问题
I am working on a code that generates PDF containing arabic texts. For each character, I am choosing the correct glyph in the presentation forms to display the text correctly. This works fine but Unicode doesn't contain presentation form of all arabic characters. For example \u067D ARABIC LETTER TEH WITH THREE DOTS ABOVE DOWNWARDS ٽ. There is no presentation form of this character even though the character has medial form, as can be seen in this string: لٽط
What is the reason that presentation forms of this and other characters are missing? Is the character not used in practice? Can the simple ARABIC LETTER TEH, which contains only one dot above and has presentation forms, be used instead? Or is it necessary to somehow build this character (e.g. by using \uFBB6 THREE DOTS ABOVE character)?
回答1:
The Arabic presentation forms should never be used for writing text. They exist only because they were needed for compatibility with older standards long ago. As such, there aren’t presentation forms for all Arabic letters in Unicode, only those necessary for this specific purpose. Many letters were also added long after the presentation forms ceased being relevant altogether. See the FAQ on Arabic for more information.
Arabic text should always be entered and stored using the regular letters (from the blocks Arabic, Arabic Supplement, and Arabic Extended-A). These letters will then automatically assume the correct shape depending on where they are situated in the word (initial, medial, or final) as can be seen in the example string you provided.
Using the character U+FBB6 ﮶ ARABIC SYMBOL THREE DOTS ABOVE would not be appropriate in this context because it is not a combining mark. It isn’t used to build new characters, but to talk about the symbol itself in isolation. From the code chart for Arabic Presentation Forms-A:
These are spacing symbols representing Arabic letter diacritics considered in isolation, as for example as in discussions about the Arabic script.
If the software you are using does not handle Arabic letter joining correctly, then there simply is no Unicode-defined way to enter the medial form of ٽ in your document. You will either have to switch to another framework entirely, or (as a last resort) encode the contextual forms you need as private-use characters in a new font, but I strongly recommend against that solution.
来源：https://stackoverflow.com/questions/57143803/missing-presentation-forms-glyphs-of-some-arabic-characters-in-unicode