I removed the willSendRequest method and set the cache model and disk
capacity ( without setting the memory capacity ) with the same
results. Is this enough information to file a bug report with or is
there some more work I could do to track this down?
Thanks again,
-Adam
On Apr 11, 2009, at 11:51 PM, Darin Adler wrote:
On Apr 11, 2009, at 6:54 PM, Adam Thorsen wrote:
I also noticed recently that there is a "Disable Caches" option
under the Develop menu on Safari. I tested it using the same
procedure and it indeed did not cache the javascript, so I know
this is possible one way or another.
It’s true, Disable Caches does this.
In fact, I read over the implementation of that menu item before
answering your original question. All that Safari does when that
menu item is selected is:
[[WebPreferences standardPreferences]
setCacheModel:WebCacheModelDocumentViewer];
[[NSURLCache sharedURLCache] setDiskCapacity:0];
I’m not sure why that works better in Safari than in your test
application.
NSURL * url = [aRequest URL];
NSURLRequest * cachelessRequest = [NSURLRequest
requestWithURL:url cachePolicy:NSURLRequestReloadIgnoringCacheData
timeoutInterval:100];
return cachelessRequest;
This code in your willSendRequest delegate method may be causing a
problem. It discards the request and creates a new one, taking only
the URL from the original request. That’s a problem because there
are other important fields in the request, such as the HTTP method,
HTTP body, and HTTP header fields. If you want to change the cache
policy (and you should not need to do this, see my original
message), you would need code more like this:
if ([request cachePolicy] == NSURLRequestReloadIgnoringCacheData)
return request;
NSURLRequest *cachelessRequest = [aRequest mutableCopy];
[cachelessRequest
setCachePolicy:NSURLRequestReloadIgnoringCacheData];
return cachelessRequest;
Hope that helps.
-- Darin
_______________________________________________
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev
