Future changes in the new package and new branch processes

[Pierre-Yves Chibon]

On Fri, Sep 05, 2014 at 06:35:45PM +0200, Haïkel wrote:
> 2014-09-05 17:08 GMT+02:00 Pierre-Yves Chibon <pingou at pingoured.fr>:
> > Dear all,
> >
> > In the last months, Till and I together with infrastructure and
> > release-engineering have been thinking and working on how we could improve the
> > current workflow for new package and new branch.
> >
> > To give you an idea, this is the current workflow:
> >
> > Current new-package procedure:
> > ==============================
> >
> > * packager opens a review-request on bugzilla
> > * reviewer sets the fedora-review flag to ?
> > * reviewer does the review
> > * reviewer sets the fedora-review flag to +
> > * packager creates the scm-request and set fedora-cvs flag to ?
> > * cvsadmin checks the review (check reviewer is a packager, check if reviewee
> >   required a sponsor or not, check if there was a review)
> > * cvsadmin processes the scm-request:
> >   - Create git repo
> >   - Create package in pkgdb
> > * cvsadmin sets fedora-cvs flag to +
> >
> > We thought that a number of these steps could be automated. For example,
> > creating the git repos themselve could be automated using information from pkgdb.
> >
> > Our idea has been to port part of this procedure to pkgdb itself.
> >
> > This is our proposal:
> >
> 
> overall, kudos for this proposal, it looks better than the current workflow.
> 
> One question, when do we close the review ticket ?
> Currently, it is supposed to be handled by the reviewee after the
> first build but it is often forgotten. Off course, it's the
> responsability of the reviewer to check this but it's an impediment.
> Astute people use bodhi to automatically close the ticket when it's
> build on a branched release, but it's not possible on rawhide.
> 
> From what I understand, pkgdb2 would be able to link a package to its
> review, so it should be possible to automatically close the review
> ticket, am I right ?

We can go several ways there:
a) let it the way it is now
   - Closed manualy
   - Closed via bodhi
   - Closed after a while when somone goes through his/her list of tickets and
     see that some can be closed
b) let pkgdb close the ticket (once the package is created)
c) let pkgdb close the ticket when the only branch requested is rawhide (once the
   package is created)

Obviously, a) requires the less work, b) is easy but c) kind of make more sense
as in theory it should be bodhi that closes the ticket (cf a)).

> > New new-branch procedure:
> > =========================
> >
> > * packager requests new branch in pkgdb (2 clicks)
> >      => requests added to the scm admin queue
> > * cvsadmin checks the request/package (check if package exists in the RHEL for
> >   EPEL branch request - check if the user is a packager done in pkgdb itself)
> > * cvsadmin approves the creation of the new branch in pkgdb
> >      => branch creation broadcasted on fedmsg
> > * git adjusted automatically
> >
> 
> *nods*
> Silly question, some of those cvsadmins checks could be automated too,
> at least, providing hints to the scm admins.
> Something like a script listening to fedmsg, doing these checks and
> then sending the results that could be consumed by pkgdb2 (maybe
> displaying them as flash messages ?)

Since the checks will be performed by cvsadmins, manually, I think it make more
sense to do them when the admin runs the script rather than automatically based
on a fedmsg. Integrated back from fedmsg to pkgdb this type of information might
not be easiest thing for no clear advantage, I think.


Thanks for the feedbacks and ideas,

Pierre