Name decoding is broken for already encoded characters.

[pschupp01]

I was facing this issue today, and managed to implement a simple failing test.

If a directory has a name containing encoded character it will fail to get the correct directory name.

I have added a directory with name two%20words, containing a single file.
And testing getDirectoryContents on this folder.

Step to reproduce:
• Add ./test/serverContents/two%20words/file2.txt
• Add the following test in getDirectoryContents.spec.js

   it("returns only the directory content when encoded special characters in name", function() {
     return this.client.getDirectoryContents("/two%20words").then(function(contents) {
         console.log(contents);
         expect(contents).to.have.lengthOf(1);
         expect(contents[0].basename).to.equal("file2.txt");
     });
 });

• npm t

[perry-mitchell]

Thanks @pschupp01! This helps alot :)

[perry-mitchell]

Added the failing test here: #94

Will try to address both this and #91

[perry-mitchell]

It's weird though, because it requests the correct URL: http://localhost:9988/webdav/server/two%2520words, but the result is wrong. The result should contain contents that include file2.txt, but instead I get http://localhost:9988/webdav/server/two%20words/file.txt in the response.. Looks like the server might be treating this incorrectly as well.

[perry-mitchell]

It looks like the decoding done on the filenames when the response is received is breaking the process.

EDIT: To be honest, I'm not quite sure where to start with this. It seems that both my test server and home server serve similar results - They decode themselves and the %20 encoding is ignored. Would be happy for someone to take a look at this.