Remove special handling of ::before and ::after on RUBY elements.

Remove special handling of ::before and ::after on RUBY elements.

Special handling of ::before and ::after in RUBY was introduced in
https://bugs.webkit.org/show_bug.cgi?id=41040 , followed by some crash fixes,
most notably https://bugs.webkit.org/show_bug.cgi?id=55930 (which is what
caused this issue - crbug.com/484438 )

No spec requires such special handling, and neither IE nor Firefox does it, and
it just complicates the code.

Just treat them as they were a first or last DOM child of the RUBY parent.
That's how ::before and ::after normally works. This means that now ::before
and ::after will be part of, or even establish, ruby runs.

Added a test for the crasher.

Also rewrote three tests as reftests, since the expectations started to fail
anyway. One test was invalid (ruby-beforeafter.html), since it required special
handling of ::before and ::after, while the two others just got different
rendering due to this change.

BUG=484438
R=leviw@chromium.org,ojan@chromium.org

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=195506

[mstensho (USE GERRIT), 2015-05-14 21:49:17]

One might say that this is one hell of a way to kill a tick, and that I could
have filed a much smaller CL that just had the child insertion code cope with
anonymous inline-blocks. However, since it seems wrong to treat ::before and
::after in a special way in RUBY anyway, why not go the other way and remove
code instead? :)

Cc'ed the original author, although he doesn't seem to be around anymore.

Special handling of ::before and ::after was introduced in
https://bugs.webkit.org/show_bug.cgi?id=41040 , followed by some crash fixes,
most notably https://bugs.webkit.org/show_bug.cgi?id=55930 (which I don't have
access to, but it's referred to in a commit comment).

[ojan, 2015-05-18 10:05:20]

ojan@chromium.org changed reviewers:
+ ojan@chromium.org

[ojan, 2015-05-18 10:05:21]

<3 changes like this. Any chance you could find the patch that added this code
and include it in the change description. It'd help to understand what we're
breaking by removing this. I expect we'll approve regardless...just want to be
confident about the change.

[mstensho (USE GERRIT), 2015-05-18 19:45:33]

On 2015/05/18 10:05:21, ojan wrote:
> <3 changes like this. Any chance you could find the patch that added this code
> and include it in the change description. It'd help to understand what we're
> breaking by removing this. I expect we'll approve regardless...just want to be
> confident about the change.

Done.

[leviw_travelin_and_unemployed, 2015-05-18 22:24:22]

Apologies for the late turnaround -- was at/traveling for BlinkON.

When I asked if you'd look into this, I didn't expect such awesome to come back!
LGTM! Thanks!

[ojan, 2015-05-19 01:07:12]

The CQ bit was checked by ojan@chromium.org

[ojan, 2015-05-19 01:07:13]

lgtm

\o/

[commit-bot: I haz the power, 2015-05-19 01:07:39]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1124313004/1

[commit-bot: I haz the power, 2015-05-19 02:51:58]

Committed patchset #1 (id:1) as
https://src.chromium.org/viewvc/blink?view=rev&revision=195506