Update RTCRtpScriptTransform to the latest version of the spec
Created attachment 422716 [details] Patch
Created attachment 422821 [details] Patch
Comment on attachment 422821 [details] Patch Wincairo might be a unified build issue
Comment on attachment 422821 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=422821&action=review > Source/WebCore/ChangeLog:10 > + Add support for options parameter provided in RTCRtpScriptTrtansform constructor. s/RTCRtpScriptTrtansform/RTCRtpScriptTransform/ > Source/WebCore/Modules/mediastream/RTCRtpScriptTransformer.cpp:55 > + auto transformer = adoptRef(*new RTCRtpScriptTransformer(context, WTFMove(options), WTFMove(port), readable.releaseReturnValue(), WTFMove(readableSource))); Is there any reason to not have the constructor get the SimpleReadableStreamSource from the ReadableStream? > Source/WebCore/Modules/mediastream/RTCRtpScriptTransformer.cpp:104 > + return writable; I thought this was wrong at first because of the variable name, something like `writableOrException` might make it clearer.
(In reply to Eric Carlson from comment #4) > Comment on attachment 422821 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=422821&action=review > > > Source/WebCore/ChangeLog:10 > > + Add support for options parameter provided in RTCRtpScriptTrtansform constructor. > > s/RTCRtpScriptTrtansform/RTCRtpScriptTransform/ OK > > Source/WebCore/Modules/mediastream/RTCRtpScriptTransformer.cpp:55 > > + auto transformer = adoptRef(*new RTCRtpScriptTransformer(context, WTFMove(options), WTFMove(port), readable.releaseReturnValue(), WTFMove(readableSource))); > > Is there any reason to not have the constructor get the > SimpleReadableStreamSource from the ReadableStream? ReadableStream can take any ReadableStreamSource but we want to keep the fact it is a SimpleReadableStreamSource. > > Source/WebCore/Modules/mediastream/RTCRtpScriptTransformer.cpp:104 > > + return writable; > > I thought this was wrong at first because of the variable name, something > like `writableOrException` might make it clearer. OK
Created attachment 423043 [details] Patch for landing
Created attachment 423052 [details] Win fix try
Don, Hironori-san, can you help me with wincairo failure? Might be a unified build issue.
Thank you for letting me know that. It looks a simple problem. Will check.
Created attachment 423080 [details] Patch
It turned out this isn't a simple problem. I'll look into into it again next week. I added some header inclusions to WebGLRenderingContextBase.h, and fixed recursive inclusion of WebGLTransformFeedback.h. This patch can compile for WinCairo. Hmm, but, it can't compile for other ports.
commit-queue failed to commit attachment 423052 [details] to WebKit repository. To retry, please set cq+ flag again.
Committed r274385: <https://commits.webkit.org/r274385> All reviewed patches have been landed. Closing bug and clearing flags on attachment 423052 [details].
<rdar://problem/75392479>