Currently, the USE(PROTECTION_SPACE_AUTH_CALLBACK) code is only implemented in ResourceHandleMac.mm. In order to use the CFNetwork based loader on Mac, we need to implement that call back in ResourceHandleCFNet.cpp as well.
<rdar://problem/9842424>
Created attachment 102025 [details] Patch
Comment on attachment 102025 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=102025&action=review r=me, but it would be good to have a bug filed for the FIXME > Source/WebCore/platform/network/cf/ResourceHandleCFNet.cpp:921 > + // FIXME: We should ask FrameLoaderClient. Is there a bug filed for this?
(In reply to comment #3) > (From update of attachment 102025 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=102025&action=review > > r=me, Thanks! > but it would be good to have a bug filed for the FIXME > > > Source/WebCore/platform/network/cf/ResourceHandleCFNet.cpp:921 > > + // FIXME: We should ask FrameLoaderClient. > > Is there a bug filed for this? It's the same in ResourceHandleMac.mm. We have lots of FIXMEs in the synchronous loader case (in both ResourceHandleCFNet.cpp and ResourceHandleMac.mm). Let me look around to see if we have a bug on that.
(In reply to comment #4) > (In reply to comment #3) > > Is there a bug filed for this? > > It's the same in ResourceHandleMac.mm. We have lots of FIXMEs in the synchronous loader case (in both ResourceHandleCFNet.cpp and ResourceHandleMac.mm). Let me look around to see if we have a bug on that. So there's bug 8342 about didReceiveAuthenticationChallenge but I couldn't find any for the protection space callback. Field <https://bugs.webkit.org/show_bug.cgi?id=65196>. I'll put a reference to it in the comments.
Comment on attachment 102025 [details] Patch Attachment 102025 [details] did not pass mac-ews (mac): Output: http://queues.webkit.org/results/9253371
(In reply to comment #6) > (From update of attachment 102025 [details]) > Attachment 102025 [details] did not pass mac-ews (mac): > Output: http://queues.webkit.org/results/9253371 Pretty sure this is because UNUSED_PARAM() should be inside #if LOG_DISABLED, not #if !LOG_DISABLED. If only the EWS bot let me see the full build result.... Will confirm with a release build (since debug doesn't have any problems compiling ResourceHandleMac.mm) and commit.
Committed r91799: <http://trac.webkit.org/changeset/91799>