Add Webm sanitization.
Created attachment 350106 [details] Patch
Created attachment 350108 [details] Patch
Comment on attachment 350106 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=350106&action=review > Source/WebCore/ChangeLog:10 > + see https://www.w3.org/TR/encrypted-media/#dom-mediakeysession-generaterequest Capital at the beginning, period at the end. > Source/WebCore/Modules/encryptedmedia/InitDataRegistry.cpp:160 > + if (buffer.isEmpty() || buffer.size() > kWebmMaxContentEncKeyIDSize) > + return std::nullopt; > + keyIDs.append(buffer.copy()); Instead of this, I'd try to sanitize and if the result is null, bailout. Otherwise, move the WTFMove the result inside the append
Created attachment 350184 [details] Patch
Comment on attachment 350184 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=350184&action=review > Source/WebCore/Modules/encryptedmedia/InitDataRegistry.cpp:157 > + RefPtr<SharedBuffer> sanitazedBuffer = sanitizeWebM(buffer); You probably mean "sanitized"
Comment on attachment 350184 [details] Patch Clearing flags on attachment: 350184 Committed r236256: <https://trac.webkit.org/changeset/236256>
All reviewed patches have been landed. Closing bug.
<rdar://problem/44636894>
(In reply to Olivier Blin from comment #5) > > Source/WebCore/Modules/encryptedmedia/InitDataRegistry.cpp:157 > > + RefPtr<SharedBuffer> sanitazedBuffer = sanitizeWebM(buffer); > > You probably mean "sanitized" Very true! Good catch. Sad that the patch landed already. Can you file a follow up please?
Reopening to attach new patch.
Created attachment 350208 [details] Patch
Created attachment 350213 [details] Patch
I pushed the attached patch 350208 by mistake, instead of pushing it on the bug 189789, I pushed it on this one. I pushed again the original patch of this bug for reference.