Klaus Frankagowa338@chaos.social
Aug 30, 2026, 4:01 AMOh and finally the filesystem all of this played out on is #btrfs, so if it does have a way to dump the raw name that would also help in this case.
Replies
- Klaus Frankagowa338@chaos.socialAug 30, 2026, 4:02 AM
So is this folder now called:
* 0x20C29C0D0A
* 0x20C2
* 0x2000
* 0x0020who knows.
Mark Whybirdwhybird@aus.socialAug 30, 2026, 4:49 AM@agowa338 If it were typable in more places, it would almost be fun* to put it in passwords.
*probably not really fun
Aug 30, 2026, 10:01 AM@agowa338 Who in their right mind uses anything other than alphanumeric characters in filenames anyway?
- Klaus Frankagowa338@chaos.socialAug 30, 2026, 10:44 AM
People in Asia apparently...
- Aug 30, 2026, 11:12 AM
@agowa338 :D
Jernej Simončič �jernej__s@infosec.exchangeAug 30, 2026, 3:07 PM
Jakob Voßnichtich@openbiblio.socialAug 30, 2026, 2:57 PM@agowa338 POSIX filenames are sequences of bytes, exluding slash (2F) or null (00) https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap03.html everything else, in particular how to get characters from/to this sequences is ¯\_(ツ)_/¯ by definition.
- Klaus Frankagowa338@chaos.socialAug 30, 2026, 3:36 PM
Hmm, how can it be that some approaches to query the path name "raw" then result in there being additional null bytes even?
There is some inconsistency I smell it and I'm not good enough to debug it further. I still have that folder here, so if you'd like to help be my guest. But otherwise the above is all I can see. And it is weird.
- Klaus Frankagowa338@chaos.socialAug 30, 2026, 4:23 PM
If I read that correctly it is not consistently defined as null-byte. But sometimes also as null-character. And with UTF-16 encoding a null character would be two null bytes.
But as I said before I'm not that familiar in debugging these kinds of issues, so I may just be misreading things here.
https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap03.html#tag_03_271
and here
https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap03.html#tag_03_445
- Jakob Voßnichtich@openbiblio.socialAug 30, 2026, 4:50 PM
@agowa338 my experiments with filenames was limited to git repositories: https://github.com/nichtich/badnames
joatjoat@mastodon.scotAug 30, 2026, 12:18 PM@agowa338 we've had similar problems with a large commercial backup tool at work. The Linux kernel and filesystems treat filenames mostly like file data, i.e. byte streams, except null and / are forbidden. It's userspace that has a problem. If you switch your encoding at a shell prompt to something 8-bit clean like ISO-8859-1 then you should be able to work with the file.