You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In working on the Unicode BIDI implementation in issue #9, we find a couple of characters which have extra behaviour outside of what BIDI provides: ZWNJ (0x200c) and ZWJ (0x200d) (§6.1). They control whether a ligature is allowed or expected between certain characters. We don't currently even consider ligatures, and maybe not all fonts will support them, but we should do some exploration as to whether we could use them. Some scripts seem to almost require them; Arabic and Indic scripts seem to change meanings based on some of these.
There do exist Unicode codepoints which correspond with commonly-used ligatures in some languages. They may be a valid starting point, which should be supported in the current code, but may not satisfy the needs of, say, direct input into text boxes.
In working on the Unicode BIDI implementation in issue #9, we find a couple of characters which have extra behaviour outside of what BIDI provides: ZWNJ (0x200c) and ZWJ (0x200d) (§6.1). They control whether a ligature is allowed or expected between certain characters. We don't currently even consider ligatures, and maybe not all fonts will support them, but we should do some exploration as to whether we could use them. Some scripts seem to almost require them; Arabic and Indic scripts seem to change meanings based on some of these.
There do exist Unicode codepoints which correspond with commonly-used ligatures in some languages. They may be a valid starting point, which should be supported in the current code, but may not satisfy the needs of, say, direct input into text boxes.
See possibly: https://www.freetype.org/freetype2/docs/reference/ft2-properties.html#glyph-to-script-map
The text was updated successfully, but these errors were encountered: