On 8/10/07, Martin Albrecht <[EMAIL PROTECTED]> wrote: > there have been some complaints about the quality of SAGE code recently, so > this is a proposal how to tackle this problem.
Yes! > 1.) Everybody fills a bug report about _anything_ annoying, wrong, broken in > SAGE. Everything! Really, everything! Its easy to close a bug with 'wontfix' > or 'cannotreproduce' and no one will be mad. The only condition is to search > the bug reports first to check whether it has been reported already. Definitely! > I don't know how you handle this, but I often don't fill bug reports because > (a) I don't care enough or (b) I'll have to fix the bug myself anyway. But I > usually make notes in the code I am writing. This should change: lots and > lots of bugreports. Also, fill in the features you are planing to implement > as a feature request and assign yourself. This way it is transparent who > wants to do what and it is easier to get in touch. Definitely. > 2.) We agree on - lets say - one Sunday a month when we all meet on IRC and > sit down fixing those bugs. No new features only bug squashing! Many projects > hold events called 'Bug Squashing Party' and this would be our 'party. I.e. > we go through the bug reports, assign people, fix them and close them (many > obsolete bug reports on Trac were not properly closed). That's an excellent idea. Could you organize it? I've long wanted to have something like this. > Potential refinements: > > 3.1) The KDE team has the concept of a "Junior Job". That is a bugfix or a > feature request someone new to the project could be able to handle. E.g. > finding an obscure SEGFAULT in code related to libSINGULAR on AMD64 machines > only is definitely no Junior Job, fixing > > http://www.sagemath.org:9002/sage_trac/ticket/415 Great idea. Specifically marking some bugs so that fixing them is a good "entry point" to SAGE. By the way, I can update the list of bug categories in trac. What would people like added? Here's a bug that I haven't reported on trac, which has "bugged" me forever. If you do: rank4:~/s was$ sage -t devel/sage-main/sage/matrix/constructor.py sage -t devel/sage-main/sage/matrix/constructor.py 3 then there's a *mysterious* 3 that appears. I have no idea why. [...] wait it just occurred to me as I reported this to type: $ cd SAGE_ROOT/local/bin/; grep 'print 3' sage-* which quickly yielded sage-doctest: print 3 so I've fixed that bug. This just illustrated the value of reporting bugs! > (ZZ.random_element(0) --> crash) is. Btw. I've just fixed and closed it. > > The KDE devs simply tag junior jobs with ("JJ:") in the bug description and > have a nice way to tell people how to get involved easily. Great idea. > 3.2) Trac should be configured to send out e-mails if something changes for a > bug that is assigned to me. Good. Does anybody here know a lot about configuring trac? -- William --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/ -~----------~----~----~----~------~----~------~--~---
