feat(Component): add wrapElement parameter so that Components can be…

[Pajn]

…placed in SVG

I'm submitting this to get some response. I don't know how to test this
as the template is loaded anyway, it just isn't rendered.

Closes #987

[mhevery]

Thanks for your contribution! In order for us to be able to accept it, we ask you to sign our CLA (contributor's license agreement). CLA is important for us to be able to avoid legal troubles down the road.

For individuals (a simple click-through form): http://code.google.com/legal/individual-cla-v1.0.html

[Pajn]

I have signed to CLA using my primary email. Should I sign with the gmail instead?

[vicb]

I'm submitting this to get some response

Why is the question ?

Should I sign with the gmail instead?

You should sign the CLA for the email associated with the commit which apparently is ok

[Pajn]

Why is the question ?
Sorry.
The namespaceUri works in Chrome (and Dartium) but not in IE. IE also requires svg to be used as tagname in Pajn@6e655d1#diff-b8155758b618de44582b59036af50e64R145
This would require that both namespaceUri and tagname is specified (as well as useShadowDom: false as SVG doesn't support shadow dom).
Just specifying the tagname (as in angular/angular.js@f0e12ea) doesn't work either (this is tried with dom.document.createElement("svg"), new dom.Element.tag("svg") and new dom.Element.html("<svg></svg>").

The alternative would be to add specific SVG support that sets tagname, namespaceUrl and shadown dom. In this case, should MathML be investigated as well?

EDIT: I found the wrap technique of AngularJS work if done like them and put it inside of the div.

[vicb]

I wonder how this work on the compiled version, don't the transformers need to be updated to take wrapElement into account ?

[Pajn]

ie: indent by 4ws, onyl specified what's needed

Fixed.

is the di pkg really needed ?

Fixed.

Hi -> SVG based component

Fixed.

What prevent this PR from being tested by a regular test (rather an an e2e test which is more expensive)

As the test rather test how the browser handles the component rather than Angular I thought it were needed, but I will investigate if a unit test will do.

I wonder how this work on the compiled version, don't the transformers need to be updated to take wrapElement into account ?

It's tested in Chrome, Chromium and IE11 (all compiled), it's not tested in Firefox as I experience the same issue with this as with shadow_dom_components example (#1337). I don't have any mac so I can't test Safari.

[mhevery]

Achievement unlocked: CLA signature found!

[naomiblack]

SVG is not a high priority in pre-1.0. @vicb can you please take a quick look and see if this can be included without a lot of work elsewhere, or close?

[vicb]

@Pajn makes sense, we have no clear policy about removing els but it's better to be on the safe side.
LGTM.

[rkirov]

This requires a trivial rebasing. Punting to post v1.0.

[naomiblack]

It turns out that that this touches a lot of AngularDart code and the rebasing was non-trivial.
We're going to close this issue to focus on 2.0.