File “…” contains non-NFC UTF-8 sequences and cannot be synced. Consider renaming.
I have files, what are hashed along with the filename, that I’d consider practically immutable — unless I use a meta filesystem, that performs the translation hot, it can’t be changed.
Ironically, I have git repos with tests including files with intentionally wrong encodings.
I support the following argument:
Syncthing will refuse to synchronize files with names not encoded in UTF-8.
Yet, it’s causing me problems.
There are likely alternatives (“don’t use syncthing for syncing”), that work, but it’d add to the complexity and overhead of my already complex and large system(s?).
One option is to manually transfer the files, and ignore them one-by-one, (as I have done before), but now it’s impractical to do this as well.
I don’t see it smart even asking permission- / for a workaround. It’d be read in sPonGeBoB.
I guess this instead is a rant of how computers and humans are broken, and attempts to fix them are with downsides.
Best bet would be to write an overlaying filesystem, likely.