[WGSL] The lexer should use the functions from wtf/ASCIICType.h

• Component: WebGPU
• Status: RESOLVED FIXED

[Robin Morisset]

As suggested by Darin in a post-landing review of https://bugs.webkit.org/show_bug.cgi?id=233276.

[Robin Morisset]

Created attachment 452367 [details]
Patch

[Darin Adler]

Comment on attachment 452367 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=452367&action=review

> Source/WebGPU/WGSL/Lexer.cpp:163
> +        if (isASCIIDigit (m_current)) {

Stray space here after isASCIIDigit.

> Source/WebGPU/WGSL/Lexer.cpp:191
> -        } else if (isIdentifierStart(m_current)) {
> +        } else if (isASCIIAlpha(m_current)) {

Not sure this change is an improvement. Might want isIdentifierStart so the intent of the code is more apparent. The implementation of that function can use isASCIIAlpha.

> Source/WebGPU/WGSL/Lexer.cpp:194
> -            while (isValidIdentifierCharacter(m_current))
> +            while (isASCIIAlphanumeric(m_current) || m_current == '_')

Ditto.

> Source/WebGPU/WGSL/Lexer.cpp:271
> -    while (isWhiteSpace(m_current)) {
> +    while (isASCIISpace(m_current)) {

It can be ambiguous sometimes which definition of whitespace to use in various contexts. Might still want a named function for that.

> Source/WebGPU/WGSL/Lexer.cpp:301
> -        value += readDecimal(m_current);
> +        value += static_cast<uint64_t>(m_current - '0');

This code looks less elegant than the old.

[Robin Morisset]

Comment on attachment 452367 [details]
Patch

Thank you for the review, I've visibly overcorrected in eliminating my code.
I'll fix it and update the patch.

[Radar WebKit Bug Importer]

<rdar://problem/89418026>

[Robin Morisset]

Created attachment 454058 [details]
Patch for landing

[EWS]

Committed r291004 (248180@main): <https://commits.webkit.org/248180@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 454058 [details].