Don't give 'pathLength' semantic errors special treatment

Don't give 'pathLength' semantic errors special treatment

Just set the parse status to the error code for "negative forbidden"
and let it propagate. the fidelity of the message is roughly the same.

BUG=231612
Committed: https://crrev.com/c7f23651516d0962a2da6e124e8a6e9134e2cc52
Cr-Commit-Position: refs/heads/master@{#368160}

[fs, 2015-12-23 16:45:39]

Description was changed from

==========
Don't give 'order' and 'pathLength' semantic errors special treatment

Just set the parse status to the error code for "negative forbidden"
and let it propagate.
This makes SVGDocumentExtensions::reportWarning dead, so remove it.

BUG=231612
==========

to

==========
Don't give 'order' and 'pathLength' semantic errors special treatment

Just set the parse status to the error code for "negative forbidden"
and let it propagate. the fidelity of the message is roughly the same.
This makes SVGDocumentExtensions::reportWarning dead, so remove it.

BUG=231612
==========

[fs, 2016-01-07 08:40:57]

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

[pdr., 2016-01-07 19:27:37]

https://codereview.chromium.org/1548933002/diff/1/third_party/WebKit/Source/c...
File third_party/WebKit/Source/core/svg/SVGFEConvolveMatrixElement.cpp (right):

https://codereview.chromium.org/1548933002/diff/1/third_party/WebKit/Source/c...
third_party/WebKit/Source/core/svg/SVGFEConvolveMatrixElement.cpp:62: if
(parseStatus == NoError && (firstInteger()->baseValue()->value() < 1 ||
secondInteger()->baseValue()->value() < 1))
Pedantic but it looks like a value of zero incorrectly result in a negative
value forbidden error.

Could you add a test of this too? Maybe just modify
svg/filters/feConvolveMatrix-error-handling.html?

[fs, 2016-01-07 19:57:39]

Description was changed from

==========
Don't give 'order' and 'pathLength' semantic errors special treatment

Just set the parse status to the error code for "negative forbidden"
and let it propagate. the fidelity of the message is roughly the same.
This makes SVGDocumentExtensions::reportWarning dead, so remove it.

BUG=231612
==========

to

==========
Don't give 'pathLength' semantic errors special treatment

Just set the parse status to the error code for "negative forbidden"
and let it propagate. the fidelity of the message is roughly the same.

BUG=231612
==========

[fs, 2016-01-07 20:03:50]

https://codereview.chromium.org/1548933002/diff/1/third_party/WebKit/Source/c...
File third_party/WebKit/Source/core/svg/SVGFEConvolveMatrixElement.cpp (right):

https://codereview.chromium.org/1548933002/diff/1/third_party/WebKit/Source/c...
third_party/WebKit/Source/core/svg/SVGFEConvolveMatrixElement.cpp:62: if
(parseStatus == NoError && (firstInteger()->baseValue()->value() < 1 ||
secondInteger()->baseValue()->value() < 1))
On 2016/01/07 at 19:27:37, pdr wrote:
> Pedantic but it looks like a value of zero incorrectly result in a negative
value forbidden error.
> 
> Could you add a test of this too? Maybe just modify
svg/filters/feConvolveMatrix-error-handling.html?

Hmm, this was just me being more blind than usual... These should indeed be > 0,
so I'll leave that for a later fix (and keep reportWarning around until then)
and try to remember to add the test at that point.

[pdr., 2016-01-07 20:30:50]

On 2016/01/07 at 20:03:50, fs wrote:
>
https://codereview.chromium.org/1548933002/diff/1/third_party/WebKit/Source/c...
> File third_party/WebKit/Source/core/svg/SVGFEConvolveMatrixElement.cpp
(right):
> 
>
https://codereview.chromium.org/1548933002/diff/1/third_party/WebKit/Source/c...
> third_party/WebKit/Source/core/svg/SVGFEConvolveMatrixElement.cpp:62: if
(parseStatus == NoError && (firstInteger()->baseValue()->value() < 1 ||
secondInteger()->baseValue()->value() < 1))
> On 2016/01/07 at 19:27:37, pdr wrote:
> > Pedantic but it looks like a value of zero incorrectly result in a negative
value forbidden error.
> > 
> > Could you add a test of this too? Maybe just modify
svg/filters/feConvolveMatrix-error-handling.html?
> 
> Hmm, this was just me being more blind than usual... These should indeed be >
0, so I'll leave that for a later fix (and keep reportWarning around until then)
and try to remember to add the test at that point.

LGTM

[fs, 2016-01-07 21:07:57]

The CQ bit was checked by fs@opera.com

[commit-bot: I haz the power, 2016-01-07 21:08:49]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1548933002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1548933002/20001

[commit-bot: I haz the power, 2016-01-07 21:21:06]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-01-07 21:22:04]

Description was changed from

==========
Don't give 'pathLength' semantic errors special treatment

Just set the parse status to the error code for "negative forbidden"
and let it propagate. the fidelity of the message is roughly the same.

BUG=231612
==========

to

==========
Don't give 'pathLength' semantic errors special treatment

Just set the parse status to the error code for "negative forbidden"
and let it propagate. the fidelity of the message is roughly the same.

BUG=231612

Committed: https://crrev.com/c7f23651516d0962a2da6e124e8a6e9134e2cc52
Cr-Commit-Position: refs/heads/master@{#368160}
==========

[commit-bot: I haz the power, 2016-01-07 21:22:04]

Patchset 2 (id:??) landed as
https://crrev.com/c7f23651516d0962a2da6e124e8a6e9134e2cc52
Cr-Commit-Position: refs/heads/master@{#368160}