On 2017-08-08 19:24:38 +0200, Christian Boltz wrote:
Am Dienstag, 8. August 2017, 18:31:00 CEST schrieb james:
Vojtěch Zeisek:
Dne úterý 8. srpna 2017 18:19:00 CEST, james napsal(a):
Vojtěch Zeisek:
Dne úterý 8. srpna 2017 13:44:00 CEST, james napsal(a): I install as many as the ones you mentioned plus a few from some web sites and I still get this:
This is weird. What is the complete command? Do all backed up files have names with only ASCII characters?
It is just a GUI program. No commands that I'm currently aware of. The files I want are from my /home dir.
Reading the backtrace indicates that Vojtěch's question points in the right direction - duplicity probably fails at a filename that is not valid utf-8.
Unfortunately the error handler also breaks at the invalid utf-8 which makes debugging this harder :-(
You can try comment out line 79 of /usr/lib64/python2.7/site-packages/ duplicity/util.py by prepending a # # e = unicode(e).encode('utf-8')
This won't fix your problem, but with some luck you'll at least get a useful error message ;-)
Hmm... IMHO, the error message is quite useful:) ... File "/usr/lib64/python2.7/site-packages/duplicity/util.py", line 79, in uexc e = unicode(e).encode('utf-8') UnicodeDecodeError: 'ascii' codec can't decode byte 0xe2 in position 746: ordinal not in range(128) tells you that 0xe2 cannot be decoded using the "ascii" encoding, which is implicitly used (in this case) when executing "unicode(e)". (However, 0xe2 is a valid unicode (see, for instance, "print unichr(0xe2)")) In this case, unicode(e, 'utf8') should do the trick.
Please open a bugreport for duplicity - assuming that filenames are always valid utf-8 is quite optimistic. Handling them as binary is a better idea.
Marcus -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org