2011-12-22 12:52:56 +00:00
|
|
|
/** Implementation of NSRegularExpression for GNUStep
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
Copyright (C) 2010 Free Software Foundation, Inc.
|
|
|
|
|
|
|
|
This file is part of the GNUstep Base Library.
|
|
|
|
|
|
|
|
This library is free software; you can redistribute it and/or
|
|
|
|
modify it under the terms of the GNU Lesser General Public
|
|
|
|
License as published by the Free Software Foundation; either
|
|
|
|
version 2 of the License, or (at your option) any later version.
|
|
|
|
|
|
|
|
This library is distributed in the hope that it will be useful,
|
|
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
2019-12-09 23:36:00 +00:00
|
|
|
Lesser General Public License for more details.
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
You should have received a copy of the GNU Lesser General Public
|
|
|
|
License along with this library; if not, write to the Free
|
2024-11-07 13:37:59 +00:00
|
|
|
Software Foundation, Inc., 31 Milk Street #960789 Boston, MA 02196 USA.
|
2010-12-16 10:59:50 +00:00
|
|
|
*/
|
|
|
|
|
2011-08-13 15:28:22 +00:00
|
|
|
|
|
|
|
#define EXPOSE_NSRegularExpression_IVARS 1
|
2010-12-16 10:59:50 +00:00
|
|
|
#import "common.h"
|
|
|
|
|
2010-12-16 10:09:43 +00:00
|
|
|
#if GS_USE_ICU == 1
|
2021-07-16 12:31:38 +00:00
|
|
|
#if defined(HAVE_UNICODE_UREGEX_H)
|
|
|
|
#include <unicode/uregex.h>
|
2021-08-06 07:29:11 +00:00
|
|
|
#elif defined(HAVE_ICU_H)
|
2021-07-16 12:31:38 +00:00
|
|
|
#include <icu.h>
|
|
|
|
#endif
|
2011-12-22 12:52:56 +00:00
|
|
|
|
2012-02-23 17:57:50 +00:00
|
|
|
/* FIXME It would be nice to use autoconf for checking whether uregex_openUText
|
|
|
|
* is defined. However the naive check using AC_CHECK_FUNCS(uregex_openUText)
|
2017-02-11 14:24:58 +00:00
|
|
|
* won't work because libicu internally renames all entry points with some cpp
|
2012-02-23 17:57:50 +00:00
|
|
|
* magic.
|
|
|
|
*/
|
2024-11-21 21:05:34 +00:00
|
|
|
#if !defined(HAVE_UREGEX_OPENUTEXT)
|
2022-02-18 16:00:14 +00:00
|
|
|
#if U_ICU_VERSION_MAJOR_NUM > 4 || (U_ICU_VERSION_MAJOR_NUM == 4 && U_ICU_VERSION_MINOR_NUM >= 4) || defined(HAVE_ICU_H)
|
2011-12-22 12:52:56 +00:00
|
|
|
#define HAVE_UREGEX_OPENUTEXT 1
|
|
|
|
#endif
|
2024-11-21 21:05:34 +00:00
|
|
|
#endif
|
|
|
|
|
|
|
|
/* Until the uregex_replaceAllUText() and uregex_replaceFirstUText() work
|
|
|
|
* without leaking memory, we can't use them :-(
|
|
|
|
* Preoblem exists on Ubuntu in 2024 with icu-74.2
|
|
|
|
*/
|
|
|
|
#if defined(HAVE_UREGEX_OPENUTEXT)
|
|
|
|
#undef HAVE_UREGEX_OPENUTEXT
|
|
|
|
#endif
|
2011-08-03 12:04:27 +00:00
|
|
|
|
2011-07-22 13:23:11 +00:00
|
|
|
#define NSRegularExpressionWorks
|
2011-08-03 12:04:27 +00:00
|
|
|
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
#define GSREGEXTYPE URegularExpression
|
|
|
|
#import "GSICUString.h"
|
2023-11-13 19:45:16 +00:00
|
|
|
#import "Foundation/NSDictionary.h"
|
|
|
|
#import "Foundation/NSException.h"
|
2011-08-02 08:33:33 +00:00
|
|
|
#import "Foundation/NSRegularExpression.h"
|
|
|
|
#import "Foundation/NSTextCheckingResult.h"
|
2011-08-03 12:04:27 +00:00
|
|
|
#import "Foundation/NSArray.h"
|
|
|
|
#import "Foundation/NSCoder.h"
|
2016-06-17 09:04:04 +00:00
|
|
|
#import "Foundation/NSUserDefaults.h"
|
|
|
|
#import "Foundation/NSNotification.h"
|
2023-11-13 18:55:49 +00:00
|
|
|
#import "Foundation/FoundationErrors.h"
|
|
|
|
#import "Foundation/NSError.h"
|
2011-08-03 12:04:27 +00:00
|
|
|
|
2025-01-04 15:05:14 +00:00
|
|
|
typedef struct {
|
|
|
|
GSRegexEnumerationCallback h; // The handler callback function
|
|
|
|
void *c; // Context for this enumeration
|
|
|
|
} GSRegexContext;
|
|
|
|
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
|
|
|
|
/**
|
|
|
|
* To be helpful, Apple decided to define a set of flags that mean exactly the
|
|
|
|
* same thing as the URegexpFlags enum in libicu, but have different values.
|
|
|
|
* This was completely stupid, but we probably have to live with it. We could
|
|
|
|
* in theory use the libicu values directly (that would be sensible), but that
|
|
|
|
* would break any code that didn't correctly use the symbolic constants.
|
|
|
|
*/
|
2010-12-16 10:59:50 +00:00
|
|
|
uint32_t
|
|
|
|
NSRegularExpressionOptionsToURegexpFlags(NSRegularExpressionOptions opts)
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
uint32_t flags = 0;
|
|
|
|
|
|
|
|
if (opts & NSRegularExpressionCaseInsensitive)
|
|
|
|
{
|
|
|
|
flags |= UREGEX_CASE_INSENSITIVE;
|
|
|
|
}
|
|
|
|
if (opts & NSRegularExpressionAllowCommentsAndWhitespace)
|
|
|
|
{
|
|
|
|
flags |= UREGEX_COMMENTS;
|
|
|
|
}
|
|
|
|
if (opts & NSRegularExpressionIgnoreMetacharacters)
|
|
|
|
{
|
2012-02-23 17:57:50 +00:00
|
|
|
flags |= UREGEX_LITERAL;
|
2010-12-16 10:59:50 +00:00
|
|
|
}
|
|
|
|
if (opts & NSRegularExpressionDotMatchesLineSeparators)
|
|
|
|
{
|
|
|
|
flags |= UREGEX_DOTALL;
|
|
|
|
}
|
|
|
|
if (opts & NSRegularExpressionAnchorsMatchLines)
|
|
|
|
{
|
|
|
|
flags |= UREGEX_MULTILINE;
|
|
|
|
}
|
|
|
|
if (opts & NSRegularExpressionUseUnixLineSeparators)
|
|
|
|
{
|
|
|
|
flags |= UREGEX_UNIX_LINES;
|
|
|
|
}
|
|
|
|
if (opts & NSRegularExpressionUseUnicodeWordBoundaries)
|
|
|
|
{
|
|
|
|
flags |= UREGEX_UWORD;
|
|
|
|
}
|
|
|
|
return flags;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
@implementation NSRegularExpression
|
2010-12-16 10:59:50 +00:00
|
|
|
|
2025-01-04 15:05:14 +00:00
|
|
|
/* Callback method to invoke a block
|
|
|
|
*/
|
|
|
|
static void
|
2025-01-04 16:50:13 +00:00
|
|
|
blockCallback(
|
2025-01-04 15:05:14 +00:00
|
|
|
void *context, NSTextCheckingResult *match,
|
|
|
|
NSMatchingFlags flags, BOOL *shouldStop)
|
|
|
|
{
|
|
|
|
GSRegexBlock block = (GSRegexBlock)context;
|
|
|
|
|
|
|
|
if (block)
|
|
|
|
{
|
|
|
|
CALL_BLOCK(block, match, flags, shouldStop);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2010-12-16 10:59:50 +00:00
|
|
|
+ (NSRegularExpression*) regularExpressionWithPattern: (NSString*)aPattern
|
2023-11-13 19:45:16 +00:00
|
|
|
options: (NSRegularExpressionOptions)opts
|
|
|
|
error: (NSError**)e
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2023-11-13 19:45:16 +00:00
|
|
|
NSRegularExpression *r;
|
|
|
|
|
|
|
|
r = [[self alloc] initWithPattern: aPattern
|
|
|
|
options: opts
|
|
|
|
error: e];
|
|
|
|
return AUTORELEASE(r);
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
2012-03-27 10:03:40 +00:00
|
|
|
|
2011-12-22 12:52:56 +00:00
|
|
|
#if HAVE_UREGEX_OPENUTEXT
|
2010-12-16 10:59:50 +00:00
|
|
|
- (id) initWithPattern: (NSString*)aPattern
|
|
|
|
options: (NSRegularExpressionOptions)opts
|
|
|
|
error: (NSError**)e
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
uint32_t flags = NSRegularExpressionOptionsToURegexpFlags(opts);
|
|
|
|
UText p = UTEXT_INITIALIZER;
|
|
|
|
UParseError pe = {0};
|
|
|
|
UErrorCode s = 0;
|
|
|
|
|
2023-11-13 19:45:16 +00:00
|
|
|
// Raise an NSInvalidArgumentException to match macOS behaviour.
|
|
|
|
if (!aPattern)
|
|
|
|
{
|
|
|
|
NSException *exp;
|
|
|
|
|
|
|
|
exp = [NSException exceptionWithName: NSInvalidArgumentException
|
|
|
|
reason: @"nil argument"
|
|
|
|
userInfo: nil];
|
2024-11-16 21:11:56 +00:00
|
|
|
RELEASE(self);
|
2023-11-13 19:45:16 +00:00
|
|
|
[exp raise];
|
|
|
|
}
|
|
|
|
|
2012-03-27 10:03:40 +00:00
|
|
|
#if !__has_feature(blocks)
|
2025-01-04 15:05:14 +00:00
|
|
|
GSOnceMLog(@"Warning: this implementation of NSRegularExpression uses"
|
|
|
|
@" -enumerateMatchesInString:options:range:callback:context: as a"
|
|
|
|
@" primitive method rather than the blocks-dependtent method used"
|
|
|
|
@" by Apple. If you must subclass NSRegularExpression, you must"
|
|
|
|
@" bear that difference in mind");
|
2012-03-27 10:03:40 +00:00
|
|
|
#endif
|
|
|
|
|
2010-12-16 10:59:50 +00:00
|
|
|
UTextInitWithNSString(&p, aPattern);
|
|
|
|
regex = uregex_openUText(&p, flags, &pe, &s);
|
|
|
|
utext_close(&p);
|
|
|
|
if (U_FAILURE(s))
|
|
|
|
{
|
2023-11-13 19:45:16 +00:00
|
|
|
/* Match macOS behaviour if the pattern is invalid.
|
|
|
|
* Example:
|
|
|
|
* Domain=NSCocoaErrorDomain
|
|
|
|
* Code=2048 "The value “<PATTERN>” is invalid."
|
|
|
|
* UserInfo={NSInvalidValue=<PATTERN>}
|
|
|
|
*/
|
|
|
|
if (e)
|
|
|
|
{
|
|
|
|
NSDictionary *userInfo;
|
|
|
|
NSString *description;
|
2023-11-13 18:55:49 +00:00
|
|
|
|
2024-11-21 21:05:34 +00:00
|
|
|
description = [NSString
|
|
|
|
stringWithFormat: @"The value “%@” is invalid.", aPattern];
|
2023-11-13 18:55:49 +00:00
|
|
|
|
2023-11-13 19:45:16 +00:00
|
|
|
userInfo = [NSDictionary dictionaryWithObjectsAndKeys:
|
|
|
|
aPattern, @"NSInvalidValue",
|
|
|
|
description, NSLocalizedDescriptionKey,
|
|
|
|
nil];
|
2023-11-13 18:55:49 +00:00
|
|
|
|
2023-11-13 19:45:16 +00:00
|
|
|
*e = [NSError errorWithDomain: NSCocoaErrorDomain
|
|
|
|
code: NSFormattingError
|
|
|
|
userInfo: userInfo];
|
|
|
|
}
|
2023-11-13 18:55:49 +00:00
|
|
|
|
2023-11-13 19:45:16 +00:00
|
|
|
DESTROY(self);
|
|
|
|
return self;
|
2010-12-16 10:59:50 +00:00
|
|
|
}
|
|
|
|
options = opts;
|
|
|
|
return self;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
2016-06-17 07:43:51 +00:00
|
|
|
- (BOOL) isEqual: (id)obj
|
|
|
|
{
|
|
|
|
if ([obj isKindOfClass: [NSRegularExpression class]])
|
|
|
|
{
|
|
|
|
if (self == obj)
|
|
|
|
{
|
|
|
|
return YES;
|
|
|
|
}
|
|
|
|
else if (options != ((NSRegularExpression*)obj)->options)
|
|
|
|
{
|
|
|
|
return NO;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
UErrorCode myErr = 0;
|
|
|
|
UErrorCode theirErr = 0;
|
|
|
|
const UText *myText = uregex_patternUText(regex, &myErr);
|
|
|
|
const UText *theirText =
|
|
|
|
uregex_patternUText(((NSRegularExpression*)obj)->regex, &theirErr);
|
|
|
|
if (U_FAILURE(myErr) != U_FAILURE(theirErr))
|
|
|
|
{
|
|
|
|
return NO;
|
|
|
|
}
|
|
|
|
else if (U_FAILURE(myErr) && U_FAILURE(theirErr))
|
|
|
|
{
|
|
|
|
return YES;
|
|
|
|
}
|
|
|
|
return utext_equals(myText, theirText);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
return [super isEqual: obj];
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-12-16 10:59:50 +00:00
|
|
|
- (NSString*) pattern
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2011-08-03 12:04:27 +00:00
|
|
|
UErrorCode s = 0;
|
|
|
|
UText *t = uregex_patternUText(regex, &s);
|
|
|
|
GSUTextString *str = NULL;
|
|
|
|
|
|
|
|
if (U_FAILURE(s))
|
|
|
|
{
|
|
|
|
return nil;
|
|
|
|
}
|
|
|
|
str = [GSUTextString new];
|
|
|
|
utext_clone(&str->txt, t, FALSE, TRUE, &s);
|
2023-11-13 19:45:16 +00:00
|
|
|
return AUTORELEASE(str);
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2011-12-22 12:52:56 +00:00
|
|
|
#else
|
|
|
|
- (id) initWithPattern: (NSString*)aPattern
|
|
|
|
options: (NSRegularExpressionOptions)opts
|
|
|
|
error: (NSError**)e
|
|
|
|
{
|
|
|
|
int32_t length = [aPattern length];
|
|
|
|
uint32_t flags = NSRegularExpressionOptionsToURegexpFlags(opts);
|
|
|
|
UParseError pe = {0};
|
|
|
|
UErrorCode s = 0;
|
|
|
|
TEMP_BUFFER(buffer, length);
|
|
|
|
|
2012-03-27 10:03:40 +00:00
|
|
|
#if !__has_feature(blocks)
|
2025-01-04 15:05:14 +00:00
|
|
|
GSOnceMLog(@"Warning: this implementation of NSRegularExpression uses"
|
|
|
|
@" -enumerateMatchesInString:options:range:callback:context: as a"
|
|
|
|
@" primitive method rather than the blocks-dependtent method used"
|
|
|
|
@" by Apple. If you must subclass NSRegularExpression, you must"
|
|
|
|
@" bear that difference in mind");
|
2012-03-27 10:03:40 +00:00
|
|
|
#endif
|
|
|
|
|
2024-11-21 21:05:34 +00:00
|
|
|
// Raise an NSInvalidArgumentException to match macOS behaviour.
|
|
|
|
if (!aPattern)
|
|
|
|
{
|
|
|
|
NSException *exp;
|
|
|
|
|
|
|
|
exp = [NSException exceptionWithName: NSInvalidArgumentException
|
|
|
|
reason: @"nil argument"
|
|
|
|
userInfo: nil];
|
|
|
|
RELEASE(self);
|
|
|
|
[exp raise];
|
|
|
|
}
|
|
|
|
|
2011-12-22 12:52:56 +00:00
|
|
|
[aPattern getCharacters: buffer range: NSMakeRange(0, length)];
|
|
|
|
regex = uregex_open(buffer, length, flags, &pe, &s);
|
|
|
|
if (U_FAILURE(s))
|
|
|
|
{
|
2024-11-21 21:05:34 +00:00
|
|
|
/* Match macOS behaviour if the pattern is invalid.
|
|
|
|
* Example:
|
|
|
|
* Domain=NSCocoaErrorDomain
|
|
|
|
* Code=2048 "The value “<PATTERN>” is invalid."
|
|
|
|
* UserInfo={NSInvalidValue=<PATTERN>}
|
|
|
|
*/
|
|
|
|
if (e)
|
|
|
|
{
|
|
|
|
NSDictionary *userInfo;
|
|
|
|
NSString *description;
|
|
|
|
|
|
|
|
description = [NSString
|
|
|
|
stringWithFormat: @"The value “%@” is invalid.", aPattern];
|
|
|
|
|
|
|
|
userInfo = [NSDictionary dictionaryWithObjectsAndKeys:
|
|
|
|
aPattern, @"NSInvalidValue",
|
|
|
|
description, NSLocalizedDescriptionKey,
|
|
|
|
nil];
|
|
|
|
|
|
|
|
*e = [NSError errorWithDomain: NSCocoaErrorDomain
|
|
|
|
code: NSFormattingError
|
|
|
|
userInfo: userInfo];
|
|
|
|
}
|
|
|
|
|
2023-11-13 19:45:16 +00:00
|
|
|
DESTROY(self);
|
|
|
|
return self;
|
2011-12-22 12:52:56 +00:00
|
|
|
}
|
|
|
|
options = opts;
|
|
|
|
return self;
|
|
|
|
}
|
|
|
|
|
2016-06-17 07:43:51 +00:00
|
|
|
- (BOOL) isEqual: (id)obj
|
|
|
|
{
|
|
|
|
if ([obj isKindOfClass: [NSRegularExpression class]])
|
|
|
|
{
|
|
|
|
if (self == obj)
|
|
|
|
{
|
|
|
|
return YES;
|
|
|
|
}
|
|
|
|
else if (options != ((NSRegularExpression*)obj)->options)
|
|
|
|
{
|
|
|
|
return NO;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
UErrorCode myErr = 0;
|
|
|
|
UErrorCode theirErr = 0;
|
|
|
|
int32_t myLen = 0;
|
|
|
|
int32_t theirLen = 0;
|
|
|
|
const UChar *myText = uregex_pattern(regex, &myLen, &myErr);
|
|
|
|
const UChar *theirText = uregex_pattern(
|
|
|
|
((NSRegularExpression*)obj)->regex,
|
|
|
|
&theirLen, &theirErr);
|
|
|
|
if (U_FAILURE(myErr) != U_FAILURE(theirErr))
|
|
|
|
{
|
|
|
|
return NO;
|
|
|
|
}
|
|
|
|
else if (U_FAILURE(myErr) && U_FAILURE(theirErr))
|
|
|
|
{
|
|
|
|
return YES;
|
|
|
|
}
|
|
|
|
if (myLen != theirLen)
|
|
|
|
{
|
|
|
|
return NO;
|
|
|
|
}
|
2023-11-13 19:45:16 +00:00
|
|
|
return (0 == memcmp((const void*)myText, (const void*)theirText, myLen));
|
2016-06-17 07:43:51 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
return [super isEqual: obj];
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-12-22 12:52:56 +00:00
|
|
|
- (NSString*) pattern
|
|
|
|
{
|
|
|
|
UErrorCode s = 0;
|
|
|
|
int32_t length;
|
|
|
|
const unichar *pattern = uregex_pattern(regex, &length, &s);
|
2016-06-17 07:43:51 +00:00
|
|
|
|
2011-12-22 12:52:56 +00:00
|
|
|
if (U_FAILURE(s))
|
|
|
|
{
|
|
|
|
return nil;
|
|
|
|
}
|
|
|
|
return [NSString stringWithCharacters: pattern length: length];
|
|
|
|
}
|
|
|
|
#endif
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
|
2016-06-17 07:43:51 +00:00
|
|
|
- (NSUInteger) hash
|
|
|
|
{
|
|
|
|
return [[self pattern] hash] ^ options;
|
|
|
|
}
|
|
|
|
|
2010-12-16 10:59:50 +00:00
|
|
|
static UBool
|
|
|
|
callback(const void *context, int32_t steps)
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2025-01-04 15:05:14 +00:00
|
|
|
BOOL stop = NO;
|
|
|
|
GSRegexContext *c = (GSRegexContext*)context;
|
2010-12-16 10:59:50 +00:00
|
|
|
|
2025-01-04 15:05:14 +00:00
|
|
|
if (NULL == c)
|
2010-12-16 10:59:50 +00:00
|
|
|
{
|
|
|
|
return FALSE;
|
|
|
|
}
|
2025-01-04 16:50:13 +00:00
|
|
|
(*c->h)(c->c, nil, NSMatchingProgress, &stop);
|
2025-01-04 15:05:14 +00:00
|
|
|
|
|
|
|
return (stop ? FALSE : TRUE);
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
2016-06-17 09:04:04 +00:00
|
|
|
|
|
|
|
#define DEFAULT_WORK_LIMIT 1500
|
|
|
|
/**
|
|
|
|
* The work limit specifies the number of iterations the matcher will do before
|
|
|
|
* aborting an operation. This ensures that degenerate pattern/input
|
|
|
|
* combinations don't send the application into what for all intents and
|
|
|
|
* purposes seems like an infinite loop.
|
|
|
|
*/
|
|
|
|
static int32_t _workLimit = DEFAULT_WORK_LIMIT;
|
|
|
|
|
|
|
|
+ (void) _defaultsChanged: (NSNotification*)n
|
|
|
|
{
|
2023-11-13 19:45:16 +00:00
|
|
|
NSUserDefaults *defs = [NSUserDefaults standardUserDefaults];
|
2025-01-04 15:05:14 +00:00
|
|
|
id value;
|
2023-11-13 19:45:16 +00:00
|
|
|
int32_t newLimit = DEFAULT_WORK_LIMIT;
|
|
|
|
|
2025-01-04 15:05:14 +00:00
|
|
|
value = [defs objectForKey: @"GSRegularExpressionWorkLimit"];
|
2016-06-17 09:04:04 +00:00
|
|
|
if ([value respondsToSelector: @selector(intValue)])
|
|
|
|
{
|
2023-11-13 19:45:16 +00:00
|
|
|
int32_t v = [value intValue];
|
|
|
|
|
2016-06-17 09:04:04 +00:00
|
|
|
if (v >= 0)
|
|
|
|
{
|
|
|
|
newLimit = v;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
_workLimit = newLimit;
|
|
|
|
}
|
|
|
|
|
|
|
|
+ (void) initialize
|
|
|
|
{
|
|
|
|
if (self == [NSRegularExpression class])
|
|
|
|
{
|
|
|
|
[[NSNotificationCenter defaultCenter]
|
|
|
|
addObserver: self
|
|
|
|
selector: @selector(_defaultsChanged:)
|
|
|
|
name: NSUserDefaultsDidChangeNotification
|
|
|
|
object: nil];
|
|
|
|
[self _defaultsChanged: nil];
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
/**
|
|
|
|
* Sets up a libicu regex object for use. Note: the documentation states that
|
|
|
|
* NSRegularExpression must be thread safe. To accomplish this, we store a
|
|
|
|
* prototype URegularExpression in the object, and then clone it in each
|
|
|
|
* method. This is required because URegularExpression, unlike
|
|
|
|
* NSRegularExpression, is stateful, and sharing this state between threads
|
|
|
|
* would break concurrent calls.
|
|
|
|
*/
|
2011-12-22 12:52:56 +00:00
|
|
|
#if HAVE_UREGEX_OPENUTEXT
|
2010-12-16 10:59:50 +00:00
|
|
|
static URegularExpression *
|
|
|
|
setupRegex(URegularExpression *regex,
|
|
|
|
NSString *string,
|
|
|
|
UText *txt,
|
|
|
|
NSMatchingOptions options,
|
|
|
|
NSRange range,
|
2025-01-04 15:05:14 +00:00
|
|
|
GSRegexContext *ctx)
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
UErrorCode s = 0;
|
|
|
|
URegularExpression *r = uregex_clone(regex, &s);
|
|
|
|
|
|
|
|
if (options & NSMatchingReportProgress)
|
|
|
|
{
|
2025-01-04 15:05:14 +00:00
|
|
|
uregex_setMatchCallback(r, callback, ctx, &s);
|
|
|
|
if (U_FAILURE(s)) NSLog(@"uregex_setMatchCallback() failed");
|
2010-12-16 10:59:50 +00:00
|
|
|
}
|
|
|
|
UTextInitWithNSString(txt, string);
|
|
|
|
uregex_setUText(r, txt, &s);
|
|
|
|
uregex_setRegion(r, range.location, range.location+range.length, &s);
|
|
|
|
if (options & NSMatchingWithoutAnchoringBounds)
|
|
|
|
{
|
|
|
|
uregex_useAnchoringBounds(r, FALSE, &s);
|
|
|
|
}
|
|
|
|
if (options & NSMatchingWithTransparentBounds)
|
|
|
|
{
|
|
|
|
uregex_useTransparentBounds(r, TRUE, &s);
|
|
|
|
}
|
2016-06-17 09:04:04 +00:00
|
|
|
uregex_setTimeLimit(r, _workLimit, &s);
|
2010-12-16 10:59:50 +00:00
|
|
|
if (U_FAILURE(s))
|
|
|
|
{
|
|
|
|
uregex_close(r);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
return r;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2011-12-22 12:52:56 +00:00
|
|
|
#else
|
|
|
|
static URegularExpression *
|
|
|
|
setupRegex(URegularExpression *regex,
|
|
|
|
NSString *string,
|
|
|
|
unichar *buffer,
|
|
|
|
int32_t length,
|
|
|
|
NSMatchingOptions options,
|
|
|
|
NSRange range,
|
2025-01-04 15:05:14 +00:00
|
|
|
GSRegexContext *ctx)
|
2011-12-22 12:52:56 +00:00
|
|
|
{
|
|
|
|
UErrorCode s = 0;
|
|
|
|
URegularExpression *r = uregex_clone(regex, &s);
|
|
|
|
|
|
|
|
[string getCharacters: buffer range: NSMakeRange(0, length)];
|
|
|
|
if (options & NSMatchingReportProgress)
|
|
|
|
{
|
2025-01-04 15:05:14 +00:00
|
|
|
uregex_setMatchCallback(r, callback, ctx, &s);
|
|
|
|
if (U_FAILURE(s)) NSLog(@"uregex_setMatchCallback() failed");
|
2011-12-22 12:52:56 +00:00
|
|
|
}
|
|
|
|
uregex_setText(r, buffer, length, &s);
|
|
|
|
uregex_setRegion(r, range.location, range.location+range.length, &s);
|
|
|
|
if (options & NSMatchingWithoutAnchoringBounds)
|
|
|
|
{
|
|
|
|
uregex_useAnchoringBounds(r, FALSE, &s);
|
|
|
|
}
|
|
|
|
if (options & NSMatchingWithTransparentBounds)
|
|
|
|
{
|
|
|
|
uregex_useTransparentBounds(r, TRUE, &s);
|
|
|
|
}
|
2016-06-17 09:04:04 +00:00
|
|
|
uregex_setTimeLimit(r, _workLimit, &s);
|
2011-12-22 12:52:56 +00:00
|
|
|
if (U_FAILURE(s))
|
|
|
|
{
|
|
|
|
uregex_close(r);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
return r;
|
|
|
|
}
|
|
|
|
#endif
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
static uint32_t
|
|
|
|
prepareResult(NSRegularExpression *regex,
|
|
|
|
URegularExpression *r,
|
|
|
|
NSRangePointer ranges,
|
|
|
|
NSUInteger groups,
|
|
|
|
UErrorCode *s)
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
uint32_t flags = 0;
|
|
|
|
NSUInteger i = 0;
|
|
|
|
|
|
|
|
for (i = 0; i < groups; i++)
|
|
|
|
{
|
2015-09-01 11:31:16 +00:00
|
|
|
NSInteger start = uregex_start(r, i, s);
|
|
|
|
NSInteger end = uregex_end(r, i, s);
|
2023-11-13 19:45:16 +00:00
|
|
|
|
|
|
|
/* The ICU API defines -1 as not found. Convert to
|
|
|
|
* NSNotFound if applicable.
|
|
|
|
*/
|
2015-09-01 11:31:16 +00:00
|
|
|
if (start == -1)
|
|
|
|
{
|
|
|
|
start = NSNotFound;
|
|
|
|
}
|
|
|
|
if (end == -1)
|
|
|
|
{
|
|
|
|
end = NSNotFound;
|
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
2014-04-12 04:38:13 +00:00
|
|
|
if (end < start)
|
|
|
|
{
|
|
|
|
flags |= NSMatchingInternalError;
|
|
|
|
end = start = NSNotFound;
|
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
ranges[i] = NSMakeRange(start, end-start);
|
|
|
|
}
|
|
|
|
if (uregex_hitEnd(r, s))
|
|
|
|
{
|
|
|
|
flags |= NSMatchingHitEnd;
|
|
|
|
}
|
|
|
|
if (uregex_requireEnd(r, s))
|
|
|
|
{
|
|
|
|
flags |= NSMatchingRequiredEnd;
|
|
|
|
}
|
|
|
|
if (0 != *s)
|
|
|
|
{
|
|
|
|
flags |= NSMatchingInternalError;
|
|
|
|
}
|
|
|
|
return flags;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
|
|
|
|
2025-01-04 15:05:14 +00:00
|
|
|
|
2011-12-22 12:52:56 +00:00
|
|
|
#if HAVE_UREGEX_OPENUTEXT
|
2010-12-16 10:59:50 +00:00
|
|
|
- (void) enumerateMatchesInString: (NSString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
2025-01-04 15:05:14 +00:00
|
|
|
callback: (GSRegexEnumerationCallback)handler
|
|
|
|
context: (void*)context
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2023-11-13 19:45:16 +00:00
|
|
|
UErrorCode s = 0;
|
|
|
|
UText txt = UTEXT_INITIALIZER;
|
|
|
|
BOOL stop = NO;
|
2025-01-04 16:50:13 +00:00
|
|
|
GSRegexContext ctx = { handler, context };
|
2025-01-04 15:05:14 +00:00
|
|
|
URegularExpression *r = setupRegex(regex, string, &txt, opts, range, &ctx);
|
2023-11-13 19:45:16 +00:00
|
|
|
NSUInteger groups = [self numberOfCaptureGroups] + 1;
|
|
|
|
NSRange ranges[groups];
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
// Should this throw some kind of exception?
|
|
|
|
if (NULL == r)
|
|
|
|
{
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
if (opts & NSMatchingAnchored)
|
|
|
|
{
|
|
|
|
if (uregex_lookingAt(r, -1, &s) && (0 == s))
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
// FIXME: Factor all of this out into prepareResult()
|
|
|
|
uint32_t flags;
|
|
|
|
NSTextCheckingResult *result;
|
|
|
|
|
|
|
|
flags = prepareResult(self, r, ranges, groups, &s);
|
2014-04-12 04:38:13 +00:00
|
|
|
result = (flags & NSMatchingInternalError) ? nil
|
|
|
|
: [NSTextCheckingResult
|
2010-12-16 10:59:50 +00:00
|
|
|
regularExpressionCheckingResultWithRanges: ranges
|
|
|
|
count: groups
|
|
|
|
regularExpression: self];
|
2025-01-04 16:50:13 +00:00
|
|
|
(*handler)(context, result, flags, &stop);
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
while (!stop && uregex_findNext(r, &s) && (0 == s))
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
uint32_t flags;
|
|
|
|
NSTextCheckingResult *result;
|
|
|
|
|
|
|
|
flags = prepareResult(self, r, ranges, groups, &s);
|
2014-04-12 04:38:13 +00:00
|
|
|
result = (flags & NSMatchingInternalError) ? nil
|
|
|
|
: [NSTextCheckingResult
|
2010-12-16 10:59:50 +00:00
|
|
|
regularExpressionCheckingResultWithRanges: ranges
|
|
|
|
count: groups
|
|
|
|
regularExpression: self];
|
2025-01-04 16:50:13 +00:00
|
|
|
(*handler)(context, result, flags, &stop);
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
}
|
|
|
|
if (opts & NSMatchingCompleted)
|
|
|
|
{
|
2025-01-04 16:50:13 +00:00
|
|
|
(*handler)(context, nil, NSMatchingCompleted, &stop);
|
2010-12-16 10:59:50 +00:00
|
|
|
}
|
|
|
|
utext_close(&txt);
|
|
|
|
uregex_close(r);
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2011-12-22 12:52:56 +00:00
|
|
|
#else
|
|
|
|
- (void) enumerateMatchesInString: (NSString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
2025-01-04 15:05:14 +00:00
|
|
|
callback: (GSRegexEnumerationCallback)handler
|
|
|
|
context: (void*)context
|
2011-12-22 12:52:56 +00:00
|
|
|
{
|
2023-11-13 19:45:16 +00:00
|
|
|
UErrorCode s = 0;
|
|
|
|
BOOL stop = NO;
|
|
|
|
int32_t length = [string length];
|
|
|
|
URegularExpression *r;
|
|
|
|
NSUInteger groups = [self numberOfCaptureGroups] + 1;
|
|
|
|
NSRange ranges[groups];
|
2025-01-04 16:50:13 +00:00
|
|
|
GSRegexContext ctx = { handler, context };
|
2011-12-22 12:52:56 +00:00
|
|
|
TEMP_BUFFER(buffer, length);
|
|
|
|
|
2025-01-04 15:05:14 +00:00
|
|
|
r = setupRegex(regex, string, buffer, length, opts, range, &ctx);
|
2011-12-22 12:52:56 +00:00
|
|
|
|
|
|
|
// Should this throw some kind of exception?
|
|
|
|
if (NULL == r)
|
|
|
|
{
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
if (opts & NSMatchingAnchored)
|
|
|
|
{
|
|
|
|
if (uregex_lookingAt(r, -1, &s) && (0 == s))
|
|
|
|
{
|
|
|
|
// FIXME: Factor all of this out into prepareResult()
|
|
|
|
uint32_t flags;
|
|
|
|
NSTextCheckingResult *result;
|
|
|
|
|
|
|
|
flags = prepareResult(self, r, ranges, groups, &s);
|
2014-04-12 04:38:13 +00:00
|
|
|
result = (flags & NSMatchingInternalError) ? nil
|
|
|
|
: [NSTextCheckingResult
|
2011-12-22 12:52:56 +00:00
|
|
|
regularExpressionCheckingResultWithRanges: ranges
|
|
|
|
count: groups
|
|
|
|
regularExpression: self];
|
2025-01-04 16:50:13 +00:00
|
|
|
(*handler)(context, result, flags, &stop);
|
2011-12-22 12:52:56 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
while (!stop && uregex_findNext(r, &s) && (0 == s))
|
|
|
|
{
|
|
|
|
uint32_t flags;
|
|
|
|
NSTextCheckingResult *result;
|
|
|
|
|
|
|
|
flags = prepareResult(self, r, ranges, groups, &s);
|
2014-04-12 04:38:13 +00:00
|
|
|
result = (flags & NSMatchingInternalError) ? nil
|
|
|
|
: [NSTextCheckingResult
|
2011-12-22 12:52:56 +00:00
|
|
|
regularExpressionCheckingResultWithRanges: ranges
|
|
|
|
count: groups
|
|
|
|
regularExpression: self];
|
2025-01-04 16:50:13 +00:00
|
|
|
(*handler)(context, result, flags, &stop);
|
2011-12-22 12:52:56 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
if (opts & NSMatchingCompleted)
|
|
|
|
{
|
2025-01-04 16:50:13 +00:00
|
|
|
(*handler)(context, nil, NSMatchingCompleted, &stop);
|
2011-12-22 12:52:56 +00:00
|
|
|
}
|
|
|
|
uregex_close(r);
|
|
|
|
}
|
|
|
|
#endif
|
2010-12-16 10:59:50 +00:00
|
|
|
|
2025-01-04 15:05:14 +00:00
|
|
|
|
|
|
|
- (void) enumerateMatchesInString: (NSString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
|
|
|
usingBlock: (GSRegexBlock)block
|
|
|
|
{
|
|
|
|
[self enumerateMatchesInString: string
|
|
|
|
options: opts
|
|
|
|
range: range
|
|
|
|
callback: blockCallback
|
|
|
|
context: (void*)block];
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2010-12-16 10:59:50 +00:00
|
|
|
/* The remaining methods are all meant to be wrappers around the primitive
|
|
|
|
* method that takes a block argument. Unfortunately, this is not really
|
|
|
|
* possible when compiling with a compiler that doesn't support blocks.
|
|
|
|
*/
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
#if __has_feature(blocks)
|
2010-12-16 10:59:50 +00:00
|
|
|
- (NSUInteger) numberOfMatchesInString: (NSString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
|
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
__block NSUInteger count = 0;
|
2024-05-12 09:03:15 +00:00
|
|
|
GSRegexBlock block;
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
opts &= ~NSMatchingReportProgress;
|
|
|
|
opts &= ~NSMatchingReportCompletion;
|
|
|
|
|
2024-05-12 09:03:15 +00:00
|
|
|
block =
|
2010-12-16 10:59:50 +00:00
|
|
|
^(NSTextCheckingResult *result, NSMatchingFlags flags, BOOL *stop)
|
|
|
|
{
|
|
|
|
count++;
|
|
|
|
};
|
|
|
|
[self enumerateMatchesInString: string
|
|
|
|
options: opts
|
|
|
|
range: range
|
|
|
|
usingBlock: block];
|
|
|
|
return count;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
- (NSTextCheckingResult*) firstMatchInString: (NSString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2024-05-12 09:03:15 +00:00
|
|
|
__block NSTextCheckingResult *r = nil;
|
|
|
|
GSRegexBlock block;
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
opts &= ~NSMatchingReportProgress;
|
|
|
|
opts &= ~NSMatchingReportCompletion;
|
|
|
|
|
2024-05-12 09:03:15 +00:00
|
|
|
block =
|
2010-12-16 10:59:50 +00:00
|
|
|
^(NSTextCheckingResult *result, NSMatchingFlags flags, BOOL *stop)
|
|
|
|
{
|
|
|
|
r = result;
|
|
|
|
*stop = YES;
|
|
|
|
};
|
|
|
|
[self enumerateMatchesInString: string
|
|
|
|
options: opts
|
|
|
|
range: range
|
|
|
|
usingBlock: block];
|
|
|
|
return r;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
- (NSArray*) matchesInString: (NSString*)string
|
|
|
|
options:(NSMatchingOptions)opts
|
|
|
|
range:(NSRange)range
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
NSMutableArray *array = [NSMutableArray array];
|
2024-05-12 09:03:15 +00:00
|
|
|
GSRegexBlock block;
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
opts &= ~NSMatchingReportProgress;
|
|
|
|
opts &= ~NSMatchingReportCompletion;
|
|
|
|
|
2024-05-12 09:03:15 +00:00
|
|
|
block =
|
2010-12-16 10:59:50 +00:00
|
|
|
^(NSTextCheckingResult *result, NSMatchingFlags flags, BOOL *stop)
|
|
|
|
{
|
|
|
|
[array addObject: result];
|
|
|
|
};
|
|
|
|
[self enumerateMatchesInString: string
|
|
|
|
options: opts
|
|
|
|
range: range
|
|
|
|
usingBlock: block];
|
|
|
|
return array;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
- (NSRange) rangeOfFirstMatchInString: (NSString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2024-05-12 09:03:15 +00:00
|
|
|
__block NSRange r = {NSNotFound, 0};
|
|
|
|
GSRegexBlock block;
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
opts &= ~NSMatchingReportProgress;
|
|
|
|
opts &= ~NSMatchingReportCompletion;
|
|
|
|
|
2024-05-12 09:03:15 +00:00
|
|
|
block =
|
2010-12-16 10:59:50 +00:00
|
|
|
^(NSTextCheckingResult *result, NSMatchingFlags flags, BOOL *stop)
|
|
|
|
{
|
|
|
|
r = [result range];
|
|
|
|
*stop = YES;
|
|
|
|
};
|
|
|
|
[self enumerateMatchesInString: string
|
|
|
|
options: opts
|
|
|
|
range: range
|
|
|
|
usingBlock: block];
|
|
|
|
return r;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
#else
|
2013-02-28 16:50:37 +00:00
|
|
|
# ifdef __clang__ /* FIXME ... this is blocks specific, not clang specific */
|
2012-03-27 10:03:40 +00:00
|
|
|
# warning Your compiler does not support blocks. NSRegularExpression will deviate from the documented behaviour when subclassing and any code that subclasses NSRegularExpression may break in unexpected ways. If you must subclass NSRegularExpression, you may want to use a compiler with blocks support.
|
|
|
|
# warning Your compiler would support blocks if you added -fblocks to your OBJCFLAGS
|
|
|
|
# endif
|
2011-12-22 12:52:56 +00:00
|
|
|
#if HAVE_UREGEX_OPENUTEXT
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
#define FAKE_BLOCK_HACK(failRet, code) \
|
2023-11-13 19:45:16 +00:00
|
|
|
UErrorCode s = 0;\
|
|
|
|
UText txt = UTEXT_INITIALIZER;\
|
|
|
|
BOOL stop = NO;\
|
|
|
|
URegularExpression *r = setupRegex(regex, string, &txt, opts, range, 0);\
|
|
|
|
\
|
2010-12-16 10:59:50 +00:00
|
|
|
if (NULL == r) { return failRet; }\
|
|
|
|
if (opts & NSMatchingAnchored)\
|
|
|
|
{\
|
|
|
|
if (uregex_lookingAt(r, -1, &s) && (0==s))\
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{\
|
2010-12-16 10:59:50 +00:00
|
|
|
code\
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}\
|
2010-12-16 10:59:50 +00:00
|
|
|
}\
|
|
|
|
else\
|
|
|
|
{\
|
|
|
|
while (!stop && uregex_findNext(r, &s) && (s == 0))\
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{\
|
2010-12-16 10:59:50 +00:00
|
|
|
code\
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}\
|
2010-12-16 10:59:50 +00:00
|
|
|
}\
|
|
|
|
utext_close(&txt);\
|
|
|
|
uregex_close(r);
|
2011-12-22 12:52:56 +00:00
|
|
|
#else
|
|
|
|
#define FAKE_BLOCK_HACK(failRet, code) \
|
|
|
|
UErrorCode s = 0;\
|
|
|
|
BOOL stop = NO;\
|
|
|
|
uint32_t length = [string length];\
|
|
|
|
URegularExpression *r;\
|
|
|
|
TEMP_BUFFER(buffer, length);\
|
|
|
|
r = setupRegex(regex, string, buffer, length, opts, range, 0);\
|
|
|
|
if (NULL == r) { return failRet; }\
|
|
|
|
if (opts & NSMatchingAnchored)\
|
|
|
|
{\
|
|
|
|
if (uregex_lookingAt(r, -1, &s) && (0==s))\
|
|
|
|
{\
|
|
|
|
code\
|
|
|
|
}\
|
|
|
|
}\
|
|
|
|
else\
|
|
|
|
{\
|
|
|
|
while (!stop && uregex_findNext(r, &s) && (s == 0))\
|
|
|
|
{\
|
|
|
|
code\
|
|
|
|
}\
|
|
|
|
}\
|
|
|
|
uregex_close(r);
|
|
|
|
#endif
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
- (NSUInteger) numberOfMatchesInString: (NSString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
|
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
NSUInteger count = 0;
|
|
|
|
|
|
|
|
FAKE_BLOCK_HACK(count,
|
|
|
|
{
|
|
|
|
count++;
|
|
|
|
});
|
|
|
|
return count;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
- (NSTextCheckingResult*) firstMatchInString: (NSString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
NSTextCheckingResult *result = nil;
|
|
|
|
NSUInteger groups = [self numberOfCaptureGroups] + 1;
|
|
|
|
NSRange ranges[groups];
|
|
|
|
|
|
|
|
FAKE_BLOCK_HACK(result,
|
|
|
|
{
|
2014-04-12 04:38:13 +00:00
|
|
|
uint32_t flags;
|
|
|
|
|
|
|
|
flags = prepareResult(self, r, ranges, groups, &s);
|
|
|
|
result = (flags & NSMatchingInternalError) ? nil
|
|
|
|
: [NSTextCheckingResult
|
2010-12-16 10:59:50 +00:00
|
|
|
regularExpressionCheckingResultWithRanges: ranges
|
|
|
|
count: groups
|
|
|
|
regularExpression: self];
|
|
|
|
stop = YES;
|
|
|
|
});
|
|
|
|
return result;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
- (NSArray*) matchesInString: (NSString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
NSMutableArray *array = [NSMutableArray array];
|
|
|
|
NSUInteger groups = [self numberOfCaptureGroups] + 1;
|
|
|
|
NSRange ranges[groups];
|
|
|
|
|
|
|
|
FAKE_BLOCK_HACK(array,
|
|
|
|
{
|
|
|
|
NSTextCheckingResult *result = NULL;
|
2014-04-12 04:38:13 +00:00
|
|
|
uint32_t flags;
|
2010-12-16 10:59:50 +00:00
|
|
|
|
2014-04-12 04:38:13 +00:00
|
|
|
flags = prepareResult(self, r, ranges, groups, &s);
|
|
|
|
result = (flags & NSMatchingInternalError) ? nil
|
|
|
|
: [NSTextCheckingResult
|
2010-12-16 10:59:50 +00:00
|
|
|
regularExpressionCheckingResultWithRanges: ranges
|
|
|
|
count: groups
|
|
|
|
regularExpression: self];
|
2014-04-12 04:38:13 +00:00
|
|
|
if (nil != result)
|
|
|
|
{
|
|
|
|
[array addObject: result];
|
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
});
|
|
|
|
return array;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
- (NSRange) rangeOfFirstMatchInString: (NSString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2011-12-23 14:59:14 +00:00
|
|
|
NSRange result = {NSNotFound, 0};
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
FAKE_BLOCK_HACK(result,
|
|
|
|
{
|
|
|
|
prepareResult(self, r, &result, 1, &s);
|
|
|
|
stop = YES;
|
|
|
|
});
|
|
|
|
return result;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
#endif
|
2010-12-16 10:59:50 +00:00
|
|
|
|
2011-12-22 12:52:56 +00:00
|
|
|
#if HAVE_UREGEX_OPENUTEXT
|
2010-12-16 10:59:50 +00:00
|
|
|
- (NSUInteger) replaceMatchesInString: (NSMutableString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
|
|
|
withTemplate: (NSString*)template
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
// FIXME: We're computing a value that is most likely ignored in an
|
|
|
|
// expensive way.
|
2024-11-21 21:05:34 +00:00
|
|
|
NSInteger results = [self numberOfMatchesInString: string
|
|
|
|
options: opts
|
|
|
|
range: range];
|
|
|
|
UErrorCode s = 0;
|
|
|
|
UText txt = UTEXT_INITIALIZER;
|
|
|
|
UText replacement = UTEXT_INITIALIZER;
|
|
|
|
GSUTextString *ret = [GSUTextString new];
|
|
|
|
URegularExpression *r = setupRegex(regex, string, &txt, opts, range, 0);
|
|
|
|
UText *output = NULL;
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
UTextInitWithNSString(&replacement, template);
|
|
|
|
|
2024-11-21 21:05:34 +00:00
|
|
|
output = uregex_replaceAllUText(r, &replacement, NULL, &s);
|
2016-10-12 13:43:32 +00:00
|
|
|
if (0 != s)
|
|
|
|
{
|
2024-11-21 21:05:34 +00:00
|
|
|
uregex_close(r);
|
|
|
|
utext_close(&replacement);
|
|
|
|
utext_close(&txt);
|
|
|
|
DESTROY(ret);
|
2016-10-12 13:43:32 +00:00
|
|
|
return 0;
|
|
|
|
}
|
2024-11-21 21:05:34 +00:00
|
|
|
utext_clone(&ret->txt, output, TRUE, TRUE, &s);
|
|
|
|
[string setString: ret];
|
|
|
|
RELEASE(ret);
|
|
|
|
uregex_close(r);
|
|
|
|
|
|
|
|
utext_close(&txt);
|
|
|
|
utext_close(output);
|
|
|
|
utext_close(&replacement);
|
2010-12-16 10:59:50 +00:00
|
|
|
return results;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
|
|
|
|
2010-12-16 10:59:50 +00:00
|
|
|
- (NSString*) stringByReplacingMatchesInString: (NSString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
|
|
|
withTemplate: (NSString*)template
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2024-11-21 21:05:34 +00:00
|
|
|
UErrorCode s = 0;
|
|
|
|
UText txt = UTEXT_INITIALIZER;
|
|
|
|
UText replacement = UTEXT_INITIALIZER;
|
|
|
|
UText *output = NULL;
|
|
|
|
GSUTextString *ret = [GSUTextString new];
|
|
|
|
URegularExpression *r = setupRegex(regex, string, &txt, opts, range, 0);
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
UTextInitWithNSString(&replacement, template);
|
|
|
|
|
2024-11-21 21:05:34 +00:00
|
|
|
output = uregex_replaceAllUText(r, &replacement, NULL, &s);
|
2016-10-12 13:43:32 +00:00
|
|
|
if (0 != s)
|
|
|
|
{
|
2024-11-21 21:05:34 +00:00
|
|
|
uregex_close(r);
|
|
|
|
utext_close(&replacement);
|
|
|
|
utext_close(&txt);
|
|
|
|
DESTROY(ret);
|
2016-10-12 13:43:32 +00:00
|
|
|
return nil;
|
|
|
|
}
|
2024-11-21 21:05:34 +00:00
|
|
|
utext_clone(&ret->txt, output, TRUE, TRUE, &s);
|
|
|
|
uregex_close(r);
|
|
|
|
|
|
|
|
utext_close(&txt);
|
|
|
|
utext_close(output);
|
|
|
|
utext_close(&replacement);
|
|
|
|
return AUTORELEASE(ret);
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
|
|
|
|
2010-12-16 10:59:50 +00:00
|
|
|
- (NSString*) replacementStringForResult: (NSTextCheckingResult*)result
|
|
|
|
inString: (NSString*)string
|
|
|
|
offset: (NSInteger)offset
|
|
|
|
template: (NSString*)template
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2024-11-21 21:05:34 +00:00
|
|
|
UErrorCode s = 0;
|
|
|
|
UText txt = UTEXT_INITIALIZER;
|
|
|
|
UText replacement = UTEXT_INITIALIZER;
|
|
|
|
UText *output = NULL;
|
|
|
|
GSUTextString *ret = [GSUTextString new];
|
|
|
|
NSRange range = [result range];
|
|
|
|
URegularExpression *r = setupRegex(regex,
|
2010-12-16 10:59:50 +00:00
|
|
|
[string substringWithRange: range],
|
|
|
|
&txt,
|
|
|
|
0,
|
|
|
|
NSMakeRange(0, range.length),
|
|
|
|
0);
|
|
|
|
|
|
|
|
UTextInitWithNSString(&replacement, template);
|
|
|
|
|
2024-11-21 21:05:34 +00:00
|
|
|
output = uregex_replaceFirstUText(r, &replacement, NULL, &s);
|
2016-10-12 13:43:32 +00:00
|
|
|
if (0 != s)
|
|
|
|
{
|
2024-11-21 21:05:34 +00:00
|
|
|
uregex_close(r);
|
|
|
|
utext_close(&replacement);
|
|
|
|
utext_close(&txt);
|
|
|
|
DESTROY(ret);
|
2016-10-12 13:43:32 +00:00
|
|
|
return nil;
|
|
|
|
}
|
2024-11-21 21:05:34 +00:00
|
|
|
utext_clone(&ret->txt, output, TRUE, TRUE, &s);
|
|
|
|
utext_close(output);
|
|
|
|
uregex_close(r);
|
|
|
|
utext_close(&txt);
|
|
|
|
utext_close(&replacement);
|
|
|
|
return AUTORELEASE(ret);
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2011-12-22 12:52:56 +00:00
|
|
|
#else
|
|
|
|
- (NSUInteger) replaceMatchesInString: (NSMutableString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
|
|
|
withTemplate: (NSString*)template
|
|
|
|
{
|
|
|
|
// FIXME: We're computing a value that is most likely ignored in an
|
|
|
|
// expensive way.
|
|
|
|
NSInteger results = [self numberOfMatchesInString: string
|
|
|
|
options: opts
|
|
|
|
range: range];
|
2024-11-21 21:05:34 +00:00
|
|
|
if (results > 0)
|
|
|
|
{
|
|
|
|
UErrorCode s = 0;
|
|
|
|
uint32_t length = [string length];
|
|
|
|
uint32_t replLength = [template length];
|
|
|
|
unichar replacement[replLength];
|
|
|
|
int32_t outLength;
|
|
|
|
URegularExpression *r;
|
|
|
|
TEMP_BUFFER(buffer, length);
|
|
|
|
|
|
|
|
r = setupRegex(regex, string, buffer, length, opts, range, 0);
|
|
|
|
[template getCharacters: replacement range: NSMakeRange(0, replLength)];
|
|
|
|
|
|
|
|
outLength = uregex_replaceAll(r, replacement, replLength, NULL, 0, &s);
|
|
|
|
if (0 == s || U_BUFFER_OVERFLOW_ERROR == s)
|
|
|
|
{
|
|
|
|
unichar *output;
|
|
|
|
|
|
|
|
s = 0; // May have been set to a buffer overflow error
|
|
|
|
|
|
|
|
output = NSZoneMalloc(0, (outLength + 1) * sizeof(unichar));
|
|
|
|
uregex_replaceAll(r, replacement, replLength,
|
|
|
|
output, outLength + 1, &s);
|
|
|
|
if (0 == s)
|
|
|
|
{
|
|
|
|
NSString *out;
|
|
|
|
|
|
|
|
out = [[NSString alloc] initWithCharactersNoCopy: output
|
|
|
|
length: outLength
|
|
|
|
freeWhenDone: YES];
|
|
|
|
[string setString: out];
|
|
|
|
RELEASE(out);
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
NSZoneFree(0, output);
|
|
|
|
results = 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
results = 0;
|
|
|
|
}
|
|
|
|
uregex_close(r);
|
|
|
|
}
|
2011-12-22 12:52:56 +00:00
|
|
|
return results;
|
|
|
|
}
|
|
|
|
|
|
|
|
- (NSString*) stringByReplacingMatchesInString: (NSString*)string
|
|
|
|
options: (NSMatchingOptions)opts
|
|
|
|
range: (NSRange)range
|
|
|
|
withTemplate: (NSString*)template
|
|
|
|
{
|
|
|
|
UErrorCode s = 0;
|
|
|
|
uint32_t length = [string length];
|
|
|
|
URegularExpression *r;
|
|
|
|
uint32_t replLength = [template length];
|
|
|
|
unichar replacement[replLength];
|
|
|
|
int32_t outLength;
|
2024-11-21 21:05:34 +00:00
|
|
|
NSString *result = nil;
|
2011-12-22 12:52:56 +00:00
|
|
|
TEMP_BUFFER(buffer, length);
|
|
|
|
|
|
|
|
r = setupRegex(regex, string, buffer, length, opts, range, 0);
|
|
|
|
[template getCharacters: replacement range: NSMakeRange(0, replLength)];
|
|
|
|
|
|
|
|
outLength = uregex_replaceAll(r, replacement, replLength, NULL, 0, &s);
|
2024-11-21 21:05:34 +00:00
|
|
|
if (0 == s || U_BUFFER_OVERFLOW_ERROR == s)
|
|
|
|
{
|
|
|
|
unichar *output;
|
|
|
|
|
|
|
|
s = 0; // may have been set to a buffer overflow error
|
|
|
|
|
|
|
|
output = NSZoneMalloc(0, (outLength + 1) * sizeof(unichar));
|
|
|
|
uregex_replaceAll(r, replacement, replLength, output, outLength + 1, &s);
|
|
|
|
if (0 == s)
|
|
|
|
{
|
|
|
|
result = AUTORELEASE([[NSString alloc]
|
|
|
|
initWithCharactersNoCopy: output
|
|
|
|
length: outLength
|
|
|
|
freeWhenDone: YES]);
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
NSZoneFree(0, output);
|
|
|
|
}
|
|
|
|
}
|
2011-12-22 12:52:56 +00:00
|
|
|
|
2024-11-21 21:05:34 +00:00
|
|
|
uregex_close(r);
|
|
|
|
return result;
|
2011-12-22 12:52:56 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
- (NSString*) replacementStringForResult: (NSTextCheckingResult*)result
|
|
|
|
inString: (NSString*)string
|
|
|
|
offset: (NSInteger)offset
|
|
|
|
template: (NSString*)template
|
|
|
|
{
|
|
|
|
UErrorCode s = 0;
|
|
|
|
NSRange range = [result range];
|
|
|
|
URegularExpression *r;
|
|
|
|
uint32_t replLength = [template length];
|
|
|
|
unichar replacement[replLength];
|
|
|
|
int32_t outLength;
|
2024-11-21 21:05:34 +00:00
|
|
|
NSString *str = nil;
|
2011-12-22 12:52:56 +00:00
|
|
|
TEMP_BUFFER(buffer, range.length);
|
|
|
|
|
|
|
|
r = setupRegex(regex,
|
|
|
|
[string substringWithRange: range],
|
|
|
|
buffer,
|
|
|
|
range.length,
|
|
|
|
0,
|
|
|
|
NSMakeRange(0, range.length),
|
|
|
|
0);
|
|
|
|
[template getCharacters: replacement range: NSMakeRange(0, replLength)];
|
|
|
|
|
|
|
|
outLength = uregex_replaceFirst(r, replacement, replLength, NULL, 0, &s);
|
2024-11-21 21:05:34 +00:00
|
|
|
if (0 == s || U_BUFFER_OVERFLOW_ERROR == s)
|
|
|
|
{
|
|
|
|
unichar *output;
|
|
|
|
|
|
|
|
s = 0;
|
|
|
|
output = NSZoneMalloc(0, (outLength + 1) * sizeof(unichar));
|
|
|
|
uregex_replaceFirst(r, replacement, replLength,
|
|
|
|
output, outLength + 1, &s);
|
|
|
|
if (0 == s)
|
|
|
|
{
|
|
|
|
str = AUTORELEASE([[NSString alloc]
|
|
|
|
initWithCharactersNoCopy: output
|
|
|
|
length: outLength
|
|
|
|
freeWhenDone: YES]);
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
NSZoneFree(0, output);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
uregex_close(r);
|
|
|
|
return str;
|
2011-12-22 12:52:56 +00:00
|
|
|
}
|
|
|
|
#endif
|
2010-12-16 10:59:50 +00:00
|
|
|
|
2024-05-12 09:03:15 +00:00
|
|
|
+ (NSString*) escapedPatternForString: (NSString *)string
|
|
|
|
{
|
|
|
|
/* https://unicode-org.github.io/icu/userguide/strings/regexp.html
|
|
|
|
* Need to escape * ? + [ ( ) { } ^ $ | \ .
|
|
|
|
*/
|
2023-01-04 02:33:47 +00:00
|
|
|
return [[NSRegularExpression
|
|
|
|
regularExpressionWithPattern: @"([*?+\\[(){}^$|\\\\.])"
|
|
|
|
options: 0
|
|
|
|
error: NULL]
|
|
|
|
stringByReplacingMatchesInString: string
|
|
|
|
options: 0
|
|
|
|
range: NSMakeRange(0, [string length])
|
|
|
|
withTemplate: @"\\\\$1"
|
|
|
|
];
|
|
|
|
}
|
|
|
|
|
2010-12-16 10:59:50 +00:00
|
|
|
- (NSRegularExpressionOptions) options
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
return options;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
- (NSUInteger) numberOfCaptureGroups
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
UErrorCode s = 0;
|
|
|
|
return uregex_groupCount(regex, &s);
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
- (void) dealloc
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
uregex_close(regex);
|
|
|
|
[super dealloc];
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
- (void) encodeWithCoder: (NSCoder*)aCoder
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
if ([aCoder allowsKeyedCoding])
|
|
|
|
{
|
|
|
|
[aCoder encodeInteger: options forKey: @"options"];
|
|
|
|
[aCoder encodeObject: [self pattern] forKey: @"pattern"];
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
[aCoder encodeValueOfObjCType: @encode(NSRegularExpressionOptions)
|
|
|
|
at: &options];
|
|
|
|
[aCoder encodeObject: [self pattern]];
|
|
|
|
}
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
- (id) initWithCoder: (NSCoder*)aCoder
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2011-08-03 12:04:27 +00:00
|
|
|
NSString *pattern;
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
if ([aCoder allowsKeyedCoding])
|
|
|
|
{
|
2011-08-03 12:04:27 +00:00
|
|
|
options = [aCoder decodeIntegerForKey: @"options"];
|
|
|
|
pattern = [aCoder decodeObjectForKey: @"pattern"];
|
2010-12-16 10:59:50 +00:00
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
[aCoder decodeValueOfObjCType: @encode(NSRegularExpressionOptions)
|
2011-08-03 12:04:27 +00:00
|
|
|
at: &options];
|
|
|
|
pattern = [aCoder decodeObject];
|
2010-12-16 10:59:50 +00:00
|
|
|
}
|
2011-08-03 12:04:27 +00:00
|
|
|
return [self initWithPattern: pattern options: options error: NULL];
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
2010-12-16 10:59:50 +00:00
|
|
|
|
|
|
|
- (id) copyWithZone: (NSZone*)aZone
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
{
|
2010-12-16 10:59:50 +00:00
|
|
|
NSRegularExpressionOptions opts = options;
|
|
|
|
UErrorCode s = 0;
|
|
|
|
URegularExpression *r = uregex_clone(regex, &s);
|
|
|
|
|
|
|
|
if (0 != s)
|
|
|
|
{
|
|
|
|
return nil;
|
|
|
|
}
|
|
|
|
|
|
|
|
self = [[self class] allocWithZone: aZone];
|
|
|
|
if (nil == self)
|
|
|
|
{
|
|
|
|
return nil;
|
|
|
|
}
|
|
|
|
options = opts;
|
|
|
|
regex = r;
|
|
|
|
return self;
|
Added implementation of NSRegularExpression, from iOS 4 Foundation.
This class is a thin wrapper around libicu regular expressions, so if we don't
have libicu we simply don't compile it at all. This will give people a linker
failure, rather than a nonfunctional class if they try to use GNUstep without
ICU with code that requires it.
The Apple documentation says that this class has a primitive method that takes
a block as an argument and that this method is called by others, so subclasses
can replace that block method without touching the convenience methods. We
mimic this behaviour when compiling with block, but when compiling without them
it's a problem. The current code contains some ugly hacks that will work in
normal usage but break with subclassing when not compiling with blocks.
This commit also includes a partial implementation of NSTextCheckingResult,
implementing the subset of its functionality required for NSRegularExpression
to work.
It also includes numerous fixes to GSICUString. This is heavily used by
NSRegularExpression, to avoid copying strings when mapping between UText for
libicu and NSString for GNUstep.
Note: I don't have a copy of iOS anywhere to test this against, so it's
entirely possible that there are significant discrepancies between this
implementation of NSRegularExpression and the iOS version. This version should
function exactly as the iOS one is described as functioning, but I think we've
all seen that Apple documentation refers more to hopes than facts. Any testing
that someone who does have an ip{hone,od,ad} can do is very welcome.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@31635 72102866-910b-0410-8b05-ffd578937521
2010-11-19 22:06:18 +00:00
|
|
|
}
|
|
|
|
@end
|
2011-12-22 12:52:56 +00:00
|
|
|
#endif //GS_ICU == 1
|
2011-08-03 12:04:27 +00:00
|
|
|
|
|
|
|
#ifndef NSRegularExpressionWorks
|
|
|
|
#import "Foundation/NSRegularExpression.h"
|
|
|
|
#import "Foundation/NSZone.h"
|
|
|
|
#import "Foundation/NSException.h"
|
|
|
|
@implementation NSRegularExpression
|
|
|
|
+ (id)allocWithZone: (NSZone*)aZone
|
|
|
|
{
|
|
|
|
[NSException raise: NSInvalidArgumentException
|
|
|
|
format: @"NSRegularExpression requires ICU 4.4 or later"];
|
|
|
|
return nil;
|
|
|
|
}
|
|
|
|
- (id) copyWithZone: (NSZone*)zone
|
|
|
|
{
|
|
|
|
return nil;
|
|
|
|
}
|
|
|
|
- (void) encodeWithCoder: (NSCoder*)aCoder
|
|
|
|
{
|
|
|
|
}
|
|
|
|
- (id) initWithCoder: (NSCoder*)aCoder
|
|
|
|
{
|
|
|
|
return nil;
|
|
|
|
}
|
|
|
|
@end
|
|
|
|
#endif
|