Commit Graph

1164 Commits (8bbe14c052248ae87606dbdfba79ea65d05889c2)

Author SHA1 Message Date
Léo Terziman 8bbe14c052 Merge commit '44773b8f9e47d9fe7e80946a1a9eb3d764fc8af2' into contrib 2014-01-17 11:52:08 +01:00
Léo Terziman 7bc074857e Missing material functions for aiUVTransform added 2014-01-17 11:51:43 +01:00
Léo Terziman d080eb493e Merge commit 'acca16c82fb9fa37c40475a423e979e9c7aa0d5a' into contrib 2014-01-17 11:49:30 +01:00
Léo Terziman bac6b34504 Removing null characters from xml parser input to avoid parsing failures 2014-01-17 11:48:42 +01:00
Léo Terziman e2676ec176 Implementing UTF8 to ISO-8859-1 conversion to improve chances of finding files in zip archives 2014-01-17 11:48:27 +01:00
Léo Terziman 0bf4aea9d5 Improved fast_atof (bis) 2014-01-17 11:47:37 +01:00
Léo Terziman 3e55dca943 Improved fast_atof 2014-01-17 11:46:44 +01:00
Léo Terziman d28fc25c5c Merge commit 'cc3f8b82f9f6fb4b295a39a3f7ac55154145cc04' into contrib
Conflicts:
	CMakeLists.txt
2014-01-17 11:45:57 +01:00
Léo Terziman 4e7b59be55 Improving homogeneity of DEBUG macro for improved portability with compilers other than Visual Studio 2014-01-17 11:43:48 +01:00
Léo Terziman 29c125e51d Minor improvement of ParsingUtils 2014-01-17 11:42:51 +01:00
Léo Terziman 44f1a1fb5d Huge optimization of Q3BSPZipArchive to reduce 3DXML importing time by a factor of 15~20. (However, the zip archive is now entirely decompressed in memory, so it takes twice as much memory) 2014-01-17 11:41:46 +01:00
Léo Terziman cf817270fc Reformating Q3BSPZipArchive 2014-01-17 11:39:51 +01:00
Léo Terziman bb9288fa26 Improved Q3BSPZipArchive + added material merging 2014-01-17 11:38:01 +01:00
Léo Terziman 51bf836db4 Fixed bugs in Q3BSPZipArchive & DefaultLogger 2014-01-17 11:36:47 +01:00
Léo Terziman 053df6907a Removed precompiled header because of a bug when switching configuration in visual studio 2014-01-17 11:35:30 +01:00
Léo Terziman 2cca861526 Merge commit '1f012384265e0dcc777e23369470391324d840e5' into contrib 2014-01-17 11:25:01 +01:00
Léo Terziman bfd3a206ea Improved CMakeLists.txt source groups 2014-01-17 11:24:16 +01:00
Léo Terziman 8ba5fa33af Improving ColladaExporter to support scenes with only one node 2014-01-17 11:21:32 +01:00
Léo Terziman 365b3aa412 Improved math operation with epsilon support 2014-01-17 11:21:15 +01:00
Léo Terziman 6c0ebb679a Merge commit '87a0e7703258576f1a7e4ba763b961da70c1cd91' into contrib
Conflicts:
	code/ColladaExporter.cpp
2014-01-17 11:19:29 +01:00
Léo Terziman a940102d70 Debugging ColladaExporter function for writing header information 2014-01-17 10:38:59 +01:00
Léo Terziman 3317f667e5 Improving ColladaExporter to add shading model, bump mapping, transparent & refraction properties 2014-01-17 10:38:42 +01:00
Léo Terziman 78389e2e86 Improved collada exporter 2014-01-17 10:38:17 +01:00
Léo Terziman 9120689f0d Improving ColladaExporter to play nice with models imported from any format. 2014-01-17 10:36:18 +01:00
Léo Terziman b4b39e026d Improving ColladaExporter to allow non-destructive cycles of import/export 2014-01-17 10:36:01 +01:00
Léo Terziman 30c6a74e1d Improving constructors of aiNode & aiScene 2014-01-17 10:31:27 +01:00
Léo Terziman 5e603d9ed8 Correcting a bug where the constructor & destructor of aiScene were not exported correctly in the assim library 2014-01-17 10:26:48 +01:00
Léo Terziman df700b998f Another small bug correction in Q3BSPZipArchive 2014-01-17 10:24:14 +01:00
Léo Terziman 2ef0f0f0e1 Correcting some bugs in Q3BSPZipArchive 2014-01-17 10:23:49 +01:00
gongwei 0b88f72be2 refactor the code, also parse texture option explictly 2013-12-02 22:58:15 +08:00
gongwei 0860254a2b Fix Obj texture parsing error: where obj texture map statement contains texture option, it will fail to parse the correct image url, as a result, the texture won't be shown. Besides, we also add clamp texture mode for obj 2013-12-02 22:55:51 +08:00
Kim Kulling ccd5b30a56 Merge pull request #199 from jdduke/obj_mg_parse
Properly parse 'mg' keyword in .obj files
2013-11-28 06:49:00 -08:00
Alexander Gessler e27d452770 Merge pull request #200 from jdduke/stl_gray_diffuse
Default to light gray diffuse for STL
2013-11-27 15:16:37 -08:00
Jared Duke 931f0489bd Properly parse 'mg' keyword in .obj files
The 'mg' keyword is currently being interpreted as a material library keyword,
when it really refers to the merging group.  Handle this case, in effect ignoring
the keyword as merging groups are currently unsupported.
2013-11-26 14:24:15 -08:00
Jared Duke 00717b6474 Default to light gray diffuse for STL
For consistency, use a light gray default diffuse color for STL files.
This provides consistency with other geometric types, like PLY, that use a
light-gray default material.
2013-11-26 14:11:24 -08:00
Alexander Gessler acca16c82f Merge pull request #196 from AIM360/UpdatedCodeComments
Updated some code comments in DefaultIOStream.cpp
2013-11-26 06:04:54 -08:00
henrikbuchholz 5983300422 Updated some code comments in DefaultIOStream.cpp
There was a misleading TODO comment that encouraged to use fseek/ftell instead of fstat. However, fstat has been used intionally because fseek/ftell is potentially unsafe.
So I replaced the TODO and added some explanation why fstat is being used instead.
2013-11-26 14:52:53 +01:00
Alexander Gessler d6a3df0acf Merge pull request #192 from AIM360/FaultToleranceFor3DSLoader
3DSLoader: Skipped chunks of size 0
2013-11-25 09:46:00 -08:00
henrikbuchholz 35fa7cb441 3DSLoader: Skipped chunks of size 0 2013-11-25 11:53:15 +01:00
henrikbuchholz d7dbdd6c3b SortByPType Posprocessing crashed for crappy models with degenerated geometry
Which models crashed before the fix?:
The crash was observed for files with the following properties:
1. They contain >=1 meshes
2. They were loaded with SortByPType option
3. They only contained degenerated meshes, so that these were skipped SortByPType

What is improved by the fix?:
Obviously, the affected models were crappy anyway and will still produce empty output after the fix. However, the fix avoids the heap-corruption, which couldn't be solved by try/catch from outside and had the annoying effect that a whole scene with hundreds of individual models could crash due to a single crappy one.

Why did it crash before?
The SortByPType deleted some exluded meshes, but didn't reset the pointers in pScene. After throwing the DeadlyImportException (no remaining meshes), remaining meshes were deleted => Excluded meshes were deleted twice.
2013-11-25 10:50:15 +01:00
acgessler d080f92241 BlenderTesselator: Fix OffsetOf() macro 2013-11-22 19:54:47 +01:00
acgessler 857e50ba40 BlenderTesselator: disable GLU tesselation code by default. 2013-11-22 13:58:40 +01:00
acgessler fbb1a79248 Update CMakeLists 2013-11-22 13:57:50 +01:00
Alice 01a6ec9059 Adding tessellation support for new Blender 2.6 polygons
- Added BlenderBMesh.cpp/h which contains a class to convert a BMesh to an old style tri/quad mesh
- Added BlenderTessellator.cpp/h which contains a class to tessellate the poly loops contained within a BMesh
2013-11-22 13:17:21 +01:00
Eric Haines f2997a562b typo fix: loose -> lose 2013-11-14 10:09:10 -08:00
Hiroyuki Inoue 3a3ac4fea2 Fix to support FBX 2011 (7100) 2013-11-12 23:25:40 +09:00
Kim Kulling 685612b9f4 - refactoring : remove not used mesh index from private method in obj-importer.
- refactoring : remove commented code.

Signed-off-by: Kim Kulling <kim.kulling@googlemail.com>
2013-10-11 15:29:17 +02:00
Calvin Hsu a9a881cde6 Blender: map texture influence to aiTextureType
Previously assimp mapped all blender image textures as
aiTextureType_DIFFUSE.  This change interprets the "mapto" property
which corresponds to the Blender "Influence" in the properties editor.

* Blender's Normal influence with "Normal Map" unchecked maps to
  aiTextureType_HEIGHT.
* Blender's Normal influence with "Normal Map" checked maps to
  aiTextureType_NORMAL.
* Influence scale is placed into AI_MATKEY_BUMPSCALING.
2013-10-08 19:01:02 -07:00
Kim Kulling e3a5480811 bugfix: fix typo's in obj-doxygen documentation.
Signed-off-by: Kim Kulling <kim.kulling@googlemail.com>
2013-10-08 16:51:59 +02:00
Kim Kulling e4fca83237 refactoring : remove dead code.
Signed-off-by: Kim Kulling <kim.kulling@googlemail.com>
2013-10-08 16:30:17 +02:00