Input type Number maximum value increase

Input type Number maximum value increase

Number input type maximum value increased from float to double. Float allowed only max value of 3.402823466e38, it has now been increased to 1.7976931348623158e+308, by using double.

InvalidStateError exception is removed from NumberInputType::setValueAsDouble and NumberInputType::setValueAsDecimal, as HTMLInputType::setValueAsNumber checks if value is infinite.
As NumberInputType functions will not be called, they have been removed.

HTMLParserIdioms has also been updated as the functions are triggered in NumberInputType functions and has check for max values, thus updated.

R=tkent, ch.dumez@samsung.com
BUG=336020
TEST=Max value number can hold and exception thrown in case higher value than double is set

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=168513

[Habib Virji, 2014-02-19 11:08:44]

Number input type limit increased by using double. 

Invalid state error removed as check exists in HTMLInputType and thus not
required.

HTMLParseIdioms too has changes as NumberInputType calls this function.

Test include increase in MAX value number input type can hold. Due to invalid
state error not getting executed, exception tests has been modified based on
input received from HTMLInputElement.

[Inactive, 2014-02-19 14:23:17]

The change looks sane but I'll let tkent review this.

https://codereview.chromium.org/172223003/diff/1/Source/core/html/parser/HTML...
File Source/core/html/parser/HTMLParserIdioms.cpp (right):

https://codereview.chromium.org/172223003/diff/1/Source/core/html/parser/HTML...
Source/core/html/parser/HTMLParserIdioms.cpp:111: // FIXME: We should use
numeric_limits<double>::max for number input type.
Please remove FIXME comment.

[Habib Virji, 2014-02-19 15:19:44]

Thanks for the review. Removed FIXME comment.

https://codereview.chromium.org/172223003/diff/1/Source/core/html/parser/HTML...
File Source/core/html/parser/HTMLParserIdioms.cpp (right):

https://codereview.chromium.org/172223003/diff/1/Source/core/html/parser/HTML...
Source/core/html/parser/HTMLParserIdioms.cpp:111: // FIXME: We should use
numeric_limits<double>::max for number input type.
On 2014/02/19 14:23:18, Chris Dumez wrote:
> Please remove FIXME comment.

Done.

[tkent, 2014-02-19 23:55:53]

+keishi and yosin

[yosin_UTC9, 2014-02-24 03:44:38]

NACK

According to HTML5 specification,
http://www.whatwg.org/specs/web-apps/2009-10-27/multipage/common-microsyntaxe...

Real numbers should be IEEE 754 single-precision floating point values.

[Habib Virji, 2014-02-24 09:22:24]

On 2014/02/24 03:44:38, Yoshi wrote:
> NACK
> 
> According to HTML5 specification,
>
http://www.whatwg.org/specs/web-apps/2009-10-27/multipage/common-microsyntaxe...
> 
> Real numbers should be IEEE 754 single-precision floating point values.

As per updated specs:
http://www.whatwg.org/specs/web-apps/current-work/#floating-point-numbers. 

Point 15: says IEEE 754 double-precision floating-point, max value is 383, which
is max value of double.

[yosin_UTC9, 2014-02-25 02:18:40]

On 2014/02/24 09:22:24, Habib Virji wrote:
> On 2014/02/24 03:44:38, Yoshi wrote:
> > NACK
> > 
> > According to HTML5 specification,
> >
>
http://www.whatwg.org/specs/web-apps/2009-10-27/multipage/common-microsyntaxe...
> > 
> > Real numbers should be IEEE 754 single-precision floating point values.
> 
> As per updated specs:
> http://www.whatwg.org/specs/web-apps/current-work/#floating-point-numbers. 
> 
> Point 15: says IEEE 754 double-precision floating-point, max value is 383,
which
> is max value of double.

Oh, my apology, I referred old spec.
ACK!

[yosin_UTC9, 2014-02-25 02:20:30]

https://codereview.chromium.org/172223003/diff/10010/Source/core/html/parser/...
File Source/core/html/parser/HTMLParserIdioms.cpp (right):

https://codereview.chromium.org/172223003/diff/10010/Source/core/html/parser/...
Source/core/html/parser/HTMLParserIdioms.cpp:110: // See HTML5 2.5.4.3 `Real
numbers.'
Please update this comment. Now we refer "2.4.4.3 Floating-point numbers" and
using IEEE 754 double-precision floating-point values

[Habib Virji, 2014-02-25 10:09:27]

Thanks Yoshi for reviewing.

Updated comments referring to new section in HTML5 spec (2.4.4.3) and updated
single to double precision point comment.

https://codereview.chromium.org/172223003/diff/10010/Source/core/html/parser/...
File Source/core/html/parser/HTMLParserIdioms.cpp (right):

https://codereview.chromium.org/172223003/diff/10010/Source/core/html/parser/...
Source/core/html/parser/HTMLParserIdioms.cpp:110: // See HTML5 2.5.4.3 `Real
numbers.'
On 2014/02/25 02:20:31, Yoshi wrote:
> Please update this comment. Now we refer "2.4.4.3 Floating-point numbers" and
> using IEEE 754 double-precision floating-point values

Done.

[yosin_UTC9, 2014-02-26 01:31:19]

https://codereview.chromium.org/172223003/diff/150001/Source/core/html/parser...
File Source/core/html/parser/HTMLParserIdioms.cpp (right):

https://codereview.chromium.org/172223003/diff/150001/Source/core/html/parser...
Source/core/html/parser/HTMLParserIdioms.cpp:98: // See HTML5 2.4.4.3
`Floating-point numbers.' and parseToDoubleForNumberType
I think URI for the spec is better. And, I would like to suggest to remove
section number. These numbers will be changed by version-to-version of specs.

tkent@?

[tkent, 2014-02-26 03:49:15]

We need to update acceptableError() and stepMismatch() in
core/html/forms/StepRange.cpp.

[Habib Virji, 2014-02-26 08:49:22]

On 2014/02/26 01:31:19, Yoshi wrote:
>
https://codereview.chromium.org/172223003/diff/150001/Source/core/html/parser...
> File Source/core/html/parser/HTMLParserIdioms.cpp (right):
> 
>
https://codereview.chromium.org/172223003/diff/150001/Source/core/html/parser...
> Source/core/html/parser/HTMLParserIdioms.cpp:98: // See HTML5 2.4.4.3
> `Floating-point numbers.' and parseToDoubleForNumberType
> I think URI for the spec is better. And, I would like to suggest to remove
> section number. These numbers will be changed by version-to-version of specs.
> 
> tkent@?

Done. As well as updates in StepRange.

[Habib Virji, 2014-02-26 17:11:12]

@tkent: I added back in StepRange, FLT_MANT_DIG as found using DBL_MANT_DIG in
acceptableError is breaking fraction tests. All test though pass and it does not
seem to be relevant to max value but rather to handle to stepMismatch.

[tkent, 2014-02-28 01:11:17]

On 2014/02/26 17:11:12, Habib Virji wrote:
> @tkent: I added back in StepRange, FLT_MANT_DIG as found using DBL_MANT_DIG in
> acceptableError is breaking fraction tests. All test though pass and it does
not
> seem to be relevant to max value but rather to handle to stepMismatch.

We should update acceptableError.  The current acceptableError is too large for
double-precision calculation.
But you're right.  We can update acceptableError in another CL.

[tkent, 2014-02-28 01:12:03]

https://codereview.chromium.org/172223003/diff/190001/Source/core/html/forms/...
File Source/core/html/forms/StepRange.cpp (right):

https://codereview.chromium.org/172223003/diff/190001/Source/core/html/forms/...
Source/core/html/forms/StepRange.cpp:29: #include <stdio.h>
This change is unnecessary.

https://codereview.chromium.org/172223003/diff/190001/Source/core/html/forms/...
Source/core/html/forms/StepRange.cpp:161: //
http://www.whatwg.org/specs/web-apps/current-work/#the-step-attribute:
We don't need to update this now.

https://codereview.chromium.org/172223003/diff/190001/Source/core/html/forms/...
Source/core/html/forms/StepRange.cpp:165: // Accepts errors in lower fractional
part which IEEE 754 double-precision
Ditto.

[Habib Virji, 2014-03-04 09:33:44]

Updated as per review comments.

https://codereview.chromium.org/172223003/diff/190001/Source/core/html/forms/...
File Source/core/html/forms/StepRange.cpp (right):

https://codereview.chromium.org/172223003/diff/190001/Source/core/html/forms/...
Source/core/html/forms/StepRange.cpp:29: #include <stdio.h>
On 2014/02/28 01:12:04, tkent wrote:
> This change is unnecessary.

Done. Apologies missed it.

https://codereview.chromium.org/172223003/diff/190001/Source/core/html/forms/...
Source/core/html/forms/StepRange.cpp:161: //
http://www.whatwg.org/specs/web-apps/current-work/#the-step-attribute:
On 2014/02/28 01:12:04, tkent wrote:
> We don't need to update this now.

Done.

https://codereview.chromium.org/172223003/diff/190001/Source/core/html/forms/...
Source/core/html/forms/StepRange.cpp:165: // Accepts errors in lower fractional
part which IEEE 754 double-precision
On 2014/02/28 01:12:04, tkent wrote:
> Ditto.

Done.

[tkent, 2014-03-04 22:19:30]

lgtm.  Thank you!

[tkent, 2014-03-04 22:19:34]

The CQ bit was checked by tkent@chromium.org

[commit-bot: I haz the power, 2014-03-04 22:19:56]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/habib.virji@samsung.com/172223003/210001

[Habib Virji, 2014-03-05 08:39:34]

On 2014/03/04 22:19:30, tkent wrote:
> lgtm.  Thank you!

Thanks.

[commit-bot: I haz the power, 2014-03-05 18:44:21]

Change committed as 168513