Yeah, the main issue is that we didn't have access to all the platforms (e.g., Qt, Gtk, Tiger) so we needed help from the bots. Adam On Wed, Jun 16, 2010 at 9:57 AM, Cameron Zwarich <cwzwarich@uwaterloo.ca> wrote:

Was anything preventing you from running these tests before you turned it on?

Cameron

On Jun 16, 2010, at 8:32 AM, Adam Barth <abarth@webkit.org> wrote:

...

Apologies for the red tree.  There are still a handful of LayoutTests we're trying to understand.  Hopefully we'll have everything cleared up by noon.

Thanks, Adam

On Wed, Jun 16, 2010 at 2:20 AM, Eric Seidel <eric@webkit.org> wrote:

...

See https://docs.google.com/document/edit?id=1as5xYjyMSCph4960iz0-Kb7hZKf_L6f2vt... for a description of expected differences.

On Wed, Jun 16, 2010 at 2:18 AM, Adam Barth <abarth@webkit.org> wrote:

...

As of early this morning, we've enabled the HTML5 tokenizer on trunk. I expect we've broken a lot of things.  Please file bugs and CC Eric, Tonyg, and myself.  Thanks for your patience.

You can get the old parser back by passing --legacy-parser to DumpRenderTree or typing the command "defaults write com.apple.Safari WebKitHTML5ParserEnabled -bool NO"

Adam

PS: The new code has a lot of ASSERTs.  If you're think parsing in debug builds is too slow, we can remove some of these. _______________________________________________ webkit-dev mailing list webkit-dev@lists.webkit.org http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

_______________________________________________ webkit-dev mailing list webkit-dev@lists.webkit.org http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev