magnum

Commit Graph

Author	SHA1	Message	Date
fauder	d53dced46d	debuggers: display when a Vector2/3/4 is a unit vector Added Float & Double overrides, so the other variants (Half & integer variants) are effectively untouched. Added Dot() & IsNormalized() intrinsic functions for each entry. The reason we need separate entries for Float & Double is because the epsilon values used for Dot() are different. Epsilon values are taken directly from Magnum::Math::TypeTraits & they are added to & subtracted from 1.0, in order to avoid using abs(), which is prohibited in .natvis.	3 years ago
Vladimír Vondruš	3b1fb704ca	Platform: add a TODO for a NVidia bug triggered by GLFW.	3 years ago
Burak Canik	f79a9dfecf	Math: fix & clarify a Deg conversion snippet. Got messed up when making the documentation snippets compiled back in `673caa23bf`.	3 years ago
Vladimír Vondruš	06e731e83a	MeshTools: remove declaration of no-longer-present test cases. These are now in GenerateLinesTest.	3 years ago
Vladimír Vondruš	3bebea4b5e	TgaImageConverter: what's up with those strange includes??	3 years ago
Vladimír Vondruš	8e132b1ee2	MeshTools: fix a unused variable warning on a no-assert build.	3 years ago
Vladimír Vondruš	ee353e1f6a	CMake: add a MAGNUM_BUILD_STATIC_UNIQUE_GLOBALS_DLL_NAME option. Same as the corresponding Corrade change. Should make MAGNUM_BUILD_STATIC_UNIQUE_GLOBALS working also in cases where Magnum is linked to just DLLs but not the main application executable. Such as various plugins or native Python modules. It isn't a nice UX to force users to hardcode a DLL name of their own application when building a dependency library, but is a simple enough middle ground between global symbol deduplication not working at all and having to loop through all possible DLLs with EnumProcessModules() until a symbol is found.	3 years ago
Vladimír Vondruš	60a348cf55	Build WindowsWeakSymbol.cpp only with unique globals enabled. Otherwise this function is not used from any code. Same as the corresponding change in Corrade.	3 years ago
Vladimír Vondruš	d982ba9ec4	MeshTools: support morph targets in concatenate(). While extremely easy to do with the linear lookup, I bet it'd be something utterly complex like trying to hash a std::pair with the original hashmap-based implementation.	3 years ago
Vladimír Vondruš	31787e687e	MeshTools: stop using std::unordered_multimap in concatenate(). Let's guess what would be faster -- making a ton of tiny allocations, hashing a bunch of numbers in a very complex way and then jumping through all those indirections and cache misses of a multimap linked list for each and every attribute of each mesh, OR going linearly through a two tiny arrays for every mesh? I don't have any benchmarks ready but I bet the latter will win.	3 years ago
Vladimír Vondruš	25190e3207	MeshTools: minor comment / TODO update in concatenate().	3 years ago
Vladimír Vondruš	eb83dd5a8f	MeshTools: clarify confusing comment.	3 years ago
Vladimír Vondruš	02e0d85f51	MeshTools: support morph targets in transform().	3 years ago
Vladimír Vondruš	e076e29b6b	MeshTools: avoid unnecessary double lookup in transform*().	3 years ago
Vladimír Vondruš	9b4ee91239	MeshTools: recognize & ignore morph targets in compile(). No builtin support for these at the moment.	3 years ago
Vladimír Vondruš	02a2986912	MeshTools: clean up MeshData construction formatting in a test.	3 years ago
Vladimír Vondruš	6b7b5f426b	MeshTools: preserve morph target ID in interleavedLayout(). And thus in all APIs that use it, or rely on related helpers.	3 years ago
Vladimír Vondruš	0607c476f7	sceneconverter: show morph target ID in --info output, if present.	3 years ago
Vladimír Vondruš	81b237ba14	sceneconverter: silly typo in a test.	3 years ago
Vladimír Vondruš	f84c3679c3	Trade: support morph targets in MeshData. Attributes that are morph targets are marked with a morph target ID, those then get treated separately in name-based lookups.	3 years ago
Vladimír Vondruš	75265d79a7	Trade: update MeshData test to the usual workflow. Back when I wrote the test I didn't know what would be the most convenient way to use the API yet, so it was unnecessarily verbose in various places. Now I do. Also using Utility::copy() to populate the arrays. No need to suffer this much for no reason.	3 years ago
Vladimír Vondruš	85906f55b2	Trade: match test case declaration order with definitons.	3 years ago
Vladimír Vondruš	c43f6bf043	MeshTools: test combineIndexedAttributes() with an offset-only attrib. Just to be sure. It doesn't cause any issues.	3 years ago
Vladimír Vondruš	62b954e258	MeshTools: implement combine*() using interleavedLayout(). So the same logic isn't duplicated in two places, plus future-proofing for when interleavedLayout() gains alignment-aware attribute reordering.	3 years ago
Vladimír Vondruš	5db4aa121e	MeshTools: minor code motion.	3 years ago
Vladimír Vondruš	2dce9a1c63	MeshTools: don't zero init an array that gets fully overwritten. I really need to deprecate the "implicit value init" Array constructor, far too many cases like this.	3 years ago
Vladimír Vondruš	b1d1d65e5a	MeshTools: don't create tons of strided views from scratch in combine*(). Instead create it once and then slice it up. Significantly less error prone and faster as well as the assertions don't need to check that much.	3 years ago
Vladimír Vondruš	758b4ad05c	MeshTools: pass an ArrayView instead of an Array reference. Huh this whole code smells ... rotten.	3 years ago
Vladimír Vondruš	e0a1f435d7	SceneTools: minor.	3 years ago
Vladimír Vondruš	ec3241f425	MeshTools: common helper for remapping MeshAttributeData to new arrays. So all the locations don't need to be explicitly tested when anything happens with this -- already way too complex -- MeshAttributeData constructor.	3 years ago
Vladimír Vondruš	f50c88b392	{Material,Mesh,SceneTools}: use Utility::copyMasked() in filter*(). It's not really implemented in a fast way yet, but that doesn't mean the custom loops were any better.	3 years ago
Vladimír Vondruš	1d9190a3c6	Trade: minor cleanup in a test. Last time I generally used () for construction was maybe 2014. Why is it still here?	3 years ago
Vladimír Vondruš	414e80bbe2	Trade: avoid double lookup in template MeshData and SceneData accessors. The second lookup was only needed for a type compatibility assertion, which is a bit unfortunate. It wouldn't be a problem on a no-assert build, but for various reasons people shouldn't be using these. Too dangerous.	3 years ago
Vladimír Vondruš	5beefc3b1c	Trade: use a colored diff comparison for large strings.	3 years ago
Vladimír Vondruš	e509f02da8	doc: update changelog.	3 years ago
Pablo Escobar	e1583700e5	debuggers: add MSVC natvis files.	3 years ago
Vladimír Vondruš	4f1a56c463	GL: fix a link to glTextureUnit().	3 years ago
Vladimír Vondruš	bbb54a9df8	It's 2023 already!	3 years ago
Vladimír Vondruš	21cff30b27	modules: update Find modules from Corrade repository.	3 years ago
Vladimír Vondruš	11fa89c36e	Test: can now do just .prefix(0) instead of some crazy shit.	3 years ago
Vladimír Vondruš	c7a4c96072	MeshTools: show an example sequence for each generate*Indices() API. Makes it easier to see what these are doing.	3 years ago
Vladimír Vondruš	05dda3f832	DebugTools: minor coverage fix. Looks like the switch cases are handled differently now, I'm pretty sure the original way worked before.	3 years ago
Vladimír Vondruš	8150e06616	DebugTools: support half floats in CompareImage for pixel views. For images it was supported already, but here I forgot somehow.	3 years ago
Vladimír Vondruš	f589aa1a5d	DebugTools: improve pixel view format autodetection in CompareImage. Originally it was just assuming that any Vector3ub or Color3ub is a normalized format. That was kinda enough for many cases, but it started to get annoying with sRGB image comparisons, as those had to be manually reinterpret with a sRGB-less format in order to pass. Now the pixel format detection looks at the expected image format as well, and if the underlying type and component count matches, it inherits the sRGB and normalized property from it as well. If not, it falls back to an integer format for vectors, and normalized format for colors. For vectors this is different from previous behavior but shouldn't cause any problem in practice -- the only result will be that the image comparison fails with a different message for pixel format mismatch than before. This now also properly and fully tests the pixelFormatFor() helper, and adds a missing Color3 specialization of it.	3 years ago
Vladimír Vondruš	e5a2b29f39	Math: disable KTX Y-flip on load in a test data generator script.	3 years ago
Vladimír Vondruš	410441791c	Math: implement BC2 flipping as well. It's a weird format nobody uses, but since I have the full context in my head now, it's easy to do.	3 years ago
Vladimír Vondruš	ccc5a45c4f	Math: can load just 6 bytes instead of 8 for BC4.	3 years ago
Vladimír Vondruš	3ccb7a9fc9	Math: utilities for flipping BC1/3/4/5 blocks. Most of the testing scaffolding here is a preparation for the actually complex formats like BC6/7 or ASTC. Also, it's great to be able to use Magnum from Python to prepare data for testing the C++ Magnum APIs.	3 years ago
Vladimír Vondruš	e02104df22	Math: print a note in the test if char is unsigned. To make it easier to discover such platforms.	3 years ago
Vladimír Vondruš	3113306d82	package/archlinux: adapt to lcov 2.0 changes.	3 years ago

1 2 3 4 5 ...

10525 Commits (d53dced46daebf532fbd7f222fda875e4262cf51) All Branches Search

10525 Commits (d53dced46daebf532fbd7f222fda875e4262cf51)

All Branches