813936 - Update nsHTMLInputElement::GetStepBase() to return the default value if there is no min

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect)

Description

[Mounir Lamouri (:mounir)]

See http://html5.org/tools/web-apps-tracker?from=7516&to=7517

It makes this valid:
<input type='number' value='1' step='2'>

otherwise, '1' wouldn't be a valid value because the step base is 0 and all pair values would have been allowed. The developer would have had to set @min to an odd number to have the expected behaviour.

Comment 1

[Mounir Lamouri (:mounir)]

Attachment: Patch

More work with the tests than the patch :'(

Comment 2

[Olli Pettay [:smaug][[e-mail]]]

(In reply to Mounir Lamouri (:mounir) from comment #1)
> More work with the tests than the patch :'(
Isn't that the norm.

Comment 3

[Olli Pettay [:smaug][[e-mail]]]

Comment on attachment 684739 [details] [diff] [review]
Patch


> nsHTMLInputElement::GetStepBase() const
> {
>   double stepBase = GetMinAsDouble();
> 
>+  // If @min is not a double, we should use defaultValue.
>+  if (MOZ_DOUBLE_IS_NaN(stepBase)) {
>+    nsAutoString stringValue;
>+    nsresult ec;
>+
>+    GetAttr(kNameSpaceID_None, nsGkAtoms::value, stringValue);
>+    stepBase = stringValue.ToDouble(&ec);
nsresult ec; should be right before 
stepBase = stringValue.ToDouble(&ec)


>-var input = document.createElement("input");
>-document.getElementById('content').appendChild(input);
>+function getFreshElement(data) {
>+  var elmt = document.createElement('input');
>+  elmt.type = data[0];
>+  return elmt;
>+}
This is somewhat odd. Don't pass array to this function, but just type.

Comment 4

[Mounir Lamouri (:mounir)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/bf5194cff8e1

Comment 5

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/bf5194cff8e1

Comment 6

[Florian Scholz (Open Web Docs)]

See bug 866457 #c1 for documentation.