Vladimír Vondruš

Nope, m.css just passes through a typing exception. The issue is on your side, [PEP604](https://www.python.org/dev/peps/pep-0604/) (`|` instead of `Union`) is only since Python 3.10.

Yes, but *I'm* relying on the interpreter being able to understand the types, so I can link them etc. So if you run m.css on 3.7, it'll fail. You'd have...

> quite significantly (With my skeptical hat on.) Do you have some numbers to back this? I'm interested in how much this helps in a codebase of this size.

It's not as simple as "enabling SSE2" since WASM has to work on ARM as well -- and that's why I'm skeptical, because different platforms have different instructions and what...

The cases we've been running into with @pezcode (in [Magnum's cgltf-based importer](https://doc.magnum.graphics/magnum/classMagnum_1_1Trade_1_1CgltfImporter.html)) were mostly due to various syntax or out-of-bounds errors, which caused the initial `cgltf_parse()` to fail. Which means...

> Do you think it would make sense to keep the functionality in the macros and instead just let the user add the error handling? Yes, that would be even...

Just FYI, and sorry for the very late comment -- as our use case ended up diverging even further from cgltf's goals, we ended up making our own parser from...

Closing as resolved, forgot this issue was even here. Just for the record, the first patch was integrated with various version-dependent modification as mosra/toolchains@d5d74307fd81c84596b3575c9692681cc61952c2, and the toolchains submodule was updated...

Eh, in the end I actually _did_ hit a similar problem, although in my case I had to add the _non-cache_ include dir there: https://github.com/mosra/toolchains/commit/6a9e0821b4ccd578feee1f30cef245cbd36cdd38 Because FindZstd installed into the...

For anyone still subscribed to this PR, a variant of the above together with structured bindings for `StridedDimensions` was added in 087ead32ce60655676ab0cb9f001e182aba1d4af and 5f4fc52ad2f07c4e022e8ba8e770144becb69b17.