Bug 245859
| Summary: | [WGSL] Don't parse literals in lexer | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Dan Glastonbury <djg> |
| Component: | WebGPU | Assignee: | Nobody <webkit-unassigned> |
| Status: | NEW | ||
| Severity: | Normal | CC: | webkit-bug-importer |
| Priority: | P2 | Keywords: | InRadar |
| Version: | WebKit Nightly Build | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
Dan Glastonbury
WGSL current parses numeric literals during lexical analysis into `double`s. This might be fine for 32-bit integers but isn't sufficient if WGSL adds 64-bit integers.
We should instead pattern match on the format for numeric literals, classify them and defer the validation of literals to a later pass after type checking has been performed. This also requires handling of half floats.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Radar WebKit Bug Importer
<rdar://problem/100592948>