roozbehp
commented
3 years ago Unicode even recommends against such invalidation for Arabic. From https://unicode.org/reports/tr53/#Dotted_circles:
“Some rendering engines will insert a dotted circle for what it understands to be an invalid sequence. This is a problem in Arabic script because something that appears invalid may actually be valid text in some lesser known orthography of a minority language or in the Quran. For example, the Microsoft Windows text rendering engine, described in [Microsoft], inserts a dotted circle in combinations of certain Quranic marks that are known to appear with each other in the Quran.
Such spell-checking processes are best implemented at a higher level than a rendering engine. Also, a dotted circle insertion algorithm that displays all canonically equivalent sequences identically is hard to design and the result may be counter-intuitive for its users.”
behdad
tiroj
The spec has an “Invalid Combining Marks” section but it is entirely something invented by Microsoft, there no Arabic orthographic rules or Unicode encoding requirements that dictates what a “valid consonant base” for a given mark or what mark combinations are “invalid”.
I keep regularly getting bug reports about people trying some uncommon mark combination and getting the mysterious dotted circles inserted between the marks (in Microsoft products). There are many reasons one would want to use an uncommon or unusual mark combination, it is not the engine’s rule to invent its own orthographic rules an impose it upon its users.
This is mostly a bug report against Microsoft implementation since it is the only known implementation that has this misfeature (thankfully), but since this text is present in the spec, I figure here is a goo place to report it as well.
Document Details
⚠ Do not edit this section. It is required for docs.microsoft.com ➟ GitHub issue linking.