[PATCH 0 of 2] keyword extension in contrib

Christian Ebert blacktrash at gmx.net
Wed Mar 7 17:32:33 CST 2007


* Thomas Arendsen Hein on Wednesday, March 07, 2007 at 23:07:58 +0100:
> One thing isn't possible with this, at least I haven't found out:
> The node hashes always list the newest version of the changeset
> which last affected the file,

Yes. That was/is the purpose. To work file-wise, not
repository-wide. Which is why this goes against the spirit of a
DSCM ;)

> but you can't use this to do something like version.py which
> never is edited, but regularly gets a new $Rev$.

I'm not sure I understand. You mean the node corresponding to
"hg identify"? I think this would be easier to implement, almost
with a decode filter. But that would only make sense for
"hg archive", otherwise all configured files would be rewritten
on a commit.

Hm. version.py mainly writes __version__.py. Do you mean
something like that?

> I have some ideas for this in my secret todo list, but currently no
> time to publish them.

Ah, not as shameless as me :)

>> I promise to stop patchbombing now ;)
> 
> No :)  Even if I didn't have time to look at it before, I consider
> this an important thing to remind us that you still are interested
> in this.

That's what I wanted to hear, hehe.

> The test isn't executable (use --git format),

oops, will do.

> and for python2.3 you need the following small patch:

Thanks. Applied at my repos.

c
-- 
_B A U S T E L L E N_ lesen! --->> <http://www.blacktrash.org/baustellen.html>


More information about the Mercurial-devel mailing list