Add TypeChecking=Unrestricted to SVGPathSeg interfaces and "constructors"

Add TypeChecking=Unrestricted to SVGPathSeg interfaces and "constructors"

"constructors" are the SVGPathElement.createSVGPathSeg* functions.
These methods have low usage, and this matches Geckos behavior.

BUG=354298, 311246
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=190173

[fs, 2015-02-13 16:12:10]

fs@opera.com changed reviewers:
+ ed@opera.com, jl@opera.com, pdr@chromium.org

[Jens Widell, 2015-02-13 16:32:41]

Thanks for working on this!

[pdr., 2015-02-13 22:47:59]

The CQ bit was checked by pdr@chromium.org

[pdr., 2015-02-13 22:48:00]

LGTM, just in time for the SVGPathSeg deprecation! :)

[commit-bot: I haz the power, 2015-02-13 22:48:45]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/925873005/1

[commit-bot: I haz the power, 2015-02-13 22:52:17]

Committed patchset #1 (id:1) as
https://src.chromium.org/viewvc/blink?view=rev&revision=190173

[kouhei (in TOK), 2015-02-16 05:06:35]

lgtm.

Just to confirm: This CL allows methods defined in those interfaces to allow
"+-Inf" values, and the CL isn't related to SVGPathSeg deprecation. Is this
correct?

[fs, 2015-02-16 09:09:26]

On 2015/02/16 05:06:35, kouhei wrote:
> lgtm.
> 
> Just to confirm: This CL allows methods defined in those interfaces to allow
> "+-Inf" values, and the CL isn't related to SVGPathSeg deprecation. Is this
> correct?

Calling SVGPathElement.createSVGPathSeg* or setting any (float) attribute on
SVGPathSeg* to NaN/+/-Inf will now throw a TypeError. That it happened to occur
around the same time as the deprecation is merely due to coincidence. (One could
say that there was a weak connection though...)

[kouhei (in TOK), 2015-02-16 10:15:40]

On 2015/02/16 09:09:26, fs wrote:
> On 2015/02/16 05:06:35, kouhei wrote:
> > lgtm.
> > 
> > Just to confirm: This CL allows methods defined in those interfaces to allow
> > "+-Inf" values, and the CL isn't related to SVGPathSeg deprecation. Is this
> > correct?
> 
> Calling SVGPathElement.createSVGPathSeg* or setting any (float) attribute on
> SVGPathSeg* to NaN/+/-Inf will now throw a TypeError. That it happened to
occur
> around the same time as the deprecation is merely due to coincidence. (One
could
> say that there was a weak connection though...)

Thanks for the clarification!