Currently there is a common implementation of ENABLE(RESOURCE_LOAD_STATISTICS) within NetworkStorageSessionCFNet.cpp that should go into NetworkStorageSession.
Created attachment 357054 [details] Patch
Created attachment 357060 [details] Patch
Yay!
Comment on attachment 357060 [details] Patch r=me You should be aware that the high-level organization of resource load statistics is not where we want it to be. I think right now it uses some strange combination of process pool and globals, and it needs some love and polish.
Comment on attachment 357060 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=357060&action=review > Source/WebCore/platform/network/NetworkStorageSession.cpp:106 > +bool NetworkStorageSession::shouldBlockCookies(const ResourceRequest& request, std::optional<uint64_t> frameID, std::optional<uint64_t> pageID) const Preexisting problem, but I presume passing std::optional by value is not best-practice? Should be const ref?
(In reply to Michael Catanzaro from comment #5) > Comment on attachment 357060 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=357060&action=review > > > Source/WebCore/platform/network/NetworkStorageSession.cpp:106 > > +bool NetworkStorageSession::shouldBlockCookies(const ResourceRequest& request, std::optional<uint64_t> frameID, std::optional<uint64_t> pageID) const > > Preexisting problem, but I presume passing std::optional by value is not > best-practice? Should be const ref? I think we're OK with it for small types.
Comment on attachment 357060 [details] Patch Clearing flags on attachment: 357060 Committed r239081: <https://trac.webkit.org/changeset/239081>
All reviewed patches have been landed. Closing bug.
<rdar://problem/46637456>