How to work around a unicode problem?

[Peter Otten]

tinnews at isbd.co.uk wrote:

> I have a small python program that uses the pyexiv2 package to view
> exif data in image files.
> 
> I've hit a problem because I have a filename with accented characters
> in its path and the pyexiv2 code traps as follows:-
> 
>     Traceback (most recent call last):
>       File "/home/chris/bin/eview.py", line 87, in <module>
>         image = pyexiv2.ImageMetadata(filepath)
>       File "/usr/lib/python2.7/dist-packages/pyexiv2/metadata.py", line
>       65, in __init__
>         self.filename = filename.encode(sys.getfilesystemencoding())
>     UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position
>     38: ordinal not in range(128)
> 
> Without digging deep into pyexiv2 is there any way I can work around
> this error?  The accented characters aren't in the filename itself,
> they're in the directory path.  I.e. it's:-
> 
>     ./1977/04 April/#09 - Monaco, inc. Musée de Poupée/p77_08_011.jpg
> 
> I could of course remove the accents but I'd much prefer not to do so.
 
Try passing a unicode filename. A quickfix:

filepath = filepath.decode(sys.getfilesystemencoding())
image = pyexiv2.ImageMetadata(filepath)

If you are using os.listdir() or glob.glob() to produce the filepath -- they 
will return unicode filenames if you invoke them with a unicode argument.