These two pieces of code seem to have a lot in common. It's probably a good idea to use Frame::createView to avoid needlessly diverging in behavior over time resulting in port specific bugs.
Yes, please! I never noticed when Frame::createView was added. Wow, 3 years ago: http://trac.webkit.org/changeset/40435
Created attachment 112345 [details] Patch
Created attachment 112348 [details] Patch
Please note: Fixed layout mode probably needs to be disabled for subframes or else you end up with scrollbars on iframe content that was meant to fit the the iframe. How the viewport meta tag and fixed layout should interact with frames and iframes is a complex question, and I'll leave that for another bug report.
Comment on attachment 112348 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=112348&action=review > Source/WebKit/chromium/src/WebFrameImpl.cpp:-2023 > - if (webView->isTransparent()) It looks like you are losing this step, which I think matters for chrome extensions.
Created attachment 113341 [details] Patch
Comment on attachment 112348 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=112348&action=review >> Source/WebKit/chromium/src/WebFrameImpl.cpp:-2023 >> - if (webView->isTransparent()) > > It looks like you are losing this step, which I think matters for chrome extensions. Oops. I totally missed those two lines. Thanks! I've uploaded a new patch to fix this.
*** Bug 70555 has been marked as a duplicate of this bug. ***
Comment on attachment 113341 [details] Patch Clearing flags on attachment: 113341 Committed r99208: <http://trac.webkit.org/changeset/99208>
All reviewed patches have been landed. Closing bug.
Created attachment 113537 [details] Patch
Comment on attachment 113537 [details] Patch Oops. Sorry. Copied a bug line, but forgot to edit it.