Fixed #612 - added cache control headers (thanks, hugo)

git-svn-id: http://code.djangoproject.com/svn/django/trunk@1020 bcc190cf-cafb-0310-a4f2-bffc1f526a37

django/utils/cache.py

@@ -21,6 +21,45 @@ import datetime, md5, re
 from django.conf import settings
 from django.core.cache import cache
 
+cc_delim_re = re.compile(r'\s*,\s*')
+def patch_cache_control(response, **kwargs):
+    """
+    This function patches the Cache-Control header by adding all
+    keyword arguments to it. The transformation is as follows:
+
+    - all keyword parameter names are turned to lowercase and
+      all _ will be translated to -
+    - if the value of a parameter is True (exatly True, not just a
+      true value), only the parameter name is added to the header
+    - all other parameters are added with their value, after applying
+      str to it.
+    """
+
+    def dictitem(s):
+        t = s.split('=',1)
+        if len(t) > 1:
+            return (t[0].lower().replace('-', '_'), t[1])
+        else:
+            return (t[0].lower().replace('-', '_'), True)
+
+    def dictvalue(t):
+        if t[1] == True:
+            return t[0]
+        else:
+            return t[0] + '=' + str(t[1])
+
+    if response.has_header('Cache-Control'):
+        print response['Cache-Control']
+        cc = cc_delim_re.split(response['Cache-Control'])
+        print cc
+        cc = dict([dictitem(el) for el in cc])
+    else:
+        cc = {}
+    for (k,v) in kwargs.items():
+        cc[k.replace('_', '-')] = v
+    cc = ', '.join([dictvalue(el) for el in cc.items()])
+    response['Cache-Control'] = cc
+
 vary_delim_re = re.compile(r',\s*')
 
 def patch_response_headers(response, cache_timeout=None):
@@ -43,8 +82,7 @@ def patch_response_headers(response, cache_timeout=None):
         response['Last-Modified'] = now.strftime('%a, %d %b %Y %H:%M:%S GMT')
     if not response.has_header('Expires'):
         response['Expires'] = expires.strftime('%a, %d %b %Y %H:%M:%S GMT')
-    if not response.has_header('Cache-Control'):
+    patch_cache_control(response, max_age=cache_timeout)
-        response['Cache-Control'] = 'max-age=%d' % cache_timeout
 
 def patch_vary_headers(response, newheaders):
     """

django/views/decorators/cache.py

@@ -10,8 +10,24 @@ example, as that is unique across a Django project.
 Additionally, all headers from the response's Vary header will be taken into
 account on caching -- just like the middleware does.
 """
+import re
 
 from django.utils.decorators import decorator_from_middleware
+from django.utils.cache import patch_cache_control
 from django.middleware.cache import CacheMiddleware
 
 cache_page = decorator_from_middleware(CacheMiddleware)
+
+def cache_control(**kwargs):
+
+    def _cache_controller(viewfunc):
+
+        def _cache_controlled(request, *args, **kw):
+            response = viewfunc(request, *args, **kw)
+            patch_cache_control(response, **kwargs)
+            return response
+
+        return _cache_controlled
+
+    return _cache_controller
+

docs/cache.txt

@@ -270,6 +270,40 @@ and a list/tuple of header names as its second argument.
 
 .. _`HTTP Vary headers`: http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.44
 
+Controlling cache: Using Vary headers
+=====================================
+
+Another problem with caching is the privacy of data, and the question where data can
+be stored in a cascade of caches. A user usually faces two kinds of caches: his own
+browser cache (a private cache) and his providers cache (a public cache). A public cache
+is used by multiple users and controlled by someone else. This poses problems with private
+(in the sense of sensitive) data - you don't want your social security number or your
+banking account numbers stored in some public cache. So web applications need a way
+to tell the caches what data is private and what is public.
+
+Other aspects are the definition how long a page should be cached at max, or wether the
+cache should allways check for newer versions and only deliver the cache content when
+there were no changes (some caches might deliver cached content even if the server page
+changed - just because the cache copy isn't yet expired).
+
+So there are a multitude of options you can control for your pages. This is where the
+Cache-Control header (more infos in `HTTP Cache-Control headers`_) comes in. The usage
+is quite simple::
+
+    @cache_control(private=True, must_revalidate=True, max_age=3600)
+    def my_view(request):
+        ...
+
+This would define the view as private, to be revalidated on every access and cache
+copies will only be stored for 3600 seconds at max.
+
+The caching middleware already set's this header up with a max-age of the CACHE_MIDDLEWARE_SETTINGS
+setting. And the cache_page decorator does the same. The cache_control decorator correctly merges
+different values into one big header, though. But you should take into account that middlewares
+might overwrite some of your headers or set their own defaults if you don't give that header yourself.
+
+.. _`HTTP Cache-Control headers`: http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.9
+
 Other optimizations
 ===================