When encoding a type to a qfo file, the type's encoding string is written
and thus needs to be valid prior to actually doing the encoding. The
problem occurs mostly in self-referential structs (particularly, obj_class)
because the struct is being encoded prior to the pointer to the struct.
This is similar to the problem with infinite recursion when encoding types.
The problem is with structs with self-referential pointers (eg, struct foo
{struct foo *bar}). The solution is to copy the type data to a buffer and
mark the buffer as transfered before actually processing the type. Further
processing of the type is done via the buffer.
As id and Class do not point to real objects as such, trying to get the
class from their types doesn't work, so instead send the message to a
"null" class that skips the method checks.
Now the classes are built "properly" (using the same tools as the parser
itself), and the structs (obj_object, obj_class and obj_protocol) are built
separately, but using the class ivars.
Even just before, type_obj_object, type_obj_class and type_obj_protocol
were a bit bogus (still are), but now the arrays used to list their ivars
are correct. I plan to create the above mentioned types using
class_to_struct to do it properly.
Type names are cleaned up, as is the creation. Also, the class pointer in
the type encoding now gets emitted. However, Still need to actually create
_OBJ_CLASS_Class and fix the type encoding reloc handling in the linker.
With the previous commit, the structures were being created before a valid
source file name was available and thus qfcc would segfault when trying to
generate a tag. Now, the tags look better anyway :).
Now, for each compilation, or before linking, only InitData needs to be
called. Fixes the double chaining internal error when compiling and linking
in the same command.
This required throwing out the primary rules that snax did up to help me
with conflicts many years ago, but they were now getting in the way. Now
the productions from primary are merged in with unary_expr.
It turns out no code was being generated for x = *y. Ouch. I suspect I need
to take a better look at expr_deref at some time in the not too distant
future.
Conflicts:
tools/qfcc/source/statements.c
The base of the type encodings block is given by the .type_encodings def.
The block begins with a "null" type (4 words of 0), followed by the first
type encoding.
At some stage, I will need to add information for extended def information
(32 bit offset, type encoding, other?), but this is good for initial
testing.
Structures (especially hard-coded ones) can be really nasty as they can
refer to themselves. Avoid the recursion by setting the type_def field of
the type before doing the recursive encodings in the structure encoder.
The encodings of static types were getting corrupted because their defs
were not necessarily in the same places between compilations when compiling
multiple files.
The result of parse_params needs to be passed through find_type before
actually being used. I guess I'd missed this back when I got things working
for qc.
Since gnu bison and flex are required anyway, no harm in using their api
prefix options. Now, qfcc can compile both QC/Ruamoko and Pascal files
(Pascal is (currently?) NOT supported in progs.src mode), selecting the
language based on the extension: .r, .qc and .c select QC/Ruamoko, .pas and
.p select Pascal, while anything else is treated as an object file (as
before).
smart leak files now produce many points, not just one per portal. Normal
leak files now center the trail on the portal (instead of some weird
weighted average).
There is no error checking, texture alignment or support for non-point
lights, but it's a start. Also, it's assumed all brushes are convex and
manifold.
Blender doesn't support multiline text or word wrapping, so the tooltips
were rather ugly. The double spacing used by layout.label() is still a
little ugly, but the resulting text box just below the entity classname is
much more usable.
All of the nastiness is hidden in bspfile.c (including the old bsp29
specific data types). However, the conversions between bsp29 and bsp2 are
implemented but not yet hooked up properly. This commit just gets the data
structures in place and the obvious changes necessary to the rest of the
engine to get it to compile, plus a few obvious "make it work" changes.
The qc source path is specified via the dirpath property in the QF Entity
Classes panel of the scene data. The scanned entity classes are stored in a
plist in a blender text file for persistent storage (so the directory
doesn't need to be scanned every time). Also, so the data doesn't have to
be parsed every time, the data is stored in a normal python class hanging
off the properties class (evil hack?).
Slightly cleaner EntityClass building, and now the directory scanner is
part of the EntityClassDict class, which also supports reading/writing
plists (for persistent storage in blender).
They are written as normal string items, so anything using the plist later
on will need to know the context, but at least now there's no need to first
convert int or float data to strings before writing a plist.
More or less.
This is a bit of an experiment (which seems to work nicely) in that qfcc
and pak are build natively in one build tree, then the full system is
cross-built in another tree using the natively build qfcc and pak. Both
trees are created by cross-configure.sh as sub-directories of the current
directory. However, cross-configure.sh still assumes it is being run in a
subdirectory of the main quakeforge directory.
cross-make.sh checks if the native tree is in the current directory and if
so, builds it, otherwise it just runs make for the cross-build tree (this
allows for running cross-make in a sub-directory for things like sorting
out build issues).
CC_FOR_BUILD is the recommended name these days (HOST_CC was from an old
gcc version, and is confusing anyway). Also, CC_FOR_BUILD should be set by
configure.
Unfortunately, just because the header is there doesn't mean anything will
actually work :(. Also, the check is based on the host vendor/os for now.
Yes, it's rather lame but it will do for now.
With this, QF will build on an almost fresh ps3toolchain install. Only two
"fixes" are needed:
o In $PS3DEV/ppu/powerpc64-ps3-elf: ln -s ../include sys-include
o libsamplerate cross-built and installed.
One common use for a mesh having multiple UV maps is when combining several
mesh objects into one: the base UV map is the result of joining the meshes
(and will be a right mess of overlapping UV islands), but an additional UV
map is then setup as a copy of the first but with the islands re-packed so
nothing overlaps. The export script now searches for the active UV map and
uses that for both UV coordinates and the skin texture (when none is
specified).
If there's no export script, or the export script has no frame information,
animation data will be collected by running through blender frames 1 to the
current frame (inclusive). Each frame will be exported as a single frame
rather than as members of a frame group.
I'd forgotten I hadn't implemented exporting vertex normals. While I've
modified things for making better use of blender's tools and avoiding the
unnecessary use of objects, the code is taken from the ajmdl blender 2.4
export script.
Return statements never flow to the next block (or any other block, for
that matter), so drawing arrows leaving them not only messes up dot's
graphs, but is quite missleading.
When mering if/goto (ie, if skipping a goto), the rest of the dead code
remover is used to delete the goto. That part of the code unuses the goto's
label. The if was getting the goto's label without the lable's used count
being incremented (the usaged temporarily increases by one). I have no idea
why the problem showed up randomly, but this seems to fix it (it fixes /a/
bug, anyway).
The naive implementation of the if/goto merging was letting the old target
of the if get dropped because the block would lose its label and thus be
judged unreachable because the preceeding goto block was still in the list.
Instead, when the if/goto are "merged", mark the goto block as unreachable,
the following block as reachable, and break out of the analysis loop to
force the removal of the goto block. Since the dead block removal function
loops until no action is taken, all other dead blocks will be removed.
The output can be controlled via --block-dot (not yet documented). The
files a named <sourcefile>.<function>.<stage>.dot. Currently, stage will be
one of "initial" (after expression to statement conversion), "thread"
(after jump threading), "dead" (after dead block removal), "final" (final
state before actual code emission).
Labels can be shared between multiple flow-control instructions, so use the
label's used counter to determine when to remove the label. This was
causing problems with the jump threading.
The common cause seems to be casting a cast (very common, and I'm not sure
just realiasing the expression would be right). It does't cause any harm
(particularly, it doesn't trigger alias def chains), so I won't worry about
it.
The actual bug might still be elsewhere, but at least now I know the alias
chains were coming from accessing .return and .param_N, which are unions
(not directly usable by the progs engine). Emitting a reference to a union
(or struct) would create an alias def, but an alias expression was created
in the expression tree to simplify return/param access. The double layer
(sometimes 3 or 4) alias isn't really neaded, so rather than layering the
aliases, just re-alias the alaised def.
It is inteded for flagging buggy conditions in the compiler, particularly
after having fixed the original bug (in case something comes back from the
dead).
v6 progs expects .zero to be only 1 word. The code actually tried to keep
vector out of .zero, but it seems I'd rearranged the structure defintion
without updating the code that kills the vector field. Problem spotted by
divVerent.
Not sure if it actually works as the clients don't render the result
properly (can't see anything where the model should be), but the output
model does import back into blender properly.
Since qf does linear interpolation of verts, this seems to be reasonable.
Certaintly better than the rose-thorns I got because I haven't figured out
how to kick the auto-clamp.
I've decided to use property lists to define mdl control scripts. Some
names will probably get changed, and I still need to write code for writing
a plist, but the hard part is pretty much done :)
Note that this is the data block that holds the list of actual shape-keys,
rather than the shape-keys themselves. I'm not sure what it's correct name
is (it's just "Key" in RNA).
I really dislike this method of setting the name, but the use of "Key" as
the datablock name is actually hard-coded into blender's C code :/
Without fakeuser set, blender will toss out the actions on save and reload.
Converting to an nla strip might take care of that, but I haven't figured
out how to do that yet, so avoid any nasty surprised for the user.
Eye position, auto rotation, sync type and particle effects can now all be
edited in blender: both import and export do the right thing. The settings
can be found in the "QF MDL" panel of the "Object" tab of the properties
view.
Just about to do a release, and I realized windows users wouldn't have any
way of checking out the new renderer. I'll add wglsl when I get a chance to
do some testing.
o All instances of LIBADD/LDADD have a corresponding DEPENDENCIES
specificatiion.
o libraries now use a lib_ldflags macro to keep things consistent
o duplication of source/lib names has been minimized (particularly in
the libraries; more work needs to be done for the executables)
o automake spec blocks have been organized (again, more work needs to be
done for the executables)
Despair has things locked down such that running qfcc during a build fails
due to lack of read access to /usr/local/lib. This is actually a good
thing as accidentally hitting old includes/libs (when a file gets deleted
in the tree) hides bugs. Thus, --no-default-paths to turn off default
search paths.
Despair has things locked down such that running qfcc during a build fails
due to lack of read access to /usr/local/lib. This is actually a good
thing as accidentally hitting old includes/libs (when a file gets deleted
in the tree) hides bugs. Thus, --no-default-paths to turn off default
search paths.
The special token __INFINITY__, like __FILE__ and friends, will expand to
a floating-point expression containing a value the C compiler considers
infinite. Obviously, this assumes that the system has relatively modern
float hardware -- but if it doesn't, having Ruamoko be able to represent
float infinity is the least of your problems. :)
I got rather tired of there being multiple definitions of mostly compatible
plane types (and I need a common type anyway). dplane_t still exists for
now because I want to be careful when messing with the actual bsp format.
The params are eye position, flags and synctype. Provision is made for
reading them from a text block on export, but nothing is done other than
retrieving the text block.
The biggest part of the speedup is reading from blender's image only once
(it seems that every read does so from GL rather than memory: ouch). Also,
cache the results for each color.
The size is actually the average area in quake units of the mesh's
triangles. Again, my results are slightly smaller (0.025).
With this, all calculable fields are set. Only eye position, flags and
synctype remain.
The calculated radius is a smidge (0.05) smaller than the original
(invisibl.mdl), but I think that's due to the difference in source data: id
used the original models, I'm using their output.
Blender must have an active shape key before shape key animation will work.
This fixes the models being locked to the first frame until a shape key is
selected via the UI.
I /did/ see the warning about vertex index 0 in the obj importer script,
but I didn't take it seriously enough. This fixes both the twisted texture
on a couple of faces, and the truly mangled tris when exporting (using
invisibl.mdl for testing).
There seems to be some problems with the UVs, only one frame is exported,
and various model params don't get set (eye position, size, bounding
radius, synctype, flags), but the size and shape look right in qf :).
Except for the normal index in the frame bounds (and potentially frame
names with junk after the terminating nul), the output is identical to the
input for:
mdl=MDL().read("invisibl.mdl")
mdl.write("test.mdl")
Bill Currie
Jeff Teunissen