I try to show Hebrew vowels without a consonant as base character. The official Unicode documents directs me to use no-break space as substitution base character (https://www.unicode.org/versions/Unicode12.0.0/ch07.pdf »Marks as Spacing Characters.« and https://www.unicode.org/versions/Unicode12.0.0/ch02.pdf »Exhibiting Nonspacing Marks in Isolation«) I inserted tildes instead of no-break spaces, as Paratext wants. After that, when I change chapters, Paratext asks me if I want to save changes and when I agree, the tildes are gone and there’s a normal space there instead of it. Do I need to enable using tildes for my project first?
Example: JDG 20:13
\w ְ ֵ֣\w*
You give an example that is hard to see. If it is something like this \w ~…\w*. Know that the no-break space will be deleted since all spaces after a open tag are compressed to one and are considered part of the tag. What you should use is this ~\w …\w*
It’s good to know about the collapsing of spaces (even if they are no-break spaces). It is just, that in this extreme case I actually do need multiple no-break spaces one after the other as base characters for the Hebrew vowels.
Without knowing what exactly you’re trying to achieve…, have you considered using ◌ (U+25CC) as a base character instead?
Thank you for your suggestion. I will give it a try.