Unicode combining characters are off-center?

I’m using LibreOffice to write up a linguistics paper, and precision is pretty important when it comes to where the diacritics go on the characters so it doesn’t look like I’m trying to attribute a feature to multiple phonemes at once in my transcriptions. For a couple of combining characters, however, they’re automatically shifted slightly right of the character they’re applied to.

It only seems to be a problem with LibreOffice. The same exact characters are absolutely fine when copy+pasted into Google Docs. Shifting around the order of the Unicode only moves them to the left instead (and splits them into multiple characters in the process, which also adds gaps into the strings).


I’ve been trying to ignore it because LibreOffice has really great functionality with everything else but I am a little bit at my wit’s end right now. Please let me know if there is a way to fix this.

If it helps at all, the specific combining characters I am having trouble with are as follows:

  • U+030a (ring above)
  • U+0325 (ring below)
  • U+031e (down tack below)
  • U+0346 (bridge above)
  • U+032A (bridge below)

Try with another font. Liberation Serif don’t works well here.

1 Like

@rhalloway02 U+0061 is LATIN SMALL LETTER A. COMBINING BRIDGE BELOW is U+032A. I fixed the question.
But, add to it OS name, LO version and save format.

Following on @LeroyG’s clue, I experimented with the fonts installed on my system. Very few fonts accept correctly several combining marks, more tolerate a single diacritic. According to the various visual glitches, I suppose the font renderer looks for a precomposed glyph in its font collection and adds extra diacritics on this precomposed character if one is found. The clue is the fact that some “sans” glyphs are use inside “serif” faces and some times not at the same size or weight (clearly indicating a substitution).
On my Linux box, only the Free family gives satisfactory results, e.g. Free Sans, Free Serif or Free Mono. Liberation faces fail when there are several diacritics on the same base character.

This is obviously an issue with the font renderer. I think there are already bug reports about it.