I've updated the patches and added tickets. As I've said in one of the patches, currently I'd go very far trying to avoid tickets simply because of the state the web interface is in: - it's very slow - the HTTPS certificate is unsigned - searching only the tickets is cumbersome - navigation on the page is broken, when I've opened a ticket and hit the back button, I land on the main project page, not the tickets page
Also, in the tracking view, there's no easy way to request a re-pull (apart from adding a comment), and there's no way to add comments to TODO items. All in all, this make creating tickets and adding tickets a very tedious process, on top of the heavyweight patch workflow (as compared to a simple pull request on bitbucket/github). There is also a lot of duplication of patch metadata going on, since the changelog line is effectively the same as the patch header (plus tickets). I prefer sending as many patches as possible upstream (from our patchset), because the whole community benefits and it makes our lives easier when switching to new upstream version. That being said, the whole process that is currently in place turns this into a needlessly high time commitment, especially given the latencies that we had for the last couple times. Sorry for the longish rant, but I'd really prefer what little time I can spent on pylint/astng to be spent on bugfixes and features instead of process. // Torsten On Mon, Mar 18, 2013 at 4:41 AM, Julien Cristau <julien.cris...@logilab.fr>wrote: > On Sun, Mar 17, 2013 at 22:44:07 -0700, Torsten Marek wrote: > > > Hi all, > > > > greetings from PyCon! I only realized today that I should > checked/initiated > > a style-checking/linting/static analysis BoF yesterday. > > > > There's a couple of new logilab.astng patches in my patch queue at > > https://bitbucket.org/tmarek/logilab-astng-patchqueue > > > > They are all fairly small and non-controversial: > > > > brain_hashlib.diff > > > Make sure that set literals are treated as inference leaves. > > > > brain_urlparse.diff > > > Add support for getitem inference on dict literals. > > > > dict_getitem_inference.diff > > > Add support for inference of subscript operations on dict literals. > > > > set_literal_inference.diff > > > Set literals should be treated as inference leaves. > > > > yield_in_lambda.diff > > > Fix inference for generator methods to correctly handle yields in > lambdas. > > > > > > Those are almost all the patches that we have for logilab.astng. I have a > > larger patchset that adds support for having the standard library in a > zip > > file (turns out almost nobody has that). However, I need to come up with > a > > good way to test that, which is complicated by exactly that. > > > > I also have a number of pylint patches (7 to be exact), but one of them > > depends on the patches here, and they're also not completely done yet > > (cleanup/testing). > > > Thanks Torsten. I've pushed this set to > http://hg.logilab.org/review/logilab/astng > and > https://www.logilab.org/project/logilab-astng?tab=vcreview_patches_tab > > Cheers, > Julien > -- > Julien Cristau <julien.cris...@logilab.fr> > Logilab http://www.logilab.fr/ > Informatique scientifique & gestion de connaissances > -- Site Reliability Engineer ∘ Google ✚ ∘ ⬕
_______________________________________________ Python-Projects mailing list Python-Projects@lists.logilab.org http://lists.logilab.org/mailman/listinfo/python-projects
