Reviewing spkgs used to be really painful--you had to unpack the new spkg and the old one and compare them manually, make sure the repo was up-to-date, the src/ was unmodified upstream, SPKG.txt had a (redundant) changelog entry, look at the patch files, etc. Also, rebuilding Sage to test a package (or risking contaminating your working copy) was bothersome as well. As a result they often languished.
This is something that should be much better in the git world: now a package update is just a git commit like any other change to Sage. Ideally it's just a version bump, in which case it's an easy review, otherwise you look at the build script modifications (and the effort spent is proportional to the size of the change, as it should be). Sometimes there is new and/or changed code and doctests in the Sage library itself, which should be reviewed like usual. As before, we more-or-less defer to the upstream's release process for the actual upstream code. This is orthogonal to installing the package and seeing that all tests still pass--that's not the reviewer's job (it can be automated, looking at the code can't be). If the new tarball is in upstream/, the patchbot can actually do this now, though we should probably enable downloading from a url in the ticket itself too (ideally this is a pointer to the actual upstream's release tarballs, so if Folker Brown bumps the number and says the new tarball is at http://www.gap-system.org/pub/gap/gap47/tar.bz2/gap4r7p2_2013_12_01-10_17.tar.bz2 that's easy to review). - Robert On Wed, Dec 25, 2013 at 10:45 AM, Francois Bissey <[email protected]> wrote: > Going back to the general subject. Could we have a shortcut on the ticket > report page > grabbing update/upgrade of packages. Unless the potential reviewer(s) are cc > automatically > that would any review easier, as in more easy to find. > > François > ________________________________________ > From: [email protected] [[email protected]] on behalf of > Francois Bissey [[email protected]] > Sent: Wednesday, 25 December 2013 12:15 > To: [email protected] > Subject: RE: [sage-devel] Reviewing package version updates (in particular, > gap) > > No. And yes that means I volunteer. Of course that means that sometimes I'll > just be a second pair of eyes to spot any glaring stuff we don't want. > The buildbot reviews a number of things far better than a single reviewer can > do. > > François > ________________________________________ > From: [email protected] [[email protected]] on behalf of > Volker Braun [[email protected]] > Sent: Wednesday, 25 December 2013 12:01 > To: [email protected] > Subject: [sage-devel] Reviewing package version updates (in particular, gap) > > The current process of updating third-party packages is broken: > > a) There isn't really anything to review beyond running it on the buildbot > b) Nobody dares to press the "positive review" button > c) Frustation to see your work waste away on trac > > for example, out of the last three Gap package updates that I packaged only > one was ever looked at. So I'm proposing that I'll just merge package updates > if they *work* as long as there isn't too much wonkyness introduced into the > build script. If you have any opinion on that, please post in this thread :-) > > [ ] Yes, go ahead > [ ] No, I want to volunteer as the reviewer for package upgrades > > See also: http://trac.sagemath.org/15582 > > -- > You received this message because you are subscribed to the Google Groups > "sage-devel" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/sage-devel. > For more options, visit https://groups.google.com/groups/opt_out. > > This email may be confidential and subject to legal privilege, it may > not reflect the views of the University of Canterbury, and it is not > guaranteed to be virus free. If you are not an intended recipient, > please notify the sender immediately and erase all copies of the message > and any attachments. > > Please refer to http://www.canterbury.ac.nz/emaildisclaimer for more > information. > > -- > You received this message because you are subscribed to the Google Groups > "sage-devel" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/sage-devel. > For more options, visit https://groups.google.com/groups/opt_out. > > This email may be confidential and subject to legal privilege, it may > not reflect the views of the University of Canterbury, and it is not > guaranteed to be virus free. If you are not an intended recipient, > please notify the sender immediately and erase all copies of the message > and any attachments. > > Please refer to http://www.canterbury.ac.nz/emaildisclaimer for more > information. > > -- > You received this message because you are subscribed to the Google Groups > "sage-devel" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/sage-devel. > For more options, visit https://groups.google.com/groups/opt_out. -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/groups/opt_out.
