Bug#519377: Integrate yaclc in devscripts?
On Mon, Jul 20, 2009 at 03:37:19PM +0200, Stefano Zacchiroli wrote:
> On Mon, Jul 20, 2009 at 03:28:34PM +0200, Patrick Schoenfeld wrote:
> > Its an already existing tool. But indeed an interesting question.
>
> It is an existing tool, ... which was going to disappear. So I think
> it is reasonable to ask whether it should really be saved or not.
Yep, I figured that you eventually meant that. But it wasn't
unambigious to say "do we need _yet another_ tool", so I thought
making this clear wouldn't be bad.
> > > If the check is all it does, it looks like a simple "--check" option
> > > can be added to tagpending to achieve the same effect.
> >
> > I don't know if this matches the use-case. AFAICS yacls is meant to
> > be used as a dput hook or so, to tell you "OH! WAIT! You are gonna
> > close the wrong bugs if you really upload now"
>
> Yep, that's clear. The reasoning for associating the two is that
> blatantly they deal with the same entities: last changelog entry, BTS
> inquiry about open bugs and owning packages. Before joining the burden
> of maintaining both, now in the same package, I cannot help thinking
> whether there is an alternative :-)
ACK.
> Regarding your dput hook use case, which is a really interesting one,
> I don't see a problem in that hook being encoded as "tagpending
> --check", you write once, you stop thinking about it forever.
Well, its not mine. I just heard about that idea, when I've thrown a
short look into the package ;)
> That said, I do agree that the name of the tool (tagpending) could
> become non-intuitive. So my proposal is to actually add the option to
> tagpending *and* to have an additional symlink under /usr/bin,
> pointing to tagpending, which would just mimic the (future) behavior
> of the tagpending check. How does that sound?
Sounds reasoable. And in consequence RM yacls, once this is done.
Best Regards,
Patrick
Reply to: