2008-6-13
| 22:32 | brosner | none of the core developers are here right now to make the design decision |
| 22:33 | Gulopin1 | yeah, i can't speak on behalf of anyone either, but i'm pretty sure robustapply is going away with the refactor |
| 22:33 | fijal | I *know* that |
| 22:33 | fijal | it just takes a while |
| 22:34 | Gulopine | trust me, i hear that |
| 22:34 | brosner | :( |
| 22:36 | fijal | brosner: I'm pretty sure that I lack expertise helping in that |
| 22:36 | Gulopine | interestingly, the pypy patch and the jython patch look dramatically different as well |
| 22:36 | Gulopine | so i'm not sure it's clear which should be committed anyway |
| 22:37 | fijal | Gulopine: I can produce pypy-compatible patch looking like jython |
| 22:37 | fijal | it's just instead of adding cases |
| 22:37 | fijal | we provided minimal subset that satisfies everybody |
| 22:37 | fijal | they're not different how they work |
| 22:37 | [530]_ | there are som fixes that are needed for double registration in the createsuperuser management function too, is this related to robustally? |
| 22:37 | Gulopine | okay |
| 22:37 | [530]_ | apply |
| 22:37 | fijal | they have just dramatically different if paths |
| 22:38 | Gulopine | i just really trust leo's judgement when it comes to jython, and i want to make sure you're not trashing his work :) |
| 22:38 | fijal | I'm not |
| 22:38 | fijal | well |
| 22:38 | fijal | you know |
| 22:38 | fijal | it's software, not psycology |
| 22:38 | fijal | we've got tests proving stuff |
| 22:38 | Gulopine | heh |
| 22:38 | fijal | if they can run on jython and pypy, it really doesn't matter who you trust |
| 22:39 | Gulopine | except that when it comes to jython and pypy, our tests are often based on erroneous assumptions |
| 22:39 | Gulopine | so sometimes the tests need to be adjusted as well, based on an intimate knowledge of the VM |
| 22:39 | Gulopine | i'm not saying this is such a case |
| 22:39 | Gulopine | but just that i know leo and i trust his knowledge |
| 22:39 | Gulopine | so i just wanted to make sure your patch took his work into account as well |
| 22:39 | Gulopine | if it does, we're good |
| 22:40 | Gulopine | that's all |
| 22:40 | [530]_ | this http://code.djangoproject.com/browser/django/tr... is pretty nasty |
| 22:41 | Gulopine | [530] hahahah, that is pretty terrible |
| 22:41 | fijal | Gulopine: really, that's why we wrote a test for that |
| 22:41 | [530]_ | partly my fault too |
| 22:41 | Gulopine | that double registration stuff really sucks |
| 22:41 | fijal | (attached to a ticket right now) |
| 22:42 | Gulopine | fijal, sorry, i won't pretend to know what's attached to every ticket in trac :( |
| 22:42 | Gulopine | i'm good, but i'm not that good |
| 22:42 | fijal | that's why I'm telling you |
| 22:42 | Gulopine | okay |
| 22:42 | Gulopine | thanks for letting me know |
| 23:14 | siudesign | There are quite a few refences to 0.97 and 0.98 in current django docs. I was thinking we should probably remove those, given the next move is to 1.0 in September |
| 23:16 | mattmcc | Hmm. I see one reference to each. |
| 23:17 | mattmcc chuckles at the 0.98 reference. | |
| 23:17 | mattmcc | "Either 0.98 or 1.0, whichever comes first." |
| 23:17 | siudesign | indeed, that would have been nice.... a while ago ;) |
| 23:17 | siudesign | glad 1.0 is se for september though |
| 23:18 | siudesign | *set |
Page 2 of 2
← Previous
(150 total)