natecook1000
commented
2 years ago For reference, here's the list of unsupported properties (as seen a bit obliquely in this test):
- Block
- Hangul_Syllable_Type
- Identifier_Status
- Identifier_Type
- Equivalent_Unified_Ideograph
- Simple_Case_Folding
- Decomposition_Type
- NFC_Quick_Check
- NFKC_Quick_Check
- NFD_Quick_Check
- NFKD_Quick_Check
- NFKC_Casefold
- Emoji_Component
- Extended_Pictographic
- Joining_Group
- Joining_Type
- Vertical_Orientation
- Line_Break
- Grapheme_Cluster_Break
- Sentence_Break
- Word_Break
- East_Asian_Width
- Prepended_Concatenation_Mark
- Bidi_Class
- Bidi_Mirroring_Glyph
- Bidi_Paired_Bracket
- Bidi_Paired_Bracket_Type
These apply to character and/or string more than UnicodeScalar, so while it would be nice to support them, there's a bit more design work required:
- Basic_Emoji*
- Emoji_Keycap_Sequence*
- RGI_Emoji_Modifier_Sequence*
- RGI_Emoji_Flag_Sequence*
- RGI_Emoji_Tag_Sequence*
- RGI_Emoji_ZWJ_Sequence*
- RGI_Emoji*
In the future, we'd like to include accessors to get all of the Unicode properties listed here: https://unicode.org/reports/tr18/#Full_Properties whether that be in the stdlib itself or a new Unicode module.