Return the correct value for NPNVsupportsCarbonBool
Created attachment 105226 [details] Patch
Can we make a regression test for this? What’s the real world impact?
(In reply to comment #2) > Can we make a regression test for this? We could, but it would involve making run-webkit-tests aware of whether Carbon plug-ins are available and look for different results depending on that. I don't think it's worth it. > > What’s the real world impact? Probably none - Carbon plug-ins just assume that the Carbon event model is supported. I just saw that it was wrong and it bothered me.
Comment on attachment 105226 [details] Patch This really should have a test associated with it.
Created attachment 105236 [details] Patch
Comment on attachment 105236 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=105236&action=review > Tools/DumpRenderTree/TestNetscapePlugIn/Tests/mac/SupportsCarbonEventModel.cpp:56 > + bool testConvert(double x, double y, NPCoordinateSpace sourceSpace, NPCoordinateSpace destSpace) > + { > + // First convert from src to dest. > + double destX, destY; > + if (!NPN_ConvertPoint(x, y, sourceSpace, &destX, &destY, destSpace)) > + return false; > + > + // Then convert back to src > + double srcX, srcY; > + if (!NPN_ConvertPoint(destX, destY, destSpace, &srcX, &srcY, sourceSpace)) > + return false; > + > + > + // Then compare. > + if (srcX != x || srcY != y) > + return false; > + > + return true; > + } This function never seems to be called. > Tools/DumpRenderTree/TestNetscapePlugIn/Tests/mac/SupportsCarbonEventModel.cpp:58 > + bool runTest() This can be private.
Committed r93814: <http://trac.webkit.org/changeset/93814>