HTMLTableSectionElement.insertRow()'s argument default value should be -1

HTMLTableSectionElement.insertRow()'s argument default value should be -1

HTMLTableSectionElement.insertRow()'s argument default value should be -1, not
0, as per the specification:
http://www.whatwg.org/specs/web-apps/current-work/multipage/tabular-data.html#the-tbody-element

The new behavior matches the one of Firefox 29 and IE11.

This CL also cleans up the implementation of HTMLTableSectionElement::insertRow()
a bit.

R=arv@chromium.org, tkent@chromium.org
BUG=369803
TEST=fast/dom/HTMLTableSectionElement/insertRow-default-argument.html

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=173339

[eseidel, 2014-05-05 15:30:51]

I hope we're convincing Safari to change too. :)

[Inactive, 2014-05-05 15:32:56]

On 2014/05/05 15:30:51, eseidel wrote:
> I hope we're convincing Safari to change too. :)

Good point, I am going to propose the same change to WebKit today.

[eseidel, 2014-05-05 15:33:42]

lgtm

https://codereview.chromium.org/263193002/diff/1/Source/core/html/HTMLTableSe...
File Source/core/html/HTMLTableSectionElement.cpp (left):

https://codereview.chromium.org/263193002/diff/1/Source/core/html/HTMLTableSe...
Source/core/html/HTMLTableSectionElement.cpp:65: if (index < -1 || index >
numRows) {
Lame that indicies don't work like python here. :)  -2 should be totally valid
IMO. :)

https://codereview.chromium.org/263193002/diff/1/Source/core/html/HTMLTableSe...
File Source/core/html/HTMLTableSectionElement.idl (right):

https://codereview.chromium.org/263193002/diff/1/Source/core/html/HTMLTableSe...
Source/core/html/HTMLTableSectionElement.idl:27: [RaisesException] HTMLElement
insertRow(optional long index);
Should this have a return value of HTMLTableRowElement?

[Inactive, 2014-05-05 15:35:48]

https://codereview.chromium.org/263193002/diff/1/Source/core/html/HTMLTableSe...
File Source/core/html/HTMLTableSectionElement.idl (right):

https://codereview.chromium.org/263193002/diff/1/Source/core/html/HTMLTableSe...
Source/core/html/HTMLTableSectionElement.idl:27: [RaisesException] HTMLElement
insertRow(optional long index);
On 2014/05/05 15:33:42, eseidel wrote:
> Should this have a return value of HTMLTableRowElement?

It could but the specification says "HTMLElement":
http://www.whatwg.org/specs/web-apps/current-work/multipage/tabular-data.html...

[Inactive, 2014-05-05 18:16:56]

On 2014/05/05 15:32:56, Chris Dumez wrote:
> On 2014/05/05 15:30:51, eseidel wrote:
> > I hope we're convincing Safari to change too. :)

I uploaded the patch to WebKit as well:
https://bugs.webkit.org/show_bug.cgi?id=132570

[Inactive, 2014-05-05 21:31:56]

The CQ bit was checked by ch.dumez@samsung.com

[commit-bot: I haz the power, 2014-05-05 21:32:14]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/ch.dumez@samsung.com/263193002/1

[commit-bot: I haz the power, 2014-05-05 21:45:26]

Change committed as 173339

[arv (Not doing code reviews), 2014-05-06 17:53:41]

https://codereview.chromium.org/263193002/diff/1/Source/core/html/HTMLTableSe...
File Source/core/html/HTMLTableSectionElement.cpp (right):

https://codereview.chromium.org/263193002/diff/1/Source/core/html/HTMLTableSe...
Source/core/html/HTMLTableSectionElement.cpp:79: Node* n = index ?
children->item(index) : firstChild();
fistChild does not seem right here. It should be firstElementChild. You should
be able to test this with a whitespace text node as the firstChild.

"Otherwise, the method must create a tr element, insert it as a child of the
table section element, immediately before the indexth tr element in the rows
collection, and finally must return the newly created tr element."

[Inactive, 2014-05-06 17:57:54]

https://codereview.chromium.org/263193002/diff/1/Source/core/html/HTMLTableSe...
File Source/core/html/HTMLTableSectionElement.cpp (right):

https://codereview.chromium.org/263193002/diff/1/Source/core/html/HTMLTableSe...
Source/core/html/HTMLTableSectionElement.cpp:79: Node* n = index ?
children->item(index) : firstChild();
On 2014/05/06 17:53:42, arv wrote:
> fistChild does not seem right here. It should be firstElementChild. You should
> be able to test this with a whitespace text node as the firstChild.
> 
> "Otherwise, the method must create a tr element, insert it as a child of the
> table section element, immediately before the indexth tr element in the rows
> collection, and finally must return the newly created tr element."

Oh, I thought this looked weird as well when I refactored this part of the code.
I will take investigate and fix if needed in a follow-up patch. Thanks for
noticing this.

[Inactive, 2014-05-06 18:55:45]

https://codereview.chromium.org/263193002/diff/1/Source/core/html/HTMLTableSe...
File Source/core/html/HTMLTableSectionElement.cpp (right):

https://codereview.chromium.org/263193002/diff/1/Source/core/html/HTMLTableSe...
Source/core/html/HTMLTableSectionElement.cpp:79: Node* n = index ?
children->item(index) : firstChild();
On 2014/05/06 17:57:54, Chris Dumez wrote:
> On 2014/05/06 17:53:42, arv wrote:
> > fistChild does not seem right here. It should be firstElementChild. You
should
> > be able to test this with a whitespace text node as the firstChild.
> > 
> > "Otherwise, the method must create a tr element, insert it as a child of the
> > table section element, immediately before the indexth tr element in the rows
> > collection, and finally must return the newly created tr element."
> 
> Oh, I thought this looked weird as well when I refactored this part of the
code.
> I will take investigate and fix if needed in a follow-up patch. Thanks for
> noticing this.

Yes, I confirmed we behave differently thatn Firefox and the spec here:
http://jsfiddle.net/JS4VU/1/