Mingye Wang
A good-ish polyfill for our desired functionality is found in [gnulib euidaccess(3) (GPL!!)](https://github.com/coreutils/gnulib/blob/master/lib/euidaccess.c). It gives two implementations: the correct but non-reentrant one using a setre{u,g}id-access-setre{u,g}id pattern and the traditional stat...
Yeah, someone will need to wipe their brain with only the phrase "implement an access check based on stat() or setuid and access(2)" left if they decided to implement it...
Makes sense. An `open()` is needed for the perm check anyways. The `chmod()` stuff is a bit extra, but ehh... let it slide.
Meh, reasoning through the open fds and fallbacks with `clonefile(2)` is still too hairy for me. * I need to `unlink` the file for `clonefile` to work, or it would...
There isn't much I can change right now, so please just throw more comments at me.
That wouldn't be possible since fcopyfile() does not perform cloning. I am not sure if it's in the documentation, since copyfile.c is a bit easier to Google. (If this was...
I don't think there is anything more I can do.
In light of an updated PR being available I am closing this one. Just don't have time for code when I am all stuck at McD, I am afraid.
Using ZIM files directly still has the issue of the built-in clustering and compression likely reducing the dedupe power. Anything we can reap from running a rolling hash on an...