David Neto

I prefer "yes", to a limited extent. https://github.com/gpuweb/gpuweb/issues/3255

What do standard libaries do for decimal floats parsing?

'override' is in that section, which is wrong too

The other thing that probably needs addressing (in #3280) is that now the '[callable](https://gpuweb.github.io/gpuweb/wgsl/#syntax-callable)' grammar element is ambiguous: callable : | [ident](https://gpuweb.github.io/gpuweb/wgsl/#syntax-ident) | [type_decl_without_ident](https://gpuweb.github.io/gpuweb/wgsl/#syntax-type_decl_without_ident) | [vec_prefix](https://gpuweb.github.io/gpuweb/wgsl/#syntax-vec_prefix) | [mat_prefix](https://gpuweb.github.io/gpuweb/wgsl/#syntax-mat_prefix) | [array](https://gpuweb.github.io/gpuweb/wgsl/#syntax-array)...

Also #3146 trunc domain includes infinity? Also #3135 Maybe related to #2888 defining "undefined value"

Removed label:editorial because of the "what happens if you're outside the domain". IMHO should be dynamic error. User doing nonsensical math.

C library docs say something like this (where it's already shown the overload returning double and float types. ``` SPECIAL VALUES sqrt(-0) returns -0. sqrt(x) returns a NaN and generates...

Ecmascript does a great job referencing IEEE 754, describing intent for fixed values and approximations, and describing edge cases and NaNs etc. https://tc39.es/ecma262/#sec-function-properties-of-the-math-object

Motivation: https://github.com/KhronosGroup/SPIRV-Tools/pull/4874

It would be nice to rely on library functions. * C99 strtod can parse a superset of WGSL hex floats * C++11 std::strtod has the same effect as C's strod...