magnum

Commit Graph

Author	SHA1	Message	Date
Vladimír Vondruš	1dbc633bc8	Trade: doc++ Somehow I forgot to cross-reference the SceneFieldData constructor and mention which are usable for strings and which not, and that no builtin fields are strings.	3 years ago
Vladimír Vondruš	83ee04b362	Trade: don't do an unnecessary cast for a MeshAttributeData view. Same as previous commit, would cause ambiguity with an upcoming StridedArrayView "updimensioning" constructor.	3 years ago
Vladimír Vondruš	0401c4ded8	Trade,Primitives: use a correct type-erased view for MeshAttributeData. This got probably implemented long before the change in `c74b4c6b90`. Or actually maybe not at all. In any case, it'd cause an ambiguity with the 2D char view constructor when the "updimensioning" StridedArrayView constructor gets introduced.	3 years ago
Vladimír Vondruš	360bcc443e	Shaders: don't use std::pair here.	3 years ago
Vladimír Vondruš	02df65f688	SceneTools: add a note why std::pair has to be used here.	3 years ago
Vladimír Vondruš	826329258a	Math: adapt to changes in Corrade's TweakableParser also for Angle. Forgot to do this in `4ef33236eb`.	3 years ago
Vladimír Vondruš	ffcba3912b	package/ci: reduce parallelism even further. Damnit, I wonder where is it using so much memory and why.	3 years ago
Vladimír Vondruš	2b53e8b974	doc: regenerate SVGs for Animation easings. I did that just to be sure that `3c8fd70c12` didn't break anything, but to my surprise arrived at a difference somewhere completely different. The problem is that the SVGs were generated before Math::pack() was fixed in `e62ce4faa6` (Feb 2019) to perform correct rounding, and so most color values were off by one. Heh.	3 years ago
Vladimír Vondruš	32398ffadd	Animation: be less strict about bounds for double easings. Happens only in release builds. Interesting.	3 years ago
Vladimír Vondruš	3c8fd70c12	Animation: make Easing a template struct, add an Easingd typedef. Similarly to how Constants and Constantsd are made.	3 years ago
Vladimír Vondruš	cb12b669a5	Animation: indent the Easing namespace. In preparation for turning it into a class templated over floats and doubles -- no functional change, just formatting.	3 years ago
Vladimír Vondruš	6ab446f7ed	Trade: document that MaterialAttributeData expect a non-empty name. It's good to have that explicitly mentioned to not have to guess (or, worse, browse the code).	3 years ago
Vladimír Vondruš	ac34c9c7e1	Shaders: list also private headers in project view of IDEs. Other libraries do this already, here it wasn't done for some reason.	3 years ago
Vladimír Vondruš	9c7545ec5e	Shaders: fix copypasted doc nonsense.	3 years ago
Vladimír Vondruš	1bdae5fddd	Shaders: properly check for EXT_gpu_shader4 if ObjectId is enabled. I was checking many things but not this one.	3 years ago
Vladimír Vondruš	f0b4459ce3	Trade: doc++ I spent some time wondering if as<PbrClearCoatMaterialData>() can be used on a material that doesn't neccessarily have that layer. It can, so hint that in the docs.	3 years ago
Vladimír Vondruš	dfb784b5ce	sceneconverter: remove a test workaround for a now-fixed ufbx issue.	3 years ago
Vladimír Vondruš	7de3704d5a	Vk: try a significantly larger count in DescriptorPool allocation test. It was creating the pool with 7 buffers, which the driver could as well just round up to 8 and then allocating 8 would not fail. The test used to work on Mesa before but not anymore. So I'm creating the pool with 8 and askin for 80 which should definitely fail. There's a lot more failures on NVidia, I "fixed" those by deleting the NVidia Vulkan ICD file for now.	3 years ago
Vladimír Vondruš	592548e220	sceneconverter: update outdated comment.	3 years ago
Vladimír Vondruš	67be997af5	AnySceneImporter: add a TODO for recognizing `*.mtl`. While UfbxImporter knows to import these directly, AssimpImporter tries to load them as FBX and fails (heh!), and ObjImporter has no idea about materials at all, so recognizing this extension would only add more harm than good at the moment.	3 years ago
Vladimír Vondruš	ee9e2e96be	sceneconverter: implement a --phong-to-pbr option. Exposes MaterialTools::phongToPbrMetallicRoughness() that got added some time ago. Most of the code and tests is scaffolding needed for direct material import and processing outside of the addImporterContents() automagic, similar to what was already done for meshes and images. In particular, adding more material conversion options such as canonicalization or deduplication will be significantly easier now that the basics are done and tested.	3 years ago
Vladimír Vondruš	b38b05cae1	sceneconverter: verify also the unimplemented case for 1D images.	3 years ago
Vladimír Vondruš	fee4de16fe	sceneconverter: test also behavior with ignored contents.	3 years ago
Vladimír Vondruš	c8d009f0a9	Trade: sort addSupporterImporterContents() warnings by dependency order. To match the order (and thus also verbose log) in which they're added inside addImporterContents(), and subsequently also the order magnum-sceneconverter processes them manually. There's no reason to have it different and cause confusion.	3 years ago
Vladimír Vondruš	79bccf6bf8	sceneconverter: reserve the mesh array before appending to it. The number of items is known beforehand, so avoid unnecessary reallocations. When mesh LODs are supported this won't be that clear anymore, nevertheless it's never bad to pre-reserve.	3 years ago
Vladimír Vondruš	427d6be514	sceneconverter: make the test instances easier to read. Using the InPlaceInit constructor instead of an unnecessarily verbose array<T>({...}) helper, indenting the command-line arguments so they're easier to distinguish from the rest.	3 years ago
Vladimír Vondruš	6dd78afd28	sceneconverter: improve some comments in the test.	3 years ago
Vladimír Vondruš	632bf93a4e	Shaders: fix the non-deprecated ES2 build. The initializer list is needed for Phong lights as well, not just (ES3+ only) skinning stuff.	3 years ago
Vladimír Vondruš	5306542876	TgaImageConverter: add a TODO. No, I'm not going to add a workaround here.	3 years ago
Vladimír Vondruš	af3e364b35	AnyImageConverter: adapt test to TgaImageConverter changes.	3 years ago
Vladimír Vondruš	1101712b16	imageconverter: adapt test to TgaImageConverter changes.	3 years ago
Vladimír Vondruš	8fd6d9e8af	TgaImageConverter: use RLE only if uncompressed output isn't smaller. Makes sense, especially for LUTs and other things that don't have any repeated pixels at all.	3 years ago
Vladimír Vondruš	52c918bcb4	TgaImageConverter: test the verbose message before a SKIP(). So it's still verified even if TgaImporter isn't built.	3 years ago
Vladimír Vondruš	0bf32ae1d2	doc: mark TgaImageConverter as having no known caveats. While StbImageConverter has now a minor caveat as it produces larger files for no apparent reason.	3 years ago
Vladimír Vondruš	5796b597f3	TgaImageConverter: implement RLE encoding. Because I got tired of having to manually postprocess ground truth files for shader tests. A bit strange / funny that the first ever version of my code manages to produce smaller files than both stb_image and ImageMagick (with -compress RunTimeLengthEncoded), for some reason both pick some strange inefficient run combinations in various scenarios, such as "copy 2" + "repeat 3" where I pick "repeat 5". And the difference is not insignificant -- when testing with some shader test files, it resulted in a ~17% smaller size! The plugin follows the stricter variant of the spec by default (i.e., splitting runs across scanline boundaries) -- so that's not the reason for the weird differences between my code and theirs -- but provides an option to not do that for even smaller files. Which I'm going to use for shader ground truth test files, because there every byte counts. This option together with the above difference causes files to be ~25% smaller, which is quite a lot.	3 years ago
Vladimír Vondruš	7ccf22e1d0	Math: remove swizzle() that was deprecated since 2019.10.	3 years ago
Vladimír Vondruš	35c02023a0	TgaImageConverter: properly test also the single-channel case. For some reason I never tested this, although code coverage says it was called from ... somewhere.	3 years ago
Vladimír Vondruš	cdb33c8723	TgaImageConverter: put test data closer to the actual test case.	3 years ago
Vladimír Vondruš	ae88ea44cf	TgaImageConverter: move verbose output to the pixel format switch. So I don't need to duplicate it for the RLE branch.	3 years ago
Vladimír Vondruš	9903e23af9	TgaImageConverter: don't zero-init all output for no reason.	3 years ago
Vladimír Vondruš	0db8e6c4a3	TgaImporter: simplify RLE detection.	3 years ago
Vladimír Vondruš	5113315cf8	doc: add info about iOS launch screens needed for native resolution. Yet another "one of the frustrating secret things about iOS". Co-authored-by: hsdk123 <daegon.dhsk@outlook.com>	3 years ago
Pablo Escobar	3e15133772	Trade: fix doc typo.	3 years ago
Vladimír Vondruš	350531f2ab	Shaders: define enum set operators right next to the typedef. And not somewhere completely different.	3 years ago
Vladimír Vondruš	b6a31a0af4	Fix double-precision Bezier typedefs to be actually double-precision.	3 years ago
Vladimír Vondruš	f75449240e	Shaders: what, C++, forward declarations work here but not there?! Who designed this misfeature, ugh.	3 years ago
Vladimír Vondruš	963faea23d	Shaders: doc++	3 years ago
Vladimír Vondruš	afa2b37389	Shaders: group whole vertex/fragment shader setup together.	3 years ago
Vladimír Vondruš	d5a03b7b12	package/ci: reduce parallelism for the ES2 build as well. Ugh...	3 years ago
Vladimír Vondruš	63806e6ce6	doc: updated changelog.	3 years ago

1 2 3 4 5 ...

10177 Commits (1dbc633bc83b9f8d6bb4ee5db4609911bc0217ac) All Branches Search

10177 Commits (1dbc633bc83b9f8d6bb4ee5db4609911bc0217ac)

All Branches