Revert of Switch the page-capturing machinery to use the new hooks.

Revert of Switch the page-capturing machinery to use the new hooks. (patchset #16 id:300001 of https://codereview.chromium.org/1398823004/ )

Reason for revert:
Caused https://code.google.com/p/chromium/issues/detail?id=583261.

Original issue's description:
> Switch the page-capturing machinery to use the new hooks.
>
> This is a timing-changing patch. Here are the details:
>
> * The page text capture machinery no longer uses archaic timers to determine the best time to grab the text from the the document. Instead, it relies on the MeaningfulLayout signals that are coming from Blink.
>
> * These signals are keyed off the compositor putting up frames, which has an interesting side effect -- at least on some platforms (CrOS and Linux), this signal will not be issued until document/frame/tab is active, which is perfectly fine for our purposes.
>
> * As a side comment, we should probably unify our position on whether compositor puts up frames for inactive tabs.
>
> * As another side comment, this would help simplify the logic in translate bubbles that watches for tab to become active.
>
> * In many cases, the timing is much shorter than before, especially for smaller pages. This is good (faster!), but also means that some tests had to be tweaked to start listening for language detection notifications a bit earlier.
>
> * Fundamentally, this is a change in timing, not behavior, so the existing test coverage seemed sufficient.
>
> * The new plumbing is still incompatible with out-of-process iframes (OOPIF) and will need to be fixed up accordingly.
>
> BUG=521166
> R=jam,andrewhayden,jochen
>
> Committed: https://crrev.com/afc918de00a2f2895b4d8eaf3a043e654ba5e8fb
> Cr-Commit-Position: refs/heads/master@{#372891}

TBR=jochen@chromium.org,andrewhayden@chromium.org,jam@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=521166
Committed: https://crrev.com/7d0ddfc32e7425fb179415319476c9447d785059
Cr-Commit-Position: refs/heads/master@{#373175}

[dglazkov, 2016-02-03 04:46:38]

Created Revert of Switch the page-capturing machinery to use the new hooks.

[commit-bot: I haz the power, 2016-02-03 04:47:04]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1659393003/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1659393003/1

[commit-bot: I haz the power, 2016-02-03 04:48:08]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-02-03 04:49:05]

Description was changed from

==========
Revert of Switch the page-capturing machinery to use the new hooks. (patchset
#16 id:300001 of https://codereview.chromium.org/1398823004/ )

Reason for revert:
Caused https://code.google.com/p/chromium/issues/detail?id=583261.

Original issue's description:
> Switch the page-capturing machinery to use the new hooks.
> 
> This is a timing-changing patch. Here are the details:
> 
> * The page text capture machinery no longer uses archaic timers to determine
the best time to grab the text from the the document. Instead, it relies on the
MeaningfulLayout signals that are coming from Blink.
> 
> * These signals are keyed off the compositor putting up frames, which has an
interesting side effect -- at least on some platforms (CrOS and Linux), this
signal will not be issued until document/frame/tab is active, which is perfectly
fine for our purposes.
> 
> * As a side comment, we should probably unify our position on whether
compositor puts up frames for inactive tabs.
> 
> * As another side comment, this would help simplify the logic in translate
bubbles that watches for tab to become active.
> 
> * In many cases, the timing is much shorter than before, especially for
smaller pages. This is good (faster!), but also means that some tests had to be
tweaked to start listening for language detection notifications a bit earlier.
> 
> * Fundamentally, this is a change in timing, not behavior, so the existing
test coverage seemed sufficient.
> 
> * The new plumbing is still incompatible with out-of-process iframes (OOPIF)
and will need to be fixed up accordingly.
> 
> BUG=521166
> R=jam,andrewhayden,jochen
> 
> Committed: https://crrev.com/afc918de00a2f2895b4d8eaf3a043e654ba5e8fb
> Cr-Commit-Position: refs/heads/master@{#372891}

TBR=jochen@chromium.org,andrewhayden@chromium.org,jam@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=521166
==========

to

==========
Revert of Switch the page-capturing machinery to use the new hooks. (patchset
#16 id:300001 of https://codereview.chromium.org/1398823004/ )

Reason for revert:
Caused https://code.google.com/p/chromium/issues/detail?id=583261.

Original issue's description:
> Switch the page-capturing machinery to use the new hooks.
>
> This is a timing-changing patch. Here are the details:
>
> * The page text capture machinery no longer uses archaic timers to determine
the best time to grab the text from the the document. Instead, it relies on the
MeaningfulLayout signals that are coming from Blink.
>
> * These signals are keyed off the compositor putting up frames, which has an
interesting side effect -- at least on some platforms (CrOS and Linux), this
signal will not be issued until document/frame/tab is active, which is perfectly
fine for our purposes.
>
> * As a side comment, we should probably unify our position on whether
compositor puts up frames for inactive tabs.
>
> * As another side comment, this would help simplify the logic in translate
bubbles that watches for tab to become active.
>
> * In many cases, the timing is much shorter than before, especially for
smaller pages. This is good (faster!), but also means that some tests had to be
tweaked to start listening for language detection notifications a bit earlier.
>
> * Fundamentally, this is a change in timing, not behavior, so the existing
test coverage seemed sufficient.
>
> * The new plumbing is still incompatible with out-of-process iframes (OOPIF)
and will need to be fixed up accordingly.
>
> BUG=521166
> R=jam,andrewhayden,jochen
>
> Committed: https://crrev.com/afc918de00a2f2895b4d8eaf3a043e654ba5e8fb
> Cr-Commit-Position: refs/heads/master@{#372891}

TBR=jochen@chromium.org,andrewhayden@chromium.org,jam@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=521166

Committed: https://crrev.com/7d0ddfc32e7425fb179415319476c9447d785059
Cr-Commit-Position: refs/heads/master@{#373175}
==========

[commit-bot: I haz the power, 2016-02-03 04:49:06]

Patchset 1 (id:??) landed as
https://crrev.com/7d0ddfc32e7425fb179415319476c9447d785059
Cr-Commit-Position: refs/heads/master@{#373175}