Iñigo Martínez

Yes, which also affects the `patch_filename` key value in wrap file format, which is always `.zip`. This issue affected the [new `mocklibc` wrap file](https://github.com/mesonbuild/mocklibc/pull/1) because it contains [two](https://github.com/mesonbuild/mocklibc/pull/1/commits/233c1b55673e12fc08d93ba311b7c2e2dd87dc5a#diff-cb03e8d01b388fcaee2733206ee201a0) [files](https://github.com/mesonbuild/mocklibc/pull/1/commits/233c1b55673e12fc08d93ba311b7c2e2dd87dc5a#diff-3b24a45a8f8b39b8951169b8c4c0e8f2) that...

[A workaround seem to be available](https://bugs.python.org/issue15795#msg311328). Adding support for different archiving utilities/compression algorithms should also help avoiding this issue while also offering a nice feature. What do you think?

The reason looks very sensible to me. I don't really have a strong opinion about supporting other formats, it was just a way to workaround the permission issue.

> Would be nice to have a test that uses generated files, or that even uses expand_content_files at all. > > Also this parameter to gnome.gtkdoc() is completely undocumented. :(...