Make it possible to not include IPC Messages headers in other headers
Created attachment 381174 [details] Patch
Comment on attachment 381174 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=381174&action=review > Source/WebKit/Scripts/webkit/messages.py:287 > +def forward_declarations_and_headers_for_replies(receiver): I think we can de-duplicate a bunch of this code. Need to think about how I want to make it look.
I guess I have some CMake fixing to do too.
Comment on attachment 381174 [details] Patch Please also update Source/WebKit/Scripts/Makefile and run it to update the generator tests.
Created attachment 381510 [details] Patch
Created attachment 381522 [details] Patch
(In reply to Alex Christensen from comment #4) > Comment on attachment 381174 [details] > Patch > > Please also update Source/WebKit/Scripts/Makefile and run it to update the > generator tests. Man, that would have been nice to know about when I was writing this :(
Comment on attachment 381522 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=381522&action=review > Source/WebKit/DerivedSources.make:212 > + @echo Generating messages receiver for $*... huh, don’t know why I pluralized that > Source/WebKit/Scripts/webkit/LegacyMessages-expected.h:364 > + using DelayedReply = GetPluginProcessConnectionDelayedReply; static void send(std::unique_ptr<IPC::Encoder>&&, IPC::Connection&, const IPC::Connection::Handle& connectionHandle); This is a bug > Source/WebKit/UIProcess/WebPageProxy.cpp:259 > + String message; Shouldn’t do this, gotta give it its own file
Comment on attachment 381522 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=381522&action=review >> Source/WebKit/Scripts/webkit/LegacyMessages-expected.h:364 >> + using DelayedReply = GetPluginProcessConnectionDelayedReply; static void send(std::unique_ptr<IPC::Encoder>&&, IPC::Connection&, const IPC::Connection::Handle& connectionHandle); > > This is a bug Hooray tests!
Created attachment 381557 [details] Patch
Created attachment 381564 [details] Patch
Created attachment 381568 [details] Patch
Comment on attachment 381568 [details] Patch r=me
Comment on attachment 381568 [details] Patch Clearing flags on attachment: 381568 Committed r251445: <https://trac.webkit.org/changeset/251445>
All reviewed patches have been landed. Closing bug.
<rdar://problem/56511583>
Comment on attachment 381568 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=381568&action=review > Source/WebKit/Scripts/Makefile:2 > + python ./generate-message-receiver.py webkit/test-superclass-messages.in --implementation webkit/MessageReceiverSuperclass-expected.cpp --header webkit/MessagesSuperclass-expected.h --reply-header webkit/MessagesRepliesSuperclassReplies-expected.h ... I typoed the word "Receiver" as "Replies" in the last filename here
(In reply to Tim Horton from comment #17) > Comment on attachment 381568 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=381568&action=review > > > Source/WebKit/Scripts/Makefile:2 > > + python ./generate-message-receiver.py webkit/test-superclass-messages.in --implementation webkit/MessageReceiverSuperclass-expected.cpp --header webkit/MessagesSuperclass-expected.h --reply-header webkit/MessagesRepliesSuperclassReplies-expected.h > > ... I typoed the word "Receiver" as "Replies" in the last filename here It looks like this change may have caused 4 webkitpy test failures: https://build.webkit.org/builders/Apple%20Mojave%20Release%20WK2%20%28Tests%29/builds/7595
I'll take a peek!
(In reply to Tim Horton from comment #19) > I'll take a peek! Fixed in r251464