If any cache keys you generate include user (staff or public) supplied data, they may: be too long, contain invalid characters (at least by memcache's standards), or both.
This helper will sub out any invalid characters and md5 the key if it's too long.
Additionally, it'll insert the CACHE_MIDDLEWARE_KEY_PREFIX from django.conf.settings for you. In your memcache instances are used by multiple applications (Django or otherwise) this can help ensure your keys are unique to that a particular app on a particular site.
A subclassed version of the standard Django Paginator (django.core.paginator.Paginator) that automatically caches pages as they are requested. Very useful if your object list is expensive to compute.
MIT licensed.
Request-phase cache middleware that checks to make sure the Cache server is running. Starting it if it is not. This is run for every request, it checks to see if it can get a defined item out of the cache, if that fails it tries to set it. Failing that it decides the server is probably crashed, so goes though and attempts to connect to the server. Failing a connection it will launch a new server.
This is probably not useful on large scale multi server deployments as they likely have their own testing for when services crash, but I am using it in a shared hosting environment where I have to run my own copy of memcache manually and cannot setup proper services testing, so I use this to just make sure the cache server is still running.
This decorator does automatic key generation and simplifies caching. Can be used with any class, not just model subclasses.
Also see [Stales Cache Decorator](http://www.djangosnippets.org/snippets/1131/).
Provides an efficient means of looking up multiple related model instances for a range of objects by pre-filling the cache attribute used by `SingleRelatedObjectDescriptor` with either complete model instances or a dict containing only specified fields, looking up all required data with a single query.
Example usage:
C:\django_projects\soclone>django-admin.py shell
>>> from soclone.models import Question
>>> from soclone.views import populate_fk_caches
>>> from django.db import connection
>>> from django.contrib.auth.models import User
>>> q = Question.objects.get(id=1)
>>> a = list(q.answers.all())
>>> connection.queries = []
>>> populate_fk_caches(User, (
... ((q,), ('author', 'last_edited_by', 'closed_by')),
... (a, ('author', 'last_edited_by')),
... ),
... fields=('username', 'gravatar', 'reputation', 'gold', 'silver',
... 'bronze'))
>>> connection.queries
[{'time': '0.000', 'sql': u'SELECT "auth_user"."id", "auth_user"."username", "au
th_user"."gravatar", "auth_user"."reputation", "auth_user"."gold", "auth_user"."
silver", "auth_user"."bronze" FROM "auth_user" WHERE "auth_user"."id" IN (1, 2)'
}]
The @cache_holding decorator works in a per-function base, you can specify a list of models or just a model and the second argument is the time in seconds to be cached. You can modify the proxy class by a keyword argument so you can do cache to all kind of things, also if you want to use a prefix as key + the hash value you can specify the prefix keyword argument.
This is an (overly simple) example of how to manually delete something from the cache.
Usage is simply `delete_url_cache('/some_view/')`
This most likely doesn't work when using Vary headers but does seem to work fine for the most general case of simply needing to programmatically flush the cache after certain actions.
A simple helper function that clears the cache for a given URL, assuming no headers. Probably best used when wired up to a model's post_save event via signals. See [message to django-users mailing list](http://groups.google.com/group/django-users/msg/b077ec2e97697601) for background.
This code improves on Django's render_to_response shortcut function in the following ways:
1. If the caller does not provide a MIME type, and the caller is passing a RequestContext, it interrogates the request to determine if the HTTP client supports application/xhtml+xml encoding. If so, it sets the request type to application/xhtml+xml to override the HttpRequest class's default mime type of text/html.
2. It caches parsed templates in its own cache to improve performance.
If you aren't using XHTML in your templates, you may choose to comment out the code that tests for and sets the application/xhtml+xml MIME type.
I place this code in a file named "util.py" in my application. In my views, I write "from app.util import render_to_response" where I used to write "from django.shortcuts import render_to_response".
Note that the caching functionality provided by this code means that you will need to recycle your Django instance when you make template changes. Instructions for doing this depend on how you have deployed your Django application.
This is intended as an alternative to http://www.djangosnippets.org/snippets/155/
Put this in your own cache.py and import it instead of django.core.cache and use it the same way. We left out the "add" function but it shouldn't be too hard to make if you want it.
From the above post: "The purpose of this caching scheme is to avoid the dog-pile effect. Dog-piling is what normally happens when your data for the cache takes more time to generate than your server is answering requests per second. In other words if your data takes 5 seconds to generate and you are serving 10 requests per second, then when the data expires the normal cache schemes will spawn 50 attempts a regenerating the data before the first request completes. The increased load from the 49 redundant processes may further increase the time it takes to generate the data. If this happens then you are well on your way into a death spiral
MintCache works to prevent this scenario by using memcached to to keep track of not just an expiration date, but also a stale date The first client to request data past the stale date is asked to refresh the data, while subsequent requests are given the stale but not-yet-expired data as if it were fresh, with the undertanding that it will get refreshed in a 'reasonable' amount of time by that initial request."