A proposal on solve encoding problem on Windows.

[Andrey]

Since I do not have a clear picture myself, it is not simple to answer the 
questions...

> Please note, that the encoding is not the property of the repository but 
> the
> > property of the working directory.
> No.  BOTH they have this property and they might be different.
>
> You can guess or make a stab a the enviromnent encoding. But you can't 
> guess the
> repo encoding as this is the encoding on the environment where the commit
> occured.
>
>
> My initial thought was that only UTF-8 shall be the encoding for the 
repository. (that is why one setting is enough) Now I see that for backwards 
compatibility we have to keep the information about the file names encoding 
for the existing repositories (which might be other then UTF-8) to allow old 
Mercurial clients to continue to work.
Indeed, it must be both.

> 3) the old Mercurial versions should work either the same or better (no
> > regression)
>
> Unfortuantely I don't see how that is possible
>
I also do not see now how it is possible. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mercurial-scm.org/pipermail/mercurial/attachments/20111021/9c2e5f7d/attachment-0002.html>