Unicode support request.

[罗勇刚(Yonggang Luo)]

The right, but utf8 support is also needed. My point of view is leave
all old repo as is. When new repo created, we can still create old
fashion by option. For rest condition, one is new repo, we can setting
the repo encoding to UTF8, the second is upgrading from existed old
repo by clearly specify the old repo. I think that's all, any
supplement?

2011/10/21, Lester Caine <lester at lsces.co.uk>:
> 罗勇刚(Yonggang Luo)  wrote:
>> I will verify it. But it's should be better you send me a simple Mercurial
>> created under linux utf8,
>> which contains a single file that the name contains 'accent' characters.
>> Then I
>> can study on it:)
>> Because my method of generating 'accent' character may different to you, I
>> will
>> directly copy it from internet.
>> But you may directly typing it. That's may differ:)
>
> THIS is exactly the problem!
> What is being loaded into mercurial are files from any source, and it is the
> 'unreliability' of that process, cross platform, which is the problem here.
>
> I don't have regular contact with the problem, but hit it quite regularly
> when
> some of my European windows based contacts supply material to work on.
> Yesterday
> I simply created a file in an hg repo in linux called 'áéíóúñ.txt' accessed
> it
> from windows, played with the encoding in windows and copied back ( without
> using hg! ). Now 'type áéíóúñ.txt' from the command line in Linux says 'file
> not
> found'. I'm not familiar enough WITH the problem to go much further than
> that,
> so personally I avoid anything complex in file names and there is no
> problem.
> Not the right approach, but until PHP is able to work with Unicode file and
> function names my hands are tied ... all of my own development work is in
> PHP
> and while UTF8 data is not a problem moving beyond that is a minefield.
>
> The other threads on this are detailing just how complicated the problem is
> and
> the difficulty of backwards compatibility ... I've some 15 years of data in
> hg
> repo's which on the whole has been tidied to loose the windows 'codepage'
> problems, but still bites from time to time. Until UTF8 is transparently
> working
> in ALL software there will not be a clean fix that everybody is happy with?
>
> --
> Lester Caine - G8HFL
> -----------------------------
> Contact - http://lsces.co.uk/wiki/?page=contact
> L.S.Caine Electronic Services - http://lsces.co.uk
> EnquirySolve - http://enquirysolve.com/
> Model Engineers Digital Workshop - http://medw.co.uk//
> Firebird - http://www.firebirdsql.org/index.php
> _______________________________________________
> Mercurial mailing list
> Mercurial at selenic.com
> http://selenic.com/mailman/listinfo/mercurial
>

-- 
从我的移动设备发送

         此致
礼
罗勇刚
Yours
    sincerely,
Yonggang Luo