Split normalized reading from non-normalized in SVGPathParser

Split normalized reading from non-normalized in SVGPathParser

Move to parseAndNormalizePath() and parsePath() respectively and move
state to the corresponding one as appropriate.

Default inline parsePathDataFromSource. Since callsite usually pass a
constant PathParsingMode and checkForInitialMoveTo, this allows for
some specialization.

BUG=467592
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=192901

[fs, 2015-03-31 18:43:28]

fs@opera.com changed reviewers:
+ fmalita@chromium.org, pdr@chromium.org, schenney@chromium.org

[f(malita), 2015-04-01 03:59:20]

The CQ bit was checked by fmalita@chromium.org

[f(malita), 2015-04-01 03:59:21]

lgtm

https://codereview.chromium.org/1045343002/diff/1/Source/core/svg/SVGPathPars...
File Source/core/svg/SVGPathParser.cpp (right):

https://codereview.chromium.org/1045343002/diff/1/Source/core/svg/SVGPathPars...
Source/core/svg/SVGPathParser.cpp:333: bool
SVGPathParser::parseAndNormalizePath()
These two are quite similar - would it be feasible to consolidate as a function
template (with a consumer arg) at some point?

[commit-bot: I haz the power, 2015-04-01 03:59:54]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1045343002/1

[commit-bot: I haz the power, 2015-04-01 04:03:40]

Committed patchset #1 (id:1) as
https://src.chromium.org/viewvc/blink?view=rev&revision=192901

[fs, 2015-04-01 08:55:29]

https://codereview.chromium.org/1045343002/diff/1/Source/core/svg/SVGPathPars...
File Source/core/svg/SVGPathParser.cpp (right):

https://codereview.chromium.org/1045343002/diff/1/Source/core/svg/SVGPathPars...
Source/core/svg/SVGPathParser.cpp:333: bool
SVGPathParser::parseAndNormalizePath()
On 2015/04/01 03:59:21, f(malita) wrote:
> These two are quite similar - would it be feasible to consolidate as a
function
> template (with a consumer arg) at some point?

Yes, I hope to get to that point eventually. Haven't quite made up my mind on
whether it's better to let consumers do normalization explicitly (aggregating
the NormalizingConsumer) or leave it to the caller - or both =). Leaning towards
the latter ATM, (early) opinions welcome.