“Mac-style” unzipping on linux

Question

I'm using Linux Mint MATE edition, in case that's relevant.

I don't know if this is Dropbox specific, but when i download a zip of a folder that was originally produced on a Mac (and that has lots of nested subfolders) from dropbox, and unzip it on my Linux machine, it's completely messed up: all folders have a duplicate with a downcased version of their name, and the subfiles appear in the duplicate instead of the original. If one of my Mac-owning colleagues unzips that same zip, then it's fine.

This is what i should get:

And this is what i actually get:

My guess is that this is somehow related to the fact that Macs are non-case-sensitive with file names, and thus there's some weird incompatibility between file system metadata, or something.

Does anyone know of a linuz unzip tool which replicates the behaviour of the Mac unzipper, or otherwise produces the correct result? I don't mind if it leaves the_MACOSX folder in there (i can just delete these), as long as the main folder is fine.

Answer 1

It does look like the problem is that the zip file uses inconsistent case variations for directory names. This is exactly what the ciopfs filesystem is for: it provides a case-insensitive view of a directory tree. There's an Ubuntu package. Create a view of an existing directory by mounting the ciopfs filesystem on a temporary directory, unpack the zip inside that view, and you'll get the archive extracted in the original directory.

mkdir ~/case-insensitive
ciopfs ~/temp ~/case-insensitive
cd ~/case-insensitive
unzip -x __MACOSX -x '__MACOSX/*' /path/to/zip
cd ~/temp
fusermount -u ~/case-insensitive

If that doesn't work for some reason, my next candidate would be the Python zipfile library (example).