When I download a folder with nested folders as a Zip archive, my directory structure is partially unwound and duplicates are created of each recursive folder with alternate case conventions and some of my files in one folder and some in another.
Let me show you a real world example. Let's say I have this directory structure in Dropbox:
Taxes
-> 2015 Taxes
-> 2015 Tax Documents
-> 2015 Morgan Stanley Tax Documents
-> 2015 Morgan Stanley 1099.pdf
-> 2015 Fidelity 1099.pdf
-> 2015 Wells Fargo 1099.pdf
-> 2015 Final Tax Return.pdf
If I click the "Taxes" folder and select "download", I'll get a zip archive file on my local machine. If I then extract that zip file, the extracted contents will have the following (REALLY messed up) directory structure:
Taxes
-> Taxes (extra level of indirection, WTF?)
-> 2015 Taxes (empty, WTF?)
-> taxes (duplicated and lowercased, WTF?)
-> 2015 Taxes (duplicated, lowercased, and empty, WTF?)
-> 2015 taxes (duplicated and lowercased, WTF?)
-> 2015 Tax Documents (duplicated, WTF?)
-> 2015 Morgan Stanley Tax Documents (duplicated and empty, WTF?)
-> 2015 Fidelity 1099.pdf
-> 2015 Wells Fargo 1099.pdf
-> 2015 tax documents (duplicated, lowercased, and missing files, WTF?)
-> 2015 Morgan Stanley Tax Documents
-> 2015 Morgan Stanley 1099.pdf
-> 2015 Final Tax Return.pdf
I now get to go through the fun exercise of reconstructing my original folder structure, which I took great pains to get the way I wanted it in the first place -- never mind the fact that I'll have to deal with this *every* time I download a zip archive of a folder from Dropbox.
So yeah, this is a pretty crappy user experience. I think you guys have some problems with your directory tree traversal algorithms that you use when you build your zip files.
I think you should get some engineers looking at this, ASAP.
