| 2016-08-31 22:03:21 |
Göran Uddeborg |
description |
When I try to start calibre, I get a backtrace that ends with
File "/usr/lib64/calibre/calibre/utils/ipc/__init__.py", line 43, in gui_socket_address
ADDRESS = (u'\0%s-calibre-gui.socket' % user).encode('ascii')
UnicodeEncodeError: 'ascii' codec can't encode character u'\xf6' in position 2: ordinal not in range(128)
My login is "göran", so this isn't really surprising given the above code. I'm not sure why the address needs to be converted to ascii. But if it has to, perhaps the conversion could be made less strict, e.g.: encode('ascii','replace')
Non-ASCII login names are probably violating some standard somewhere, and a strict reasoning might claim this is not a bug. But does calibre really have to be that restrictive? |
When I try to start calibre, I get a backtrace that ends with
File "/usr/lib64/calibre/calibre/utils/ipc/__init__.py", line 43, in gui_socket_address
ADDRESS = (u'\0%s-calibre-gui.socket' % user).encode('ascii')
UnicodeEncodeError: 'ascii' codec can't encode character u'\xf6' in position 2: ordinal not in range(128)
My login is "göran", so this isn't really surprising given the above code. I'm not sure why the address needs to be converted to ascii. But if it has to, perhaps the conversion could be made less strict, e.g.: encode('ascii','replace')
Non-ASCII login names are probably violating some standard somewhere, and a strict reasoning might claim this is not a bug. But does calibre really have to be that restrictive?
Environment:
Fedora Linux
calibre-2.65.1-1.fc25.x86_64 |
|
