Implement DOM methods: prepend, append, after, before and replaceWith.

Implement DOM methods: prepend, append, after, before and replaceWith.

As per https://dom.spec.whatwg.org/#childnode ChildNode interface
should contain after(), before() and replaceWith() API.

As per https://dom.spec.whatwg.org/#parentnode ParentNode interface
should contain append(), prepend() API.

Intent to Implement and ship link: https://groups.google.com/a/chromium.org/forum/#!searchin/blink-dev/paritosh/blink-dev/efUPtYm1PP8/MGoTi17AYpcJ

( From paritosh.in@samsung.com's original CL https://codereview.chromium.org/1085843002 )

R=mkwst
BUG=255482
Committed: https://crrev.com/c5fb1b6a7e7fc6eccd8355211c43cb6396f86d9a
Cr-Commit-Position: refs/heads/master@{#391196}

[sof, 2016-04-30 15:58:20]

Description was changed from

==========
Implement DOM methods: prepend, append, after, before and replaceWith.

R=
BUG=255482
==========

to

==========
Implement DOM methods: prepend, append, after, before and replaceWith.

As per https://dom.spec.whatwg.org/#childnode ChildNode interface
should contain after(), before() and replaceWith() API.

As per https://dom.spec.whatwg.org/#parentnode ParentNode interface
should contain append(), prepend() API.

Intent to Implement and ship link:
https://groups.google.com/a/chromium.org/forum/#!searchin/blink-dev/paritosh/...

( From paritosh.in@samsung.com's original CL
https://codereview.chromium.org/1085843002 )

R=
BUG=255482
==========

[sof, 2016-04-30 16:13:44]

sigbjornf@opera.com changed reviewers:
+ haraken@chromium.org, mkwst@chromium.org, rbyers@chromium.org

[sof, 2016-04-30 16:13:44]

please take a look.

The CL this is based off of didn't get the final push across the finishing line
sometime ago, hence this.

wrt the "intent to implement and ship" involved here -- shipping will have to be
visited later & isn't being done here. But do notice that empty @@unscopables
are additionally exposed by having RuntimeEnabled-conditional [Unscopeable]s.

[Mike West, 2016-05-02 09:23:29]

This looks good, but based on the previous bug, I'd like to understand how this
resolved the issued that cropped up in
https://bugs.chromium.org/p/chromium/issues/detail?id=509461. Could you help me
out there?

[Mike West, 2016-05-02 09:26:12]

I'd also suggest just pinging the intent to ship thread and dropping the runtime
flag from this CL; this should be pretty straightforward now that Safari and
Firefox are already shipping the bits and pieces here, so I'd expect you to just
be able to expose it from the get-go.

[sof, 2016-05-02 09:39:49]

On 2016/05/02 09:26:12, Mike West wrote:
> I'd also suggest just pinging the intent to ship thread and dropping the
runtime
> flag from this CL; this should be pretty straightforward now that Safari and
> Firefox are already shipping the bits and pieces here, so I'd expect you to
just
> be able to expose it from the get-go.

Unclean uses of ref pointers would be my guess, but I haven't investigated. With
Oilpan, pointer handling is more regular & simple.

[sof, 2016-05-02 09:43:21]

On 2016/05/02 09:39:49, sof wrote:
> On 2016/05/02 09:26:12, Mike West wrote:
> > I'd also suggest just pinging the intent to ship thread and dropping the
> runtime
> > flag from this CL; this should be pretty straightforward now that Safari and
> > Firefox are already shipping the bits and pieces here, so I'd expect you to
> just
> > be able to expose it from the get-go.
> 
> Unclean uses of ref pointers would be my guess, but I haven't investigated.
With
> Oilpan, pointer handling is more regular & simple.

(That comment was meant as a reply to #4, sorry.)

One milestone behind experimental would be my suggestion, just in case -- FF
hasn't shipped this yet, just appeared in nightlies.

[Rick Byers, 2016-05-02 14:52:34]

rbyers@chromium.org changed reviewers:
- rbyers@chromium.org

[Rick Byers, 2016-05-02 14:57:37]

On 2016/05/02 09:43:21, sof wrote:
> On 2016/05/02 09:39:49, sof wrote:
> > On 2016/05/02 09:26:12, Mike West wrote:
> > > I'd also suggest just pinging the intent to ship thread and dropping the
> > runtime
> > > flag from this CL; this should be pretty straightforward now that Safari
and
> > > Firefox are already shipping the bits and pieces here, so I'd expect you
to
> > just
> > > be able to expose it from the get-go.
> > 
> > Unclean uses of ref pointers would be my guess, but I haven't investigated.
> With
> > Oilpan, pointer handling is more regular & simple.
> 
> (That comment was meant as a reply to #4, sorry.)
> 
> One milestone behind experimental would be my suggestion, just in case -- FF
> hasn't shipped this yet, just appeared in nightlies.

Yeah if the CL is ready to go, there's probably no sense in blocking on an
intent thread before landing.  So might as well follow the normal pattern for
non-trivial new APIs by landing first as 'experimental'.  This also may give us
a bit more chance to collect any developer feedback.  Also having the
RuntimeEnabledFeature here is even as status=stable for a milestone is
convenient in case we discover compat implications and need to flip it back to
'experimental' on the branch.

Still, if the intent thread goes quickly then I wouldn't hesitate to flip this
to 'stable' in M52 (no need to actually be shipping for a a milestone as
'experimental' first).

Moving myself to cc (and adding tkent) as I'm not an expert in DOM APIs.

[sof, 2016-05-02 20:04:46]

On 2016/05/02 09:39:49, sof wrote:
> On 2016/05/02 09:26:12, Mike West wrote:
> > I'd also suggest just pinging the intent to ship thread and dropping the
> runtime
> > flag from this CL; this should be pretty straightforward now that Safari and
> > Firefox are already shipping the bits and pieces here, so I'd expect you to
> just
> > be able to expose it from the get-go.
> 
> Unclean uses of ref pointers would be my guess, but I haven't investigated.
With
> Oilpan, pointer handling is more regular & simple.

I reckon it is the use of a raw pointer for |viableNextSibling| in
|replaceWith()|

 https://codereview.chromium.org/1934123002/patch/20001/30022

that's the problem across the appendChild() calls that convertNodesIntoNode()
performs -- mutation event dispatching can then remove |viableNextSibling| from
the tree & de-ref it..leaving a dangling pointer. Which is a non-issue in an
Oilpan-based setting, pointers being stack reachable is sufficient.

[Mike West, 2016-05-03 05:19:08]

Got it, thank you and LGTM. I'm happy either way; landing this as experimental
is fine with me. It seems straightforward enough to simply ship, but I'm happy
to leave that up to you and our friendly OWNERS. :)

-mike

[sof, 2016-05-03 05:58:33]

The CQ bit was checked by sigbjornf@opera.com

[commit-bot: I haz the power, 2016-05-03 05:58:44]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1934123002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1934123002/20001

[commit-bot: I haz the power, 2016-05-03 06:05:29]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-05-03 06:05:30]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[sof, 2016-05-03 06:14:01]

On 2016/05/03 05:19:08, Mike West wrote:
> Got it, thank you and LGTM. I'm happy either way; landing this as experimental
> is fine with me. It seems straightforward enough to simply ship, but I'm happy
> to leave that up to you and our friendly OWNERS. :)
> 
> -mike

Thanks Mike,

I'll re-initiate the "intent" blink-dev thread once landed & we'll work out how
to ship these Node additions from that, hopefully.

Safari did ship this last summer (at the same time the original CL here got
close to landing), and it apparently was hiccup-free.

[sof, 2016-05-03 06:14:53]

Description was changed from

==========
Implement DOM methods: prepend, append, after, before and replaceWith.

As per https://dom.spec.whatwg.org/#childnode ChildNode interface
should contain after(), before() and replaceWith() API.

As per https://dom.spec.whatwg.org/#parentnode ParentNode interface
should contain append(), prepend() API.

Intent to Implement and ship link:
https://groups.google.com/a/chromium.org/forum/#!searchin/blink-dev/paritosh/...

( From paritosh.in@samsung.com's original CL
https://codereview.chromium.org/1085843002 )

R=
BUG=255482
==========

to

==========
Implement DOM methods: prepend, append, after, before and replaceWith.

As per https://dom.spec.whatwg.org/#childnode ChildNode interface
should contain after(), before() and replaceWith() API.

As per https://dom.spec.whatwg.org/#parentnode ParentNode interface
should contain append(), prepend() API.

Intent to Implement and ship link:
https://groups.google.com/a/chromium.org/forum/#!searchin/blink-dev/paritosh/...

( From paritosh.in@samsung.com's original CL
https://codereview.chromium.org/1085843002 )

R=mkwst
BUG=255482
==========

[sof, 2016-05-03 06:19:22]

sigbjornf@opera.com changed reviewers:
+ jochen@chromium.org

[sof, 2016-05-03 06:19:23]

aww, yet another case where I attribute too many superpowers to reviewers :)

webexposed/ changes needs API owner approval - jochen@ could you take a look,
please? (tkent@ is not available this week)

[jochen (gone - plz use gerrit), 2016-05-03 11:00:52]

lgtm

[sof, 2016-05-03 11:02:28]

The CQ bit was checked by sigbjornf@opera.com

[commit-bot: I haz the power, 2016-05-03 11:02:42]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1934123002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1934123002/20001

[commit-bot: I haz the power, 2016-05-03 11:07:42]

Description was changed from

==========
Implement DOM methods: prepend, append, after, before and replaceWith.

As per https://dom.spec.whatwg.org/#childnode ChildNode interface
should contain after(), before() and replaceWith() API.

As per https://dom.spec.whatwg.org/#parentnode ParentNode interface
should contain append(), prepend() API.

Intent to Implement and ship link:
https://groups.google.com/a/chromium.org/forum/#!searchin/blink-dev/paritosh/...

( From paritosh.in@samsung.com's original CL
https://codereview.chromium.org/1085843002 )

R=mkwst
BUG=255482
==========

to

==========
Implement DOM methods: prepend, append, after, before and replaceWith.

As per https://dom.spec.whatwg.org/#childnode ChildNode interface
should contain after(), before() and replaceWith() API.

As per https://dom.spec.whatwg.org/#parentnode ParentNode interface
should contain append(), prepend() API.

Intent to Implement and ship link:
https://groups.google.com/a/chromium.org/forum/#!searchin/blink-dev/paritosh/...

( From paritosh.in@samsung.com's original CL
https://codereview.chromium.org/1085843002 )

R=mkwst
BUG=255482
==========

[commit-bot: I haz the power, 2016-05-03 11:07:44]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-05-03 11:10:10]

Description was changed from

==========
Implement DOM methods: prepend, append, after, before and replaceWith.

As per https://dom.spec.whatwg.org/#childnode ChildNode interface
should contain after(), before() and replaceWith() API.

As per https://dom.spec.whatwg.org/#parentnode ParentNode interface
should contain append(), prepend() API.

Intent to Implement and ship link:
https://groups.google.com/a/chromium.org/forum/#!searchin/blink-dev/paritosh/...

( From paritosh.in@samsung.com's original CL
https://codereview.chromium.org/1085843002 )

R=mkwst
BUG=255482
==========

to

==========
Implement DOM methods: prepend, append, after, before and replaceWith.

As per https://dom.spec.whatwg.org/#childnode ChildNode interface
should contain after(), before() and replaceWith() API.

As per https://dom.spec.whatwg.org/#parentnode ParentNode interface
should contain append(), prepend() API.

Intent to Implement and ship link:
https://groups.google.com/a/chromium.org/forum/#!searchin/blink-dev/paritosh/...

( From paritosh.in@samsung.com's original CL
https://codereview.chromium.org/1085843002 )

R=mkwst
BUG=255482

Committed: https://crrev.com/c5fb1b6a7e7fc6eccd8355211c43cb6396f86d9a
Cr-Commit-Position: refs/heads/master@{#391196}
==========

[commit-bot: I haz the power, 2016-05-03 11:10:12]

Patchset 2 (id:??) landed as
https://crrev.com/c5fb1b6a7e7fc6eccd8355211c43cb6396f86d9a
Cr-Commit-Position: refs/heads/master@{#391196}