { + res = SZ_ERROR_PROGRESS; + break; + } + } + } + LzmaEnc_Finish(pp); + return res; +} + +SRes LzmaEnc_WriteProperties(CLzmaEncHandle pp, Byte *props, SizeT *size) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + int i; + UInt32 dictSize = p->dictSize; + if (*size < LZMA_PROPS_SIZE) + return SZ_ERROR_PARAM; + *size = LZMA_PROPS_SIZE; + props[0] = (Byte)((p->pb * 5 + p->lp) * 9 + p->lc); + + for (i = 11; i <= 30; i++) + { + if (dictSize <= ((UInt32)2 << i)) + { + dictSize = (2 << i); + break; + } + if (dictSize <= ((UInt32)3 << i)) + { + dictSize = (3 << i); + break; + } + } + + for (i = 0; i < 4; i++) + props[1 + i] = (Byte)(dictSize >> (8 * i)); + return SZ_OK; +} + +SRes LzmaEnc_MemEncode(CLzmaEncHandle pp, Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen, + int writeEndMark, ICompressProgress *progress, ISzAlloc *alloc, ISzAlloc *allocBig) +{ + SRes res; + CLzmaEnc *p = (CLzmaEnc *)pp; + + CSeqOutStreamBuf outStream; + + LzmaEnc_SetInputBuf(p, src, srcLen); + + outStream.funcTable.Write = MyWrite; + outStream.data = dest; + outStream.rem = *destLen; + outStream.overflow = False; + + p->writeEndMark = writeEndMark; + res = LzmaEnc_Encode(pp, &outStream.funcTable, &p->seqBufInStream.funcTable, + progress, alloc, allocBig); + + *destLen -= outStream.rem; + if (outStream.overflow) + return SZ_ERROR_OUTPUT_EOF; + return res; +} + +SRes LzmaEncode(Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen, + const CLzmaEncProps *props, Byte *propsEncoded, SizeT *propsSize, int writeEndMark, + ICompressProgress *progress, ISzAlloc *alloc, ISzAlloc *allocBig) +{ + CLzmaEnc *p = (CLzmaEnc *)LzmaEnc_Create(alloc); + SRes res; + if (p == 0) + return SZ_ERROR_MEM; + + res = LzmaEnc_SetProps(p, props); + if (res == SZ_OK) + { + res = LzmaEnc_WriteProperties(p, propsEncoded, propsSize); + if (res == SZ_OK) + res = LzmaEnc_MemEncode(p, dest, destLen, src, srcLen, + writeEndMark, progress, alloc, allocBig); + } + + LzmaEnc_Destroy(p, alloc, allocBig); + return res; +} diff --git a/lzma/C/LzmaEnc.h b/lzma/C/LzmaEnc.h new file mode 100644 index 0000000000..e3d84fa35e --- /dev/null +++ b/lzma/C/LzmaEnc.h @@ -0,0 +1,72 @@ +/* LzmaEnc.h -- LZMA Encoder +2008-10-04 : Igor Pavlov : Public domain */ + +#ifndef __LZMAENC_H +#define __LZMAENC_H + +#include "Types.h" + +#define LZMA_PROPS_SIZE 5 + +typedef struct _CLzmaEncProps +{ + int level; /* 0 <= level <= 9 */ + UInt32 dictSize; /* (1 << 12) <= dictSize <= (1 << 27) for 32-bit version + (1 << 12) <= dictSize <= (1 << 30) for 64-bit version + default = (1 << 24) */ + int lc; /* 0 <= lc <= 8, default = 3 */ + int lp; /* 0 <= lp <= 4, default = 0 */ + int pb; /* 0 <= pb <= 4, default = 2 */ + int algo; /* 0 - fast, 1 - normal, default = 1 */ + int fb; /* 5 <= fb <= 273, default = 32 */ + int btMode; /* 0 - hashChain Mode, 1 - binTree mode - normal, default = 1 */ + int numHashBytes; /* 2, 3 or 4, default = 4 */ + UInt32 mc; /* 1 <= mc <= (1 << 30), default = 32 */ + unsigned writeEndMark; /* 0 - do not write EOPM, 1 - write EOPM, default = 0 */ + int numThreads; /* 1 or 2, default = 2 */ +} CLzmaEncProps; + +void LzmaEncProps_Init(CLzmaEncProps *p); +void LzmaEncProps_Normalize(CLzmaEncProps *p); +UInt32 LzmaEncProps_GetDictSize(const CLzmaEncProps *props2); + + +/* ---------- CLzmaEncHandle Interface ---------- */ + +/* LzmaEnc_* functions can return the following exit codes: +Returns: + SZ_OK - OK + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_PARAM - Incorrect paramater in props + SZ_ERROR_WRITE - Write callback error. + SZ_ERROR_PROGRESS - some break from progress callback + SZ_ERROR_THREAD - errors in multithreading functions (only for Mt version) +*/ + +typedef void * CLzmaEncHandle; + +CLzmaEncHandle LzmaEnc_Create(ISzAlloc *alloc); +void LzmaEnc_Destroy(CLzmaEncHandle p, ISzAlloc *alloc, ISzAlloc *allocBig); +SRes LzmaEnc_SetProps(CLzmaEncHandle p, const CLzmaEncProps *props); +SRes LzmaEnc_WriteProperties(CLzmaEncHandle p, Byte *properties, SizeT *size); +SRes LzmaEnc_Encode(CLzmaEncHandle p, ISeqOutStream *outStream, ISeqInStream *inStream, + ICompressProgress *progress, ISzAlloc *alloc, ISzAlloc *allocBig); +SRes LzmaEnc_MemEncode(CLzmaEncHandle p, Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen, + int writeEndMark, ICompressProgress *progress, ISzAlloc *alloc, ISzAlloc *allocBig); + +/* ---------- One Call Interface ---------- */ + +/* LzmaEncode +Return code: + SZ_OK - OK + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_PARAM - Incorrect paramater + SZ_ERROR_OUTPUT_EOF - output buffer overflow + SZ_ERROR_THREAD - errors in multithreading functions (only for Mt version) +*/ + +SRes LzmaEncode(Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen, + const CLzmaEncProps *props, Byte *propsEncoded, SizeT *propsSize, int writeEndMark, + ICompressProgress *progress, ISzAlloc *alloc, ISzAlloc *allocBig); + +#endif diff --git a/lzma/C/LzmaLib.c b/lzma/C/LzmaLib.c new file mode 100644 index 0000000000..3e3cf40ae4 --- /dev/null +++ b/lzma/C/LzmaLib.c @@ -0,0 +1,46 @@ +/* LzmaLib.c -- LZMA library wrapper +2008-08-05 +Igor Pavlov +Public domain */ + +#include "LzmaEnc.h" +#include "LzmaDec.h" +#include "Alloc.h" +#include "LzmaLib.h" + +static void *SzAlloc(void *p, size_t size) { p = p; return MyAlloc(size); } +static void SzFree(void *p, void *address) { p = p; MyFree(address); } +static ISzAlloc g_Alloc = { SzAlloc, SzFree }; + +MY_STDAPI LzmaCompress(unsigned char *dest, size_t *destLen, const unsigned char *src, size_t srcLen, + unsigned char *outProps, size_t *outPropsSize, + int level, /* 0 <= level <= 9, default = 5 */ + unsigned dictSize, /* use (1 << N) or (3 << N). 4 KB < dictSize <= 128 MB */ + int lc, /* 0 <= lc <= 8, default = 3 */ + int lp, /* 0 <= lp <= 4, default = 0 */ + int pb, /* 0 <= pb <= 4, default = 2 */ + int fb, /* 5 <= fb <= 273, default = 32 */ + int numThreads /* 1 or 2, default = 2 */ +) +{ + CLzmaEncProps props; + LzmaEncProps_Init(&props); + props.level = level; + props.dictSize = dictSize; + props.lc = lc; + props.lp = lp; + props.pb = pb; + props.fb = fb; + props.numThreads = numThreads; + + return LzmaEncode(dest, destLen, src, srcLen, &props, outProps, outPropsSize, 0, + NULL, &g_Alloc, &g_Alloc); +} + + +MY_STDAPI LzmaUncompress(unsigned char *dest, size_t *destLen, const unsigned char *src, size_t *srcLen, + const unsigned char *props, size_t propsSize) +{ + ELzmaStatus status; + return LzmaDecode(dest, destLen, src, srcLen, props, (unsigned)propsSize, LZMA_FINISH_ANY, &status, &g_Alloc); +} diff --git a/lzma/C/LzmaLib.h b/lzma/C/LzmaLib.h new file mode 100644 index 0000000000..849d347e95 --- /dev/null +++ b/lzma/C/LzmaLib.h @@ -0,0 +1,135 @@ +/* LzmaLib.h -- LZMA library interface +2008-08-05 +Igor Pavlov +Public domain */ + +#ifndef __LZMALIB_H +#define __LZMALIB_H + +#include "Types.h" + +#ifdef __cplusplus + #define MY_EXTERN_C extern "C" +#else + #define MY_EXTERN_C extern +#endif + +#define MY_STDAPI MY_EXTERN_C int MY_STD_CALL + +#define LZMA_PROPS_SIZE 5 + +/* +RAM requirements for LZMA: + for compression: (dictSize * 11.5 + 6 MB) + state_size + for decompression: dictSize + state_size + state_size = (4 + (1.5 << (lc + lp))) KB + by default (lc=3, lp=0), state_size = 16 KB. + +LZMA properties (5 bytes) format + Offset Size Description + 0 1 lc, lp and pb in encoded form. + 1 4 dictSize (little endian). +*/ + +/* +LzmaCompress +------------ + +outPropsSize - + In: the pointer to the size of outProps buffer; *outPropsSize = LZMA_PROPS_SIZE = 5. + Out: the pointer to the size of written properties in outProps buffer; *outPropsSize = LZMA_PROPS_SIZE = 5. + + LZMA Encoder will use defult values for any parameter, if it is + -1 for any from: level, loc, lp, pb, fb, numThreads + 0 for dictSize + +level - compression level: 0 <= level <= 9; + + level dictSize algo fb + 0: 16 KB 0 32 + 1: 64 KB 0 32 + 2: 256 KB 0 32 + 3: 1 MB 0 32 + 4: 4 MB 0 32 + 5: 16 MB 1 32 + 6: 32 MB 1 32 + 7+: 64 MB 1 64 + + The default value for "level" is 5. + + algo = 0 means fast method + algo = 1 means normal method + +dictSize - The dictionary size in bytes. The maximum value is + 128 MB = (1 << 27) bytes for 32-bit version + 1 GB = (1 << 30) bytes for 64-bit version + The default value is 16 MB = (1 << 24) bytes. + It's recommended to use the dictionary that is larger than 4 KB and + that can be calculated as (1 << N) or (3 << N) sizes. + +lc - The number of literal context bits (high bits of previous literal). + It can be in the range from 0 to 8. The default value is 3. + Sometimes lc=4 gives the gain for big files. + +lp - The number of literal pos bits (low bits of current position for literals). + It can be in the range from 0 to 4. The default value is 0. + The lp switch is intended for periodical data when the period is equal to 2^lp. + For example, for 32-bit (4 bytes) periodical data you can use lp=2. Often it's + better to set lc=0, if you change lp switch. + +pb - The number of pos bits (low bits of current position). + It can be in the range from 0 to 4. The default value is 2. + The pb switch is intended for periodical data when the period is equal 2^pb. + +fb - Word size (the number of fast bytes). + It can be in the range from 5 to 273. The default value is 32. + Usually, a big number gives a little bit better compression ratio and + slower compression process. + +numThreads - The number of thereads. 1 or 2. The default value is 2. + Fast mode (algo = 0) can use only 1 thread. + +Out: + destLen - processed output size +Returns: + SZ_OK - OK + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_PARAM - Incorrect paramater + SZ_ERROR_OUTPUT_EOF - output buffer overflow + SZ_ERROR_THREAD - errors in multithreading functions (only for Mt version) +*/ + +MY_STDAPI LzmaCompress(unsigned char *dest, size_t *destLen, const unsigned char *src, size_t srcLen, + unsigned char *outProps, size_t *outPropsSize, /* *outPropsSize must be = 5 */ + int level, /* 0 <= level <= 9, default = 5 */ + unsigned dictSize, /* default = (1 << 24) */ + int lc, /* 0 <= lc <= 8, default = 3 */ + int lp, /* 0 <= lp <= 4, default = 0 */ + int pb, /* 0 <= pb <= 4, default = 2 */ + int fb, /* 5 <= fb <= 273, default = 32 */ + int numThreads /* 1 or 2, default = 2 */ + ); + +/* +LzmaUncompress +-------------- +In: + dest - output data + destLen - output data size + src - input data + srcLen - input data size +Out: + destLen - processed output size + srcLen - processed input size +Returns: + SZ_OK - OK + SZ_ERROR_DATA - Data error + SZ_ERROR_MEM - Memory allocation arror + SZ_ERROR_UNSUPPORTED - Unsupported properties + SZ_ERROR_INPUT_EOF - it needs more bytes in input buffer (src) +*/ + +MY_STDAPI LzmaUncompress(unsigned char *dest, size_t *destLen, const unsigned char *src, SizeT *srcLen, + const unsigned char *props, size_t propsSize); + +#endif diff --git a/lzma/C/Threads.c b/lzma/C/Threads.c new file mode 100644 index 0000000000..1602183129 --- /dev/null +++ b/lzma/C/Threads.c @@ -0,0 +1,113 @@ +/* Threads.c -- multithreading library +2008-08-05 +Igor Pavlov +Public domain */ + +#include "Threads.h" +#include + +static WRes GetError() +{ + DWORD res = GetLastError(); + return (res) ? (WRes)(res) : 1; +} + +WRes HandleToWRes(HANDLE h) { return (h != 0) ? 0 : GetError(); } +WRes BOOLToWRes(BOOL v) { return v ? 0 : GetError(); } + +static WRes MyCloseHandle(HANDLE *h) +{ + if (*h != NULL) + if (!CloseHandle(*h)) + return GetError(); + *h = NULL; + return 0; +} + +WRes Thread_Create(CThread *thread, THREAD_FUNC_RET_TYPE (THREAD_FUNC_CALL_TYPE *startAddress)(void *), LPVOID parameter) +{ + unsigned threadId; /* Windows Me/98/95: threadId parameter may not be NULL in _beginthreadex/CreateThread functions */ + thread->handle = + /* CreateThread(0, 0, startAddress, parameter, 0, &threadId); */ + (HANDLE)_beginthreadex(NULL, 0, startAddress, parameter, 0, &threadId); + /* maybe we must use errno here, but probably GetLastError() is also OK. */ + return HandleToWRes(thread->handle); +} + +WRes WaitObject(HANDLE h) +{ + return (WRes)WaitForSingleObject(h, INFINITE); +} + +WRes Thread_Wait(CThread *thread) +{ + if (thread->handle == NULL) + return 1; + return WaitObject(thread->handle); +} + +WRes Thread_Close(CThread *thread) +{ + return MyCloseHandle(&thread->handle); +} + +WRes Event_Create(CEvent *p, BOOL manualReset, int initialSignaled) +{ + p->handle = CreateEvent(NULL, manualReset, (initialSignaled ? TRUE : FALSE), NULL); + return HandleToWRes(p->handle); +} + +WRes ManualResetEvent_Create(CManualResetEvent *p, int initialSignaled) + { return Event_Create(p, TRUE, initialSignaled); } +WRes ManualResetEvent_CreateNotSignaled(CManualResetEvent *p) + { return ManualResetEvent_Create(p, 0); } + +WRes AutoResetEvent_Create(CAutoResetEvent *p, int initialSignaled) + { return Event_Create(p, FALSE, initialSignaled); } +WRes AutoResetEvent_CreateNotSignaled(CAutoResetEvent *p) + { return AutoResetEvent_Create(p, 0); } + +WRes Event_Set(CEvent *p) { return BOOLToWRes(SetEvent(p->handle)); } +WRes Event_Reset(CEvent *p) { return BOOLToWRes(ResetEvent(p->handle)); } +WRes Event_Wait(CEvent *p) { return WaitObject(p->handle); } +WRes Event_Close(CEvent *p) { return MyCloseHandle(&p->handle); } + + +WRes Semaphore_Create(CSemaphore *p, UInt32 initiallyCount, UInt32 maxCount) +{ + p->handle = CreateSemaphore(NULL, (LONG)initiallyCount, (LONG)maxCount, NULL); + return HandleToWRes(p->handle); +} + +WRes Semaphore_Release(CSemaphore *p, LONG releaseCount, LONG *previousCount) +{ + return BOOLToWRes(ReleaseSemaphore(p->handle, releaseCount, previousCount)); +} +WRes Semaphore_ReleaseN(CSemaphore *p, UInt32 releaseCount) +{ + return Semaphore_Release(p, (LONG)releaseCount, NULL); +} +WRes Semaphore_Release1(CSemaphore *p) +{ + return Semaphore_ReleaseN(p, 1); +} + +WRes Semaphore_Wait(CSemaphore *p) { return WaitObject(p->handle); } +WRes Semaphore_Close(CSemaphore *p) { return MyCloseHandle(&p->handle); } + +WRes CriticalSection_Init(CCriticalSection *p) +{ + /* InitializeCriticalSection can raise only STATUS_NO_MEMORY exception */ +#ifndef __GNUC__ + __try +#endif + { + InitializeCriticalSection(p); + /* InitializeCriticalSectionAndSpinCount(p, 0); */ + } +#ifndef __GNUC__ + __except (EXCEPTION_EXECUTE_HANDLER) { return 1; } +#endif + return 0; +} + diff --git a/lzma/C/Threads.h b/lzma/C/Threads.h new file mode 100644 index 0000000000..4235c44c24 --- /dev/null +++ b/lzma/C/Threads.h @@ -0,0 +1,78 @@ +/* Threads.h -- multithreading library +2008-11-22 : Igor Pavlov : Public domain */ + +#ifndef __7Z_THRESDS_H +#define __7Z_THRESDS_H + +#ifdef _WIN32 +#include +#endif + +#ifdef _WIN32 +typedef DWORD WRes; +#else +typedef int WRes; +#endif + +#include "Types.h" + +typedef struct _CThread +{ + HANDLE handle; +} CThread; + +#define Thread_Construct(thread) (thread)->handle = NULL +#define Thread_WasCreated(thread) ((thread)->handle != NULL) + +typedef unsigned THREAD_FUNC_RET_TYPE; +#define THREAD_FUNC_CALL_TYPE __stdcall +#define THREAD_FUNC_DECL THREAD_FUNC_RET_TYPE THREAD_FUNC_CALL_TYPE + +WRes Thread_Create(CThread *thread, THREAD_FUNC_RET_TYPE (THREAD_FUNC_CALL_TYPE *startAddress)(void *), LPVOID parameter); +WRes Thread_Wait(CThread *thread); +WRes Thread_Close(CThread *thread); + +typedef struct _CEvent +{ + HANDLE handle; +} CEvent; + +typedef CEvent CAutoResetEvent; +typedef CEvent CManualResetEvent; + +#define Event_Construct(event) (event)->handle = NULL +#define Event_IsCreated(event) ((event)->handle != NULL) + +WRes ManualResetEvent_Create(CManualResetEvent *event, int initialSignaled); +WRes ManualResetEvent_CreateNotSignaled(CManualResetEvent *event); +WRes AutoResetEvent_Create(CAutoResetEvent *event, int initialSignaled); +WRes AutoResetEvent_CreateNotSignaled(CAutoResetEvent *event); +WRes Event_Set(CEvent *event); +WRes Event_Reset(CEvent *event); +WRes Event_Wait(CEvent *event); +WRes Event_Close(CEvent *event); + + +typedef struct _CSemaphore +{ + HANDLE handle; +} CSemaphore; + +#define Semaphore_Construct(p) (p)->handle = NULL + +WRes Semaphore_Create(CSemaphore *p, UInt32 initiallyCount, UInt32 maxCount); +WRes Semaphore_ReleaseN(CSemaphore *p, UInt32 num); +WRes Semaphore_Release1(CSemaphore *p); +WRes Semaphore_Wait(CSemaphore *p); +WRes Semaphore_Close(CSemaphore *p); + + +typedef CRITICAL_SECTION CCriticalSection; + +WRes CriticalSection_Init(CCriticalSection *p); +#define CriticalSection_Delete(p) DeleteCriticalSection(p) +#define CriticalSection_Enter(p) EnterCriticalSection(p) +#define CriticalSection_Leave(p) LeaveCriticalSection(p) + +#endif + diff --git a/lzma/C/Types.h b/lzma/C/Types.h new file mode 100644 index 0000000000..ebac878bbe --- /dev/null +++ b/lzma/C/Types.h @@ -0,0 +1,200 @@ +/* Types.h -- Basic types +2008-11-23 : Igor Pavlov : Public domain */ + +#ifndef __7Z_TYPES_H +#define __7Z_TYPES_H + +#include + +#define SZ_OK 0 + +#define SZ_ERROR_DATA 1 +#define SZ_ERROR_MEM 2 +#define SZ_ERROR_CRC 3 +#define SZ_ERROR_UNSUPPORTED 4 +#define SZ_ERROR_PARAM 5 +#define SZ_ERROR_INPUT_EOF 6 +#define SZ_ERROR_OUTPUT_EOF 7 +#define SZ_ERROR_READ 8 +#define SZ_ERROR_WRITE 9 +#define SZ_ERROR_PROGRESS 10 +#define SZ_ERROR_FAIL 11 +#define SZ_ERROR_THREAD 12 + +#define SZ_ERROR_ARCHIVE 16 +#define SZ_ERROR_NO_ARCHIVE 17 + +typedef int SRes; + +#ifndef RINOK +#define RINOK(x) { int __result__ = (x); if (__result__ != 0) return __result__; } +#endif + +#ifndef ZCONF_H +typedef unsigned char Byte; +#endif +typedef short Int16; +typedef unsigned short UInt16; + +#ifdef _LZMA_UINT32_IS_ULONG +typedef long Int32; +typedef unsigned long UInt32; +#else +typedef int Int32; +typedef unsigned int UInt32; +#endif + +#ifdef _SZ_NO_INT_64 + +/* define _SZ_NO_INT_64, if your compiler doesn't support 64-bit integers. + NOTES: Some code will work incorrectly in that case! */ + +typedef long Int64; +typedef unsigned long UInt64; + +#else + +#if defined(_MSC_VER) || defined(__BORLANDC__) +typedef __int64 Int64; +typedef unsigned __int64 UInt64; +#else +typedef long long int Int64; +typedef unsigned long long int UInt64; +#endif + +#endif + +#ifdef _LZMA_NO_SYSTEM_SIZE_T +typedef UInt32 SizeT; +#else +typedef size_t SizeT; +#endif + +typedef int Bool; +#define True 1 +#define False 0 + + +#ifdef _MSC_VER + +#if _MSC_VER >= 1300 +#define MY_NO_INLINE __declspec(noinline) +#else +#define MY_NO_INLINE +#endif + +#define MY_CDECL __cdecl +#define MY_STD_CALL __stdcall +#define MY_FAST_CALL MY_NO_INLINE __fastcall + +#else + +#define MY_CDECL +#define MY_STD_CALL +#define MY_FAST_CALL + +#endif + + +/* The following interfaces use first parameter as pointer to structure */ + +typedef struct +{ + SRes (*Read)(void *p, void *buf, size_t *size); + /* if (input(*size) != 0 && output(*size) == 0) means end_of_stream. + (output(*size) < input(*size)) is allowed */ +} ISeqInStream; + +/* it can return SZ_ERROR_INPUT_EOF */ +SRes SeqInStream_Read(ISeqInStream *stream, void *buf, size_t size); +SRes SeqInStream_Read2(ISeqInStream *stream, void *buf, size_t size, SRes errorType); +SRes SeqInStream_ReadByte(ISeqInStream *stream, Byte *buf); + +typedef struct +{ + size_t (*Write)(void *p, const void *buf, size_t size); + /* Returns: result - the number of actually written bytes. + (result < size) means error */ +} ISeqOutStream; + +typedef enum +{ + SZ_SEEK_SET = 0, + SZ_SEEK_CUR = 1, + SZ_SEEK_END = 2 +} ESzSeek; + +typedef struct +{ + SRes (*Read)(void *p, void *buf, size_t *size); /* same as ISeqInStream::Read */ + SRes (*Seek)(void *p, Int64 *pos, ESzSeek origin); +} ISeekInStream; + +typedef struct +{ + SRes (*Look)(void *p, void **buf, size_t *size); + /* if (input(*size) != 0 && output(*size) == 0) means end_of_stream. + (output(*size) > input(*size)) is not allowed + (output(*size) < input(*size)) is allowed */ + SRes (*Skip)(void *p, size_t offset); + /* offset must be <= output(*size) of Look */ + + SRes (*Read)(void *p, void *buf, size_t *size); + /* reads directly (without buffer). It's same as ISeqInStream::Read */ + SRes (*Seek)(void *p, Int64 *pos, ESzSeek origin); +} ILookInStream; + +SRes LookInStream_LookRead(ILookInStream *stream, void *buf, size_t *size); +SRes LookInStream_SeekTo(ILookInStream *stream, UInt64 offset); + +/* reads via ILookInStream::Read */ +SRes LookInStream_Read2(ILookInStream *stream, void *buf, size_t size, SRes errorType); +SRes LookInStream_Read(ILookInStream *stream, void *buf, size_t size); + +#define LookToRead_BUF_SIZE (1 << 14) + +typedef struct +{ + ILookInStream s; + ISeekInStream *realStream; + size_t pos; + size_t size; + Byte buf[LookToRead_BUF_SIZE]; +} CLookToRead; + +void LookToRead_CreateVTable(CLookToRead *p, int lookahead); +void LookToRead_Init(CLookToRead *p); + +typedef struct +{ + ISeqInStream s; + ILookInStream *realStream; +} CSecToLook; + +void SecToLook_CreateVTable(CSecToLook *p); + +typedef struct +{ + ISeqInStream s; + ILookInStream *realStream; +} CSecToRead; + +void SecToRead_CreateVTable(CSecToRead *p); + +typedef struct +{ + SRes (*Progress)(void *p, UInt64 inSize, UInt64 outSize); + /* Returns: result. (result != SZ_OK) means break. + Value (UInt64)(Int64)-1 for size means unknown value. */ +} ICompressProgress; + +typedef struct +{ + void *(*Alloc)(void *p, size_t size); + void (*Free)(void *p, void *address); /* address can be 0 */ +} ISzAlloc; + +#define IAlloc_Alloc(p, size) (p)->Alloc((p), size) +#define IAlloc_Free(p, a) (p)->Free((p), a) + +#endif diff --git a/lzma/CMakeLists.txt b/lzma/CMakeLists.txt new file mode 100644 index 0000000000..c522f6c356 --- /dev/null +++ b/lzma/CMakeLists.txt @@ -0,0 +1,20 @@ +cmake_minimum_required( VERSION 2.4 ) + +if( CMAKE_COMPILER_IS_GNUC ) + set( CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Wall -fomit-frame-pointer" ) +endif( CMAKE_COMPILER_IS_GNUC ) + +set( LZMA_FILES + C/Alloc.c + C/LzFind.c + C/LzFindMt.c + C/LzmaDec.c + C/LzmaEnc.c + C/LzmaLib.c ) + +if( WIN32 ) + set( LZMA_FILES ${LZMA_FILES} C/Threads.c ) +endif( WIN32 ) + +add_library( lzma ${LZMA_FILES} ) +target_link_libraries( lzma ) diff --git a/lzma/history.txt b/lzma/history.txt new file mode 100644 index 0000000000..9bed5ebbef --- /dev/null +++ b/lzma/history.txt @@ -0,0 +1,236 @@ +HISTORY of the LZMA SDK +----------------------- + +4.65 2009-02-03 +------------------------- +- Some minor fixes + + +4.63 2008-12-31 +------------------------- +- Some minor fixes + + +4.61 beta 2008-11-23 +------------------------- +- The bug in ANSI-C LZMA Decoder was fixed: + If encoded stream was corrupted, decoder could access memory + outside of allocated range. +- Some changes in ANSI-C 7z Decoder interfaces. +- LZMA SDK is placed in the public domain. + + +4.60 beta 2008-08-19 +------------------------- +- Some minor fixes. + + +4.59 beta 2008-08-13 +------------------------- +- The bug was fixed: + LZMA Encoder in fast compression mode could access memory outside of + allocated range in some rare cases. + + +4.58 beta 2008-05-05 +------------------------- +- ANSI-C LZMA Decoder was rewritten for speed optimizations. +- ANSI-C LZMA Encoder was included to LZMA SDK. +- C++ LZMA code now is just wrapper over ANSI-C code. + + +4.57 2007-12-12 +------------------------- +- Speed optimizations in Ñ++ LZMA Decoder. +- Small changes for more compatibility with some C/C++ compilers. + + +4.49 beta 2007-07-05 +------------------------- +- .7z ANSI-C Decoder: + - now it supports BCJ and BCJ2 filters + - now it supports files larger than 4 GB. + - now it supports "Last Write Time" field for files. +- C++ code for .7z archives compressing/decompressing from 7-zip + was included to LZMA SDK. + + +4.43 2006-06-04 +------------------------- +- Small changes for more compatibility with some C/C++ compilers. + + +4.42 2006-05-15 +------------------------- +- Small changes in .h files in ANSI-C version. + + +4.39 beta 2006-04-14 +------------------------- +- The bug in versions 4.33b:4.38b was fixed: + C++ version of LZMA encoder could not correctly compress + files larger than 2 GB with HC4 match finder (-mfhc4). + + +4.37 beta 2005-04-06 +------------------------- +- Fixes in C++ code: code could no be compiled if _NO_EXCEPTIONS was defined. + + +4.35 beta 2005-03-02 +------------------------- +- The bug was fixed in C++ version of LZMA Decoder: + If encoded stream was corrupted, decoder could access memory + outside of allocated range. + + +4.34 beta 2006-02-27 +------------------------- +- Compressing speed and memory requirements for compressing were increased +- LZMA now can use only these match finders: HC4, BT2, BT3, BT4 + + +4.32 2005-12-09 +------------------------- +- Java version of LZMA SDK was included + + +4.30 2005-11-20 +------------------------- +- Compression ratio was improved in -a2 mode +- Speed optimizations for compressing in -a2 mode +- -fb switch now supports values up to 273 +- The bug in 7z_C (7zIn.c) was fixed: + It used Alloc/Free functions from different memory pools. + So if program used two memory pools, it worked incorrectly. +- 7z_C: .7z format supporting was improved +- LZMA# SDK (C#.NET version) was included + + +4.27 (Updated) 2005-09-21 +------------------------- +- Some GUIDs/interfaces in C++ were changed. + IStream.h: + ISequentialInStream::Read now works as old ReadPart + ISequentialOutStream::Write now works as old WritePart + + +4.27 2005-08-07 +------------------------- +- The bug in LzmaDecodeSize.c was fixed: + if _LZMA_IN_CB and _LZMA_OUT_READ were defined, + decompressing worked incorrectly. + + +4.26 2005-08-05 +------------------------- +- Fixes in 7z_C code and LzmaTest.c: + previous versions could work incorrectly, + if malloc(0) returns 0 + + +4.23 2005-06-29 +------------------------- +- Small fixes in C++ code + + +4.22 2005-06-10 +------------------------- +- Small fixes + + +4.21 2005-06-08 +------------------------- +- Interfaces for ANSI-C LZMA Decoder (LzmaDecode.c) were changed +- New additional version of ANSI-C LZMA Decoder with zlib-like interface: + - LzmaStateDecode.h + - LzmaStateDecode.c + - LzmaStateTest.c +- ANSI-C LZMA Decoder now can decompress files larger than 4 GB + + +4.17 2005-04-18 +------------------------- +- New example for RAM->RAM compressing/decompressing: + LZMA + BCJ (filter for x86 code): + - LzmaRam.h + - LzmaRam.cpp + - LzmaRamDecode.h + - LzmaRamDecode.c + - -f86 switch for lzma.exe + + +4.16 2005-03-29 +------------------------- +- The bug was fixed in LzmaDecode.c (ANSI-C LZMA Decoder): + If _LZMA_OUT_READ was defined, and if encoded stream was corrupted, + decoder could access memory outside of allocated range. +- Speed optimization of ANSI-C LZMA Decoder (now it's about 20% faster). + Old version of LZMA Decoder now is in file LzmaDecodeSize.c. + LzmaDecodeSize.c can provide slightly smaller code than LzmaDecode.c +- Small speed optimization in LZMA C++ code +- filter for SPARC's code was added +- Simplified version of .7z ANSI-C Decoder was included + + +4.06 2004-09-05 +------------------------- +- The bug in v4.05 was fixed: + LZMA-Encoder didn't release output stream in some cases. + + +4.05 2004-08-25 +------------------------- +- Source code of filters for x86, IA-64, ARM, ARM-Thumb + and PowerPC code was included to SDK +- Some internal minor changes + + +4.04 2004-07-28 +------------------------- +- More compatibility with some C++ compilers + + +4.03 2004-06-18 +------------------------- +- "Benchmark" command was added. It measures compressing + and decompressing speed and shows rating values. + Also it checks hardware errors. + + +4.02 2004-06-10 +------------------------- +- C++ LZMA Encoder/Decoder code now is more portable + and it can be compiled by GCC on Linux. + + +4.01 2004-02-15 +------------------------- +- Some detection of data corruption was enabled. + LzmaDecode.c / RangeDecoderReadByte + ..... + { + rd->ExtraBytes = 1; + return 0xFF; + } + + +4.00 2004-02-13 +------------------------- +- Original version of LZMA SDK + + + +HISTORY of the LZMA +------------------- + 2001-2008: Improvements to LZMA compressing/decompressing code, + keeping compatibility with original LZMA format + 1996-2001: Development of LZMA compression format + + Some milestones: + + 2001-08-30: LZMA compression was added to 7-Zip + 1999-01-02: First version of 7-Zip was released + + +End of document diff --git a/lzma/lzma.txt b/lzma/lzma.txt new file mode 100644 index 0000000000..d4f4af929a --- /dev/null +++ b/lzma/lzma.txt @@ -0,0 +1,594 @@ +LZMA SDK 4.65 +------------- + +LZMA SDK provides the documentation, samples, header files, libraries, +and tools you need to develop applications that use LZMA compression. + +LZMA is default and general compression method of 7z format +in 7-Zip compression program (www.7-zip.org). LZMA provides high +compression ratio and very fast decompression. + +LZMA is an improved version of famous LZ77 compression algorithm. +It was improved in way of maximum increasing of compression ratio, +keeping high decompression speed and low memory requirements for +decompressing. + + + +LICENSE +------- + +LZMA SDK is written and placed in the public domain by Igor Pavlov. + + +LZMA SDK Contents +----------------- + +LZMA SDK includes: + + - ANSI-C/C++/C#/Java source code for LZMA compressing and decompressing + - Compiled file->file LZMA compressing/decompressing program for Windows system + + +UNIX/Linux version +------------------ +To compile C++ version of file->file LZMA encoding, go to directory +C++/7zip/Compress/LZMA_Alone +and call make to recompile it: + make -f makefile.gcc clean all + +In some UNIX/Linux versions you must compile LZMA with static libraries. +To compile with static libraries, you can use +LIB = -lm -static + + +Files +--------------------- +lzma.txt - LZMA SDK description (this file) +7zFormat.txt - 7z Format description +7zC.txt - 7z ANSI-C Decoder description +methods.txt - Compression method IDs for .7z +lzma.exe - Compiled file->file LZMA encoder/decoder for Windows +history.txt - history of the LZMA SDK + + +Source code structure +--------------------- + +C/ - C files + 7zCrc*.* - CRC code + Alloc.* - Memory allocation functions + Bra*.* - Filters for x86, IA-64, ARM, ARM-Thumb, PowerPC and SPARC code + LzFind.* - Match finder for LZ (LZMA) encoders + LzFindMt.* - Match finder for LZ (LZMA) encoders for multithreading encoding + LzHash.h - Additional file for LZ match finder + LzmaDec.* - LZMA decoding + LzmaEnc.* - LZMA encoding + LzmaLib.* - LZMA Library for DLL calling + Types.h - Basic types for another .c files + Threads.* - The code for multithreading. + + LzmaLib - LZMA Library (.DLL for Windows) + + LzmaUtil - LZMA Utility (file->file LZMA encoder/decoder). + + Archive - files related to archiving + 7z - 7z ANSI-C Decoder + +CPP/ -- CPP files + + Common - common files for C++ projects + Windows - common files for Windows related code + + 7zip - files related to 7-Zip Project + + Common - common files for 7-Zip + + Compress - files related to compression/decompression + + Copy - Copy coder + RangeCoder - Range Coder (special code of compression/decompression) + LZMA - LZMA compression/decompression on C++ + LZMA_Alone - file->file LZMA compression/decompression + Branch - Filters for x86, IA-64, ARM, ARM-Thumb, PowerPC and SPARC code + + Archive - files related to archiving + + Common - common files for archive handling + 7z - 7z C++ Encoder/Decoder + + Bundles - Modules that are bundles of other modules + + Alone7z - 7zr.exe: Standalone version of 7z.exe that supports only 7z/LZMA/BCJ/BCJ2 + Format7zR - 7zr.dll: Reduced version of 7za.dll: extracting/compressing to 7z/LZMA/BCJ/BCJ2 + Format7zExtractR - 7zxr.dll: Reduced version of 7zxa.dll: extracting from 7z/LZMA/BCJ/BCJ2. + + UI - User Interface files + + Client7z - Test application for 7za.dll, 7zr.dll, 7zxr.dll + Common - Common UI files + Console - Code for console archiver + + + +CS/ - C# files + 7zip + Common - some common files for 7-Zip + Compress - files related to compression/decompression + LZ - files related to LZ (Lempel-Ziv) compression algorithm + LZMA - LZMA compression/decompression + LzmaAlone - file->file LZMA compression/decompression + RangeCoder - Range Coder (special code of compression/decompression) + +Java/ - Java files + SevenZip + Compression - files related to compression/decompression + LZ - files related to LZ (Lempel-Ziv) compression algorithm + LZMA - LZMA compression/decompression + RangeCoder - Range Coder (special code of compression/decompression) + + +C/C++ source code of LZMA SDK is part of 7-Zip project. +7-Zip source code can be downloaded from 7-Zip's SourceForge page: + + http://sourceforge.net/projects/sevenzip/ + + + +LZMA features +------------- + - Variable dictionary size (up to 1 GB) + - Estimated compressing speed: about 2 MB/s on 2 GHz CPU + - Estimated decompressing speed: + - 20-30 MB/s on 2 GHz Core 2 or AMD Athlon 64 + - 1-2 MB/s on 200 MHz ARM, MIPS, PowerPC or other simple RISC + - Small memory requirements for decompressing (16 KB + DictionarySize) + - Small code size for decompressing: 5-8 KB + +LZMA decoder uses only integer operations and can be +implemented in any modern 32-bit CPU (or on 16-bit CPU with some conditions). + +Some critical operations that affect the speed of LZMA decompression: + 1) 32*16 bit integer multiply + 2) Misspredicted branches (penalty mostly depends from pipeline length) + 3) 32-bit shift and arithmetic operations + +The speed of LZMA decompressing mostly depends from CPU speed. +Memory speed has no big meaning. But if your CPU has small data cache, +overall weight of memory speed will slightly increase. + + +How To Use +---------- + +Using LZMA encoder/decoder executable +-------------------------------------- + +Usage: LZMA inputFile outputFile [...] + + e: encode file + + d: decode file + + b: Benchmark. There are two tests: compressing and decompressing + with LZMA method. Benchmark shows rating in MIPS (million + instructions per second). Rating value is calculated from + measured speed and it is normalized with Intel's Core 2 results. + Also Benchmark checks possible hardware errors (RAM + errors in most cases). Benchmark uses these settings: + (-a1, -d21, -fb32, -mfbt4). You can change only -d parameter. + Also you can change the number of iterations. Example for 30 iterations: + LZMA b 30 + Default number of iterations is 10. + + + + + -a{N}: set compression mode 0 = fast, 1 = normal + default: 1 (normal) + + d{N}: Sets Dictionary size - [0, 30], default: 23 (8MB) + The maximum value for dictionary size is 1 GB = 2^30 bytes. + Dictionary size is calculated as DictionarySize = 2^N bytes. + For decompressing file compressed by LZMA method with dictionary + size D = 2^N you need about D bytes of memory (RAM). + + -fb{N}: set number of fast bytes - [5, 273], default: 128 + Usually big number gives a little bit better compression ratio + and slower compression process. + + -lc{N}: set number of literal context bits - [0, 8], default: 3 + Sometimes lc=4 gives gain for big files. + + -lp{N}: set number of literal pos bits - [0, 4], default: 0 + lp switch is intended for periodical data when period is + equal 2^N. For example, for 32-bit (4 bytes) + periodical data you can use lp=2. Often it's better to set lc0, + if you change lp switch. + + -pb{N}: set number of pos bits - [0, 4], default: 2 + pb switch is intended for periodical data + when period is equal 2^N. + + -mf{MF_ID}: set Match Finder. Default: bt4. + Algorithms from hc* group doesn't provide good compression + ratio, but they often works pretty fast in combination with + fast mode (-a0). + + Memory requirements depend from dictionary size + (parameter "d" in table below). + + MF_ID Memory Description + + bt2 d * 9.5 + 4MB Binary Tree with 2 bytes hashing. + bt3 d * 11.5 + 4MB Binary Tree with 3 bytes hashing. + bt4 d * 11.5 + 4MB Binary Tree with 4 bytes hashing. + hc4 d * 7.5 + 4MB Hash Chain with 4 bytes hashing. + + -eos: write End Of Stream marker. By default LZMA doesn't write + eos marker, since LZMA decoder knows uncompressed size + stored in .lzma file header. + + -si: Read data from stdin (it will write End Of Stream marker). + -so: Write data to stdout + + +Examples: + +1) LZMA e file.bin file.lzma -d16 -lc0 + +compresses file.bin to file.lzma with 64 KB dictionary (2^16=64K) +and 0 literal context bits. -lc0 allows to reduce memory requirements +for decompression. + + +2) LZMA e file.bin file.lzma -lc0 -lp2 + +compresses file.bin to file.lzma with settings suitable +for 32-bit periodical data (for example, ARM or MIPS code). + +3) LZMA d file.lzma file.bin + +decompresses file.lzma to file.bin. + + +Compression ratio hints +----------------------- + +Recommendations +--------------- + +To increase the compression ratio for LZMA compressing it's desirable +to have aligned data (if it's possible) and also it's desirable to locate +data in such order, where code is grouped in one place and data is +grouped in other place (it's better than such mixing: code, data, code, +data, ...). + + +Filters +------- +You can increase the compression ratio for some data types, using +special filters before compressing. For example, it's possible to +increase the compression ratio on 5-10% for code for those CPU ISAs: +x86, IA-64, ARM, ARM-Thumb, PowerPC, SPARC. + +You can find C source code of such filters in C/Bra*.* files + +You can check the compression ratio gain of these filters with such +7-Zip commands (example for ARM code): +No filter: + 7z a a1.7z a.bin -m0=lzma + +With filter for little-endian ARM code: + 7z a a2.7z a.bin -m0=arm -m1=lzma + +It works in such manner: +Compressing = Filter_encoding + LZMA_encoding +Decompressing = LZMA_decoding + Filter_decoding + +Compressing and decompressing speed of such filters is very high, +so it will not increase decompressing time too much. +Moreover, it reduces decompression time for LZMA_decoding, +since compression ratio with filtering is higher. + +These filters convert CALL (calling procedure) instructions +from relative offsets to absolute addresses, so such data becomes more +compressible. + +For some ISAs (for example, for MIPS) it's impossible to get gain from such filter. + + +LZMA compressed file format +--------------------------- +Offset Size Description + 0 1 Special LZMA properties (lc,lp, pb in encoded form) + 1 4 Dictionary size (little endian) + 5 8 Uncompressed size (little endian). -1 means unknown size + 13 Compressed data + + +ANSI-C LZMA Decoder +~~~~~~~~~~~~~~~~~~~ + +Please note that interfaces for ANSI-C code were changed in LZMA SDK 4.58. +If you want to use old interfaces you can download previous version of LZMA SDK +from sourceforge.net site. + +To use ANSI-C LZMA Decoder you need the following files: +1) LzmaDec.h + LzmaDec.c + Types.h +LzmaUtil/LzmaUtil.c is example application that uses these files. + + +Memory requirements for LZMA decoding +------------------------------------- + +Stack usage of LZMA decoding function for local variables is not +larger than 200-400 bytes. + +LZMA Decoder uses dictionary buffer and internal state structure. +Internal state structure consumes + state_size = (4 + (1.5 << (lc + lp))) KB +by default (lc=3, lp=0), state_size = 16 KB. + + +How To decompress data +---------------------- + +LZMA Decoder (ANSI-C version) now supports 2 interfaces: +1) Single-call Decompressing +2) Multi-call State Decompressing (zlib-like interface) + +You must use external allocator: +Example: +void *SzAlloc(void *p, size_t size) { p = p; return malloc(size); } +void SzFree(void *p, void *address) { p = p; free(address); } +ISzAlloc alloc = { SzAlloc, SzFree }; + +You can use p = p; operator to disable compiler warnings. + + +Single-call Decompressing +------------------------- +When to use: RAM->RAM decompressing +Compile files: LzmaDec.h + LzmaDec.c + Types.h +Compile defines: no defines +Memory Requirements: + - Input buffer: compressed size + - Output buffer: uncompressed size + - LZMA Internal Structures: state_size (16 KB for default settings) + +Interface: + int LzmaDecode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, + const Byte *propData, unsigned propSize, ELzmaFinishMode finishMode, + ELzmaStatus *status, ISzAlloc *alloc); + In: + dest - output data + destLen - output data size + src - input data + srcLen - input data size + propData - LZMA properties (5 bytes) + propSize - size of propData buffer (5 bytes) + finishMode - It has meaning only if the decoding reaches output limit (*destLen). + LZMA_FINISH_ANY - Decode just destLen bytes. + LZMA_FINISH_END - Stream must be finished after (*destLen). + You can use LZMA_FINISH_END, when you know that + current output buffer covers last bytes of stream. + alloc - Memory allocator. + + Out: + destLen - processed output size + srcLen - processed input size + + Output: + SZ_OK + status: + LZMA_STATUS_FINISHED_WITH_MARK + LZMA_STATUS_NOT_FINISHED + LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK + SZ_ERROR_DATA - Data error + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_UNSUPPORTED - Unsupported properties + SZ_ERROR_INPUT_EOF - It needs more bytes in input buffer (src). + + If LZMA decoder sees end_marker before reaching output limit, it returns OK result, + and output value of destLen will be less than output buffer size limit. + + You can use multiple checks to test data integrity after full decompression: + 1) Check Result and "status" variable. + 2) Check that output(destLen) = uncompressedSize, if you know real uncompressedSize. + 3) Check that output(srcLen) = compressedSize, if you know real compressedSize. + You must use correct finish mode in that case. */ + + +Multi-call State Decompressing (zlib-like interface) +---------------------------------------------------- + +When to use: file->file decompressing +Compile files: LzmaDec.h + LzmaDec.c + Types.h + +Memory Requirements: + - Buffer for input stream: any size (for example, 16 KB) + - Buffer for output stream: any size (for example, 16 KB) + - LZMA Internal Structures: state_size (16 KB for default settings) + - LZMA dictionary (dictionary size is encoded in LZMA properties header) + +1) read LZMA properties (5 bytes) and uncompressed size (8 bytes, little-endian) to header: + unsigned char header[LZMA_PROPS_SIZE + 8]; + ReadFile(inFile, header, sizeof(header) + +2) Allocate CLzmaDec structures (state + dictionary) using LZMA properties + + CLzmaDec state; + LzmaDec_Constr(&state); + res = LzmaDec_Allocate(&state, header, LZMA_PROPS_SIZE, &g_Alloc); + if (res != SZ_OK) + return res; + +3) Init LzmaDec structure before any new LZMA stream. And call LzmaDec_DecodeToBuf in loop + + LzmaDec_Init(&state); + for (;;) + { + ... + int res = LzmaDec_DecodeToBuf(CLzmaDec *p, Byte *dest, SizeT *destLen, + const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode); + ... + } + + +4) Free all allocated structures + LzmaDec_Free(&state, &g_Alloc); + +For full code example, look at C/LzmaUtil/LzmaUtil.c code. + + +How To compress data +-------------------- + +Compile files: LzmaEnc.h + LzmaEnc.c + Types.h + +LzFind.c + LzFind.h + LzFindMt.c + LzFindMt.h + LzHash.h + +Memory Requirements: + - (dictSize * 11.5 + 6 MB) + state_size + +Lzma Encoder can use two memory allocators: +1) alloc - for small arrays. +2) allocBig - for big arrays. + +For example, you can use Large RAM Pages (2 MB) in allocBig allocator for +better compression speed. Note that Windows has bad implementation for +Large RAM Pages. +It's OK to use same allocator for alloc and allocBig. + + +Single-call Compression with callbacks +-------------------------------------- + +Check C/LzmaUtil/LzmaUtil.c as example, + +When to use: file->file decompressing + +1) you must implement callback structures for interfaces: +ISeqInStream +ISeqOutStream +ICompressProgress +ISzAlloc + +static void *SzAlloc(void *p, size_t size) { p = p; return MyAlloc(size); } +static void SzFree(void *p, void *address) { p = p; MyFree(address); } +static ISzAlloc g_Alloc = { SzAlloc, SzFree }; + + CFileSeqInStream inStream; + CFileSeqOutStream outStream; + + inStream.funcTable.Read = MyRead; + inStream.file = inFile; + outStream.funcTable.Write = MyWrite; + outStream.file = outFile; + + +2) Create CLzmaEncHandle object; + + CLzmaEncHandle enc; + + enc = LzmaEnc_Create(&g_Alloc); + if (enc == 0) + return SZ_ERROR_MEM; + + +3) initialize CLzmaEncProps properties; + + LzmaEncProps_Init(&props); + + Then you can change some properties in that structure. + +4) Send LZMA properties to LZMA Encoder + + res = LzmaEnc_SetProps(enc, &props); + +5) Write encoded properties to header + + Byte header[LZMA_PROPS_SIZE + 8]; + size_t headerSize = LZMA_PROPS_SIZE; + UInt64 fileSize; + int i; + + res = LzmaEnc_WriteProperties(enc, header, &headerSize); + fileSize = MyGetFileLength(inFile); + for (i = 0; i < 8; i++) + header[headerSize++] = (Byte)(fileSize >> (8 * i)); + MyWriteFileAndCheck(outFile, header, headerSize) + +6) Call encoding function: + res = LzmaEnc_Encode(enc, &outStream.funcTable, &inStream.funcTable, + NULL, &g_Alloc, &g_Alloc); + +7) Destroy LZMA Encoder Object + LzmaEnc_Destroy(enc, &g_Alloc, &g_Alloc); + + +If callback function return some error code, LzmaEnc_Encode also returns that code. + + +Single-call RAM->RAM Compression +-------------------------------- + +Single-call RAM->RAM Compression is similar to Compression with callbacks, +but you provide pointers to buffers instead of pointers to stream callbacks: + +HRes LzmaEncode(Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen, + CLzmaEncProps *props, Byte *propsEncoded, SizeT *propsSize, int writeEndMark, + ICompressProgress *progress, ISzAlloc *alloc, ISzAlloc *allocBig); + +Return code: + SZ_OK - OK + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_PARAM - Incorrect paramater + SZ_ERROR_OUTPUT_EOF - output buffer overflow + SZ_ERROR_THREAD - errors in multithreading functions (only for Mt version) + + + +LZMA Defines +------------ + +_LZMA_SIZE_OPT - Enable some optimizations in LZMA Decoder to get smaller executable code. + +_LZMA_PROB32 - It can increase the speed on some 32-bit CPUs, but memory usage for + some structures will be doubled in that case. + +_LZMA_UINT32_IS_ULONG - Define it if int is 16-bit on your compiler and long is 32-bit. + +_LZMA_NO_SYSTEM_SIZE_T - Define it if you don't want to use size_t type. + + +C++ LZMA Encoder/Decoder +~~~~~~~~~~~~~~~~~~~~~~~~ +C++ LZMA code use COM-like interfaces. So if you want to use it, +you can study basics of COM/OLE. +C++ LZMA code is just wrapper over ANSI-C code. + + +C++ Notes +~~~~~~~~~~~~~~~~~~~~~~~~ +If you use some C++ code folders in 7-Zip (for example, C++ code for .7z handling), +you must check that you correctly work with "new" operator. +7-Zip can be compiled with MSVC 6.0 that doesn't throw "exception" from "new" operator. +So 7-Zip uses "CPP\Common\NewHandler.cpp" that redefines "new" operator: +operator new(size_t size) +{ + void *p = ::malloc(size); + if (p == 0) + throw CNewException(); + return p; +} +If you use MSCV that throws exception for "new" operator, you can compile without +"NewHandler.cpp". So standard exception will be used. Actually some code of +7-Zip catches any exception in internal code and converts it to HRESULT code. +So you don't need to catch CNewException, if you call COM interfaces of 7-Zip. + +--- + +http://www.7-zip.org +http://www.7-zip.org/sdk.html +http://www.7-zip.org/support.html diff --git a/lzma/lzmalib.vcproj b/lzma/lzmalib.vcproj new file mode 100644 index 0000000000..f59e6d4725 --- /dev/null +++ b/lzma/lzmalib.vcproj @@ -0,0 +1,371 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt index d8d23c5d14..1647f41a8d 100644 --- a/src/CMakeLists.txt +++ b/src/CMakeLists.txt @@ -58,8 +58,8 @@ if( WIN32 ) set( FMOD_SEARCH_PATHS "C:/Program Files/FMOD SoundSystem/FMOD Programmers API ${WIN_TYPE}/api" "C:/Program Files (x86)/FMOD SoundSystem/FMOD Programmers API ${WIN_TYPE}/api" - "E:/Programs/Dev/FMOD/FMOD Programmers API ${WIN_TYPE}/api" - "E:/Program Files (x86)/FMOD SoundSystem/FMOD Programmers API ${WIN_TYPE}/api" ) + # This next one is for me. + "E:/Software/Dev/FMOD/${WIN_TYPE}/api" ) set( FMOD_INC_PATH_SUFFIXES PATH_SUFFIXES inc ) set( FMOD_LIB_PATH_SUFFIXES PATH_SUFFIXES lib ) set( NASM_NAMES nasmw nasm ) @@ -362,7 +362,7 @@ add_custom_target( revision_check ALL # Libraries ZDoom needs set( ZDOOM_LIBS ${ZDOOM_LIBS} "${ZLIB_LIBRARIES}" "${JPEG_LIBRARIES}" "${FMOD_LIBRARY}" ) -include_directories( "${ZLIB_INCLUDE_DIR}" "${JPEG_INCLUDE_DIR}" "${FMOD_INCLUDE_DIR}" ) +include_directories( "${ZLIB_INCLUDE_DIR}" "${JPEG_INCLUDE_DIR}" "${FMOD_INCLUDE_DIR}" "${BZIP2_INCLUDE_DIR}" "${LZMA_INCLUDE_DIR}" ) # Start defining source files for ZDoom @@ -690,7 +690,7 @@ add_executable( zdoom WIN32 set_source_files_properties( xlat/parse_xlat.cpp PROPERTIES OBJECT_DEPENDS "${CMAKE_CURRENT_BINARY_DIR}/xlat_parser.c" ) set_source_files_properties( sc_man.cpp PROPERTIES OBJECT_DEPENDS "${CMAKE_CURRENT_BINARY_DIR}/sc_man_scanner.h" ) -target_link_libraries( zdoom ${ZDOOM_LIBS} snes_spc gdtoa dumb ) +target_link_libraries( zdoom ${ZDOOM_LIBS} snes_spc gdtoa dumb bzip2 lzma ) include_directories( . g_doom g_heretic diff --git a/src/compatibility.cpp b/src/compatibility.cpp index 68e60b88c7..e44a672d81 100644 --- a/src/compatibility.cpp +++ b/src/compatibility.cpp @@ -144,6 +144,7 @@ void ParseCompatibility() } else { + x = 0; sc.ScriptError("MD5 signature must be a hexadecimal value"); } } diff --git a/src/files.cpp b/src/files.cpp index d7ff7875a4..08e261d1bd 100644 --- a/src/files.cpp +++ b/src/files.cpp @@ -266,6 +266,205 @@ void FileReaderZ::FillBuffer () Stream.avail_in = numread; } +//========================================================================== +// +// FileReaderZ +// +// The bzip2 wrapper +// reads data from a libbzip2 compressed stream +// +//========================================================================== + +FileReaderBZ2::FileReaderBZ2 (FileReader &file) +: File(file), SawEOF(false) +{ + int err; + + FillBuffer (); + + Stream.bzalloc = NULL; + Stream.bzfree = NULL; + Stream.opaque = NULL; + + err = BZ2_bzDecompressInit(&Stream, 0, 0); + + if (err != BZ_OK) + { + I_Error ("FileReaderBZ2: bzDecompressInit failed: %d\n", err); + } +} + +FileReaderBZ2::~FileReaderBZ2 () +{ + BZ2_bzDecompressEnd (&Stream); +} + +long FileReaderBZ2::Read (void *buffer, long len) +{ + int err; + + Stream.next_out = (char *)buffer; + Stream.avail_out = len; + + do + { + err = BZ2_bzDecompress(&Stream); + if (Stream.avail_in == 0 && !SawEOF) + { + FillBuffer (); + } + } while (err == BZ_OK && Stream.avail_out != 0); + + if (err != BZ_OK && err != BZ_STREAM_END) + { + I_Error ("Corrupt bzip2 stream"); + } + + if (Stream.avail_out != 0) + { + I_Error ("Ran out of data in bzip2 stream"); + } + + return len - Stream.avail_out; +} + +void FileReaderBZ2::FillBuffer () +{ + long numread = File.Read(InBuff, BUFF_SIZE); + + if (numread < BUFF_SIZE) + { + SawEOF = true; + } + Stream.next_in = (char *)InBuff; + Stream.avail_in = numread; +} + +//========================================================================== +// +// bz_internal_error +// +// libbzip2 wants this, since we build it with BZ_NO_STDIO set. +// +//========================================================================== + +extern "C" void bz_internal_error (int errcode) +{ + I_FatalError("libbzip2: internal error number %d\n", errcode); +} + +//========================================================================== +// +// FileReaderLZMA +// +// The lzma wrapper +// reads data from a LZMA compressed stream +// +//========================================================================== + +static void *SzAlloc(void *p, size_t size) { p = p; return malloc(size); } +static void SzFree(void *p, void *address) { p = p; free(address); } +static ISzAlloc g_Alloc = { SzAlloc, SzFree }; + +FileReaderLZMA::FileReaderLZMA (FileReader &file, size_t uncompressed_size, bool zip) +: File(file), SawEOF(false) +{ + BYTE header[4 + LZMA_PROPS_SIZE]; + int err; + + assert(zip == true); + + Size = uncompressed_size; + OutProcessed = 0; + + // Read zip LZMA properties header + if (File.Read(header, sizeof(header)) < sizeof(header)) + { + I_Error("FileReaderLZMA: File too shart\n"); + } + if (header[2] + header[3] * 256 != LZMA_PROPS_SIZE) + { + I_Error("FileReaderLZMA: LZMA props size is %d (expected %d)\n", + header[2] + header[3] * 256, LZMA_PROPS_SIZE); + } + + FillBuffer(); + + LzmaDec_Construct(&Stream); + err = LzmaDec_Allocate(&Stream, header + 4, LZMA_PROPS_SIZE, &g_Alloc); + + if (err != SZ_OK) + { + I_Error("FileReaderLZMA: LzmaDec_Allocate failed: %d\n", err); + } + + LzmaDec_Init(&Stream); +} + +FileReaderLZMA::~FileReaderLZMA () +{ + LzmaDec_Free(&Stream, &g_Alloc); +} + +long FileReaderLZMA::Read (void *buffer, long len) +{ + int err; + Byte *next_out = (Byte *)buffer; + + do + { + ELzmaFinishMode finish_mode = LZMA_FINISH_ANY; + ELzmaStatus status; + size_t out_processed = len; + size_t in_processed = InSize; + + err = LzmaDec_DecodeToBuf(&Stream, next_out, &out_processed, InBuff + InPos, &in_processed, finish_mode, &status); + InPos += in_processed; + InSize -= in_processed; + next_out += out_processed; + len = (long)(len - out_processed); + if (err != SZ_OK) + { + I_Error ("Corrupt LZMA stream"); + } + if (in_processed == 0 && out_processed == 0) + { + if (status != LZMA_STATUS_FINISHED_WITH_MARK) + { + I_Error ("Corrupt LZMA stream"); + } + } + if (InSize == 0 && !SawEOF) + { + FillBuffer (); + } + } while (err == SZ_OK && len != 0); + + if (err != Z_OK && err != Z_STREAM_END) + { + I_Error ("Corrupt LZMA stream"); + } + + if (len != 0) + { + I_Error ("Ran out of data in LZMA stream"); + } + + return (long)(next_out - (Byte *)buffer); +} + +void FileReaderLZMA::FillBuffer () +{ + long numread = File.Read(InBuff, BUFF_SIZE); + + if (numread < BUFF_SIZE) + { + SawEOF = true; + } + InPos = 0; + InSize = numread; +} + //========================================================================== // // MemoryReader @@ -320,4 +519,3 @@ char *MemoryReader::Gets(char *strbuf, int len) { return GetsFromBuffer(bufptr, strbuf, len); } - diff --git a/src/files.h b/src/files.h index 5b9d79433d..22e90f9607 100644 --- a/src/files.h +++ b/src/files.h @@ -3,6 +3,8 @@ #include #include +#include "bzlib.h" +#include "LzmaDec.h" #include "doomtype.h" #include "m_swap.h" @@ -140,6 +142,132 @@ private: FileReaderZ &operator= (const FileReaderZ &) { return *this; } }; +// Wraps around a FileReader to decompress a bzip2 stream +class FileReaderBZ2 +{ +public: + FileReaderBZ2 (FileReader &file); + ~FileReaderBZ2 (); + + long Read (void *buffer, long len); + + FileReaderBZ2 &operator>> (BYTE &v) + { + Read (&v, 1); + return *this; + } + + FileReaderBZ2 &operator>> (SBYTE &v) + { + Read (&v, 1); + return *this; + } + + FileReaderBZ2 &operator>> (WORD &v) + { + Read (&v, 2); + v = LittleShort(v); + return *this; + } + + FileReaderBZ2 &operator>> (SWORD &v) + { + Read (&v, 2); + v = LittleShort(v); + return *this; + } + + FileReaderBZ2 &operator>> (DWORD &v) + { + Read (&v, 4); + v = LittleLong(v); + return *this; + } + + FileReaderBZ2 &operator>> (fixed_t &v) + { + Read (&v, 4); + v = LittleLong(v); + return *this; + } + +private: + enum { BUFF_SIZE = 4096 }; + + FileReader &File; + bool SawEOF; + bz_stream Stream; + BYTE InBuff[BUFF_SIZE]; + + void FillBuffer (); + + FileReaderBZ2 &operator= (const FileReaderBZ2 &) { return *this; } +}; + +// Wraps around a FileReader to decompress a lzma stream +class FileReaderLZMA +{ +public: + FileReaderLZMA (FileReader &file, size_t uncompressed_size, bool zip); + ~FileReaderLZMA (); + + long Read (void *buffer, long len); + + FileReaderLZMA &operator>> (BYTE &v) + { + Read (&v, 1); + return *this; + } + + FileReaderLZMA &operator>> (SBYTE &v) + { + Read (&v, 1); + return *this; + } + + FileReaderLZMA &operator>> (WORD &v) + { + Read (&v, 2); + v = LittleShort(v); + return *this; + } + + FileReaderLZMA &operator>> (SWORD &v) + { + Read (&v, 2); + v = LittleShort(v); + return *this; + } + + FileReaderLZMA &operator>> (DWORD &v) + { + Read (&v, 4); + v = LittleLong(v); + return *this; + } + + FileReaderLZMA &operator>> (fixed_t &v) + { + Read (&v, 4); + v = LittleLong(v); + return *this; + } + +private: + enum { BUFF_SIZE = 4096 }; + + FileReader &File; + bool SawEOF; + CLzmaDec Stream; + size_t Size; + size_t InPos, InSize; + size_t OutProcessed; + BYTE InBuff[BUFF_SIZE]; + + void FillBuffer (); + + FileReaderLZMA &operator= (const FileReaderLZMA &) { return *this; } +}; class MemoryReader : public FileReader { diff --git a/src/g_shared/a_weapons.cpp b/src/g_shared/a_weapons.cpp index 7ab477b62f..a7d83aa789 100644 --- a/src/g_shared/a_weapons.cpp +++ b/src/g_shared/a_weapons.cpp @@ -756,7 +756,7 @@ AWeapon *FWeaponSlot::PickWeapon(player_t *player) } if (player->ReadyWeapon != NULL) { - for (i = 0; i < Weapons.Size(); i++) + for (i = 0; (unsigned)i < Weapons.Size(); i++) { if (Weapons[i].Type == player->ReadyWeapon->GetClass() || (player->ReadyWeapon->WeaponFlags & WIF_POWERED_UP && @@ -1590,7 +1590,7 @@ const PClass *Net_ReadWeapon(BYTE **stream) { index = (index & 0x7F) | (ReadByte(stream) << 7); } - if (index >= Weapons_ntoh.Size()) + if ((unsigned)index >= Weapons_ntoh.Size()) { return NULL; } diff --git a/src/sc_man.cpp b/src/sc_man.cpp index 6119b98fca..dfd202004a 100644 --- a/src/sc_man.cpp +++ b/src/sc_man.cpp @@ -1122,6 +1122,7 @@ void STACK_ARGS FScriptPosition::Message (int severity, const char *message, ... case MSG_FATAL: I_Error ("Script error, \"%s\" line %d:\n%s\n", FileName.GetChars(), ScriptLine, composed.GetChars()); + return; } Printf (level, "%sScript %s, \"%s\" line %d:\n%s%s\n", color, type, FileName.GetChars(), ScriptLine, color, composed.GetChars()); diff --git a/src/sound/fmod_wrap.h b/src/sound/fmod_wrap.h index d2afc6ac6b..aa298311c8 100644 --- a/src/sound/fmod_wrap.h +++ b/src/sound/fmod_wrap.h @@ -167,16 +167,14 @@ namespace FMOD FMOD_RESULT getDSPClock (unsigned int *hi, unsigned int *lo) { return FMOD_System_GetDSPClock(this, hi, lo); } // Recording API. - FMOD_RESULT setRecordDriver (int driver) { return FMOD_System_SetRecordDriver(this, driver); } - FMOD_RESULT getRecordDriver (int *driver) { return FMOD_System_GetRecordDriver(this, driver); } FMOD_RESULT getRecordNumDrivers (int *numdrivers) { return FMOD_System_GetRecordNumDrivers(this, numdrivers); } FMOD_RESULT getRecordDriverInfo (int id, char *name, int namelen, FMOD_GUID *guid) { return FMOD_System_GetRecordDriverInfo(this, id, name, namelen, guid); } FMOD_RESULT getRecordDriverCaps (int id, FMOD_CAPS *caps, int *minfrequency, int *maxfrequency) { return FMOD_System_GetRecordDriverCaps(this, id, caps, minfrequency, maxfrequency); } - FMOD_RESULT getRecordPosition (unsigned int *position) { return FMOD_System_GetRecordPosition(this, position); } + FMOD_RESULT getRecordPosition (int id, unsigned int *position) { return FMOD_System_GetRecordPosition(this, id, position); } - FMOD_RESULT recordStart (Sound *sound, bool loop) { return FMOD_System_RecordStart(this, (FMOD_SOUND *)sound, loop); } - FMOD_RESULT recordStop () { return FMOD_System_RecordStop(this); } - FMOD_RESULT isRecording (bool *recording) { FMOD_BOOL b; FMOD_RESULT res = FMOD_System_IsRecording(this, &b); *recording = b; return res; } + FMOD_RESULT recordStart (int id, Sound *sound, bool loop) { return FMOD_System_RecordStart(this, id, (FMOD_SOUND *)sound, loop); } + FMOD_RESULT recordStop (int id) { return FMOD_System_RecordStop(this, id); } + FMOD_RESULT isRecording (int id, bool *recording) { FMOD_BOOL b; FMOD_RESULT res = FMOD_System_IsRecording(this, id, &b); *recording = b; return res; } // Geometry API. FMOD_RESULT createGeometry (int maxpolygons, int maxvertices, Geometry **geometry) { return FMOD_System_CreateGeometry(this, maxpolygons, maxvertices, (FMOD_GEOMETRY **)geometry); } diff --git a/src/stats.h b/src/stats.h index dfbe261742..bb1deebeb9 100644 --- a/src/stats.h +++ b/src/stats.h @@ -131,7 +131,7 @@ inline volatile unsigned long long rdtsc() #endif { register unsigned long long tsc asm("eax"); - asm volatile ("\trdtsc\n" : : : "eax, "edx"); + asm volatile ("\trdtsc\n" : : : "eax", "edx"); return tsc; } return 0; diff --git a/src/w_wad.cpp b/src/w_wad.cpp index 06d9684a58..34e3f83939 100644 --- a/src/w_wad.cpp +++ b/src/w_wad.cpp @@ -91,6 +91,7 @@ struct FWadCollection::LumpRecord { char * fullname; // only valid for files loaded from a .zip file char name[9]; + BYTE method; // zip compression method short wadnum; WORD flags; int position; @@ -299,7 +300,7 @@ int FWadCollection::AddExternalFile(const char *filename) static DWORD Zip_FindCentralDir(FileReader * fin) { - unsigned char* buf; + unsigned char buf[BUFREADCOMMENT + 4]; DWORD FileSize; DWORD uBackRead; DWORD uMaxBack; // maximum size of global comment @@ -310,38 +311,35 @@ static DWORD Zip_FindCentralDir(FileReader * fin) FileSize = fin->Tell(); uMaxBack = MIN(0xffff, FileSize); - buf = (unsigned char*)malloc(BUFREADCOMMENT+4); - if (buf == NULL) return 0; - uBackRead = 4; while (uBackRead < uMaxBack) { DWORD uReadSize, uReadPos; int i; - if (uBackRead +BUFREADCOMMENT > uMaxBack) + if (uBackRead + BUFREADCOMMENT > uMaxBack) uBackRead = uMaxBack; else uBackRead += BUFREADCOMMENT; - uReadPos = FileSize - uBackRead ; + uReadPos = FileSize - uBackRead; - uReadSize = MIN((BUFREADCOMMENT+4) , (FileSize-uReadPos)); + uReadSize = MIN((BUFREADCOMMENT + 4), (FileSize - uReadPos)); - if (fin->Seek(uReadPos,SEEK_SET) != 0) break; + if (fin->Seek(uReadPos, SEEK_SET) != 0) break; if (fin->Read(buf, (SDWORD)uReadSize) != (SDWORD)uReadSize) break; - for (i=(int)uReadSize-3; (i--)>0;) - if (((*(buf+i))==0x50) && ((*(buf+i+1))==0x4b) && - ((*(buf+i+2))==0x05) && ((*(buf+i+3))==0x06)) + for (i = (int)uReadSize - 3; (i--) > 0;) + { + if (buf[i] == 'P' && buf[i+1] == 'K' && buf[i+2] == 5 && buf[i+3] == 6) { - uPosFound = uReadPos+i; + uPosFound = uReadPos + i; break; } + } - if (uPosFound!=0) - break; + if (uPosFound != 0) + break; } - free(buf); return uPosFound; } @@ -373,7 +371,7 @@ void FWadCollection::AddFile (const char *filename, const char * data, int lengt int startlump; wadlump_t* fileinfo = NULL, *fileinfo2free = NULL; wadlump_t singleinfo; - TArray EmbeddedWADs; + TArray EmbeddedWADs; void * directory = NULL; if (length==-1) @@ -508,10 +506,10 @@ void FWadCollection::AddFile (const char *filename, const char * data, int lengt else if (header.magic[0] == ZIP_ID) { DWORD centraldir = Zip_FindCentralDir(wadinfo); - FZipCentralInfo info; + FZipEndOfCentralDirectory info; int skipped = 0; - if (centraldir==0) + if (centraldir == 0) { Printf("\n%s: ZIP file corrupt!\n", filename); return; @@ -519,56 +517,61 @@ void FWadCollection::AddFile (const char *filename, const char * data, int lengt // Read the central directory info. wadinfo->Seek(centraldir, SEEK_SET); - wadinfo->Read(&info, sizeof(FZipCentralInfo)); + wadinfo->Read(&info, sizeof(FZipEndOfCentralDirectory)); // No multi-disk zips! - if (info.wEntryCount != info.wTotalEntryCount || - info.wNumberDiskWithCD != 0 || info.wNumberDisk != 0) + if (info.NumEntries != info.NumEntriesOnAllDisks || + info.FirstDisk != 0 || info.DiskNumber != 0) { Printf("\n%s: Multipart Zip files are not supported.\n", filename); return; } - NumLumps += LittleShort(info.wEntryCount); + NumLumps += LittleShort(info.NumEntries); LumpInfo.Resize(NumLumps); lump_p = &LumpInfo[startlump]; // Load the entire central directory. Too bad that this contains variable length entries... - directory = malloc(LittleLong(info.dwCDSize)); - wadinfo->Seek(LittleLong(info.dwCDOffset), SEEK_SET); - wadinfo->Read(directory, LittleLong(info.dwCDSize)); + directory = malloc(LittleLong(info.DirectorySize)); + wadinfo->Seek(LittleLong(info.DirectoryOffset), SEEK_SET); + wadinfo->Read(directory, LittleLong(info.DirectorySize)); - char * dirptr =(char*)directory; - for (int i = 0; i < LittleShort(info.wEntryCount); i++) + char *dirptr = (char*)directory; + for (int i = 0; i < LittleShort(info.NumEntries); i++) { - FZipFileInfo * zip_fh = (FZipFileInfo*)dirptr; + FZipCentralDirectoryInfo *zip_fh = (FZipCentralDirectoryInfo *)dirptr; + char name[256]; char base[256]; - int len = LittleShort(zip_fh->wFileNameSize); - strncpy(name, dirptr + sizeof(FZipFileInfo), MIN(len, 255)); - name[len]=0; - dirptr += sizeof(FZipFileInfo) + - LittleShort(zip_fh->wFileNameSize) + - LittleShort(zip_fh->wExtraSize) + - LittleShort(zip_fh->wCommentSize); + int len = LittleShort(zip_fh->NameLength); + strncpy(name, dirptr + sizeof(FZipCentralDirectoryInfo), MIN(len, 255)); + name[len] = 0; + dirptr += sizeof(FZipCentralDirectoryInfo) + + LittleShort(zip_fh->NameLength) + + LittleShort(zip_fh->ExtraLength) + + LittleShort(zip_fh->CommentLength); // skip Directories - if(name[len - 1] == '/' && LittleLong(zip_fh->dwSize) == 0) + if(name[len - 1] == '/' && LittleLong(zip_fh->UncompressedSize) == 0) { skipped++; continue; } - // Ignore obsolete compression formats - if(LittleShort(zip_fh->wCompression) != 0 && LittleShort(zip_fh->wCompression) != Z_DEFLATED) + // Ignore unknown compression formats + zip_fh->Method = LittleShort(zip_fh->Method); + if (zip_fh->Method != METHOD_STORED && + zip_fh->Method != METHOD_DEFLATE && + zip_fh->Method != METHOD_LZMA && + zip_fh->Method != METHOD_BZIP2) { - Printf("\n: %s: '%s' uses an unsupported compression algorithm.\n", filename, name); + Printf("\n: %s: '%s' uses an unsupported compression algorithm (#%d).\n", filename, name, zip_fh->Method); skipped++; continue; } // Also ignore encrypted entries - if(LittleShort(zip_fh->wFlags) & ZF_ENCRYPTED) + if(LittleShort(zip_fh->Flags) & ZF_ENCRYPTED) { Printf("\n%s: '%s' is encrypted. Encryption is not supported.\n", filename, name); skipped++; @@ -591,16 +594,16 @@ void FWadCollection::AddFile (const char *filename, const char * data, int lengt } //ExtractFileBase(name, base); - char * lname=strrchr(name,'/'); - if (!lname) lname=name; + char *lname = strrchr(name,'/'); + if (!lname) lname = name; else lname++; strcpy(base, lname); - char * dot = strrchr(base,'.'); - if (dot) *dot=0; + char *dot = strrchr(base, '.'); + if (dot) *dot = 0; uppercopy(lump_p->name, base); lump_p->name[8] = 0; lump_p->fullname = copystring(name); - lump_p->size = zip_fh->dwSize; + lump_p->size = LittleLong(zip_fh->UncompressedSize); // Map some directories to WAD namespaces. // Note that some of these namespaces don't exist in WADS. @@ -621,15 +624,15 @@ void FWadCollection::AddFile (const char *filename, const char * data, int lengt // Anything that is not in one of these subdirectories or the main directory // should not be accessible through the standard WAD functions but only through // the ones which look for the full name. - if (lump_p->namespc==-1) + if (lump_p->namespc == -1) { memset(lump_p->name, 0, 8); } lump_p->wadnum = (WORD)Wads.Size(); - lump_p->flags = LittleShort(zip_fh->wCompression) == Z_DEFLATED? - LUMPF_COMPRESSED|LUMPF_ZIPFILE : LUMPF_ZIPFILE; - lump_p->compressedsize = LittleLong(zip_fh->dwCompressedSize); + lump_p->flags = (zip_fh->Method != METHOD_STORED) ? LUMPF_COMPRESSED|LUMPF_ZIPFILE : LUMPF_ZIPFILE; + lump_p->method = zip_fh->Method; + lump_p->compressedsize = LittleLong(zip_fh->CompressedSize); // Since '\' can't be used as a file name's part inside a ZIP // we have to work around this for sprites because it is a valid @@ -638,15 +641,15 @@ void FWadCollection::AddFile (const char *filename, const char * data, int lengt { char * c; - while ((c=(char*)memchr(lump_p->name, '^', 8))) + while ((c = (char*)memchr(lump_p->name, '^', 8))) { - *c='\\'; + *c = '\\'; } } // The start of the file will be determined the first time it is accessed. lump_p->flags |= LUMPF_NEEDFILESTART; - lump_p->position = LittleLong(zip_fh->dwFileOffset); + lump_p->position = LittleLong(zip_fh->LocalHeaderOffset); lump_p++; } // Resize the lump record array to its actual size @@ -716,30 +719,40 @@ void FWadCollection::AddFile (const char *filename, const char * data, int lengt for(unsigned int i = 0; i < EmbeddedWADs.Size(); i++) { - FZipFileInfo * zip_fh = EmbeddedWADs[i]; - FZipLocalHeader localHeader; + FZipCentralDirectoryInfo *zip_fh = EmbeddedWADs[i]; + FZipLocalFileHeader localHeader; - *wadstr=0; - size_t len = LittleShort(zip_fh->wFileNameSize); - if (len+strlen(path) > 255) len = 255-strlen(path); - strncpy(wadstr, ((char*)zip_fh) + sizeof(FZipFileInfo), len); - wadstr[len]=0; + *wadstr = 0; + size_t len = LittleShort(zip_fh->NameLength); + if (len + strlen(path) > 255) len = 255 - strlen(path); + strncpy(wadstr, ((char*)zip_fh) + sizeof(FZipCentralDirectoryInfo), len); + wadstr[len] = 0; - DWORD size = LittleLong(zip_fh->dwSize); - char * buffer = new char[size]; + DWORD size = LittleLong(zip_fh->UncompressedSize); + char *buffer = new char[size]; - int position = LittleLong(zip_fh->dwFileOffset) ; + int position = LittleLong(zip_fh->LocalHeaderOffset) ; wadinfo->Seek(position, SEEK_SET); wadinfo->Read(&localHeader, sizeof(localHeader)); - position += LittleShort(localHeader.wExtraSize) + sizeof(FZipLocalHeader) + LittleShort(zip_fh->wFileNameSize); + position += sizeof(FZipLocalFileHeader) + LittleShort(localHeader.ExtraLength) + LittleShort(zip_fh->NameLength); wadinfo->Seek(position, SEEK_SET); - if (zip_fh->wCompression == Z_DEFLATED) + if (LittleShort(zip_fh->Method) == METHOD_DEFLATE) { FileReaderZ frz(*wadinfo, true); frz.Read(buffer, size); } + else if (LittleShort(zip_fh->Method) == METHOD_LZMA) + { + FileReaderLZMA frz(*wadinfo, size, true); + frz.Read(buffer, size); + } + else if (LittleShort(zip_fh->Method) == METHOD_BZIP2) + { + FileReaderBZ2 frz(*wadinfo); + frz.Read(buffer, size); + } else { wadinfo->Read(buffer, size); @@ -1746,16 +1759,16 @@ void FWadCollection::SetLumpAddress(LumpRecord *l) // This file is inside a zip and has not been opened before. // Position points to the start of the local file header, which we must // read and skip so that we can get to the actual file data. - FZipLocalHeader localHeader; + FZipLocalFileHeader localHeader; int skiplen; int address; WadFileRecord *wad = Wads[l->wadnum]; address = wad->Tell(); - wad->Seek (l->position, SEEK_SET); - wad->Read (&localHeader, sizeof(localHeader)); - skiplen = LittleShort(localHeader.wFileNameSize) + LittleShort(localHeader.wExtraSize); + wad->Seek(l->position, SEEK_SET); + wad->Read(&localHeader, sizeof(localHeader)); + skiplen = LittleShort(localHeader.NameLength) + LittleShort(localHeader.ExtraLength); l->position += sizeof(localHeader) + skiplen; l->flags &= ~LUMPF_NEEDFILESTART; } @@ -1792,10 +1805,27 @@ FWadLump FWadCollection::OpenLumpNum (int lump) if (l->flags & LUMPF_COMPRESSED) { // A compressed entry in a .zip file - char * buffer = new char[l->size+1]; // the last byte is used as a reference counter + char *buffer = new char[l->size + 1]; // the last byte is used as a reference counter buffer[l->size] = 0; - FileReaderZ frz(*wad, true); - frz.Read(buffer, l->size); + if (l->method == METHOD_DEFLATE) + { + FileReaderZ frz(*wad, true); + frz.Read(buffer, l->size); + } + else if (l->method == METHOD_LZMA) + { + FileReaderLZMA frz(*wad, l->size, true); + frz.Read(buffer, l->size); + } + else if (l->method == METHOD_BZIP2) + { + FileReaderBZ2 frz(*wad); + frz.Read(buffer, l->size); + } + else + { + assert(0); // Should not get here + } return FWadLump(buffer, l->size, true); } else if (l->flags & LUMPF_EXTERNAL) @@ -1875,10 +1905,27 @@ FWadLump *FWadCollection::ReopenLumpNum (int lump) { // A compressed entry in a .zip file int address = wad->Tell(); // read from the existing WadFileRecord without reopening - char * buffer = new char[l->size+1]; // the last byte is used as a reference counter + char *buffer = new char[l->size + 1]; // the last byte is used as a reference counter wad->Seek(l->position, SEEK_SET); - FileReaderZ frz(*wad, true); - frz.Read(buffer, l->size); + if (l->method == METHOD_DEFLATE) + { + FileReaderZ frz(*wad, true); + frz.Read(buffer, l->size); + } + else if (l->method == METHOD_LZMA) + { + FileReaderLZMA frz(*wad, l->size, true); + frz.Read(buffer, l->size); + } + else if (l->method == METHOD_BZIP2) + { + FileReaderBZ2 frz(*wad); + frz.Read(buffer, l->size); + } + else + { + assert(0); // Should not get here + } wad->Seek(address, SEEK_SET); return new FWadLump(buffer, l->size, true); //... but restore the file pointer afterward! } diff --git a/src/w_zip.h b/src/w_zip.h index ccfbe68f46..0189d3c94e 100644 --- a/src/w_zip.h +++ b/src/w_zip.h @@ -1,62 +1,72 @@ #ifndef __W_ZIP #define __W_ZIP -#pragma pack(1) -struct FZipCentralInfo +#pragma pack(push, 1) +// FZipCentralInfo +struct FZipEndOfCentralDirectory { - // patched together from information from Q3's unzip.c - DWORD dwSignature; - WORD wNumberDisk; - WORD wNumberDiskWithCD; - WORD wEntryCount; - WORD wTotalEntryCount; - DWORD dwCDSize; - DWORD dwCDOffset; - WORD wCommentSize; + DWORD Magic; + WORD DiskNumber; + WORD FirstDisk; + WORD NumEntries; + WORD NumEntriesOnAllDisks; + DWORD DirectorySize; + DWORD DirectoryOffset; + WORD ZipCommentLength; }; -struct FZipFileInfo +// FZipFileInfo +struct FZipCentralDirectoryInfo { - // patched together from information from Q3's unzip.c - DWORD dwSignature; - WORD wVersion; - WORD wRequiredVersion; - WORD wFlags; - WORD wCompression; - DWORD dwLastModified; - DWORD dwCrc; - DWORD dwCompressedSize; - DWORD dwSize; - WORD wFileNameSize; - WORD wExtraSize; - WORD wCommentSize; - WORD wDiskStart; - WORD wInternalAttributes; - DWORD wExternalAttributes; - DWORD dwFileOffset; + DWORD Magic; + BYTE VersionMadeBy[2]; + BYTE VersionToExtract[2]; + WORD Flags; + WORD Method; + WORD ModTime; + WORD ModDate; + DWORD CRC32; + DWORD CompressedSize; + DWORD UncompressedSize; + WORD NameLength; + WORD ExtraLength; + WORD CommentLength; + WORD StartingDiskNumber; + WORD InternalAttributes; + DWORD ExternalAttributes; + DWORD LocalHeaderOffset; // file name and other variable length info follows }; -struct FZipLocalHeader +// FZipLocalHeader +struct FZipLocalFileHeader { - // patched together from information from Q3's unzip.c - DWORD dwSignature; - WORD wRequiredVersion; - WORD wFlags; - WORD wCompression; - DWORD dwLastModified; - DWORD dwCrc; - DWORD dwCompressedSize; - DWORD dwSize; - WORD wFileNameSize; - WORD wExtraSize; + DWORD Magic; + BYTE VersionToExtract[2]; + WORD Flags; + WORD Method; + WORD ModTime; + WORD ModDate; + DWORD CRC32; + DWORD CompressedSize; + DWORD UncompressedSize; + WORD NameLength; + WORD ExtraLength; // file name and other variable length info follows }; -#pragma pack() +#pragma pack(pop) -#define Z_DEFLATED 8 +#define ZIP_LOCALFILE MAKE_ID('P','K',3,4) +#define ZIP_CENTRALFILE MAKE_ID('P','K',1,2) +#define ZIP_ENDOFDIR MAKE_ID('P','K',5,6) + +#define METHOD_STORED 0 +#define METHOD_DEFLATE 8 +#define METHOD_BZIP2 12 +#define METHOD_LZMA 14 +#define METHOD_PPMD 98 // File header flags. #define ZF_ENCRYPTED 0x1 diff --git a/src/win32/i_crash.cpp b/src/win32/i_crash.cpp index e0e9416e61..6f8963b27a 100644 --- a/src/win32/i_crash.cpp +++ b/src/win32/i_crash.cpp @@ -1004,7 +1004,7 @@ static void StackWalk (HANDLE file, void *dumpaddress, DWORD *topOfStack, DWORD const BYTE *bytep = (BYTE *)code; BYTE peekb; -#define chkbyte(x,m,v) SafeReadMemory(x, &peekb, 1) && ((peekb & m) == v) +#define chkbyte(x,m,v) (SafeReadMemory(x, &peekb, 1) && ((peekb & m) == v)) if (chkbyte(bytep - 5, 0xFF, 0xE8) || chkbyte(bytep - 5, 0xFF, 0xE9)) { diff --git a/tools/zipdir/CMakeLists.txt b/tools/zipdir/CMakeLists.txt index 177cd1eddd..7dd6d63c3d 100644 --- a/tools/zipdir/CMakeLists.txt +++ b/tools/zipdir/CMakeLists.txt @@ -1,6 +1,8 @@ cmake_minimum_required( VERSION 2.4 ) +message(STATUS "${ZLIB_INCLUDE_DIR}" ) +message(STATUS "${BZIP2_INCLUDE_DIR}" ) +message(STATUS "${LZMA_INCLUDE_DIR}" ) +include_directories( "${ZLIB_INCLUDE_DIR}" "${BZIP2_INCLUDE_DIR}" "${LZMA_INCLUDE_DIR}" ) add_executable( zipdir - zipdir.c - ioapi.c - zip.c ) -target_link_libraries( zipdir ${ZLIB_LIBRARIES} ) + zipdir.c ) +target_link_libraries( zipdir ${ZLIB_LIBRARIES} bzip2 lzma ) diff --git a/tools/zipdir/ioapi.c b/tools/zipdir/ioapi.c deleted file mode 100644 index bb585ba33d..0000000000 --- a/tools/zipdir/ioapi.c +++ /dev/null @@ -1,177 +0,0 @@ -/* ioapi.c -- IO base function header for compress/uncompress .zip - files using zlib + zip or unzip API - - Version 1.01e, February 12th, 2005 - - Copyright (C) 1998-2005 Gilles Vollant -*/ - -#include -#include -#include - -#include "../../zlib/zlib.h" -#include "ioapi.h" - - - -/* I've found an old Unix (a SunOS 4.1.3_U1) without all SEEK_* defined.... */ - -#ifndef SEEK_CUR -#define SEEK_CUR 1 -#endif - -#ifndef SEEK_END -#define SEEK_END 2 -#endif - -#ifndef SEEK_SET -#define SEEK_SET 0 -#endif - -voidpf ZCALLBACK fopen_file_func OF(( - voidpf opaque, - const char* filename, - int mode)); - -uLong ZCALLBACK fread_file_func OF(( - voidpf opaque, - voidpf stream, - void* buf, - uLong size)); - -uLong ZCALLBACK fwrite_file_func OF(( - voidpf opaque, - voidpf stream, - const void* buf, - uLong size)); - -long ZCALLBACK ftell_file_func OF(( - voidpf opaque, - voidpf stream)); - -long ZCALLBACK fseek_file_func OF(( - voidpf opaque, - voidpf stream, - uLong offset, - int origin)); - -int ZCALLBACK fclose_file_func OF(( - voidpf opaque, - voidpf stream)); - -int ZCALLBACK ferror_file_func OF(( - voidpf opaque, - voidpf stream)); - - -voidpf ZCALLBACK fopen_file_func (opaque, filename, mode) - voidpf opaque; - const char* filename; - int mode; -{ - FILE* file = NULL; - const char* mode_fopen = NULL; - if ((mode & ZLIB_FILEFUNC_MODE_READWRITEFILTER)==ZLIB_FILEFUNC_MODE_READ) - mode_fopen = "rb"; - else - if (mode & ZLIB_FILEFUNC_MODE_EXISTING) - mode_fopen = "r+b"; - else - if (mode & ZLIB_FILEFUNC_MODE_CREATE) - mode_fopen = "wb"; - - if ((filename!=NULL) && (mode_fopen != NULL)) - file = fopen(filename, mode_fopen); - return file; -} - - -uLong ZCALLBACK fread_file_func (opaque, stream, buf, size) - voidpf opaque; - voidpf stream; - void* buf; - uLong size; -{ - uLong ret; - ret = (uLong)fread(buf, 1, (size_t)size, (FILE *)stream); - return ret; -} - - -uLong ZCALLBACK fwrite_file_func (opaque, stream, buf, size) - voidpf opaque; - voidpf stream; - const void* buf; - uLong size; -{ - uLong ret; - ret = (uLong)fwrite(buf, 1, (size_t)size, (FILE *)stream); - return ret; -} - -long ZCALLBACK ftell_file_func (opaque, stream) - voidpf opaque; - voidpf stream; -{ - long ret; - ret = ftell((FILE *)stream); - return ret; -} - -long ZCALLBACK fseek_file_func (opaque, stream, offset, origin) - voidpf opaque; - voidpf stream; - uLong offset; - int origin; -{ - int fseek_origin=0; - long ret; - switch (origin) - { - case ZLIB_FILEFUNC_SEEK_CUR : - fseek_origin = SEEK_CUR; - break; - case ZLIB_FILEFUNC_SEEK_END : - fseek_origin = SEEK_END; - break; - case ZLIB_FILEFUNC_SEEK_SET : - fseek_origin = SEEK_SET; - break; - default: return -1; - } - ret = 0; - fseek((FILE *)stream, offset, fseek_origin); - return ret; -} - -int ZCALLBACK fclose_file_func (opaque, stream) - voidpf opaque; - voidpf stream; -{ - int ret; - ret = fclose((FILE *)stream); - return ret; -} - -int ZCALLBACK ferror_file_func (opaque, stream) - voidpf opaque; - voidpf stream; -{ - int ret; - ret = ferror((FILE *)stream); - return ret; -} - -void fill_fopen_filefunc (pzlib_filefunc_def) - zlib_filefunc_def* pzlib_filefunc_def; -{ - pzlib_filefunc_def->zopen_file = fopen_file_func; - pzlib_filefunc_def->zread_file = fread_file_func; - pzlib_filefunc_def->zwrite_file = fwrite_file_func; - pzlib_filefunc_def->ztell_file = ftell_file_func; - pzlib_filefunc_def->zseek_file = fseek_file_func; - pzlib_filefunc_def->zclose_file = fclose_file_func; - pzlib_filefunc_def->zerror_file = ferror_file_func; - pzlib_filefunc_def->opaque = NULL; -} diff --git a/tools/zipdir/ioapi.h b/tools/zipdir/ioapi.h deleted file mode 100644 index e73a3b2bd8..0000000000 --- a/tools/zipdir/ioapi.h +++ /dev/null @@ -1,75 +0,0 @@ -/* ioapi.h -- IO base function header for compress/uncompress .zip - files using zlib + zip or unzip API - - Version 1.01e, February 12th, 2005 - - Copyright (C) 1998-2005 Gilles Vollant -*/ - -#ifndef _ZLIBIOAPI_H -#define _ZLIBIOAPI_H - - -#define ZLIB_FILEFUNC_SEEK_CUR (1) -#define ZLIB_FILEFUNC_SEEK_END (2) -#define ZLIB_FILEFUNC_SEEK_SET (0) - -#define ZLIB_FILEFUNC_MODE_READ (1) -#define ZLIB_FILEFUNC_MODE_WRITE (2) -#define ZLIB_FILEFUNC_MODE_READWRITEFILTER (3) - -#define ZLIB_FILEFUNC_MODE_EXISTING (4) -#define ZLIB_FILEFUNC_MODE_CREATE (8) - - -#ifndef ZCALLBACK - -#if (defined(WIN32) || defined (WINDOWS) || defined (_WINDOWS)) && defined(CALLBACK) && defined (USEWINDOWS_CALLBACK) -#define ZCALLBACK CALLBACK -#else -#define ZCALLBACK -#endif -#endif - -#ifdef __cplusplus -extern "C" { -#endif - -typedef voidpf (ZCALLBACK *open_file_func) OF((voidpf opaque, const char* filename, int mode)); -typedef uLong (ZCALLBACK *read_file_func) OF((voidpf opaque, voidpf stream, void* buf, uLong size)); -typedef uLong (ZCALLBACK *write_file_func) OF((voidpf opaque, voidpf stream, const void* buf, uLong size)); -typedef long (ZCALLBACK *tell_file_func) OF((voidpf opaque, voidpf stream)); -typedef long (ZCALLBACK *seek_file_func) OF((voidpf opaque, voidpf stream, uLong offset, int origin)); -typedef int (ZCALLBACK *close_file_func) OF((voidpf opaque, voidpf stream)); -typedef int (ZCALLBACK *testerror_file_func) OF((voidpf opaque, voidpf stream)); - -typedef struct zlib_filefunc_def_s -{ - open_file_func zopen_file; - read_file_func zread_file; - write_file_func zwrite_file; - tell_file_func ztell_file; - seek_file_func zseek_file; - close_file_func zclose_file; - testerror_file_func zerror_file; - voidpf opaque; -} zlib_filefunc_def; - - - -void fill_fopen_filefunc OF((zlib_filefunc_def* pzlib_filefunc_def)); - -#define ZREAD(filefunc,filestream,buf,size) ((*((filefunc).zread_file))((filefunc).opaque,filestream,buf,size)) -#define ZWRITE(filefunc,filestream,buf,size) ((*((filefunc).zwrite_file))((filefunc).opaque,filestream,buf,size)) -#define ZTELL(filefunc,filestream) ((*((filefunc).ztell_file))((filefunc).opaque,filestream)) -#define ZSEEK(filefunc,filestream,pos,mode) ((*((filefunc).zseek_file))((filefunc).opaque,filestream,pos,mode)) -#define ZCLOSE(filefunc,filestream) ((*((filefunc).zclose_file))((filefunc).opaque,filestream)) -#define ZERROR(filefunc,filestream) ((*((filefunc).zerror_file))((filefunc).opaque,filestream)) - - -#ifdef __cplusplus -} -#endif - -#endif - diff --git a/tools/zipdir/zip.c b/tools/zipdir/zip.c deleted file mode 100644 index d1a47b49a5..0000000000 --- a/tools/zipdir/zip.c +++ /dev/null @@ -1,1304 +0,0 @@ -/* zip.c -- IO on .zip files using zlib - Version 1.01e, February 12th, 2005 - - 27 Dec 2004 Rolf Kalbermatter - Modification to zipOpen2 to support globalComment retrieval. - - Copyright (C) 1998-2005 Gilles Vollant - - Read zip.h for more info -*/ - - -#include -#include -#include -#include -#include "../../zlib/zlib.h" -#include "zip.h" - -#ifdef STDC -# include -# include -# include -#endif -#ifdef NO_ERRNO_H - extern int errno; -#else -# include -#endif - -#define NOCRYPT - -#ifndef local -# define local static -#endif -/* compile with -Dlocal if your debugger can't find static symbols */ - -#ifndef VERSIONMADEBY -# define VERSIONMADEBY (0x0) /* platform depedent */ -#endif - -#ifndef Z_BUFSIZE -#define Z_BUFSIZE (16384) -#endif - -#ifndef Z_MAXFILENAMEINZIP -#define Z_MAXFILENAMEINZIP (256) -#endif - -#ifndef ALLOC -# define ALLOC(size) (malloc(size)) -#endif -#ifndef TRYFREE -# define TRYFREE(p) {if (p) free(p);} -#endif - -/* -#define SIZECENTRALDIRITEM (0x2e) -#define SIZEZIPLOCALHEADER (0x1e) -*/ - -/* I've found an old Unix (a SunOS 4.1.3_U1) without all SEEK_* defined.... */ - -#ifndef SEEK_CUR -#define SEEK_CUR 1 -#endif - -#ifndef SEEK_END -#define SEEK_END 2 -#endif - -#ifndef SEEK_SET -#define SEEK_SET 0 -#endif - -#ifndef DEF_MEM_LEVEL -#if MAX_MEM_LEVEL >= 8 -# define DEF_MEM_LEVEL 8 -#else -# define DEF_MEM_LEVEL MAX_MEM_LEVEL -#endif -#endif -const char zip_copyright[] = - " zip 1.01 Copyright 1998-2004 Gilles Vollant - http://www.winimage.com/zLibDll"; - - -#define SIZEDATA_INDATABLOCK (4096-(4*4)) - -#define LOCALHEADERMAGIC (0x04034b50) -#define CENTRALHEADERMAGIC (0x02014b50) -#define ENDHEADERMAGIC (0x06054b50) - -#define FLAG_LOCALHEADER_OFFSET (0x06) -#define CRC_LOCALHEADER_OFFSET (0x0e) - -#define SIZECENTRALHEADER (0x2e) /* 46 */ - -typedef struct linkedlist_datablock_internal_s -{ - struct linkedlist_datablock_internal_s* next_datablock; - uLong avail_in_this_block; - uLong filled_in_this_block; - uLong unused; /* for future use and alignement */ - unsigned char data[SIZEDATA_INDATABLOCK]; -} linkedlist_datablock_internal; - -typedef struct linkedlist_data_s -{ - linkedlist_datablock_internal* first_block; - linkedlist_datablock_internal* last_block; -} linkedlist_data; - - -typedef struct -{ - z_stream stream; /* zLib stream structure for inflate */ - int stream_initialised; /* 1 is stream is initialised */ - uInt pos_in_buffered_data; /* last written byte in buffered_data */ - - uLong pos_local_header; /* offset of the local header of the file - currenty writing */ - char* central_header; /* central header data for the current file */ - uLong size_centralheader; /* size of the central header for cur file */ - uLong flag; /* flag of the file currently writing */ - - int method; /* compression method of file currenty wr.*/ - int raw; /* 1 for directly writing raw data */ - Byte buffered_data[Z_BUFSIZE];/* buffer contain compressed data to be writ*/ - uLong dosDate; - uLong crc32; - int encrypt; -} curfile_info; - -typedef struct -{ - zlib_filefunc_def z_filefunc; - voidpf filestream; /* io structore of the zipfile */ - linkedlist_data central_dir;/* datablock with central dir in construction*/ - int in_opened_file_inzip; /* 1 if a file in the zip is currently writ.*/ - curfile_info ci; /* info on the file curretly writing */ - - uLong begin_pos; /* position of the beginning of the zipfile */ - uLong add_position_when_writting_offset; - uLong number_entry; -#ifndef NO_ADDFILEINEXISTINGZIP - char *globalcomment; -#endif -} zip_internal; - - - -local linkedlist_datablock_internal* allocate_new_datablock() -{ - linkedlist_datablock_internal* ldi; - ldi = (linkedlist_datablock_internal*) - ALLOC(sizeof(linkedlist_datablock_internal)); - if (ldi!=NULL) - { - ldi->next_datablock = NULL ; - ldi->filled_in_this_block = 0 ; - ldi->avail_in_this_block = SIZEDATA_INDATABLOCK ; - } - return ldi; -} - -local void free_datablock(ldi) - linkedlist_datablock_internal* ldi; -{ - while (ldi!=NULL) - { - linkedlist_datablock_internal* ldinext = ldi->next_datablock; - TRYFREE(ldi); - ldi = ldinext; - } -} - -local void init_linkedlist(ll) - linkedlist_data* ll; -{ - ll->first_block = ll->last_block = NULL; -} -/* -local void free_linkedlist(ll) - linkedlist_data* ll; -{ - free_datablock(ll->first_block); - ll->first_block = ll->last_block = NULL; -} -*/ - -local int add_data_in_datablock(ll,buf,len) - linkedlist_data* ll; - const void* buf; - uLong len; -{ - linkedlist_datablock_internal* ldi; - const unsigned char* from_copy; - - if (ll==NULL) - return ZIP_INTERNALERROR; - - if (ll->last_block == NULL) - { - ll->first_block = ll->last_block = allocate_new_datablock(); - if (ll->first_block == NULL) - return ZIP_INTERNALERROR; - } - - ldi = ll->last_block; - from_copy = (unsigned char*)buf; - - while (len>0) - { - uInt copy_this; - uInt i; - unsigned char* to_copy; - - if (ldi->avail_in_this_block==0) - { - ldi->next_datablock = allocate_new_datablock(); - if (ldi->next_datablock == NULL) - return ZIP_INTERNALERROR; - ldi = ldi->next_datablock ; - ll->last_block = ldi; - } - - if (ldi->avail_in_this_block < len) - copy_this = (uInt)ldi->avail_in_this_block; - else - copy_this = (uInt)len; - - to_copy = &(ldi->data[ldi->filled_in_this_block]); - - for (i=0;ifilled_in_this_block += copy_this; - ldi->avail_in_this_block -= copy_this; - from_copy += copy_this ; - len -= copy_this; - } - return ZIP_OK; -} - - - -/****************************************************************************/ - -#ifndef NO_ADDFILEINEXISTINGZIP -/* =========================================================================== - Inputs a long in LSB order to the given file - nbByte == 1, 2 or 4 (byte, short or long) -*/ - -local int ziplocal_putValue OF((const zlib_filefunc_def* pzlib_filefunc_def, - voidpf filestream, uLong x, int nbByte)); -local int ziplocal_putValue (pzlib_filefunc_def, filestream, x, nbByte) - const zlib_filefunc_def* pzlib_filefunc_def; - voidpf filestream; - uLong x; - int nbByte; -{ - unsigned char buf[4]; - int n; - for (n = 0; n < nbByte; n++) - { - buf[n] = (unsigned char)(x & 0xff); - x >>= 8; - } - if (x != 0) - { /* data overflow - hack for ZIP64 (X Roche) */ - for (n = 0; n < nbByte; n++) - { - buf[n] = 0xff; - } - } - - if (ZWRITE(*pzlib_filefunc_def,filestream,buf,nbByte)!=(uLong)nbByte) - return ZIP_ERRNO; - else - return ZIP_OK; -} - -local void ziplocal_putValue_inmemory OF((void* dest, uLong x, int nbByte)); -local void ziplocal_putValue_inmemory (dest, x, nbByte) - void* dest; - uLong x; - int nbByte; -{ - unsigned char* buf=(unsigned char*)dest; - int n; - for (n = 0; n < nbByte; n++) { - buf[n] = (unsigned char)(x & 0xff); - x >>= 8; - } - - if (x != 0) - { /* data overflow - hack for ZIP64 */ - for (n = 0; n < nbByte; n++) - { - buf[n] = 0xff; - } - } -} - -/****************************************************************************/ - - -local uLong ziplocal_TmzDateToDosDate(ptm,dosDate) - const tm_zip* ptm; - uLong dosDate; -{ - uLong year = (uLong)ptm->tm_year; - if (year>1980) - year-=1980; - else if (year>80) - year-=80; - return - (uLong) (((ptm->tm_mday) + (32 * (ptm->tm_mon+1)) + (512 * year)) << 16) | - ((ptm->tm_sec/2) + (32* ptm->tm_min) + (2048 * (uLong)ptm->tm_hour)); -} - - -/****************************************************************************/ - -local int ziplocal_getByte OF(( - const zlib_filefunc_def* pzlib_filefunc_def, - voidpf filestream, - int *pi)); - -local int ziplocal_getByte(pzlib_filefunc_def,filestream,pi) - const zlib_filefunc_def* pzlib_filefunc_def; - voidpf filestream; - int *pi; -{ - unsigned char c; - int err = (int)ZREAD(*pzlib_filefunc_def,filestream,&c,1); - if (err==1) - { - *pi = (int)c; - return ZIP_OK; - } - else - { - if (ZERROR(*pzlib_filefunc_def,filestream)) - return ZIP_ERRNO; - else - return ZIP_EOF; - } -} - - -/* =========================================================================== - Reads a long in LSB order from the given gz_stream. Sets -*/ -local int ziplocal_getShort OF(( - const zlib_filefunc_def* pzlib_filefunc_def, - voidpf filestream, - uLong *pX)); - -local int ziplocal_getShort (pzlib_filefunc_def,filestream,pX) - const zlib_filefunc_def* pzlib_filefunc_def; - voidpf filestream; - uLong *pX; -{ - uLong x ; - int i; - int err; - - err = ziplocal_getByte(pzlib_filefunc_def,filestream,&i); - x = (uLong)i; - - if (err==ZIP_OK) - err = ziplocal_getByte(pzlib_filefunc_def,filestream,&i); - x += ((uLong)i)<<8; - - if (err==ZIP_OK) - *pX = x; - else - *pX = 0; - return err; -} - -local int ziplocal_getLong OF(( - const zlib_filefunc_def* pzlib_filefunc_def, - voidpf filestream, - uLong *pX)); - -local int ziplocal_getLong (pzlib_filefunc_def,filestream,pX) - const zlib_filefunc_def* pzlib_filefunc_def; - voidpf filestream; - uLong *pX; -{ - uLong x ; - int i; - int err; - - err = ziplocal_getByte(pzlib_filefunc_def,filestream,&i); - x = (uLong)i; - - if (err==ZIP_OK) - err = ziplocal_getByte(pzlib_filefunc_def,filestream,&i); - x += ((uLong)i)<<8; - - if (err==ZIP_OK) - err = ziplocal_getByte(pzlib_filefunc_def,filestream,&i); - x += ((uLong)i)<<16; - - if (err==ZIP_OK) - err = ziplocal_getByte(pzlib_filefunc_def,filestream,&i); - x += ((uLong)i)<<24; - - if (err==ZIP_OK) - *pX = x; - else - *pX = 0; - return err; -} - -#ifndef BUFREADCOMMENT -#define BUFREADCOMMENT (0x400) -#endif -/* - Locate the Central directory of a zipfile (at the end, just before - the global comment) -*/ -local uLong ziplocal_SearchCentralDir OF(( - const zlib_filefunc_def* pzlib_filefunc_def, - voidpf filestream)); - -local uLong ziplocal_SearchCentralDir(pzlib_filefunc_def,filestream) - const zlib_filefunc_def* pzlib_filefunc_def; - voidpf filestream; -{ - unsigned char* buf; - uLong uSizeFile; - uLong uBackRead; - uLong uMaxBack=0xffff; /* maximum size of global comment */ - uLong uPosFound=0; - - if (ZSEEK(*pzlib_filefunc_def,filestream,0,ZLIB_FILEFUNC_SEEK_END) != 0) - return 0; - - - uSizeFile = ZTELL(*pzlib_filefunc_def,filestream); - - if (uMaxBack>uSizeFile) - uMaxBack = uSizeFile; - - buf = (unsigned char*)ALLOC(BUFREADCOMMENT+4); - if (buf==NULL) - return 0; - - uBackRead = 4; - while (uBackReaduMaxBack) - uBackRead = uMaxBack; - else - uBackRead+=BUFREADCOMMENT; - uReadPos = uSizeFile-uBackRead ; - - uReadSize = ((BUFREADCOMMENT+4) < (uSizeFile-uReadPos)) ? - (BUFREADCOMMENT+4) : (uSizeFile-uReadPos); - if (ZSEEK(*pzlib_filefunc_def,filestream,uReadPos,ZLIB_FILEFUNC_SEEK_SET)!=0) - break; - - if (ZREAD(*pzlib_filefunc_def,filestream,buf,uReadSize)!=uReadSize) - break; - - for (i=(int)uReadSize-3; (i--)>0;) - if (((*(buf+i))==0x50) && ((*(buf+i+1))==0x4b) && - ((*(buf+i+2))==0x05) && ((*(buf+i+3))==0x06)) - { - uPosFound = uReadPos+i; - break; - } - - if (uPosFound!=0) - break; - } - TRYFREE(buf); - return uPosFound; -} -#endif /* !NO_ADDFILEINEXISTINGZIP*/ - -/************************************************************/ -extern zipFile ZEXPORT zipOpen2 (pathname, append, globalcomment, pzlib_filefunc_def) - const char *pathname; - int append; - zipcharpc* globalcomment; - zlib_filefunc_def* pzlib_filefunc_def; -{ - zip_internal ziinit; - zip_internal* zi; - int err=ZIP_OK; - - - if (pzlib_filefunc_def==NULL) - fill_fopen_filefunc(&ziinit.z_filefunc); - else - ziinit.z_filefunc = *pzlib_filefunc_def; - - ziinit.filestream = (*(ziinit.z_filefunc.zopen_file)) - (ziinit.z_filefunc.opaque, - pathname, - (append == APPEND_STATUS_CREATE) ? - (ZLIB_FILEFUNC_MODE_READ | ZLIB_FILEFUNC_MODE_WRITE | ZLIB_FILEFUNC_MODE_CREATE) : - (ZLIB_FILEFUNC_MODE_READ | ZLIB_FILEFUNC_MODE_WRITE | ZLIB_FILEFUNC_MODE_EXISTING)); - - if (ziinit.filestream == NULL) - return NULL; - ziinit.begin_pos = ZTELL(ziinit.z_filefunc,ziinit.filestream); - ziinit.in_opened_file_inzip = 0; - ziinit.ci.stream_initialised = 0; - ziinit.number_entry = 0; - ziinit.add_position_when_writting_offset = 0; - init_linkedlist(&(ziinit.central_dir)); - - - zi = (zip_internal*)ALLOC(sizeof(zip_internal)); - if (zi==NULL) - { - ZCLOSE(ziinit.z_filefunc,ziinit.filestream); - return NULL; - } - - /* now we add file in a zipfile */ -# ifndef NO_ADDFILEINEXISTINGZIP - ziinit.globalcomment = NULL; - if (append == APPEND_STATUS_ADDINZIP) - { - uLong byte_before_the_zipfile;/* byte before the zipfile, (>0 for sfx)*/ - - uLong size_central_dir; /* size of the central directory */ - uLong offset_central_dir; /* offset of start of central directory */ - uLong central_pos,uL; - - uLong number_disk; /* number of the current dist, used for - spaning ZIP, unsupported, always 0*/ - uLong number_disk_with_CD; /* number the the disk with central dir, used - for spaning ZIP, unsupported, always 0*/ - uLong number_entry; - uLong number_entry_CD; /* total number of entries in - the central dir - (same than number_entry on nospan) */ - uLong size_comment; - - central_pos = ziplocal_SearchCentralDir(&ziinit.z_filefunc,ziinit.filestream); - if (central_pos==0) - err=ZIP_ERRNO; - - if (ZSEEK(ziinit.z_filefunc, ziinit.filestream, - central_pos,ZLIB_FILEFUNC_SEEK_SET)!=0) - err=ZIP_ERRNO; - - /* the signature, already checked */ - if (ziplocal_getLong(&ziinit.z_filefunc, ziinit.filestream,&uL)!=ZIP_OK) - err=ZIP_ERRNO; - - /* number of this disk */ - if (ziplocal_getShort(&ziinit.z_filefunc, ziinit.filestream,&number_disk)!=ZIP_OK) - err=ZIP_ERRNO; - - /* number of the disk with the start of the central directory */ - if (ziplocal_getShort(&ziinit.z_filefunc, ziinit.filestream,&number_disk_with_CD)!=ZIP_OK) - err=ZIP_ERRNO; - - /* total number of entries in the central dir on this disk */ - if (ziplocal_getShort(&ziinit.z_filefunc, ziinit.filestream,&number_entry)!=ZIP_OK) - err=ZIP_ERRNO; - - /* total number of entries in the central dir */ - if (ziplocal_getShort(&ziinit.z_filefunc, ziinit.filestream,&number_entry_CD)!=ZIP_OK) - err=ZIP_ERRNO; - - if ((number_entry_CD!=number_entry) || - (number_disk_with_CD!=0) || - (number_disk!=0)) - err=ZIP_BADZIPFILE; - - /* size of the central directory */ - if (ziplocal_getLong(&ziinit.z_filefunc, ziinit.filestream,&size_central_dir)!=ZIP_OK) - err=ZIP_ERRNO; - - /* offset of start of central directory with respect to the - starting disk number */ - if (ziplocal_getLong(&ziinit.z_filefunc, ziinit.filestream,&offset_central_dir)!=ZIP_OK) - err=ZIP_ERRNO; - - /* zipfile global comment length */ - if (ziplocal_getShort(&ziinit.z_filefunc, ziinit.filestream,&size_comment)!=ZIP_OK) - err=ZIP_ERRNO; - - if ((central_pos0) - { - ziinit.globalcomment = ALLOC(size_comment+1); - if (ziinit.globalcomment) - { - size_comment = ZREAD(ziinit.z_filefunc, ziinit.filestream,ziinit.globalcomment,size_comment); - ziinit.globalcomment[size_comment]=0; - } - } - - byte_before_the_zipfile = central_pos - - (offset_central_dir+size_central_dir); - ziinit.add_position_when_writting_offset = byte_before_the_zipfile; - - { - uLong size_central_dir_to_read = size_central_dir; - size_t buf_size = SIZEDATA_INDATABLOCK; - void* buf_read = (void*)ALLOC(buf_size); - if (ZSEEK(ziinit.z_filefunc, ziinit.filestream, - offset_central_dir + byte_before_the_zipfile, - ZLIB_FILEFUNC_SEEK_SET) != 0) - err=ZIP_ERRNO; - - while ((size_central_dir_to_read>0) && (err==ZIP_OK)) - { - uLong read_this = SIZEDATA_INDATABLOCK; - if (read_this > size_central_dir_to_read) - read_this = size_central_dir_to_read; - if (ZREAD(ziinit.z_filefunc, ziinit.filestream,buf_read,read_this) != read_this) - err=ZIP_ERRNO; - - if (err==ZIP_OK) - err = add_data_in_datablock(&ziinit.central_dir,buf_read, - (uLong)read_this); - size_central_dir_to_read-=read_this; - } - TRYFREE(buf_read); - } - ziinit.begin_pos = byte_before_the_zipfile; - ziinit.number_entry = number_entry_CD; - - if (ZSEEK(ziinit.z_filefunc, ziinit.filestream, - offset_central_dir+byte_before_the_zipfile,ZLIB_FILEFUNC_SEEK_SET)!=0) - err=ZIP_ERRNO; - } - - if (globalcomment) - { - *globalcomment = ziinit.globalcomment; - } -# endif /* !NO_ADDFILEINEXISTINGZIP*/ - - if (err != ZIP_OK) - { -# ifndef NO_ADDFILEINEXISTINGZIP - TRYFREE(ziinit.globalcomment); -# endif /* !NO_ADDFILEINEXISTINGZIP*/ - TRYFREE(zi); - return NULL; - } - else - { - *zi = ziinit; - return (zipFile)zi; - } -} - -extern zipFile ZEXPORT zipOpen (pathname, append) - const char *pathname; - int append; -{ - return zipOpen2(pathname,append,NULL,NULL); -} - -extern int ZEXPORT zipOpenNewFileInZip3 (file, filename, zipfi, - extrafield_local, size_extrafield_local, - extrafield_global, size_extrafield_global, - comment, method, level, raw, - windowBits, memLevel, strategy, - password, crcForCrypting) - zipFile file; - const char* filename; - const zip_fileinfo* zipfi; - const void* extrafield_local; - uInt size_extrafield_local; - const void* extrafield_global; - uInt size_extrafield_global; - const char* comment; - int method; - int level; - int raw; - int windowBits; - int memLevel; - int strategy; - const char* password; - uLong crcForCrypting; -{ - zip_internal* zi; - uInt size_filename; - uInt size_comment; - uInt i; - int err = ZIP_OK; - - if (file == NULL) - return ZIP_PARAMERROR; - if ((method!=0) && (method!=Z_DEFLATED)) - return ZIP_PARAMERROR; - - zi = (zip_internal*)file; - - if (zi->in_opened_file_inzip == 1) - { - err = zipCloseFileInZip (file); - if (err != ZIP_OK) - return err; - } - - - if (filename==NULL) - filename="-"; - - if (comment==NULL) - size_comment = 0; - else - size_comment = (uInt)strlen(comment); - - size_filename = (uInt)strlen(filename); - - if (zipfi == NULL) - zi->ci.dosDate = 0; - else - { - if (zipfi->dosDate != 0) - zi->ci.dosDate = zipfi->dosDate; - else zi->ci.dosDate = ziplocal_TmzDateToDosDate(&zipfi->tmz_date,zipfi->dosDate); - } - - zi->ci.flag = 0; - if ((level==8) || (level==9)) - zi->ci.flag |= 2; - if ((level==2)) - zi->ci.flag |= 4; - if ((level==1)) - zi->ci.flag |= 6; - if (password != NULL) - zi->ci.flag |= 1; - - zi->ci.crc32 = 0; - zi->ci.method = method; - zi->ci.encrypt = 0; - zi->ci.stream_initialised = 0; - zi->ci.pos_in_buffered_data = 0; - zi->ci.raw = raw; - zi->ci.pos_local_header = ZTELL(zi->z_filefunc,zi->filestream) ; - zi->ci.size_centralheader = SIZECENTRALHEADER + size_filename + - size_extrafield_global + size_comment; - zi->ci.central_header = (char*)ALLOC((uInt)zi->ci.size_centralheader); - - ziplocal_putValue_inmemory(zi->ci.central_header,(uLong)CENTRALHEADERMAGIC,4); - /* version info */ - ziplocal_putValue_inmemory(zi->ci.central_header+4,(uLong)VERSIONMADEBY,2); - ziplocal_putValue_inmemory(zi->ci.central_header+6,(uLong)20,2); - ziplocal_putValue_inmemory(zi->ci.central_header+8,(uLong)zi->ci.flag,2); - ziplocal_putValue_inmemory(zi->ci.central_header+10,(uLong)zi->ci.method,2); - ziplocal_putValue_inmemory(zi->ci.central_header+12,(uLong)zi->ci.dosDate,4); - ziplocal_putValue_inmemory(zi->ci.central_header+16,(uLong)0,4); /*crc*/ - ziplocal_putValue_inmemory(zi->ci.central_header+20,(uLong)0,4); /*compr size*/ - ziplocal_putValue_inmemory(zi->ci.central_header+24,(uLong)0,4); /*uncompr size*/ - ziplocal_putValue_inmemory(zi->ci.central_header+28,(uLong)size_filename,2); - ziplocal_putValue_inmemory(zi->ci.central_header+30,(uLong)size_extrafield_global,2); - ziplocal_putValue_inmemory(zi->ci.central_header+32,(uLong)size_comment,2); - ziplocal_putValue_inmemory(zi->ci.central_header+34,(uLong)0,2); /*disk nm start*/ - - if (zipfi==NULL) - ziplocal_putValue_inmemory(zi->ci.central_header+36,(uLong)0,2); - else - ziplocal_putValue_inmemory(zi->ci.central_header+36,(uLong)zipfi->internal_fa,2); - - if (zipfi==NULL) - ziplocal_putValue_inmemory(zi->ci.central_header+38,(uLong)0,4); - else - ziplocal_putValue_inmemory(zi->ci.central_header+38,(uLong)zipfi->external_fa,4); - - ziplocal_putValue_inmemory(zi->ci.central_header+42,(uLong)zi->ci.pos_local_header- zi->add_position_when_writting_offset,4); - - for (i=0;ici.central_header+SIZECENTRALHEADER+i) = *(filename+i); - - for (i=0;ici.central_header+SIZECENTRALHEADER+size_filename+i) = - *(((const char*)extrafield_global)+i); - - for (i=0;ici.central_header+SIZECENTRALHEADER+size_filename+ - size_extrafield_global+i) = *(comment+i); - if (zi->ci.central_header == NULL) - return ZIP_INTERNALERROR; - - /* write the local header */ - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)LOCALHEADERMAGIC,4); - - if (err==ZIP_OK) - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)20,2);/* version needed to extract */ - if (err==ZIP_OK) - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)zi->ci.flag,2); - - if (err==ZIP_OK) - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)zi->ci.method,2); - - if (err==ZIP_OK) - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)zi->ci.dosDate,4); - - if (err==ZIP_OK) - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)0,4); /* crc 32, unknown */ - if (err==ZIP_OK) - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)0,4); /* compressed size, unknown */ - if (err==ZIP_OK) - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)0,4); /* uncompressed size, unknown */ - - if (err==ZIP_OK) - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)size_filename,2); - - if (err==ZIP_OK) - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)size_extrafield_local,2); - - if ((err==ZIP_OK) && (size_filename>0)) - if (ZWRITE(zi->z_filefunc,zi->filestream,filename,size_filename)!=size_filename) - err = ZIP_ERRNO; - - if ((err==ZIP_OK) && (size_extrafield_local>0)) - if (ZWRITE(zi->z_filefunc,zi->filestream,extrafield_local,size_extrafield_local) - !=size_extrafield_local) - err = ZIP_ERRNO; - - zi->ci.stream.avail_in = (uInt)0; - zi->ci.stream.avail_out = (uInt)Z_BUFSIZE; - zi->ci.stream.next_out = zi->ci.buffered_data; - zi->ci.stream.total_in = 0; - zi->ci.stream.total_out = 0; - - if ((err==ZIP_OK) && (zi->ci.method == Z_DEFLATED) && (!zi->ci.raw)) - { - zi->ci.stream.zalloc = (alloc_func)0; - zi->ci.stream.zfree = (free_func)0; - zi->ci.stream.opaque = (voidpf)0; - - if (windowBits>0) - windowBits = -windowBits; - - err = deflateInit2(&zi->ci.stream, level, - Z_DEFLATED, windowBits, memLevel, strategy); - - if (err==Z_OK) - zi->ci.stream_initialised = 1; - } - - if (err==Z_OK) - zi->in_opened_file_inzip = 1; - return err; -} - -extern int ZEXPORT zipOpenNewFileInZip2(file, filename, zipfi, - extrafield_local, size_extrafield_local, - extrafield_global, size_extrafield_global, - comment, method, level, raw) - zipFile file; - const char* filename; - const zip_fileinfo* zipfi; - const void* extrafield_local; - uInt size_extrafield_local; - const void* extrafield_global; - uInt size_extrafield_global; - const char* comment; - int method; - int level; - int raw; -{ - return zipOpenNewFileInZip3 (file, filename, zipfi, - extrafield_local, size_extrafield_local, - extrafield_global, size_extrafield_global, - comment, method, level, raw, - -MAX_WBITS, DEF_MEM_LEVEL, Z_DEFAULT_STRATEGY, - NULL, 0); -} - -extern int ZEXPORT zipOpenNewFileInZip (file, filename, zipfi, - extrafield_local, size_extrafield_local, - extrafield_global, size_extrafield_global, - comment, method, level) - zipFile file; - const char* filename; - const zip_fileinfo* zipfi; - const void* extrafield_local; - uInt size_extrafield_local; - const void* extrafield_global; - uInt size_extrafield_global; - const char* comment; - int method; - int level; -{ - return zipOpenNewFileInZip2 (file, filename, zipfi, - extrafield_local, size_extrafield_local, - extrafield_global, size_extrafield_global, - comment, method, level, 0); -} - -local int zipFlushWriteBuffer(zi) - zip_internal* zi; -{ - int err=ZIP_OK; - - if (ZWRITE(zi->z_filefunc,zi->filestream,zi->ci.buffered_data,zi->ci.pos_in_buffered_data) - !=zi->ci.pos_in_buffered_data) - err = ZIP_ERRNO; - zi->ci.pos_in_buffered_data = 0; - return err; -} - -extern int ZEXPORT zipWriteInFileInZip (file, buf, len) - zipFile file; - const void* buf; - unsigned len; -{ - zip_internal* zi; - int err=ZIP_OK; - - if (file == NULL) - return ZIP_PARAMERROR; - zi = (zip_internal*)file; - - if (zi->in_opened_file_inzip == 0) - return ZIP_PARAMERROR; - - zi->ci.stream.next_in = (void*)buf; - zi->ci.stream.avail_in = len; - zi->ci.crc32 = crc32(zi->ci.crc32,buf,len); - - while ((err==ZIP_OK) && (zi->ci.stream.avail_in>0)) - { - if (zi->ci.stream.avail_out == 0) - { - if (zipFlushWriteBuffer(zi) == ZIP_ERRNO) - err = ZIP_ERRNO; - zi->ci.stream.avail_out = (uInt)Z_BUFSIZE; - zi->ci.stream.next_out = zi->ci.buffered_data; - } - - - if(err != ZIP_OK) - break; - - if ((zi->ci.method == Z_DEFLATED) && (!zi->ci.raw)) - { - uLong uTotalOutBefore = zi->ci.stream.total_out; - err=deflate(&zi->ci.stream, Z_NO_FLUSH); - zi->ci.pos_in_buffered_data += (uInt)(zi->ci.stream.total_out - uTotalOutBefore) ; - - } - else - { - uInt copy_this,i; - if (zi->ci.stream.avail_in < zi->ci.stream.avail_out) - copy_this = zi->ci.stream.avail_in; - else - copy_this = zi->ci.stream.avail_out; - for (i=0;ici.stream.next_out)+i) = - *(((const char*)zi->ci.stream.next_in)+i); - { - zi->ci.stream.avail_in -= copy_this; - zi->ci.stream.avail_out-= copy_this; - zi->ci.stream.next_in+= copy_this; - zi->ci.stream.next_out+= copy_this; - zi->ci.stream.total_in+= copy_this; - zi->ci.stream.total_out+= copy_this; - zi->ci.pos_in_buffered_data += copy_this; - } - } - } - - return err; -} - -extern int ZEXPORT zipCloseFileInZipRaw (file, uncompressed_size, crc32) - zipFile file; - uLong uncompressed_size; - uLong crc32; -{ - zip_internal* zi; - uLong compressed_size; - int err=ZIP_OK; - - if (file == NULL) - return ZIP_PARAMERROR; - zi = (zip_internal*)file; - - if (zi->in_opened_file_inzip == 0) - return ZIP_PARAMERROR; - zi->ci.stream.avail_in = 0; - - if ((zi->ci.method == Z_DEFLATED) && (!zi->ci.raw)) - while (err==ZIP_OK) - { - uLong uTotalOutBefore; - if (zi->ci.stream.avail_out == 0) - { - if (zipFlushWriteBuffer(zi) == ZIP_ERRNO) - err = ZIP_ERRNO; - zi->ci.stream.avail_out = (uInt)Z_BUFSIZE; - zi->ci.stream.next_out = zi->ci.buffered_data; - } - uTotalOutBefore = zi->ci.stream.total_out; - err=deflate(&zi->ci.stream, Z_FINISH); - zi->ci.pos_in_buffered_data += (uInt)(zi->ci.stream.total_out - uTotalOutBefore) ; - } - - if (err==Z_STREAM_END) - err=ZIP_OK; /* this is normal */ - - if ((zi->ci.pos_in_buffered_data>0) && (err==ZIP_OK)) - if (zipFlushWriteBuffer(zi)==ZIP_ERRNO) - err = ZIP_ERRNO; - - if ((zi->ci.method == Z_DEFLATED) && (!zi->ci.raw)) - { - err=deflateEnd(&zi->ci.stream); - zi->ci.stream_initialised = 0; - } - - if (!zi->ci.raw) - { - crc32 = (uLong)zi->ci.crc32; - uncompressed_size = (uLong)zi->ci.stream.total_in; - } - compressed_size = (uLong)zi->ci.stream.total_out; - - ziplocal_putValue_inmemory(zi->ci.central_header+16,crc32,4); /*crc*/ - ziplocal_putValue_inmemory(zi->ci.central_header+20, - compressed_size,4); /*compr size*/ - if (zi->ci.stream.data_type == Z_ASCII) - ziplocal_putValue_inmemory(zi->ci.central_header+36,(uLong)Z_ASCII,2); - ziplocal_putValue_inmemory(zi->ci.central_header+24, - uncompressed_size,4); /*uncompr size*/ - - if (err==ZIP_OK) - err = add_data_in_datablock(&zi->central_dir,zi->ci.central_header, - (uLong)zi->ci.size_centralheader); - free(zi->ci.central_header); - - if (err==ZIP_OK) - { - long cur_pos_inzip = ZTELL(zi->z_filefunc,zi->filestream); - if (ZSEEK(zi->z_filefunc,zi->filestream, - zi->ci.pos_local_header + 14,ZLIB_FILEFUNC_SEEK_SET)!=0) - err = ZIP_ERRNO; - - if (err==ZIP_OK) - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,crc32,4); /* crc 32, unknown */ - - if (err==ZIP_OK) /* compressed size, unknown */ - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,compressed_size,4); - - if (err==ZIP_OK) /* uncompressed size, unknown */ - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,uncompressed_size,4); - - if (ZSEEK(zi->z_filefunc,zi->filestream, - cur_pos_inzip,ZLIB_FILEFUNC_SEEK_SET)!=0) - err = ZIP_ERRNO; - } - - zi->number_entry ++; - zi->in_opened_file_inzip = 0; - - return err; -} - -extern int ZEXPORT zipCloseFileInZip (file) - zipFile file; -{ - return zipCloseFileInZipRaw (file,0,0); -} - -extern int ZEXPORT zipAddFileToZip (file, filename, zipfi, buf, len) - zipFile file; - const char *filename; - const zip_fileinfo *zipfi; - const void *buf; - unsigned len; -{ - zip_internal* zi; - uLong compressed_size; - uLong uncompressed_size; - uLong crc32; - long data_pos; - int err = ZIP_OK; - - if (file == NULL) - return ZIP_PARAMERROR; - - if (err == ZIP_OK) - err = zipOpenNewFileInZip(file, filename, zipfi, NULL, 0, NULL, 0, NULL, Z_DEFLATED, Z_BEST_COMPRESSION); - - zi = (zip_internal*)file; - - data_pos = ZTELL(zi->z_filefunc, zi->filestream); - - if (err == ZIP_OK) - err = zipWriteInFileInZip(file, buf, len); - - if (err != ZIP_OK) - return err; - - zi->ci.stream.avail_in = 0; - - while (err==ZIP_OK) - { - uLong uTotalOutBefore; - if (zi->ci.stream.avail_out == 0) - { - if (zipFlushWriteBuffer(zi) == ZIP_ERRNO) - err = ZIP_ERRNO; - zi->ci.stream.avail_out = (uInt)Z_BUFSIZE; - zi->ci.stream.next_out = zi->ci.buffered_data; - } - uTotalOutBefore = zi->ci.stream.total_out; - err=deflate(&zi->ci.stream, Z_FINISH); - zi->ci.pos_in_buffered_data += (uInt)(zi->ci.stream.total_out - uTotalOutBefore) ; - } - - if (err==Z_STREAM_END) - err=ZIP_OK; /* this is normal */ - - if ((zi->ci.pos_in_buffered_data>0) && (err==ZIP_OK)) - if (zipFlushWriteBuffer(zi)==ZIP_ERRNO) - err = ZIP_ERRNO; - - err=deflateEnd(&zi->ci.stream); - zi->ci.stream_initialised = 0; - - crc32 = (uLong)zi->ci.crc32; - uncompressed_size = (uLong)zi->ci.stream.total_in; - compressed_size = (uLong)zi->ci.stream.total_out; - - if (uncompressed_size <= compressed_size) - { - /* Compressed size was no better than uncompressed, so redo it. */ - zi->ci.stream.avail_in = (uInt)0; - zi->ci.stream.avail_out = (uInt)Z_BUFSIZE; - zi->ci.stream.next_out = zi->ci.buffered_data; - zi->ci.stream.total_in = 0; - zi->ci.stream.total_out = 0; - zi->ci.method = 0; - ziplocal_putValue_inmemory(zi->ci.central_header+10,0,2); /* method */ - - if (ZSEEK(zi->z_filefunc,zi->filestream, - data_pos,ZLIB_FILEFUNC_SEEK_SET)!=0) - err = ZIP_ERRNO; - - err = zipWriteInFileInZip(file, buf, len); - if ((zi->ci.pos_in_buffered_data>0) && (err==ZIP_OK)) - if (zipFlushWriteBuffer(zi)==ZIP_ERRNO) - err = ZIP_ERRNO; - compressed_size = (uLong)zi->ci.stream.total_out; - } - - ziplocal_putValue_inmemory(zi->ci.central_header+16,crc32,4); /*crc*/ - ziplocal_putValue_inmemory(zi->ci.central_header+20, - compressed_size,4); /*compr size*/ - if (zi->ci.stream.data_type == Z_ASCII) - ziplocal_putValue_inmemory(zi->ci.central_header+36,(uLong)Z_ASCII,2); - ziplocal_putValue_inmemory(zi->ci.central_header+24, - uncompressed_size,4); /*uncompr size*/ - - if (err==ZIP_OK) - err = add_data_in_datablock(&zi->central_dir,zi->ci.central_header, - (uLong)zi->ci.size_centralheader); - free(zi->ci.central_header); - - if (err==ZIP_OK) - { - long cur_pos_inzip = ZTELL(zi->z_filefunc,zi->filestream); - if (ZSEEK(zi->z_filefunc,zi->filestream, - zi->ci.pos_local_header + 8,ZLIB_FILEFUNC_SEEK_SET)!=0) - err = ZIP_ERRNO; - - if (err==ZIP_OK) - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)zi->ci.method,2); - - if (err==ZIP_OK) - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)zi->ci.dosDate,4); - - if (err==ZIP_OK) - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,crc32,4); /* crc 32, unknown */ - - if (err==ZIP_OK) /* compressed size, unknown */ - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,compressed_size,4); - - if (err==ZIP_OK) /* uncompressed size, unknown */ - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,uncompressed_size,4); - - if (ZSEEK(zi->z_filefunc,zi->filestream, - cur_pos_inzip,ZLIB_FILEFUNC_SEEK_SET)!=0) - err = ZIP_ERRNO; - } - - zi->number_entry ++; - zi->in_opened_file_inzip = 0; - - return err; -} - -extern int ZEXPORT zipClose (file, global_comment) - zipFile file; - const char* global_comment; -{ - zip_internal* zi; - int err = 0; - uLong size_centraldir = 0; - uLong centraldir_pos_inzip; - uInt size_global_comment; - if (file == NULL) - return ZIP_PARAMERROR; - zi = (zip_internal*)file; - - if (zi->in_opened_file_inzip == 1) - { - err = zipCloseFileInZip (file); - } - -#ifndef NO_ADDFILEINEXISTINGZIP - if (global_comment==NULL) - global_comment = zi->globalcomment; -#endif - if (global_comment==NULL) - size_global_comment = 0; - else - size_global_comment = (uInt)strlen(global_comment); - - centraldir_pos_inzip = ZTELL(zi->z_filefunc,zi->filestream); - if (err==ZIP_OK) - { - linkedlist_datablock_internal* ldi = zi->central_dir.first_block ; - while (ldi!=NULL) - { - if ((err==ZIP_OK) && (ldi->filled_in_this_block>0)) - if (ZWRITE(zi->z_filefunc,zi->filestream, - ldi->data,ldi->filled_in_this_block) - !=ldi->filled_in_this_block ) - err = ZIP_ERRNO; - - size_centraldir += ldi->filled_in_this_block; - ldi = ldi->next_datablock; - } - } - free_datablock(zi->central_dir.first_block); - - if (err==ZIP_OK) /* Magic End */ - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)ENDHEADERMAGIC,4); - - if (err==ZIP_OK) /* number of this disk */ - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)0,2); - - if (err==ZIP_OK) /* number of the disk with the start of the central directory */ - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)0,2); - - if (err==ZIP_OK) /* total number of entries in the central dir on this disk */ - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)zi->number_entry,2); - - if (err==ZIP_OK) /* total number of entries in the central dir */ - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)zi->number_entry,2); - - if (err==ZIP_OK) /* size of the central directory */ - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)size_centraldir,4); - - if (err==ZIP_OK) /* offset of start of central directory with respect to the - starting disk number */ - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream, - (uLong)(centraldir_pos_inzip - zi->add_position_when_writting_offset),4); - - if (err==ZIP_OK) /* zipfile comment length */ - err = ziplocal_putValue(&zi->z_filefunc,zi->filestream,(uLong)size_global_comment,2); - - if ((err==ZIP_OK) && (size_global_comment>0)) - if (ZWRITE(zi->z_filefunc,zi->filestream, - global_comment,size_global_comment) != size_global_comment) - err = ZIP_ERRNO; - - if (ZCLOSE(zi->z_filefunc,zi->filestream) != 0) - if (err == ZIP_OK) - err = ZIP_ERRNO; - -#ifndef NO_ADDFILEINEXISTINGZIP - TRYFREE(zi->globalcomment); -#endif - TRYFREE(zi); - - return err; -} diff --git a/tools/zipdir/zip.h b/tools/zipdir/zip.h deleted file mode 100644 index 224c800a2a..0000000000 --- a/tools/zipdir/zip.h +++ /dev/null @@ -1,244 +0,0 @@ -/* zip.h -- IO for compress .zip files using zlib - Version 1.01e, February 12th, 2005 - - Copyright (C) 1998-2005 Gilles Vollant - - This unzip package allow creates .ZIP file, compatible with PKZip 2.04g - WinZip, InfoZip tools and compatible. - Multi volume ZipFile (span) are not supported. - Encryption compatible with pkzip 2.04g only supported - Old compressions used by old PKZip 1.x are not supported - - For uncompress .zip file, look at unzip.h - - - I WAIT FEEDBACK at mail info@winimage.com - Visit also http://www.winimage.com/zLibDll/unzip.html for evolution - - Condition of use and distribution are the same than zlib : - - This software is provided 'as-is', without any express or implied - warranty. In no event will the authors be held liable for any damages - arising from the use of this software. - - Permission is granted to anyone to use this software for any purpose, - including commercial applications, and to alter it and redistribute it - freely, subject to the following restrictions: - - 1. The origin of this software must not be misrepresented; you must not - claim that you wrote the original software. If you use this software - in a product, an acknowledgment in the product documentation would be - appreciated but is not required. - 2. Altered source versions must be plainly marked as such, and must not be - misrepresented as being the original software. - 3. This notice may not be removed or altered from any source distribution. - - -*/ - -/* for more info about .ZIP format, see - http://www.info-zip.org/pub/infozip/doc/appnote-981119-iz.zip - http://www.info-zip.org/pub/infozip/doc/ - PkWare has also a specification at : - ftp://ftp.pkware.com/probdesc.zip -*/ - -#ifndef _zip_H -#define _zip_H - -#ifdef __cplusplus -extern "C" { -#endif - -#ifndef _ZLIB_H -#include "../../zlib/zlib.h" -#endif - -#ifndef _ZLIBIOAPI_H -#include "ioapi.h" -#endif - -#if defined(STRICTZIP) || defined(STRICTZIPUNZIP) -/* like the STRICT of WIN32, we define a pointer that cannot be converted - from (void*) without cast */ -typedef struct TagzipFile__ { int unused; } zipFile__; -typedef zipFile__ *zipFile; -#else -typedef voidp zipFile; -#endif - -#define ZIP_OK (0) -#define ZIP_EOF (0) -#define ZIP_ERRNO (Z_ERRNO) -#define ZIP_PARAMERROR (-102) -#define ZIP_BADZIPFILE (-103) -#define ZIP_INTERNALERROR (-104) - -#ifndef DEF_MEM_LEVEL -# if MAX_MEM_LEVEL >= 8 -# define DEF_MEM_LEVEL 8 -# else -# define DEF_MEM_LEVEL MAX_MEM_LEVEL -# endif -#endif -/* default memLevel */ - -/* tm_zip contain date/time info */ -typedef struct tm_zip_s -{ - uInt tm_sec; /* seconds after the minute - [0,59] */ - uInt tm_min; /* minutes after the hour - [0,59] */ - uInt tm_hour; /* hours since midnight - [0,23] */ - uInt tm_mday; /* day of the month - [1,31] */ - uInt tm_mon; /* months since January - [0,11] */ - uInt tm_year; /* years - [1980..2044] */ -} tm_zip; - -typedef struct -{ - tm_zip tmz_date; /* date in understandable format */ - uLong dosDate; /* if dos_date == 0, tmu_date is used */ -/* uLong flag; */ /* general purpose bit flag 2 bytes */ - - uLong internal_fa; /* internal file attributes 2 bytes */ - uLong external_fa; /* external file attributes 4 bytes */ -} zip_fileinfo; - -typedef const char* zipcharpc; - - -#define APPEND_STATUS_CREATE (0) -#define APPEND_STATUS_CREATEAFTER (1) -#define APPEND_STATUS_ADDINZIP (2) - -extern zipFile ZEXPORT zipOpen OF((const char *pathname, int append)); -/* - Create a zipfile. - pathname contain on Windows XP a filename like "c:\\zlib\\zlib113.zip" or on - an Unix computer "zlib/zlib113.zip". - if the file pathname exist and append==APPEND_STATUS_CREATEAFTER, the zip - will be created at the end of the file. - (useful if the file contain a self extractor code) - if the file pathname exist and append==APPEND_STATUS_ADDINZIP, we will - add files in existing zip (be sure you don't add file that doesn't exist) - If the zipfile cannot be opened, the return value is NULL. - Else, the return value is a zipFile Handle, usable with other function - of this zip package. -*/ - -/* Note : there is no delete function into a zipfile. - If you want delete file into a zipfile, you must open a zipfile, and create another - Of couse, you can use RAW reading and writing to copy the file you did not want delte -*/ - -extern zipFile ZEXPORT zipOpen2 OF((const char *pathname, - int append, - zipcharpc* globalcomment, - zlib_filefunc_def* pzlib_filefunc_def)); - -extern int ZEXPORT zipOpenNewFileInZip OF((zipFile file, - const char* filename, - const zip_fileinfo* zipfi, - const void* extrafield_local, - uInt size_extrafield_local, - const void* extrafield_global, - uInt size_extrafield_global, - const char* comment, - int method, - int level)); -/* - Open a file in the ZIP for writing. - filename : the filename in zip (if NULL, '-' without quote will be used - *zipfi contain supplemental information - if extrafield_local!=NULL and size_extrafield_local>0, extrafield_local - contains the extrafield data the the local header - if extrafield_global!=NULL and size_extrafield_global>0, extrafield_global - contains the extrafield data the the local header - if comment != NULL, comment contain the comment string - method contain the compression method (0 for store, Z_DEFLATED for deflate) - level contain the level of compression (can be Z_DEFAULT_COMPRESSION) -*/ - - -extern int ZEXPORT zipOpenNewFileInZip2 OF((zipFile file, - const char* filename, - const zip_fileinfo* zipfi, - const void* extrafield_local, - uInt size_extrafield_local, - const void* extrafield_global, - uInt size_extrafield_global, - const char* comment, - int method, - int level, - int raw)); - -/* - Same than zipOpenNewFileInZip, except if raw=1, we write raw file - */ - -extern int ZEXPORT zipOpenNewFileInZip3 OF((zipFile file, - const char* filename, - const zip_fileinfo* zipfi, - const void* extrafield_local, - uInt size_extrafield_local, - const void* extrafield_global, - uInt size_extrafield_global, - const char* comment, - int method, - int level, - int raw, - int windowBits, - int memLevel, - int strategy, - const char* password, - uLong crcForCtypting)); - -/* - Same than zipOpenNewFileInZip2, except - windowBits,memLevel,,strategy : see parameter strategy in deflateInit2 - password : crypting password (NULL for no crypting) - crcForCtypting : crc of file to compress (needed for crypting) - */ - - -extern int ZEXPORT zipWriteInFileInZip OF((zipFile file, - const void* buf, - unsigned len)); -/* - Write data in the zipfile -*/ - -extern int ZEXPORT zipAddFileToZip OF((zipFile file, - const char *filename, - const zip_fileinfo *zipfi, - const void *buf, - unsigned len)); -/* - [RH] Add a file to the zipfile, using deflate or stored, whichever is best -*/ - -extern int ZEXPORT zipCloseFileInZip OF((zipFile file)); -/* - Close the current file in the zipfile -*/ - -extern int ZEXPORT zipCloseFileInZipRaw OF((zipFile file, - uLong uncompressed_size, - uLong crc32)); -/* - Close the current file in the zipfile, for fiel opened with - parameter raw=1 in zipOpenNewFileInZip2 - uncompressed_size and crc32 are value for the uncompressed size -*/ - -extern int ZEXPORT zipClose OF((zipFile file, - const char* global_comment)); -/* - Close the zipfile -*/ - -#ifdef __cplusplus -} -#endif - -#endif /* _zip_H */ diff --git a/tools/zipdir/zipdir.c b/tools/zipdir/zipdir.c index 6a2adf2800..251fb50f20 100644 --- a/tools/zipdir/zipdir.c +++ b/tools/zipdir/zipdir.c @@ -1,6 +1,6 @@ /* ** zipdir.c -** Copyright (C) 2008 Randy Heit +** Copyright (C) 2008-2009 Randy Heit ** ** This program is free software; you can redistribute it and/or modify ** it under the terms of the GNU General Public License as published by @@ -18,7 +18,7 @@ ** **************************************************************************** ** -** Usage: zipdir ... +** Usage: zipdir [-dfuq] ... ** ** Given one or more directories, their contents are scanned recursively. ** If any files are newer than the zip file or the zip file does not exist, @@ -44,7 +44,13 @@ #include #include #include -#include "zip.h" +#include "zlib.h" +#include "bzlib.h" +#include "LzmaEnc.h" +#include "7zVersion.h" +#ifdef PPMD +#include "../../ppmd/PPMd.h" +#endif // MACROS ------------------------------------------------------------------ @@ -52,12 +58,48 @@ #define __cdecl #endif +#ifndef WORDS_BIGENDIAN +#define MAKE_ID(a,b,c,d) ((a)|((b)<<8)|((c)<<16)|((d)<<24)) +#define LittleShort(x) (x) +#define LittleLong(x) (x) +#else +#define MAKE_ID(a,b,c,d) ((d)|((c)<<8)|((b)<<16)|((a)<<24)) +static unsigned short LittleShort(unsigned short x) +{ + return (x>>8) | (x<<8); +} + +static unsigned int LittleLong(unsigned int x) +{ + return (x>>24) | ((x>>8) & 0xff00) | ((x<<8) & 0xff0000) | (x<<24); +} +#endif + +#define ZIP_LOCALFILE MAKE_ID('P','K',3,4) +#define ZIP_CENTRALFILE MAKE_ID('P','K',1,2) +#define ZIP_ENDOFDIR MAKE_ID('P','K',5,6) + +#define METHOD_STORED 0 +#define METHOD_DEFLATE 8 +#define METHOD_BZIP2 12 +#define METHOD_LZMA 14 +#define METHOD_PPMD 98 + +// Buffer size for central directory search +#define BUFREADCOMMENT (0x400) + // TYPES ------------------------------------------------------------------- typedef struct file_entry_s { struct file_entry_s *next; time_t time_write; + unsigned int uncompressed_size; + unsigned int compressed_size; + unsigned int crc32; + unsigned int zip_offset; + short date, time; + short method; char path[]; } file_entry_t; @@ -75,6 +117,66 @@ typedef struct file_sorted_s char *path_in_zip; } file_sorted_t; +typedef struct compressor_s +{ + int (*compress)(Byte *out, unsigned int *outlen, const Byte *in, unsigned int inlen); + int method; +} compressor_t; + +typedef unsigned int DWORD; +typedef unsigned short WORD; +typedef unsigned char BYTE; + +#pragma pack(push,1) +typedef struct +{ + DWORD Magic; // 0 + BYTE VersionToExtract[2]; // 4 + WORD Flags; // 6 + WORD Method; // 8 + WORD ModTime; // 10 + WORD ModDate; // 12 + DWORD CRC32; // 14 + DWORD CompressedSize; // 18 + DWORD UncompressedSize; // 22 + WORD NameLength; // 26 + WORD ExtraLength; // 28 +} LocalFileHeader; + +typedef struct +{ + DWORD Magic; + BYTE VersionMadeBy[2]; + BYTE VersionToExtract[2]; + WORD Flags; + WORD Method; + WORD ModTime; + WORD ModDate; + DWORD CRC32; + DWORD CompressedSize; + DWORD UncompressedSize; + WORD NameLength; + WORD ExtraLength; + WORD CommentLength; + WORD StartingDiskNumber; + WORD InternalAttributes; + DWORD ExternalAttributes; + DWORD LocalHeaderOffset; +} CentralDirectoryEntry; + +typedef struct +{ + DWORD Magic; + WORD DiskNumber; + WORD FirstDisk; + WORD NumEntries; + WORD NumEntriesOnAllDisks; + DWORD DirectorySize; + DWORD DirectoryOffset; + WORD ZipCommentLength; +} EndOfCentralDirectory; +#pragma pack(pop) + // EXTERNAL FUNCTION PROTOTYPES -------------------------------------------- // PUBLIC FUNCTION PROTOTYPES ---------------------------------------------- @@ -92,19 +194,49 @@ dir_tree_t *add_dirs(char **argv); int count_files(dir_tree_t *trees); int __cdecl sort_cmp(const void *a, const void *b); file_sorted_t *sort_files(dir_tree_t *trees, int num_files); -void write_zip(const char *zipname, dir_tree_t *trees); -int append_to_zip(zipFile zipfile, const file_sorted_t *file); +void write_zip(const char *zipname, dir_tree_t *trees, int update); +int append_to_zip(FILE *zip_file, file_sorted_t *file, FILE *ozip, BYTE *odir); +int write_central_dir(FILE *zip, file_sorted_t *file); +void time_to_dos(struct tm *time, short *dosdate, short *dostime); +int method_to_version(int method); +const char *method_name(int method); +int compress_lzma(Byte *out, unsigned int *outlen, const Byte *in, unsigned int inlen); +int compress_bzip2(Byte *out, unsigned int *outlen, const Byte *in, unsigned int inlen); +int compress_ppmd(Byte *out, unsigned int *outlen, const Byte *in, unsigned int inlen); +int compress_deflate(Byte *out, unsigned int *outlen, const Byte *in, unsigned int inlen); +BYTE *find_central_dir(FILE *fin); +CentralDirectoryEntry *find_file_in_zip(BYTE *dir, const char *path, unsigned int len, unsigned int crc, short date, short time); +int copy_zip_file(FILE *zip, file_entry_t *file, FILE *ozip, CentralDirectoryEntry *dirent); // PRIVATE FUNCTION PROTOTYPES --------------------------------------------- +static void *SzAlloc(void *p, size_t size) { p = p; return malloc(size); } +static void SzFree(void *p, void *address) { p = p; free(address); } + // EXTERNAL DATA DECLARATIONS ---------------------------------------------- // PUBLIC DATA DEFINITIONS ------------------------------------------------- +int DeflateOnly; +int UpdateCount; +int Quiet; + // PRIVATE DATA DEFINITIONS ------------------------------------------------ static int no_mem; +static ISzAlloc Alloc = { SzAlloc, SzFree }; +static compressor_t Compressors[] = +{ + { compress_lzma, METHOD_LZMA }, + { compress_bzip2, METHOD_BZIP2 }, +#ifdef PPMD + { compress_ppmd, METHOD_PPMD }, +#endif + { compress_deflate, METHOD_DEFLATE }, + { NULL, 0 } +}; + // CODE -------------------------------------------------------------------- //========================================================================== @@ -122,7 +254,11 @@ void print_usage(const char *cmdname) cmdname = rchar+1; } #endif - fprintf(stderr, "Usage: %s ...\n", cmdname); + fprintf(stderr, "Usage: %s [options] ...\n" + "Options: -d Use deflate compression only\n" + " -f Force creation of archive\n" + " -u Only update changed files\n" + " -q Do not list files\n", cmdname); } //========================================================================== @@ -494,11 +630,14 @@ file_sorted_t *sort_files(dir_tree_t *trees, int num_files) // //========================================================================== -void write_zip(const char *zipname, dir_tree_t *trees) +void write_zip(const char *zipname, dir_tree_t *trees, int update) { + char tempname[_MAX_PATH]; + EndOfCentralDirectory dirend; int i, num_files; file_sorted_t *sorted; - zipFile zip; + FILE *zip, *ozip = NULL; + void *central_dir = NULL; num_files = count_files(trees); sorted = sort_files(trees, num_files); @@ -507,21 +646,105 @@ void write_zip(const char *zipname, dir_tree_t *trees) no_mem = 1; return; } - zip = zipOpen(zipname, APPEND_STATUS_CREATE); + if (update) + { + sprintf(tempname, "%s.temp", zipname); + ozip = fopen(zipname, "rb"); + if (ozip == NULL) + { + fprintf(stderr, "Could not open %s for updating: %s\n", zipname, strerror(errno)); + update = 0; + } + else + { + central_dir = find_central_dir(ozip); + if (central_dir == NULL) + { + fprintf(stderr, "Could not read central directory from %s. (Is it a zipfile?)\n", zipname); + update = 0; + } + } + if (!update) + { + fprintf(stderr, "Will proceed as if -u had not been specified.\n"); + } + } + if (update) + { + zip = fopen(tempname, "wb"); + } + else + { + zip = fopen(zipname, "wb"); + } if (zip == NULL) { fprintf(stderr, "Could not open %s: %s\n", zipname, strerror(errno)); } else { + // Write each file. for (i = 0; i < num_files; ++i) { - append_to_zip(zip, sorted + i); + if (append_to_zip(zip, sorted + i, ozip, central_dir)) + { + break; + } + } + if (i == num_files) + { + // Write central directory. + dirend.DirectoryOffset = ftell(zip); + for (i = 0; i < num_files; ++i) + { + write_central_dir(zip, sorted + i); + } + // Write the directory terminator. + dirend.Magic = ZIP_ENDOFDIR; + dirend.DiskNumber = 0; + dirend.FirstDisk = 0; + dirend.NumEntriesOnAllDisks = dirend.NumEntries = LittleShort(i); + dirend.DirectorySize = LittleLong(ftell(zip) - dirend.DirectoryOffset); + dirend.DirectoryOffset = LittleLong(dirend.DirectoryOffset); + dirend.ZipCommentLength = 0; + if (fwrite(&dirend, sizeof(dirend), 1, zip) != 1) + { + fprintf(stderr, "Failed writing zip directory terminator: %s\n", strerror(errno)); + } + printf("%s contains %d files (updated %d)\n", zipname, num_files, UpdateCount); + fclose(zip); + + if (ozip != NULL) + { + // Delete original, and rename temp to take its place + fclose(ozip); + ozip = NULL; + if (remove(zipname)) + { + fprintf(stderr, "Could not delete old zip: %s\nUpdated zip can be found at %s\n", + strerror(errno), tempname); + } + else if (rename(tempname, zipname)) + { + fprintf(stderr, "Could not rename %s to %s: %s\n", + tempname, zipname, strerror(errno)); + } + } + else + { + fclose(zip); + } } - zipClose(zip, NULL); - printf("Wrote %d/%d files to %s\n", i, num_files, zipname); } free(sorted); + if (ozip != NULL) + { + fclose(ozip); + } + if (central_dir != NULL) + { + free(central_dir); + } } //========================================================================== @@ -537,37 +760,35 @@ void write_zip(const char *zipname, dir_tree_t *trees) // //========================================================================== -int append_to_zip(zipFile zipfile, const file_sorted_t *file) +int append_to_zip(FILE *zip_file, file_sorted_t *filep, FILE *ozip, BYTE *odir) { + LocalFileHeader local; + uLong crc; + file_entry_t *file; char *readbuf; + char *compbuf[2]; + unsigned int comp_len[2]; + int offset[2]; + int method[2]; + int best; + int slot; FILE *lumpfile; - size_t readlen; - size_t len; - zip_fileinfo zip_inf; + unsigned int readlen; + unsigned int len; + int i; struct tm *ltime; - // clear zip_inf structure - memset(&zip_inf, 0, sizeof(zip_inf)); + file = filep->file; // try to determine local time - ltime = localtime(&file->file->time_write); - // if succeeded, - if (ltime != NULL) - { - // put it into the zip_inf structure - zip_inf.tmz_date.tm_sec = ltime->tm_sec; - zip_inf.tmz_date.tm_min = ltime->tm_min; - zip_inf.tmz_date.tm_hour = ltime->tm_hour; - zip_inf.tmz_date.tm_mday = ltime->tm_mday; - zip_inf.tmz_date.tm_mon = ltime->tm_mon; - zip_inf.tmz_date.tm_year = ltime->tm_year; - } + ltime = localtime(&file->time_write); + time_to_dos(ltime, &file->date, &file->time); // lumpfile = source file - lumpfile = fopen(file->file->path, "rb"); + lumpfile = fopen(file->path, "rb"); if (lumpfile == NULL) { - fprintf(stderr, "Could not open %s: %s\n", file->file->path, strerror(errno)); + fprintf(stderr, "Could not open %s: %s\n", file->path, strerror(errno)); return 1; } // len = source size @@ -584,7 +805,7 @@ int append_to_zip(zipFile zipfile, const file_sorted_t *file) return 1; } // read the whole source file into buffer - readlen = fread(readbuf, 1, len, lumpfile); + readlen = (unsigned int)fread(readbuf, 1, len, lumpfile); fclose(lumpfile); // if read less bytes than expected, @@ -592,30 +813,692 @@ int append_to_zip(zipFile zipfile, const file_sorted_t *file) { // diagnose and return error free(readbuf); - fprintf(stderr, "Unable to read %s\n", file->file->path); + fprintf(stderr, "Unable to read %s\n", file->path); return 1; } // file loaded - // create zip entry, giving entry name and zip_inf data, - // write data into zipfile, and close the zip entry - if (Z_OK != zipAddFileToZip(zipfile, file->path_in_zip, &zip_inf, readbuf, (unsigned)len)) + file->uncompressed_size = len; + file->compressed_size = len; + file->method = METHOD_STORED; + + // Calculate CRC32 for file. + crc = crc32(0, NULL, 0); + crc = crc32(crc, readbuf, (uInt)len); + file->crc32 = LittleLong(crc); + + // Can we save time and just copy the file from the old zip? + if (odir != NULL && ozip != NULL) { + CentralDirectoryEntry *dirent; + + dirent = find_file_in_zip(odir, filep->path_in_zip, len, crc, file->date, file->time); + if (dirent != NULL) + { + i = copy_zip_file(zip_file, file, ozip, dirent); + if (i > 0) + { + return 0; + } + if (i < 0) + { + fprintf(stderr, "Unable to write %s to zip\n", file->path); + return 1; + } + } + } + + if (!Quiet) + { + if (ozip != NULL) + { + printf("Updating %-40s", filep->path_in_zip); + } + else + { + printf("Adding %-40s", filep->path_in_zip); + } + } + UpdateCount++; + + // Allocate a buffer for compression, one byte less than the source buffer. + // If it doesn't fit in that space, then skip compression and store it as-is. + compbuf[0] = malloc(len - 1); + compbuf[1] = malloc(len - 1); + best = -1; // best slot + slot = 0; // slot we are compressing to now + + // Find best compression method. We have two output buffers. One to hold the + // best compression method, and the other to hold the compression we are trying + // now. + for (i = 0; Compressors[i].compress != NULL; ++i) + { + if (DeflateOnly && Compressors[i].method != METHOD_DEFLATE) + { + continue; + } + comp_len[slot] = len - 1; + method[slot] = Compressors[i].method; + offset[slot] = Compressors[i].compress(compbuf[slot], &comp_len[slot], readbuf, len); + if (offset[slot] >= 0) + { + if (best < 0 || comp_len[slot] <= comp_len[best]) + { + best = slot; + slot ^= 1; + } + } + } + + if (best >= 0) + { + file->method = method[best]; + file->compressed_size = comp_len[best]; + } +// printf("%s -> method %d -> slot %d\n", filep->path_in_zip, file->method, best); + + // Fill in local directory header. + local.Magic = ZIP_LOCALFILE; + local.VersionToExtract[0] = method_to_version(file->method); + local.VersionToExtract[1] = 0; + local.Flags = file->method == METHOD_DEFLATE ? LittleShort(2) : 0; + local.Method = LittleShort(file->method); + local.ModTime = file->time; + local.ModDate = file->date; + local.CRC32 = file->crc32; + local.UncompressedSize = LittleLong(file->uncompressed_size); + local.CompressedSize = LittleLong(file->compressed_size); + local.NameLength = LittleShort((unsigned short)strlen(filep->path_in_zip)); + local.ExtraLength = 0; + + file->zip_offset = ftell(zip_file); + + // Write out the header, file name, and file data. + if (fwrite(&local, sizeof(local), 1, zip_file) != 1 || + fwrite(filep->path_in_zip, strlen(filep->path_in_zip), 1, zip_file) != 1 || + (file->method ? fwrite(compbuf[best] + offset[best], 1, comp_len[best], zip_file) != comp_len[best] : + fwrite(readbuf, 1, len, zip_file) != len)) + { + if (!Quiet) + { + printf("\n"); + } + fprintf(stderr, "Unable to write %s to zip\n", file->path); free(readbuf); - fprintf(stderr, "Unable to write %s to zip\n", file->file->path); + if (compbuf != NULL) + { + free(compbuf); + } return 1; } + // all done free(readbuf); + if (compbuf[0] != NULL) + { + free(compbuf[0]); + } + if (compbuf[1] != NULL) + { + free(compbuf[1]); + } + if (!Quiet) + { + printf("%5.1f%% [%6u/%6u] %s\n", 100.0 - 100.0 * file->compressed_size / file->uncompressed_size, + file->compressed_size, file->uncompressed_size, method_name(file->method)); + } return 0; } +//========================================================================== +// +// write_central_dir +// +// Writes the central directory entry for a file. +// +//========================================================================== + +int write_central_dir(FILE *zip, file_sorted_t *filep) +{ + CentralDirectoryEntry dir; + file_entry_t *file; + + file = filep->file; + dir.Magic = ZIP_CENTRALFILE; + dir.VersionMadeBy[0] = 20; + dir.VersionMadeBy[1] = 0; + dir.VersionToExtract[0] = method_to_version(file->method); + dir.VersionToExtract[1] = 0; + dir.Flags = file->method == METHOD_DEFLATE ? LittleShort(2) : 0; + dir.Method = LittleShort(file->method); + dir.ModTime = file->time; + dir.ModDate = file->date; + dir.CRC32 = file->crc32; + dir.CompressedSize = file->compressed_size; + dir.UncompressedSize = file->uncompressed_size; + dir.NameLength = LittleShort((unsigned short)strlen(filep->path_in_zip)); + dir.ExtraLength = 0; + dir.CommentLength = 0; + dir.StartingDiskNumber = 0; + dir.InternalAttributes = 0; + dir.ExternalAttributes = 0; + dir.LocalHeaderOffset = LittleLong(file->zip_offset); + + if (fwrite(&dir, sizeof(dir), 1, zip) != 1 || + fwrite(filep->path_in_zip, strlen(filep->path_in_zip), 1, zip) != 1) + { + fprintf(stderr, "Error writing central directory header for %s: %s\n", file->path, strerror(errno)); + return 1; + } + return 0; +} + +//========================================================================== +// +// time_to_dos +// +// Converts time from struct tm to the DOS format used by zip files. +// +//========================================================================== + +void time_to_dos(struct tm *time, short *dosdate, short *dostime) +{ + if (time == NULL || time->tm_year < 80) + { + *dosdate = *dostime = 0; + } + else + { + *dosdate = LittleShort((time->tm_year - 80) * 512 + (time->tm_mon + 1) * 32 + time->tm_mday); + *dostime = LittleShort(time->tm_hour * 2048 + time->tm_min * 32 + time->tm_sec / 2); + } +} + +//========================================================================== +// +// method_to_version +// +// Given a compression method, returns the version of the ZIP appnote +// required to decompress it, for filling in the directory information. +// +//========================================================================== + +int method_to_version(int method) +{ + // Apparently, real-world programs get confused by setting the version + // to extract field to something other than 2.0. +#if 0 + if (method == METHOD_LZMA || method == METHOD_PPMD) + return 63; + if (method == METHOD_BZIP2) + return 46; +#endif + // Default anything else to PKZIP 2.0. + return 20; +} + +//========================================================================== +// +// method_name +// +// Returns the name of the compression method. If the method is unknown, +// this will point to a static buffer. +// +//========================================================================== + +const char *method_name(int method) +{ + static char unkn[16]; + + if (method == METHOD_STORED) + { + return "Stored"; + } + if (method == METHOD_DEFLATE) + { + return "Deflate"; + } + if (method == METHOD_LZMA) + { + return "LZMA"; + } + if (method == METHOD_PPMD) + { + return "PPMd"; + } + if (method == METHOD_BZIP2) + { + return "BZip2"; + } + sprintf(unkn, "Unk:%03d", method); + return unkn; +} + +//========================================================================== +// +// compress_lzma +// +// Returns non-negative offset to start of data stream on success. Barring +// any strange changes to the LZMA library in the future, success should +// always return 0. +// +//========================================================================== + +int compress_lzma(Byte *out, unsigned int *outlen, const Byte *in, unsigned int inlen) +{ + CLzmaEncProps lzma_props; + size_t props_size; + size_t comp_len; + int offset; + + if (*outlen < 1 + 4 + LZMA_PROPS_SIZE) + { + // Not enough room for LZMA properties header + compressed data. + return -1; + } + if (out == NULL || in == NULL || inlen == 0) + { + return -1; + } + + LzmaEncProps_Init(&lzma_props); +// lzma_props.level = 9; + props_size = LZMA_PROPS_SIZE; + comp_len = *outlen - 4 - LZMA_PROPS_SIZE; + + if (SZ_OK != LzmaEncode(out + 4 + LZMA_PROPS_SIZE, &comp_len, in, inlen, &lzma_props, + out + 4, &props_size, 0, NULL, &Alloc, &Alloc)) + { + return -1; + } + // Fill in LZMA properties header + offset = 0; + if (props_size != LZMA_PROPS_SIZE) + { + // Move LZMA properties to be adjacent to the compressed data, because for + // some reaseon the library didn't use all the space provided. + int i; + + offset = (int)(LZMA_PROPS_SIZE - props_size); + for (i = 4 + LZMA_PROPS_SIZE - 1; i > 4 + offset; --i) + { + out[i] = out[i - offset]; + } + } + out[offset] = MY_VER_MAJOR; + out[offset+1] = MY_VER_MINOR; + out[offset+2] = (Byte)props_size; + out[offset+3] = 0; + // Add header length to outlen + *outlen = (unsigned int)(comp_len + 4 + props_size); + return offset; +} + +//========================================================================== +// +// compress_bzip2 +// +// Returns 0 on success, negative on failure. +// +//========================================================================== + +int compress_bzip2(Byte *out, unsigned int *outlen, const Byte *in, unsigned int inlen) +{ + if (BZ_OK == BZ2_bzBuffToBuffCompress(out, outlen, in, inlen, 9, 0, 0)) + { + return 0; + } + return -1; +} + +#ifdef PPMD +//========================================================================== +// +// compress_ppmd +// +// Returns 0 on success, negative on failure. +// +// Big problem here: The zip format only allows for PPMd I rev. 1. This +// version of the code is incompatible with 64-bit processors. PPMd J rev. 1 +// corrects this and also compresses slightly better, but it also changes +// the data format and is incompatible with I rev. 1. The PPMd source code +// is a tangled mass that I cannot comprehend, so fixing I rev. 1 to work +// on 64-bit processors is well beyond my means. Hence, I cannot currently +// support PPMd in zips. If the zip spec gets updated to allow J rev. 1, +// then I can, but not any sooner. +// +//========================================================================== + +int compress_ppmd(Byte *out, unsigned int *outlen, const Byte *in, unsigned int inlen) +{ + int maxorder = 8; + int sasize = 8; + int cutoff = 0; + + _PPMD_FILE ppsin = { (char *)in, inlen, 0 }; + _PPMD_FILE ppsout = { out + 2, *outlen - 2, 0}; + + if (!PPMd_StartSubAllocator(sasize)) + { + return -1; + } + PPMd_EncodeFile(&ppsout, &ppsin, maxorder, cutoff); + PPMd_StopSubAllocator(); + if (ppsout.eof) + { + return -1; + } + if (!ppsin.eof) + { + return -1; + } + + *(short *)out = LittleShort((maxorder - 1) + ((sasize - 1) << 4) + (cutoff << 12)); + *outlen = *outlen - ppsout.buffersize; + return 0; +} +#endif + +//========================================================================== +// +// compress_deflate +// +// Returns 0 on success, negative on failure. +// +//========================================================================== + +int compress_deflate(Byte *out, unsigned int *outlen, const Byte *in, unsigned int inlen) +{ + z_stream stream; + int err; + + stream.next_in = (Bytef *)in; + stream.avail_in = inlen; + stream.next_out = out; + stream.avail_out = *outlen; + stream.zalloc = (alloc_func)0; + stream.zfree = (free_func)0; + stream.opaque = (voidpf)0; + + err = deflateInit2(&stream, 9, Z_DEFLATED, -15, 9, Z_DEFAULT_STRATEGY); + if (err != Z_OK) return -1; + + err = deflate(&stream, Z_FINISH); + if (err != Z_STREAM_END) { + deflateEnd(&stream); + return -1; + } + *outlen = stream.total_out; + + err = deflateEnd(&stream); + return err == Z_OK ? 0 : -1; +} + +//========================================================================== +// +// find_central_dir +// +// Finds and loads the central directory records in the file. +// Taken from Quake3 source and modified. +// +//========================================================================== + +BYTE *find_central_dir(FILE *fin) +{ + unsigned char buf[BUFREADCOMMENT + 4]; + EndOfCentralDirectory eod; + BYTE *dir; + long file_size; + long back_read; + long max_back; // maximum size of global comment + long pos_found = 0; + + fseek(fin, 0, SEEK_END); + + file_size = ftell(fin); + max_back = 0xffff > file_size ? file_size : 0xffff; + + back_read = 4; + while (back_read < max_back) + { + DWORD read_size, read_pos; + int i; + if (back_read + BUFREADCOMMENT > max_back) + back_read = max_back; + else + back_read += BUFREADCOMMENT; + read_pos = file_size - back_read; + + read_size = (BUFREADCOMMENT + 4) < (file_size - read_pos) ? + (BUFREADCOMMENT + 4) : (file_size - read_pos); + + if (fseek(fin, read_pos, SEEK_SET) != 0) + return NULL; + + if (fread(buf, 1, read_size, fin) != read_size) + return NULL; + + for (i = (int)read_size - 3; (i--) > 0;) + { + if (buf[i] == 'P' && buf[i+1] == 'K' && buf[i+2] == 5 && buf[i+3] == 6) + { + pos_found = read_pos + i; + break; + } + } + + if (pos_found != 0) + break; + } + if (pos_found == 0 || + fseek(fin, pos_found, SEEK_SET) != 0 || + fread(&eod, sizeof(eod), 1, fin) != 1 || + fseek(fin, LittleShort(eod.DirectoryOffset), SEEK_SET) != 0) + { + return NULL; + } + dir = malloc(LittleLong(eod.DirectorySize) + 4); + if (dir == NULL) + { + no_mem = 1; + return NULL; + } + if (fread(dir, 1, LittleLong(eod.DirectorySize), fin) != LittleLong(eod.DirectorySize)) + { + free(dir); + return NULL; + } + if (*(DWORD *)dir != ZIP_CENTRALFILE) + { + free(dir); + return NULL; + } + *(DWORD *)(dir + LittleLong(eod.DirectorySize)) = ZIP_ENDOFDIR; + return dir; +} + +//========================================================================== +// +// find_file_in_zip +// +// Returns a pointer to a central directory entry to a file, if it was +// found in the zip's directory. Data endianness is in zip order. +// +//========================================================================== + +CentralDirectoryEntry *find_file_in_zip(BYTE *dir, const char *path, unsigned int len, unsigned int crc, short date, short time) +{ + int pathlen = (int)strlen(path); + CentralDirectoryEntry *ent; + int flags; + + while (*(DWORD *)dir == ZIP_CENTRALFILE) + { + ent = (CentralDirectoryEntry *)dir; + if (pathlen == LittleShort(ent->NameLength) && + strncmp((char *)(ent + 1), path, pathlen) == 0) + { + // Found something that matches by name. + break; + } + dir += sizeof(*ent) + LittleShort(ent->NameLength) + LittleShort(ent->ExtraLength) + LittleShort(ent->CommentLength); + } + if (*(DWORD *)dir != ZIP_CENTRALFILE) + { + return NULL; + } + if (crc != LittleLong(ent->CRC32)) + { + return NULL; + } + if (len != LittleLong(ent->UncompressedSize)) + { + return NULL; + } + // Should I check modification date and time here? + flags = LittleShort(ent->Flags); + if (flags & 1) + { // Don't want to deal with encryption. + return NULL; + } + if (ent->ExtraLength != 0) + { // Don't want to deal with extra data. + return NULL; + } + // Okay, looks good. + return ent; +} + +//========================================================================== +// +// copy_zip_file +// +// Copies one file from ozip to zip. Returns positive on success, zero if +// the file could not be found, and negative if it failed while writing the +// file. +// +//========================================================================== + +int copy_zip_file(FILE *zip, file_entry_t *file, FILE *ozip, CentralDirectoryEntry *ent) +{ + LocalFileHeader lfh; + BYTE *buf; + DWORD buf_size; + + if (fseek(ozip, LittleLong(ent->LocalHeaderOffset), SEEK_SET) != 0) + { + return 0; + } + if (fread(&lfh, sizeof(lfh), 1, ozip) != 1) + { + return 0; + } + // Check to make sure the local header matches the central directory. + if (lfh.Flags != ent->Flags || lfh.Method != ent->Method || + lfh.CRC32 != ent->CRC32 || lfh.CompressedSize != ent->CompressedSize || + lfh.UncompressedSize != ent->UncompressedSize || + lfh.NameLength != ent->NameLength || lfh.ExtraLength != ent->ExtraLength) + { + return 0; + } + buf_size = LittleShort(lfh.NameLength) + LittleLong(lfh.CompressedSize); + buf = malloc(buf_size); + if (buf == NULL) + { + return 0; + } + if (fread(buf, 1, buf_size, ozip) != buf_size) + { + free(buf); + return 0; + } + // Check to be sure name matches. + if (strncmp(buf, (char *)(ent + 1), LittleShort(lfh.NameLength)) != 0) + { + free(buf); + return 0; + } + // Looks good. Let's write it in. + file->zip_offset = ftell(zip); + if (fwrite(&lfh, sizeof(lfh), 1, zip) != 1 || + fwrite(buf, 1, buf_size, zip) != buf_size) + { + free(buf); + return -1; + } + file->date = lfh.ModDate; + file->time = lfh.ModTime; + file->uncompressed_size = LittleLong(lfh.UncompressedSize); + file->compressed_size = LittleLong(lfh.CompressedSize); + file->method = LittleShort(lfh.Method); + file->crc32 = lfh.CRC32; + return 1; +} + +//========================================================================== +// +// main +// +//========================================================================== + int __cdecl main (int argc, char **argv) { dir_tree_t *tree, *trees; file_entry_t *file; struct stat zipstat; int needwrite; + int i, j, k; + int force = 0; + int update = 0; + + // Find options. Options are removed from the array. + for (i = k = 1; i < argc; ++i) + { + if (argv[i][0] == '-') + { + if (argv[i][1] == '-') + { + if (argv[i][2] == '\0') + { // -- terminates option handling for the rest of the command line + break; + } + } + for (j = 1; argv[i][j] != '\0'; ++j) + { + if (argv[i][j] == 'f') + { + force = 1; + } + else if (argv[i][j] == 'd') + { + DeflateOnly = 1; + } + else if (argv[i][j] == 'u') + { + update = 1; + } + else if (argv[i][j] == 'q') + { + Quiet = 1; + } + else + { + fprintf(stderr, "Unknown option '%c'\n", argv[i][j]); + print_usage(argv[0]); + return 1; + } + } + } + else + { + argv[k++] = argv[i]; + } + } + for (; i <= argc; ++i) + { + argv[k++] = argv[i]; + } + argc -= i - k; if (argc < 3) { @@ -624,26 +1507,27 @@ int __cdecl main (int argc, char **argv) } trees = add_dirs(&argv[2]); - if(no_mem) + if (no_mem) { free_dir_trees(trees); fprintf(stderr, "Out of memory.\n"); return 1; } - needwrite = 0; + needwrite = force; if (stat(argv[1], &zipstat) != 0) { if (errno == ENOENT) { needwrite = 1; + update = 0; // Can't update what's not there. } else { fprintf(stderr, "Could not stat %s: %s\n", argv[1], strerror(errno)); } } - else + else if (!needwrite) { // Check the files in each tree. If any one of them was modified more // recently than the zip, then it needs to be recreated. @@ -659,9 +1543,9 @@ int __cdecl main (int argc, char **argv) } } } - if (needwrite) + if (force || needwrite) { - write_zip(argv[1], trees); + write_zip(argv[1], trees, update); } free_dir_trees(trees); if (no_mem) @@ -671,3 +1555,17 @@ int __cdecl main (int argc, char **argv) } return 0; } + +//========================================================================== +// +// bz_internal_error +// +// libbzip2 wants this, since we build it with BZ_NO_STDIO set. +// +//========================================================================== + +void bz_internal_error (int errcode) +{ + fprintf(stderr, "libbzip2: internal error number %d\n", errcode); + exit(3); +} diff --git a/tools/zipdir/zipdir.vcproj b/tools/zipdir/zipdir.vcproj index 61103dfb1d..d461ec5d22 100644 --- a/tools/zipdir/zipdir.vcproj +++ b/tools/zipdir/zipdir.vcproj @@ -44,6 +44,7 @@ - - - - - - - - diff --git a/wadsrc/CMakeLists.txt b/wadsrc/CMakeLists.txt index b4c6e211fd..cfe0739c42 100644 --- a/wadsrc/CMakeLists.txt +++ b/wadsrc/CMakeLists.txt @@ -6,7 +6,7 @@ add_custom_command( OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/generated/dehsupp.lmp DEPENDS dehsupp ${CMAKE_CURRENT_SOURCE_DIR}/sources/dehsupp.txt ) add_custom_command( OUTPUT ${ZDOOM_OUTPUT_DIR}/zdoom.pk3 - COMMAND ${CMAKE_BINARY_DIR}/tools/zipdir/zipdir ${ZDOOM_OUTPUT_DIR}/zdoom.pk3 ${CMAKE_CURRENT_SOURCE_DIR}/static ${CMAKE_CURRENT_BINARY_DIR}/generated + COMMAND ${CMAKE_BINARY_DIR}/tools/zipdir/zipdir -u ${ZDOOM_OUTPUT_DIR}/zdoom.pk3 ${CMAKE_CURRENT_SOURCE_DIR}/static ${CMAKE_CURRENT_BINARY_DIR}/generated DEPENDS zipdir ${CMAKE_CURRENT_BINARY_DIR}/generated/dehsupp.lmp ${CMAKE_CURRENT_SOURCE_DIR}/static ) add_custom_target( pk3 ALL diff --git a/wadsrc/wadsrc.vcproj b/wadsrc/wadsrc.vcproj index cac80d07a0..45bc588450 100644 --- a/wadsrc/wadsrc.vcproj +++ b/wadsrc/wadsrc.vcproj @@ -37,7 +37,7 @@ diff --git a/zdoom.sln b/zdoom.sln index a22f342947..7d64456b10 100644 --- a/zdoom.sln +++ b/zdoom.sln @@ -2,7 +2,11 @@ Microsoft Visual Studio Solution File, Format Version 9.00 # Visual Studio 2005 Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "zdoom", "zdoom.vcproj", "{8049475B-5C87-46F9-9358-635218A4EF18}" ProjectSection(ProjectDependencies) = postProject + {E83FD370-2E72-4D4C-9427-FF9D9DED1E88} = {E83FD370-2E72-4D4C-9427-FF9D9DED1E88} + {6EB27E78-7C7A-4F08-8E19-957E8EB3A20F} = {6EB27E78-7C7A-4F08-8E19-957E8EB3A20F} + {A7DE5C73-D623-4118-A48A-BDFD1FAE97D4} = {A7DE5C73-D623-4118-A48A-BDFD1FAE97D4} {AC3F5340-40CB-4C3A-8AA7-CB7158DB4466} = {AC3F5340-40CB-4C3A-8AA7-CB7158DB4466} + {8997289F-10BF-4678-8BAA-3BB509C84953} = {8997289F-10BF-4678-8BAA-3BB509C84953} {B68E0ABF-B627-48A3-A92F-D8F827A75054} = {B68E0ABF-B627-48A3-A92F-D8F827A75054} {1D179D4B-F008-431B-8C72-111F8372584F} = {1D179D4B-F008-431B-8C72-111F8372584F} {DA47396F-60C1-4BDE-A977-7F7DE461CF77} = {DA47396F-60C1-4BDE-A977-7F7DE461CF77} @@ -10,8 +14,6 @@ Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "zdoom", "zdoom.vcproj", "{8 {667D2EE7-C357-49E2-9BAB-0A4A45F0F76E} = {667D2EE7-C357-49E2-9BAB-0A4A45F0F76E} {6077B7D6-349F-4077-B552-3BC302EF5859} = {6077B7D6-349F-4077-B552-3BC302EF5859} {F9D9E7D4-E1A2-4866-9E85-B1B14137EE63} = {F9D9E7D4-E1A2-4866-9E85-B1B14137EE63} - {E83FD370-2E72-4D4C-9427-FF9D9DED1E88} = {E83FD370-2E72-4D4C-9427-FF9D9DED1E88} - {8997289F-10BF-4678-8BAA-3BB509C84953} = {8997289F-10BF-4678-8BAA-3BB509C84953} EndProjectSection EndProject Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "zlib", "zlib\zlib.vcproj", "{F9D9E7D4-E1A2-4866-9E85-B1B14137EE63}" @@ -46,9 +48,15 @@ Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "gdtoa", "gdtoa\gdtoa.vcproj EndProject Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "zipdir", "tools\zipdir\zipdir.vcproj", "{24A19C02-F041-4AB0-A1A1-02E1E88EDBD3}" ProjectSection(ProjectDependencies) = postProject + {6EB27E78-7C7A-4F08-8E19-957E8EB3A20F} = {6EB27E78-7C7A-4F08-8E19-957E8EB3A20F} {F9D9E7D4-E1A2-4866-9E85-B1B14137EE63} = {F9D9E7D4-E1A2-4866-9E85-B1B14137EE63} + {A7DE5C73-D623-4118-A48A-BDFD1FAE97D4} = {A7DE5C73-D623-4118-A48A-BDFD1FAE97D4} EndProjectSection EndProject +Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "lzmalib", "lzma\lzmalib.vcproj", "{6EB27E78-7C7A-4F08-8E19-957E8EB3A20F}" +EndProject +Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "bzip2", "bzip2\bzip2.vcproj", "{A7DE5C73-D623-4118-A48A-BDFD1FAE97D4}" +EndProject Global GlobalSection(SolutionConfigurationPlatforms) = preSolution Debug|Win32 = Debug|Win32 @@ -161,6 +169,22 @@ Global {24A19C02-F041-4AB0-A1A1-02E1E88EDBD3}.Release|Win32.Build.0 = Release|Win32 {24A19C02-F041-4AB0-A1A1-02E1E88EDBD3}.Release|x64.ActiveCfg = Release|x64 {24A19C02-F041-4AB0-A1A1-02E1E88EDBD3}.Release|x64.Build.0 = Release|x64 + {6EB27E78-7C7A-4F08-8E19-957E8EB3A20F}.Debug|Win32.ActiveCfg = Debug|Win32 + {6EB27E78-7C7A-4F08-8E19-957E8EB3A20F}.Debug|Win32.Build.0 = Debug|Win32 + {6EB27E78-7C7A-4F08-8E19-957E8EB3A20F}.Debug|x64.ActiveCfg = Debug|x64 + {6EB27E78-7C7A-4F08-8E19-957E8EB3A20F}.Debug|x64.Build.0 = Debug|x64 + {6EB27E78-7C7A-4F08-8E19-957E8EB3A20F}.Release|Win32.ActiveCfg = Release|Win32 + {6EB27E78-7C7A-4F08-8E19-957E8EB3A20F}.Release|Win32.Build.0 = Release|Win32 + {6EB27E78-7C7A-4F08-8E19-957E8EB3A20F}.Release|x64.ActiveCfg = Release|x64 + {6EB27E78-7C7A-4F08-8E19-957E8EB3A20F}.Release|x64.Build.0 = Release|x64 + {A7DE5C73-D623-4118-A48A-BDFD1FAE97D4}.Debug|Win32.ActiveCfg = Debug|Win32 + {A7DE5C73-D623-4118-A48A-BDFD1FAE97D4}.Debug|Win32.Build.0 = Debug|Win32 + {A7DE5C73-D623-4118-A48A-BDFD1FAE97D4}.Debug|x64.ActiveCfg = Debug|x64 + {A7DE5C73-D623-4118-A48A-BDFD1FAE97D4}.Debug|x64.Build.0 = Debug|x64 + {A7DE5C73-D623-4118-A48A-BDFD1FAE97D4}.Release|Win32.ActiveCfg = Release|Win32 + {A7DE5C73-D623-4118-A48A-BDFD1FAE97D4}.Release|Win32.Build.0 = Release|Win32 + {A7DE5C73-D623-4118-A48A-BDFD1FAE97D4}.Release|x64.ActiveCfg = Release|x64 + {A7DE5C73-D623-4118-A48A-BDFD1FAE97D4}.Release|x64.Build.0 = Release|x64 EndGlobalSection GlobalSection(SolutionProperties) = preSolution HideSolutionNode = FALSE diff --git a/zdoom.vcproj b/zdoom.vcproj index 2158b68881..dc8e6e6534 100644 --- a/zdoom.vcproj +++ b/zdoom.vcproj @@ -1,7 +1,7 @@ - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + + + + - - - + + + - - - + + + - - - + + + - - - + + + - - - + + + - - - @@ -1852,6 +1844,14 @@ Outputs="$(IntDir)/$(InputName).obj" /> + + + @@ -2017,6 +2017,14 @@ Outputs="$(IntDir)\$(InputName).obj" /> + + + @@ -2027,14 +2035,6 @@ Outputs="$(IntDir)\$(InputName).obj" /> - - - + + + - - - + + + @@ -5335,14 +5343,6 @@ AdditionalIncludeDirectories="src\win32;$(NoInherit)" /> - - - @@ -5621,7 +5621,7 @@ />