UnicodeDecodeError with "bzr fast-import"

[Paul Moore]

2008/5/18 Teemu Likonen <tlikonen at iki.fi>:
> I'm trying to fast-import two smallish git repositories to bzr.
> Unfortunately I'm facing character encoding problems which I'm not able
> to solve. I'm using the latest development revisions of both bzr and bzr
> fast-import plugin.
>
> The other repo does not convert at all. It stops almost immediately and
> displays the following error:
>
>  bzr: ERROR: exceptions.UnicodeDecodeError: 'ascii' codec can't decode
>  byte 0xc3 in position 5: ordinal not in range(128)
>
> The other repo converts nicely but in bzr commit messages individual
> bytes in UTF-8 multibyte sequenced are converted to separate characters.
> I.e. "ä" which is UTF-8 encoded as c3a4 (hex) will display as "ä" in
> bzr log output.
>
> Any ideas how to solve this?

Sounds like the same bug I hit a couple of days ago importing from
svn. I've reported it as bug number 230996 - Ian said he hopes to get
to it sometime next week.

Paul.