see patch
Created attachment 81420 [details] Patch
Attachment 81420 [details] did not build on mac: Build output: http://queues.webkit.org/results/7700652
Created attachment 81425 [details] Patch
Created attachment 81430 [details] Patch
Comment on attachment 81430 [details] Patch None of these #includes or forward-declarations is required on Windows because we pull them in through our prefix header. My guess is that adding them here will just slow the build down with no positive effect. In what situation are these #includes and forward-declarations useful?
Comment on attachment 81430 [details] Patch I suspect patrick is trying to build without a prefix header. These includes make sense to me, but I don't work on the windows port anymore so I defer to those who do.
Comment on attachment 81430 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=81430&action=review > Source/WebCore/platform/network/cf/DNSCFNet.cpp:39 > #if PLATFORM(WIN) > #include "LoaderRunLoopCF.h" > +#include <CFNetwork/CFNetwork.h> Why only do this for PLATFORM(WIN)? > Source/WebCore/platform/network/cf/ProxyServerCFNet.cpp:34 > +#if PLATFORM(WIN) > +#include <CFNetwork/CFNetwork.h> > +#endif Why only do this for PLATFORM(WIN)? > Source/WebCore/platform/network/cf/SocketStreamHandleCFNet.cpp:50 > #if PLATFORM(WIN) > #include "LoaderRunLoopCF.h" > +#include <CFNetwork/CFNetwork.h> Why only do this for PLATFORM(WIN)?
Comment on attachment 81430 [details] Patch (In reply to comment #7) > (From update of attachment 81430 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=81430&action=review > > > Source/WebCore/platform/network/cf/DNSCFNet.cpp:39 > > #if PLATFORM(WIN) > > #include "LoaderRunLoopCF.h" > > +#include <CFNetwork/CFNetwork.h> > > Why only do this for PLATFORM(WIN)? It's a private header on Mac.
Comment on attachment 81430 [details] Patch Clearing flags on attachment: 81430 Committed r77937: <http://trac.webkit.org/changeset/77937>
All reviewed patches have been landed. Closing bug.