Try this userstyle: http://userstyles.org/styles/49276
It does nothing on Nightlies.
If you remove structural pseudo-class ":only-child" from the code - the style begins to work, so the style itself is valid.
Thhis is no bug. DOM for image document has changed.
You should modify the style.
See Bug 528046 - Print / print-preview crops very tall images at page boundaries
I'm going to reopen this. The change in behavior here was unnecessary, and I think we should revert it.
And in particular, that behavior change violates http://www.whatwg.org/specs/web-apps/current-work/multipage/history.html#read-image
Requesting tracking for this regression.
I actually had no idea that this was specified.
I think we should back out bug 528046 and fix it another way.
I'm wrong, we don't need to back out the whole thing. See discussion in bug 528046.
Created attachment 552025 [details] [diff] [review]
Very sorry for the big delay on this, I had tree problems and ended up re-cloning and rebuilding the world which is a lot of trouble in a VM...
Yeah, this just moves the style node back into the head.
Would you mind adding some tests to check that this doesn't happen again? Load an image in an IFRAME and check that the <body> only has one child, the <img>, at least. Ditto for video.
Sorry for the delay, I'm still having more tree troubles :/
Roc, Michael, can I submit this test to the HTML test suite? (W3C test suite license <http://www.w3.org/Consortium/Legal/2008/04-testsuite-license.html> and 3-clause BSD <http://www.w3.org/Consortium/Legal/2008/03-bsd-license.html>.)
We need some sort of blanket policy that permits all our tests to be modified and submitted to the W3C by anyone. Would you like to explain on mozilla.governance what is required?
why should release drivers be tracking this?
This is fixed; doesn't need tracking.