Minor tweaks to HTMLCollection

• Component: DOM
• Status: RESOLVED FIXED

[Darin Adler]

Minor tweaks to HTMLCollection

[Darin Adler]

Created attachment 241278 [details]
Patch

[Chris Dumez]

Comment on attachment 241278 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=241278&action=review

nice, r=me

> Source/WebCore/html/HTMLCollection.cpp:451
>      for (unsigned i = 0; i < size; i++) {

nit: ++i

> Source/WebCore/html/HTMLFormControlsCollection.cpp:124
> +Element* HTMLFormControlsCollection::namedItem(const AtomicString& name) const

This could probably return an HTMLElement* even.

> Source/WebCore/html/HTMLFormControlsCollection.h:44
> +    virtual Element* namedItem(const AtomicString& name) const override;

Could likely return an HTMLElement* even.

> Source/WebCore/html/HTMLSelectElement.h:90
> +    Element* namedItem(const AtomicString& name);

Could likely return a HTMLOptionElement*

> Source/WebCore/html/HTMLSelectElement.h:91
> +    Element* item(unsigned index);

Ditto.

> Source/WebCore/page/scrolling/AxisScrollSnapOffsets.cpp:42
> +    for (Element* child = parent.firstElementChild(); child; child = child->nextElementSibling()) {

for (auto& child : childrenOfType<Element>(parent)) { ?

[Darin Adler]

Committed r175947: <http://trac.webkit.org/changeset/175947>

[Chris Dumez]

Seems to have caused assertions in debug builds:
https://build.webkit.org/results/Apple%20Mavericks%20Debug%20WK1%20(Tests)/r175947%20(8835)/results.html

I am looking now.

[Chris Dumez]

Comment on attachment 241278 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=241278&action=review

> Source/WebCore/html/HTMLCollection.cpp:335
> +        for (traversedCount = 0; element && traversedCount < count; ++traversedCount)

I think the issue is that traversedCount gets incremented 1 time less than it used to be, in the case where it hits the end of the list (element is null).

[Chris Dumez]

Comment on attachment 241278 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=241278&action=review

>> Source/WebCore/html/HTMLCollection.cpp:335
>> +        for (traversedCount = 0; element && traversedCount < count; ++traversedCount)
> 
> I think the issue is that traversedCount gets incremented 1 time less than it used to be, in the case where it hits the end of the list (element is null).

Sorry, I meant it gets incremented 1 time more (not less) than before when element is null.