encoding of 'hg convert' output

[Greg Ward]

On Thu, Aug 27, 2009 at 4:07 AM, Martin Geisler<mg at lazybytes.net> wrote:
> Perhaps the convert function should stop playing with encoding.encoding.
> The sources and sinks can then use their own encoding, which most of
> them seems to do anyway (I saw lots of references to self.encoding).

Perhaps there should be distinct encodings for ui and "everything
else".  I think the encoding.encoding monkey-patch in convert is done
to ensure that what we write to the output Mercurial repo is indeed
UTF-8.  (And it might also help to correctly interpret data coming
from other sane VC systems that require [or strongly encourage]
internal use of UTF-8, like Subversion and Git.)  (I believe that
handling insane VC systems, like CVS, is up to the importer.)

Greg