magnum

Commit Graph

Author	SHA1	Message	Date
Vladimír Vondruš	8dddda9c77	Trade: support zero and negative strides in mesh attribute data. This is not something the classic GPU vertex pipeline can handle (except maybe Vulkan, which can handle zero strides for instanced attributes?), but useful for other scenarios. This means existing code needs to be aware of and handle the new corner case.	4 years ago
Vladimír Vondruš	344dd3fc17	MeshTools: fix compile() with a non-trivial index offset. Haha, what the heck, how did nobody hit this yet.	4 years ago
Vladimír Vondruš	fce0756fc7	MeshTools: merge isInterleaved() and interleavedData() tests together. Because it tested the exact same thing for the most part.	4 years ago
Vladimír Vondruš	e496241290	MeshTools: unconditionally repack the input in removeDuplicates(). Since the main speed advantage of the function is that it hashes a whole vertex together instead of going through the attribute arrays one by one, it can't really operate on whatever funny interleaved layout it was given as there may be padding bytes with random content, breaking the deduplication. Since checking that a layout is really padding-less is rather complex, the repacking is now performed always. This also means the && overload makes no sense anymore and thus it was dropped. This makes the test added in the previous commit not assert anymore, and behave the same as the padding-less case.	4 years ago
Vladimír Vondruš	594393d842	MeshTools: test removeDuplicates() with various attribute padding. Leads to an assertion inside StridedArrayView constructor, because I just pass through the original attribute offsets/strides even though interleavedMutableData() gives me a much smaller stride. Moreover, because I just perform the deduplication on the original vertex data including all random padding, the duplicate removal won't always be able to find all duplicates. So test that case here as well -- turns out we just have to repack the data completely, after all.	4 years ago
Vladimír Vondruš	b6dd52eb62	MeshTools: add a flag to toggle layout preserving in interleave() etc. This took me quite a while to realize -- not always it's desirable to have the original layout unconditionally preserved, especially if for example filtering a MeshData to just a subset of attributes.	4 years ago
Vladimír Vondruš	2e90a93c75	MeshTools: properly verify all concatenate()d attribute offsets.	4 years ago
Vladimír Vondruš	0b6cb663cf	MeshTools: typo.	4 years ago
Vladimír Vondruš	7c17557d15	MeshTools: test impl-specific vert fmats directly in interleave() also. Hrm th mssags r gttng t lng t ft on a single line.	4 years ago
Vladimír Vondruš	073ef55f08	MeshTools: don't put const on arguments in functions defined in headers. Doxygen takes that literally and shows that in the output. Useless noise.	4 years ago
Vladimír Vondruš	29da22326e	doc: yep, this order is wrong. Doesn't help that the function is named in a LTR manner but the multiplication order is RTL.	4 years ago
Vladimír Vondruš	01660a7538	package/ci: ah yes, Xcode images rotting away right under my very hands. TBF I have to thank CircleCI for sending me no less than three warning e-mails about Xcode image deprecation. I just didn't bother updating until it started actually failing.	4 years ago
Vladimír Vondruš	f693550b66	MeshTools: document what interleavedData() really returns.	4 years ago
Vladimír Vondruš	26503041f7	Platform: test and document exit() behavior in {Glfw,Sdl2}Application. It doesn't fire an ExitEvent.	4 years ago
Vladimír Vondruš	4ae49595f9	Trade: minor test cleanup.	4 years ago
Vladimír Vondruš	a7573babc4	MeshTools: doc++	4 years ago
Vladimír Vondruš	ed489a1112	doc: ew, a std::vector! Funny how even doing the insanely complex operation of extending a MeshData with one extra attribute is still shorter than manually populating the GL::Mesh.	4 years ago
Vladimír Vondruš	a93d221c12	Trade: this comment no longer makes sense. Should have been dropped alongside the changes in `270e93e134` already.	4 years ago
Vladimír Vondruš	36c23db9af	Trade: doc++	4 years ago
Vladimír Vondruš	96f4dd077f	Trade: the comma in this assert feels weird.	4 years ago
Vladimír Vondruš	68d15bd8d0	Trade: fix SceneData field containment checks for negative strides. Why didn't I think of extracting the common code earlier?	4 years ago
Vladimír Vondruš	f685673be9	Trade: verify negative SceneData strides with offset-only fields also. Yay, works!	4 years ago
Vladimír Vondruš	cad13c9f40	GL,Vk: clarify behavior with implementation-specific type/format values. It may soon start happening that those will have an implementation-specific value that is neither a GL or a Vulkan identifier, so people should not just pass them around without checking.	4 years ago
Vladimír Vondruš	69b6ebd2f5	MeshTools: assert on impl-specific vert format in transform*().	4 years ago
Vladimír Vondruš	ce2ca47a81	MeshTools: assert on impl-specific vert format in removeDuplicates*(). It was already done in removeDuplicatesFuzzy(), so just make it consistent with the rest (shorter message, listing also the offending attribute index).	4 years ago
Vladimír Vondruš	0119d614a5	MeshTools: test that owned() works with impl-specific vert formats. Yes it does!	4 years ago
Vladimír Vondruš	8ae8603d7a	MeshTools: assert on impl-specific vert format in interleave*(). Haha, I even had a TODO here. An exception in this case is when the mesh is already interleaved -- then the layout is kept intact and thus it's not needed to know what vertex format sizes to repack.	4 years ago
Vladimír Vondruš	c489ea738f	MeshTools: assert on impl-specific vert format in duplicate().	4 years ago
Vladimír Vondruš	40acab4e37	MeshTools: assert on impl-specific vert format in concatenate*().	4 years ago
Vladimír Vondruš	cef7f44903	MeshTools: assert on impl-specific vert format in combine*Attributes(). This asserted on some random nondescript hard-to-track-down place before, which wasn't great. Also explicitly listing the requirement in the docs.	4 years ago
Vladimír Vondruš	d5927a9fb6	MeshTools: fix outdated naming in a test. VertexFormat used to be an (excessively long) MeshAttributeType and apparently I didn't clean up all cases yet.	4 years ago
Vladimír Vondruš	a72b4039f8	MeshTools: clarify behavior of impl-specific MeshPrimitive in compile().	4 years ago
Vladimír Vondruš	e75c089f10	Vk: doc++	4 years ago
Vladimír Vondruš	db5ac5f8c1	Document MeshIndexType equivalents in D3D and Metal. It was done for PixelFormat, VertexFormat and MeshPrimitive already, not sure why not here.	4 years ago
Vladimír Vondruš	812e93ae94	Doc++	4 years ago
Vladimír Vondruš	80a679ba93	Minor code motion.	4 years ago
Vladimír Vondruš	042f28a90e	GL: hook up EXT_texture_sRGB_RG8 for desktop GL as well. Was browsing the extension registry looking for something else and found this instead. It used to be ES catching up with desktop, now it's the other way around.	4 years ago
Vladimír Vondruš	b7876ed1cf	package/ci: reduce parallelism in the linux static build. Uh oh, do I have my files too large now?	4 years ago
Vladimír Vondruš	6d6676e1cd	package/ci: drop Gitter webhook notifications. It never really worked, was either too verbose or too quiet, and now it 404s on AppVeyor.	4 years ago
Vladimír Vondruš	8088ed8cda	package/ci: GitHub says I can't use git:// anymore. I kinda get the point, to avoid rogue third parties tainting my ŕepositories with nasty backdoors, yes, but unless everything on the way to the server including DNS queries is encrypted, then it's still just a half of the solution and any silly corporate firewall can still prevent me from cloning stuff from github dot com.	4 years ago
Vladimír Vondruš	a146f9f74b	sceneconverter: add an --concatenate-meshes option. Highly experimental and very assert-y, as it doesn't handle mesh primitive / attribute type incompatibility or conversion of loops/strips/fans to indexed meshes yet.	4 years ago
Vladimír Vondruš	86db988a0c	sceneconverter: skip refcounting in --info when there's no refeeree. Should have been done in `f10d74041b` already, but somehow I failed to test for this case -- if there are no scenes (materials/textures), it doesn't make sense to print reference count for cameras/meshes/lights (textures/images).	4 years ago
Vladimír Vondruš	44cce23595	SceneTools: hint at MeshTools::transform*D() in flattenMeshHierarchy().	4 years ago
Vladimír Vondruš	efa206b9f4	MeshTools: doc++	4 years ago
Vladimír Vondruš	dae4a35fd9	package/archlinux: let the coverage build go through even with failures.	4 years ago
Vladimír Vondruš	6ba8695668	AnyImageConverter: whoops, again didn't test the correct path here. Good thing the coverage reports exist.	4 years ago
Vladimír Vondruš	8ed08d7733	SceneTools: use a single temp allocation in flattenMeshHierarchy().	4 years ago
Vladimír Vondruš	eb2e4ff25d	SceneTools: make orderClusterParents() use one allocation instead of 4. Mmm. I wonder how much will future mosra hate me for making this code so obfuscated.	4 years ago
Vladimír Vondruš	f053ffeb15	SceneTools: remove one more allocation in orderClusterParents().	4 years ago
Vladimír Vondruš	99d82f5e62	SceneTools: we don't need growable arrays in orderClusterParents().	4 years ago

1 2 3 4 5 ...

9310 Commits (8dddda9c77ea282811baa540b4a6c8217cc465ab) All Branches Search

9310 Commits (8dddda9c77ea282811baa540b4a6c8217cc465ab)

All Branches