2008-8-8
| 21:53 | Magus- | uh, no |
| 21:54 | eRAZOR | thought so because cache_control also allows to specify the timeout |
| 21:54 | doctormo | Hmm it doesn't look like it's possible to get the form or field from a widget during render... that kind of makes things hard to do context based widgets. |
| 21:54 | Magus- | eRAZOR: cache_control is for headers sent to the browser |
| 21:54 | eRAZOR | a I see |
| 21:54 | Magus- | like don't cache this for X time |
| 21:54 | eRAZOR | yep |
| 21:54 | Magus- | or whatever |
| 21:55 | Magus- | doctormo: correct, widget does not know of form or field |
| 21:55 | mattmcc | doctormo: You can make your own widget & field to pass whatever information you need into the widget. |
| 21:56 | doctormo | mattmcc: yes but the form context is needed for the data available to the form to be available to the widget. It doesn't work logically |
| 21:56 | eRAZOR | so it looks like I need to use both decorators to be on the safe side |
| 21:59 | d00gs | what's the best way to do an ordered many-to-many? order_with_respect_to on an intermediary model? |
| 21:59 | Magus- | to do an ordered m2m you must define the join table yourself and not use the m2m field, last I checked |
| 21:59 | d00gs | so doing it with a through won't work? |
| 22:00 | mattmcc | If it doesn't, that'd be an interesting feature to attempt, since ordered m2m's are a fairly common need. |
| 22:00 | d00gs | yeah, i figured it would be |
| 22:00 | d00gs | ok, i'll play around, just wondering if there was a "best practice" |
| 22:01 | Magus- | through? |
| 22:01 | Prometheus | looks like I'm in for quite a bit of reading :) |
| 22:01 | Magus- | did I miss that ticket merging again? |
| 22:01 | d00gs | it merged like a week or two ago |
| 22:01 | d00gs | i think |
| 22:01 | mattmcc | You blinked. :) |
| 22:06 | doctormo | Hmm, I fixed my problem by getting ajax to make further requests on page load, although it'd very wasteful, it seems the only way to get around the context restrictions. |
| 22:07 | mattmcc | That's unlikely. |
| 22:07 | mattmcc | I have custom forms that pass all sorts of extra data down to the widget. |
| 22:12 | _Gast_84_ | |
| 22:12 | _Gast_84_ | |
| 22:12 | _Gast_84_ | |
| 22:13 | hydrapheetz | *ker BONK* |
| 22:13 | thunderbolt | Yay! |
| 22:19 | quantumsummers | anyone know a nice way to iterate over form fields? |
| 22:20 | quantumsummers | in python that is |
| 22:20 | mattmcc | The form.fields collection? |
| 22:20 | Magus- | a normal for loop over it like any dict? |
| 22:20 | Magus- | (regarding form.fields) |
| 22:21 | quantumsummers | giving us unsubscriptable error on a commented out line in the loop |
| 22:21 | Magus- | perhaps if you tried showing us your code |
| 22:21 | mattmcc | You're doing it wrong. |
| 22:21 | Magus- | instead of wanting us to guess at it |
| 22:22 | quantumsummers | BoundField is unsubscriptable |
| 22:22 | mattmcc | It's true, it's not. |
| 22:22 | Magus- | well yah, it isn't |
| 22:22 | Magus- | so stop trying to do somefield[0] or such |
| 22:24 | quantumsummers | form_as_div ftw |
| 22:24 | Magus- | ? |
| 22:28 | Glenjamin | if thats all you want just make an as_div mixin for your forms |
| 22:28 | Eloff | I've read in several places now that cherrypy 3 has a faster pure python WSGI server than mod_python, could django be run on that? |
| 22:28 | mattmcc | Um, mod_wsgi? :) |
| 22:29 | Eloff | mattmcc, good point, I'd be surprised if cherrypy is faster than mod_wsgi, but I guess anything is possible until you measure |
| 22:29 | doctormo | mattmcc: then can you perlpaste your code so I can see them, I can't quite see how you can get variables out of scope to be submitted to widgets/fields |
| 22:30 | HenrikV | has tried to have a named url say name="home", where {% url home %} throws an exception ? |
| 22:31 | mattmcc | doctormo: Well, maybe if you mentioned what exactly you're trying to access.. Since you haven't done that yet. |
| 22:31 | HenrikV | form_as_dl > all others :) |
| 22:31 | doctormo | mattmcc: No I did, I said I wanted to access form data at widget rendering time. |
| 22:35 | eRAZOR | is there a convenient way to monitor the status of the django cache system? |
| 22:35 | mattmcc | doctormo: So from a form's __init__, you can grab the data argument and pass it through self.fields['foo'] |
| 22:35 | mattmcc | ...or not. |
| 22:36 | Magus- | eRAZOR: what status? |
| 22:36 | mattmcc | eRAZOR: That'd depend on the backend. I imagine there are lots of monitoring tools for memcached. |
| 22:36 | mattmcc | And since that's the only backend worth using in production, that should suffice. ;) |
| 22:36 | eRAZOR | magus: number of items in the cache .. perhaps with a summery per cache_key domain (string based) |
| 22:36 | eRAZOR | something like that |
| 22:37 | Magus- | eRAZOR: django keeps no such record like that |
| 22:37 | Magus- | so its up to your backend, as mattmcc said |
| 22:37 | eRAZOR | nothing fancy .. just to get a grasp on whats going on |
| 22:37 | eRAZOR | ok |
| 22:38 | eRAZOR | I'm pretty certain that there are monitoring solutions for memcached but since I'm currently developing with the django builtin cache backend, I thought I ask |
| 22:38 | coulix | Hi, i get a strange error with no debugging template on this small piece of "simple" code http://dpaste.com/70277/ any hints on what could it be ? |
| 22:39 | Magus- | you're returning a string |
| 22:39 | Magus- | obviously that won't work |
| 22:39 | Magus- | line 13 |
| 22:39 | mattmcc | coulix: Looks like you meant to use a redirect response. |
| 22:40 | coulix | please hit me .... time to sleep |
| 23:19 | Ceddy | hey all, slightly strange question, but i have a model, and want a method which lists all the methods attached to that model, any ideas how to access this? |
| 23:19 | mattmcc | dir() |
| 23:20 | mattmcc | But wouldn't reading the documentation be easier? |
| 23:20 | Glenjamin | will django-admin work if its in a different directory? |
| 23:20 | Glenjamin | (ie. can i copy it to python\scripts) |
| 23:21 | Ceddy | mattmcc: is dir() directed at me? |
| 23:21 | horza | Hmm, if you try and extend a template using the name it doesn't exclude itself and goes into infinite loop |
| 23:21 | mattmcc | horza: Right, only one template can exist with a given name. |
| 23:22 | horza | glenjamin: Why don't you put a symbolic link in python/scripts? |
| 23:22 | mattmcc | Ceddy: Yep. |
| 23:22 | horza | mattmcc: So how do I over-ride an admin template, when all I want to do is add a couple of lines? |
| 23:23 | mattmcc | horza: That depends. Some admin views look for a series of templates, such as the model change forms. |
| 23:23 | Glenjamin | hrm, symlink will work |
| 23:23 | mattmcc | For others, you'd just have to copy the template somewhere and override the admin one entirely. |
| 23:24 | Ceddy | thanks, essentially we are expanding the no. of fields all the time, and i dont want to keep sotring all the info twice, so just want a list of all the fields as i need it else where in the code.... |
| 23:24 | mattmcc blinks. | |
| 23:24 | horza | mattmcc: over-riding entirely makes maintenance a bit of a nightmare. I guess I could create a whole load of change_form_xxx and put includes in each |
| 23:26 | stp2007 | I have an app being accessed directly and through Facebook. I'm trying to figure out how to allow the direct urls to be app\view\ and FB urls to be FB\app\view\. Tried a regular expression in root.py during the include of the app.urls but that didn't work although my regex foo sucks so that could be it. Thoughts? |
| 23:29 | horza | mattmcc: No, admin/objectname/change_form.html still goes into infinite loop |
| 23:29 | horza | mattmcc: I'll hack it into base.html :-( |
| 23:30 | Killarny | I have three models that are sharing one abstract model, and one model which relates to all three - how can I give that model a foreignkey to three other models, dependant upon which of the three is in use? |
| 23:30 | Killarny | I've tried linking to the abstract model, but that doesn't work .... am I being dense? |
| 23:30 | trhaynes | Is there an easy way to deploy two instances of my project such that they share the same code but use different databases? |
| 23:31 | mattmcc | trhaynes: Sure, it's just a matter of settings. |
| 23:31 | stp2007 | Killarny, try http://www.djangoproject.com/documentation/cont... |