[darcs-users] Re: [BK] upgrade will be needed

Andrea Arcangeli andrea at suse.de
Tue Feb 22 14:05:27 UTC 2005


Hi Ralph,

On Tue, Feb 22, 2005 at 10:25:24AM +0000, Ralph Corderoy wrote:
> 
> Hi Andrea,
> 
> > >     (c) Directly corresponds to traditional diff 'n' patch, easy to
> > >     think about, no surprises
> > 
> > Nobody is supposed to touch the repo with an editor anyway, all it
> > matters is how fast the command works.
> 
> It's nice for it to be available for reading though.  grep, awk, & Co.
> can often ask queries that the SCM doesn't allow.

Well that's because the SCM is missing features ;)

anyway what I normally ask is related to the history of a single file,
so having all modification of the same file in one palce is more
efficient for that too.


the thing about the disk and nfs corrupting data over time doesn't seem
very good argument. First I'd never export any repo with nfs and I'd
always use ssh for all commits. Second if you do incremental backups
with diff -urN you'll be in the same situation as with the patchset
per-file approach (i.e. that you may have future patches not applying
anymore after fixing up the longstanding corruption). The only way I see
a protection against those corruption sources is to store a md5 hash of
the new contents of the file atomically in each commit. So before doing
a new commit you can check the old copy in the repo is still coherent.

> lkml dropped.

correct ;)




More information about the darcs-users mailing list