2008-3-3
| 23:15 | Zalamander | visik7 personally, I think it's best to keep them separated, yes. |
| 23:15 | Zalamander | you can get RequestContext with render_to_response as well, but it's more convoluted |
| 23:19 | Emilian | tess, it is preference..but I am not sure if I understand..a plugin's size depends on what it is written, if it is compressed etc. not on JQuery itself. Regardless, it is a moot point. |
| 23:21 | tesseracter | Emilian, actually, i now remember the exact reason i dont use jQuery. the sortables serialize function. mootools is the only serializer that looks like [ "id", ["foo", "bar"], "baz"]. everyone else looks like [ "id_1", "bar_3", "foo_2", "baz_4"] and that is a pain. |
| 23:23 | pedjango | what's the proper way of ensuring certain URLs are always HTTPS? |
| 23:23 | tesseracter | i mean, whats with the numbers? i would have had to make my own serializer just to format my sortables. with mootools i just try them in as a list. no numbers. pretty silly reason, but oh well. |
| 23:24 | Emilian | whatever floats your boat |
| 23:24 | EyePulp | probably only of interest to myself, but it seems like you can feed strings as date values for object filtering. It seems to correctly convert to date vales for criteria (which SQL does internally anyway, but I couldn't figure out if django did too) |
| 23:25 | EyePulp | s/vales/values |
| 23:26 | Emilian | no more s/// D: |
| 23:30 | AzMoo | In the docs for modelforms it says: "The form field's label is set to the verbose_name of the model field, with the first character capitalized." but this doesn't work for the label attached to the errors. Is that a bug? |
| 23:30 | Emilian | aZmOO i LOOKED INTO IT |
| 23:30 | Emilian | IT IS BY DESIGN |
| 23:30 | Emilian | haha caps |
| 23:30 | AzMoo | !! |
| 23:30 | ubernostrum | AzMoo: error keys never use the verbose_name or the label. |
| 23:30 | Emilian | I was waiting for you to come online |
| 23:30 | Magus- | and no, that isn't a bug |
| 23:30 | AzMoo | By design? Why? |
| 23:30 | ubernostrum | Error keys always use the actual Python-level genuine name of the field. |
| 23:30 | Emilian | the key in the ErrorDict is the field name not the field.label |
| 23:30 | ubernostrum | AzMoo: well, what happens when someone translates your form class? |
| 23:30 | Magus- | one normally shows errors next to a field |
| 23:31 | ubernostrum | AzMoo: suddenly the key changes based on active language. |
| 23:31 | ubernostrum | Whoops. |
| 23:31 | Emilian | it could still pass the field.label though |
| 23:31 | Magus- | not to mention that keys like "some label with text" wouldn't be valid in templates :) |
| 23:31 | ubernostrum | Indeed. |
| 23:31 | Magus- | er, 'with spaces' should've been there |
| 23:31 | ubernostrum | So, for quite good reasons, the actual honest to goodness name of the field is used. |
| 23:32 | Emilian | right but you can still pass the field.label in the errors |
| 23:32 | Emilian | and keep the key as the field name |
| 23:32 | Emilian | and then just display the field.label |
| 23:32 | AzMoo | Emilian: That's what I was thinking |
| 23:32 | AzMoo | damn, I was about to follow up on that with him. |
| 23:33 | Emilian | I put in a ticket, maybe it will get changed |
| 23:33 | AzMoo | Magus-: I'm not talking keys, I'm talking about the label displayed by errors.as_ul |
| 23:33 | Emilian | #6705 |
| 23:33 | DjangoBot | |
| 23:33 | AzMoo | Emilian: Thanks for that mate, much appreciated. |
| 23:33 | Emilian | no problem |
| 23:35 | AzMoo | Magus-: Although looking at the actual code for that part of it I see what you mean, but I think there'd be a better way to do that. |
| 23:35 | FunkyBob | DRY... you already have the field label in the field list... |
| 23:35 | AzMoo | Yep, it's just longer. |
| 23:36 | AzMoo | Wait, what does DRY mean? |
| 23:36 | Magus- just uses {{ form.foo.errors }} next to the field location | |
| 23:36 | brehaut | dont repeat yourself |
| 23:36 | AzMoo | Magus-: Yeah, it doesn't fit in this layout though, so I figured I'd put them all on top. |
| 23:37 | Emilian | same here |
| 23:37 | Emilian | it shouldn't be that hard to pass it in along with e.messages |
| 23:37 | Magus- | Emilian: time to write a patch then ;) |
| 23:37 | Emilian | I've thought about it |
| 23:37 | Emilian | lol |
| 23:38 | Emilian | the way I'd do it is by having the value be a dictionary with a label key and an errors key |
| 23:38 | brehaut | does anyone know what file in django actually loads the settings file? |
| 23:38 | AzMoo | Now, this ticket: http://code.djangoproject.com/ticket/4046 <-- Is the reason that ubernostrum's patch hasn't been included because of the lack of tests? |
| 23:39 | Magus- | brehaut: django.conf would likely give you a hint to that |
| 23:39 | AzMoo | Emilian: I think that'd be a great idea, but I'm thinking totally backwards incompatible. |
| 23:39 | Magus- | since you do "from django.conf import settings" to get at them |
| 23:40 | brehaut | yes |
| 23:40 | brehaut | i had missed the __init__ |
| 23:40 | brehaut | i havent been in python or django land for a while |
| 23:42 | Emilian | I don't see what it has to be compatible with |
| 23:42 | necrite | in the doc of django auth it say that we have to use login( request, user ) but when I run the code I get login() takes exactly 1 argument (2 given) .. any idea? |
| 23:43 | Magus- | necrite: don't name your view login, of course |
| 23:43 | Magus- | ;) |
| 23:43 | Emilian | that one got me too when I started as well |
| 23:43 | Emilian | ha ha |
| 23:43 | Emilian | WTF IS GOING ON?! |
| 23:43 | Emilian | D: |
| 23:44 | AzMoo | Emilian: Existing code? Such a change would break anybody using ErrorDict, wouldn't it? |
| 23:44 | bem | anyone using django.contrib.gis with the admin application? |
| 23:44 | necrite | Magus-, heheh is the most standar method :P |
| 23:45 | brehaut | anyone know if there is a way to get a reference to the module that statements are executing in ? |
| 23:45 | Magus- | brehaut: why? |
| 23:46 | brehaut | magus: so i can dynamically add attributes to it |
| 23:46 | Magus- | why? |
| 23:46 | brehaut | so i can have attributes injected into a settings file dynamically |
| 23:47 | Magus- | why? |
| 23:47 | amerinese | can i specify a custom widget for just one _instance_ of a newform? |
| 23:47 | Magus- | jump to the end result already |
| 23:47 | Magus- | there's likely an easier way to do things |
| 23:47 | FunkyBob | brehaut: enough with the how... tell us the WHY |
| 23:48 | polpak | brehaut: you have a problem you're attempting to solve via this method. What is the problem. |
| 23:48 | Sunmaster_14 | I'm having the following problem: http://dpaste.com/37833/ I checked the db, and the value I entered is definitely in there. |
| 23:48 | brehaut | as much as anything its to see if i can. currently the project ive become involved in uses a bunch of if switches based on host name to let devs configure the settings (mostly DB etc) to their local machine configurations |
| 23:49 | brehaut | i want to bust those out into their own domain based files |
| 23:49 | Magus- | there's a far easier method for that |
| 23:49 | brehaut | then import the relevant ones |
| 23:49 | Magus- | insanely easier |
| 23:49 | Magus- | Sunmaster_14: you need to set it with a date object, not a string |
| 23:50 | Sunmaster_14 | oh, ok |
| 23:50 | Magus- | brehaut: http://dpaste.com/37834/ |
| 23:50 | Sunmaster_14 | but it didn't complain at me when I saved it. |
| 23:50 | Magus- | put that at the end of the file, then devs can have their own "local_settings.py" |
| 23:50 | Magus- | it imports and overrides and boom |
| 23:50 | Magus- | "the file" == settings.py, btw |
| 23:50 | brehaut | yes i was aware of that |
| 23:50 | brehaut | thats what i do myself |
| 23:50 | Magus- | just don't put local_settings.py in source control |