On Wed, 3 Oct 2018 19:15:14 +0200 Mattia Rizzolo <mattia@debian.org> wrote: > Hi! > > On Wed, Oct 03, 2018 at 05:20:59PM +0100, Neil Williams wrote: > > https://release.debian.org/transitions/html/python3.7.html > > > > At what point does a package listed as unknown get processed to > > determine if it is good or bad? > > python3 transitions are annoyingly hard to track due to many case > corners. > > Yesterday I failed at least 6 bugs against packages that were marked > as unknown due to the wrong build-depends line (IIRC you were amongst > the maintainers of those packages). > > > What is the trigger for that process? > > that's enterely manual, and the sad bit is that there aren't "notes" > nor anybody is going to manually exclude packages from the tracker. > Except people fixing their wrong build-depends (but that still leaves > cases of packages correctly build-depending on e.g. python3-all-)bg > for tests but still being arch:all and so creating a package with a > depends on only a possibly unversioned python3). > > > How do arch:all packages affect the tracker? > > like all other packages... I don't understand your question. > > > https://tracker.debian.org/pkg/nuitka is marked as good but all of > > the other arch:all packages are unknown. > > most of arch:all packages shouldn't appear in the tracker at all (and > indeed they don't). Thanks, that answered my question above. > > > Equally, packages are listed as unknown with a highlight denoting > > "Dependencies" on another package. If that package is marked as > > good, why is the unknown package not checked? > > What highlighting are you talking about? It's a highlight shown on mouseover of the package name in the transition page. Turns out it's not relevant to the problem itself. > > > The package I care about most is at Dependency level 7 and has a > > highlight Dependencies: pyyaml which is in the good list. How do I > > identify what (if anything) I can do about this being listed as > > unknown? As far as I can tell, the package doesn't depend on any of > > the packages currently listed as bad (most of which are sid-only). > > I assume you are talking about src:lava. > > It's weird, I thought I had sent a bug to that, I wonder why I > didn't... It's ok, I can see the problem from your bug report on black. Thank you, that has made it much clearer. I've updated black and uploaded with a build-depends on python3 but not the rest of the list of alternatives I had included in -1. I'm preparing an update of src:lava as well. > > At any rate, the issue is like this: > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=910094 > I must have messed up my list while running mass-bug.... > > > One of my packages is at Dependency level 1 and unknown but I can't > > tell if I have done anything wrong or how the package affects the > > transition. > > I assume you are talking about src:black. > And indeed I did open a bug for that one yesterday: > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=910094 > (incidentally, the same bug I reported a few lines above :P) > > > At any rate, those false positives only cause noise in the tracker, > they aren't actually hindering the transition if not for people like > you now and me yesterday that wested their time looking at them. All the more reason to fix the false positive. Sorry to have taken up your time and I am grateful for the help. > > > Currently proper overview of the transition is blocked by > src:python3.7 not migrating due to src:openssl blocking the world. OK, I understand. > > -- > regards, > Mattia Rizzolo > > GPG Key: 66AE 2B4A FCCF 3F52 DA18 4D18 4B04 3FCD B944 4540 .''`. > more about me: > https://mapreri.org : :' : Launchpad > user: https://launchpad.net/~mapreri `. `'` Debian > QA page: https://qa.debian.org/developer.php?login=mattia `- -- Neil Williams home@codehelp.co.uk
Attachment:
pgpoLYPf7Ct6h.pgp
Description: OpenPGP digital signature