I think I have a good idea how to fix this. in lldbsuite.support we can
add unicode_file.py. Make it implement the file interface so that it looks
just like an io object so it can be used with print(file=), and with the
'with' keyword, etc. Make the open method call io.open() with an encoding,
then make the read / write methods do a python version check before sending
the read / write call to the underlying io object. Then you basically just
do
self.session_file = unicode_file.open(path, mode='w', encoding='utf-8')
and then you can write however you want.
print(u"", self.session_file)
print("", self.session_file)
will both work in any python version
On Thu, Jan 28, 2016 at 10:21 AM Tamas Berghammer <tbergham...@google.com>
wrote:
> tberghammer added a comment.
>
> The traceback calculation is done inside unittest2 so I think we shouldn't
> change it and we don't have access to the session variable either.
>
>
> http://reviews.llvm.org/D16680
>
>
>
>
_______________________________________________
lldb-commits mailing list
lldb-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
