[PATCH] Check case-folding clashes during addremove

Andrei Vermel avermel at mail.ru
Thu Mar 29 02:43:31 CDT 2007

Okay, here goes.

I do the check and issue warning message regardless of filesystem.
Add is only aborted on a casefolding filesystem.


----- Original Message ----- 
From: "Matt Mackall" <mpm at selenic.com>
To: "Andrei Vermel" <avermel at mail.ru>
Cc: "Alexis S. L. Carvalho" <alexis at cecm.usp.br>; 
<mercurial-devel at selenic.com>
Sent: Thursday, March 29, 2007 9:49 AM
Subject: Re: [PATCH] Check case-folding clashes during addremove

> On Thu, Mar 29, 2007 at 09:58:56AM +0400, Andrei Vermel wrote:
>> >This really should go into dirstate.py, so that hg add can also benefit
>> >from it.  If you look at the dirstate.update function, you'll see that
>> >we already do some checks when state == 'a' (i.e. we're adding a file).
>> There's a small problem with this. Although dirstate.update() takes 
>> 'files',
>> turns out it is called for each file separately. Rebuilding the name 
>> clash
>> map for each added file would be an overhead for big number of added
>> files.
>> As an option, the clash map can persist in dirstate and get updated
>> incrementally with dirstate.map. What do you think?
> Yes, definitely.
> -- 
> Mathematics is the supreme nostalgia of our time.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: addremove_clashes
Type: application/octet-stream
Size: 4072 bytes
Desc: not available
Url : http://www.selenic.com/pipermail/mercurial-devel/attachments/20070329/8d45b119/addremove_clashes.obj

More information about the Mercurial-devel mailing list