2009-7-19
| 22:35 | vIkSiT | hehe |
| 22:36 | vIkSiT | hmm or not possibly. I refreshed caches, and it still looks the same |
| 22:36 | jaidan | now if I remember correctly you mentioned firebug which I think has a javascript profiler built in |
| 22:37 | Seamus | vIkSiT: cool, glad to see you figured it out. at first, I was under the impression that the delay was before rendering started |
| 22:37 | vIkSiT | Seamus, thats my impression still. As I mentioned above. I'm going to try in another browser to see how it goes |
| 22:38 | vIkSiT | Seamus, ah, and it looks like its still pre-rending. I've disabled JS from my template itself, rather than use noscript. And rendering through safari rather than FF |
| 22:38 | vIkSiT | rendering* issue |
| 22:39 | vIkSiT | Safari shows "loading n/7 items" |
| 22:39 | wooboodoo | is there anybody experienced with photologue? |
| 22:46 | kikoodu84 | t las ? |
| 22:46 | Anonyme3437 | kikoo |
| 22:46 | Anonyme3437 | ya qqun |
| 22:46 | kikoodu84 | cool=eir�f |
| 22:46 | Anonyme3437 | bjr |
| 22:47 | Anonyme3437 | ya qqun |
| 22:53 | jeremy_c | when starting a new django project, should I use the latest dev version or 1.0.2? |
| 22:55 | vIkSiT | I'd say latest SVN |
| 22:55 | vIkSiT | I"m using it without any problems whatsoever |
| 22:56 | kojiro | vIkSiT: do you svn up every day? |
| 22:56 | vIkSiT | not every day, once a week perhaps |
| 22:57 | SmileyChris | jeremy_c: definitely dev |
| 22:57 | tel0 | why my saved instance doesnt show in admin? |
| 22:58 | tel0 | it's only when i specify list_display |
| 22:58 | tel0 | when i removed list_display, it appears again |
| 22:58 | evmar | what's the proper way to make an command-line utility that interacts with my db via models? is it adding subcommands to manage.py or is there a better way? |
| 22:58 | SmileyChris | evmar: yeah, management commands is the best way |
| 22:59 | tel0 | looks like a bug in django |
| 23:00 | evmar | SmileyChris: feels kinda heavyweight, so i wanted to doublecheck. thanks! |
| 23:00 | tel0 | well... anybody got such problem |
| 23:00 | tel0 | ? |
| 23:06 | tel0 | guys |
| 23:06 | tel0 | |
| 23:06 | tel0 | pls help |
| 23:10 | jeremy_c | Ok, thanks, I just did a svn co. |
| 23:11 | tel0 | SmileyChris: please, can you help me? |
| 23:11 | SmileyChris | tel0: you're not doing anything funky with managers? |
| 23:12 | tel0 | no =) |
| 23:12 | kojiro | interns, then? |
| 23:12 | SmileyChris | tel0: yes, it sounds buggy then |
| 23:12 | tel0 | just 2 simple models, i've broken my mind trying to understand what's the reason |
| 23:12 | SmileyChris | tel0: try removing your verbose names |
| 23:13 | SmileyChris | tel0: do you need to remove both fields from list_display for it to show? |
| 23:13 | tel0 | SmileyChris: yes |
| 23:13 | tel0 | only both |
| 23:14 | tel0 | i use 1.1 beta... |
| 23:14 | SmileyChris | tel0: both those fk models have __unicode__ methods? |
| 23:15 | tel0 | and they're correct |
| 23:15 | tel0 | btw, removing verbose names didnt help |
| 23:15 | tel0 | yep |
| 23:16 | SmileyChris | |
| 23:18 | webesayz | any idea why I would be getting an apache error AttributeError: 'SafeUnicode' object has no attribute 'has_header' - when trying to generate a PDF with Pisa?? - full error at http://pastebin.com/d1814d10d - this is a fresh server and the script works on another site though only difference I can find so far is using Python2.6 vs Python2.5 |
| 23:18 | tel0 | SmileyChris: hm... looks similar, but i got only valid data |
| 23:18 | SmileyChris | or so you think ;) |
| 23:18 | SmileyChris | tel0: if you really want to debug, get django-debug-toolbar and look at the sql calls |
| 23:18 | SmileyChris | then check them yourself |
| 23:19 | seemant | SmileyChris: all I want is my User model to return get_full_name() in its __unicode__() method, instead of username |
| 23:19 | seemant | shouldn't be this hard :( |
| 23:19 | seemant | hey kojiro |
| 23:19 | kojiro | yo, seemant |
| 23:19 | SmileyChris | seemant: huh? |
| 23:20 | kojiro | seemant: have you some code to pastebin? |
| 23:20 | tel0 | SmileyChris: K, thanks. But the data is really correct, i can see it, save etc - but only if list_display doesn't contain fk |
| 23:20 | seemant | kojiro: http://dpaste.com/68955/ |
| 23:21 | seemant | SmileyChris: sorry, that was out of the blue, I know |
| 23:21 | kojiro | heh |
| 23:21 | seemant | SmileyChris: but really it's all I want |
| 23:21 | kojiro | seemant: may I see a bit more of the class? |
| 23:21 | SmileyChris | seemant: so what's the problem? |
| 23:21 | seemant | kojiro: sure, just look in /usr/lib/python/site-packages/django/contrib/auth/models for the whole class |
| 23:21 | seemant | SmileyChris: how do you do it? |
| 23:22 | seemant | kojiro: you're looking for User |
| 23:22 | SmileyChris | you can't override the unicode method for some third party class |
| 23:22 | kojiro | seemant: I see, so you're overriding that class and you really only want to override one method? |
| 23:22 | seemant | no no no |
| 23:22 | seemant | I just want the simplest route to being able to do that |
| 23:23 | seemant | at the moment, I have a frigging proxy model pointing to User, and its only method is that one |
| 23:23 | kojiro | heh, sounds moderately familiar |
| 23:23 | seemant | but this leads to auth.groups and permissions being messed up for all sorts of users |
| 23:26 | seemant | so, like, I could just patch django's models.py class to change its __unicode__ function, but that is probably non-preferred (ie, I'll have to do it, everytime I update django) |
| 23:26 | seemant | or I could subclass User, which means making a custom backend, adding some middleware, etc |
| 23:26 | seemant | all for the pleasure of showing a user's full name |
| 23:26 | seemant | seems a bit hacky |
| 23:28 | joshuajonah got site redesign/clean up done. Woo! | |
| 23:28 | kojiro | |
| 23:30 | kojiro | seemant: I can commiserate with you -- we have a similar frustration. But I don't know the solution... yet |
| 23:32 | kojiro | The right way to fix this would be for Django to implement User in an abstract enough way that it could be subclassed easily, without having to jump through hoops. |
| 23:32 | seemant | ubernostrum: got any suggestions for that, perchance? |
| 23:34 | mib_jago | How would you guys model a "could have" relationship e.g a sale "could have" a cancellation? |
| 23:35 | dstufft | FK with null=True, blank=True |
| 23:35 | dstufft | i think |
| 23:36 | mib_jago | dstufft: so you mean in sale class, have a cancellation field which is FK null=True, blank = True? |
| 23:36 | kojiro | is a Cancellation an object? |
| 23:37 | kojiro | I guess it could be; I was thinking NullBoolean |
| 23:38 | mib_jago | kojiro I have it as a class because sales and cancellations are fed in from spreadsheet and each is its own row. I think it makes it easier for me this way. Especially for the accounts girls to match things up to whats on screen |
| 23:38 | kojiro nods | |
| 23:39 | mib_jago | originally i think I had sale extending cancellation which was horribly wrong... |
| 23:40 | kojiro | heh |
| 23:40 | kojiro | "A cancellation is a type of sale" |
| 23:40 | kojiro | no wait... |
| 23:40 | dstufft | you mean a sale is a type of cancellation :p |