bzr 0.13.0 Add Error: Path is not unicode normalized

[Alexander Belchenko]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dongsheng Song пишет:
> Here is my  .bzr.log related to this error:
> InvalidNormalization: Path
> u'\u56fd\u5bb6\u8f6f\u4ef6\u4ea7\u4e1a\u57fa\u5730\uff08\u5e7f\u5dde\uff09\u4f7f\u7528\u6307\u5357\uff0d\u4e34\u65f6\u7248.doc'
>
> is not unicode normalized
>
> return code 3

Thank you for log.

The problem itself not with your CJK, but with punctuation marks in filename
and unicode normalization itself.
Because normalized form change symbols '()-' to FULL WIDTH ones. I compare char
to char (original and normalized form) and this characters is different:

8th: u'\uff08' != u'('
11th: u'\uff09' != u')'
16th: u'\uff0d' != u'-'

I need John Meinel opinion on this problem.

Actually unicode normalization initially was written for support
of MacOS. But IIRC it's related to characters with diacritic signs?

2Dongsheng Song:
as short-term workaround: you could remove '()-'
from filename that cause error. After file was added,
you could try to rename it to desired form. If rename fails
you will rename this file later, when 0.15 is out
(I'm sure we fix this problem in next release).

Alexander

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFv0x7zYr338mxwCURAqxxAJ4kMSrTLzXmIXdmqB8is6OLowCGNQCdFN7u
BCAOTi9WsLezVXbnAkWlzLU=
=AuJy
-----END PGP SIGNATURE-----