Question

I have a nautilus script to copy tunes that I like into a special folder which I sync to my phone and my car. It fails on paths with funny characters like á in them. I’m fixing it incrementally with stuff like:

temp = temp.replace('%20', ' ')
temp = temp.replace('%5B', '[')
temp = temp.replace('%5D', ']')

but I’m getting tired of these bandaid solutions, and I’m sure there is a better way to do this with str.encode or str.decode.

Does anyone recognise this strange encoding and how I can handle it properly? The problem is, for example, I have a folder such as

/media/music/kálmán balogh and the gipsy cimbalom band/aven shavale

on my disk, but when I get it using os.getenv('NAUTILUS_SCRIPT_CURRENT_URI'), i.e. the currently selected folder in nautilus, it appears in python as

/media/music/k%C3%A1lm%C3%A1n balogh and the gipsy cimbalom band/aven shavale

and then other actions such as renaming or copying the file don’t work because it doesn’t find the file on disk.

Answer 1

You are looking at url encoding. Use urllib.unquote() to interpret these to UTF-8 encoded text, then decode to unicode:

>>> import urllib
>>> urllib.unquote('/media/music/k%C3%A1lm%C3%A1n balogh and the gipsy cimbalom band/aven shavale').decode('utf8')
u'/media/music/k\xe1lm\xe1n balogh and the gipsy cimbalom band/aven shavale'
>>> print urllib.unquote('/media/music/k%C3%A1lm%C3%A1n balogh and the gipsy cimbalom band/aven shavale').decode('utf8')
/media/music/kálmán balogh and the gipsy cimbalom band/aven shavale

In Python 3, you need to use urllib.parse.unquote(); the function was moved.