Files silently skipped when extracting from 7z archive

[stephane352]

Hello,
Extract is working fine with zip files.
Unfortunately, when working with 7z files, if the archive contains file names with accents like Café.txt or Stéphane.txt, the file is not extracted (and sometimes other files in the archive as well).
This is a serious issue because the user is not notified that some files were skipped and only the administrator can see an error in the Nextcloud server logs (e.g. "Warning core normalizing failed for "Café.txt"").
This happens in Nextcloud 24.0.9 as well as in 26.0.4, with the latest version of Extract.
Would it be possible to fix this please ?
Thanks in advance.
Kind regards,
Stéphane

[Fantu]

This issue is still present in nextcloud 30 with app added manually from #114
I specify that it does not happen with all accented characters, because if you make file names with accented characters (as I did now from Windows) they are compressed (with 7zip on Windows) and then decompressed on NextCloud they are all there without problems
So it seems to be a problem related only to accented characters but with different encoding (perhaps similar to the encoding errors I see on some file names in scans).
@AntDavidLima is possible to solve or workaround it in extract? If you can't find and convert the encoding (of the filename), it might be acceptable also to change the file names by removing the characters that are not recognized correctly and that would prevent you from saving them. It's better to have 1-2 characters less than having no files at all, and since it doesn't show any errors, you won't notice the missing extracted files.

[AntDavidLima]

Thank you @stephane352 and @Fantu for the report, we will be looking into this and other important issues as soon as possible.

Sinse we have some clients that have interest in this app, and they are resposible for financing our work on the LibreCode Cooperative, we need to give them priority for they use cases.

But this kind of issue is very important to us, and as soon we have time, or someone contributes financeraly to the project wanting this issue fixed, we will work to resolve this problem.