monotone-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Monotone-devel] Re: some (negative) feedback -- useful reading


From: Bruce Stephens
Subject: [Monotone-devel] Re: some (negative) feedback -- useful reading
Date: Wed, 25 Jan 2006 11:15:16 +0000
User-agent: Gnus/5.11 (Gnus v5.11) Emacs/22.0.50 (gnu/linux)

Richard Levitte - VMS Whacker <address@hidden> writes:

[...]

> Read Koen Kooi's second message (21 Jan 01:19).  It looks like there
> was some issue with a revision coming from another -unknown- branch
> or so.  And then he took the opportunity to fiddle with the
> database.  Which part was corruption and what caused what is a bit
> difficult to decipher.

Yeah, <http://thread.gmane.org/gmane.comp.handhelds.openembedded/8657>
doesn't make it sound like anything I'd call "database corruption".
It indicates a problem, that anybody with write access to a repository
can write anything, including stuff that nobody wants there, but I'm
not sure what a good solution would be.  

However, the message doesn't really make complete sense to me.  I
don't see why having a random branch in a repository would cause a
problem, even if some revision becomes a descendent---if a revision
becomes an ancestor of something in that branch, then you'll get the
necessary revisions anyway (even if you don't get the branch certs).

Maybe they do "pull '*'" and things, and allow permission only on some
branches?

Ah, here's the thread, I think:
<http://thread.gmane.org/gmane.comp.handhelds.openembedded/7675>.  I
still don't see what happened, but they give more information, so
probably you can follow it.

> njs> Does anyone know what they mean by git having cherrypicking?  I
> njs> assume this must be just the equivalent of a nice wrapper around
> njs> diff+patch?
>
> I looked around a bit, but couldn't find anything better than this:
>
> http://www.kernel.org/pub/software/scm/git/docs/git-cherry-pick.html
>
> So it looks like your guess is correct.

Probably the previously discussed "monotone patch" would do?  GNU Arch
does a bit more than that: it records that particular changesets have
been applied, and so it can avoid doing so again (if you ultimately
merge the whole branch).  I'm not too sure that that matters much,
though.  

It's odd that one of the negative points against subversion is that it
doesn't have cherry picking---maybe they mean it doesn't have anything
except cherry picking?





reply via email to

[Prev in Thread] Current Thread [Next in Thread]