Russian Accents Type In The Answer Error

When I type in the answer for Russian words with accents, the accent itself becomes a separate character when I flip the card over. See attached:

The purpose of the accents is to emphasize which vowel sounds have more emphasis when speaking Russian.

It used to be that the accents were on top of the character rather than beside it.

I’ve tried looking on the forums and googling but I couldn’t find a solution to my problem.

I want to fix the problem so that the accent is above the Russian characters rather than beside them.

I can not reproduce this. Did you try using the latest version? I think the problem might be in the card template. Can you share that?

I am using the latest version. Sure. See attached:



An important question so we can figure out where this might need to be fixed – Is this the same or different from studying this card on Anki Desktop?


Your templates aren’t the problem.

I believe this is related to Unicode normalization. The way Anki does it is described in the manual – Adding/Editing - Anki Manual. I’m not familiar with Russian accenting at all, but does that sound similar?

How are you typing the character –

(a) non-normalized (a character plus an added accent)? – I think it won’t be able to match, and you’ll need to enter it as a normalized character instead.

(b) normalized (single character that includes accent)? – then I suspect the type-answer logic is stripping that out somehow, when it strips out the formatting.

It is the same studying this card on Anki Desktop. See attached:

I am copy pasting the character from a language learning app.
“Э́то бана́н” Sorry, I’m not sure if it’s non-normalized or normalized.

Since it’s happening in Desktop as well, it probably needs Desktop-side attention.

According to this handy tool – What Unicode character is this ? – both of those stress marks are separate characters. [Or at least they were by the time they made it through the forum to me.]
image

Am I right in understanding that this stress mark can be applied to pretty much any character – or at least any vowel character – in Russian? I’m not sure that stress marks like this are able be normalized in the way we usually think of it happening. If letter-with-the-stress-mark isn’t a separate character that exists in Unicode, there’s nothing to normalize that to – it just needs to be matched-as-entered.

@dae This has got to be a text-matching hurdle that the general programming community has already faced, right? :sweat_smile:

It is handled correctly on things like an accented ‘a’, which has its own entry in the unicode tables. Э́ does not appear to have its own unicode entry, so it can only be made from two separate codepoints.

Any ideas what is happening to this type-answer then? It seems like if it is separate codepoints throughout, it should (1) match as separate codepoints, and (2) remain displayed as a composed character.

I can’t seem to reproduce the issue. I tested with 23.06.3, pasted Э́то бана́н into that card, and got the following:

Screenshot 2024-07-06 at 8.26.07 PM

1 Like