Follow-up to #3492.
The current subset closure deliberately covers locale-independent Unicode upper/lower case variants and capitalize. CSS can synthesize additional glyphs through locale/context-sensitive casing and non-case transforms that a code-point closure cannot infer exhaustively.
Scope:
- cover Turkish/Azeri dotted-I, German sharp-S, and other locale/context-sensitive case behavior using authored
lang context;
- cover
text-transform: full-width and full-size-kana;
- preserve the existing encoded-URL budget and full-font fallback;
- add positive browser fixtures for the transformed glyphs;
- add a realistic payload-size benchmark so correctness hardening cannot silently turn common subsets into full-font downloads.
Until this lands, #3492 documents its case closure as locale-independent rather than exhaustive.
Follow-up to #3492.
The current subset closure deliberately covers locale-independent Unicode upper/lower case variants and
capitalize. CSS can synthesize additional glyphs through locale/context-sensitive casing and non-case transforms that a code-point closure cannot infer exhaustively.Scope:
langcontext;text-transform: full-widthandfull-size-kana;Until this lands, #3492 documents its case closure as locale-independent rather than exhaustive.