[Webkit-unassigned] [Bug 52268] Setting outerText should convert CR/LF to <br>

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=52268





--- Comment #4 from Eric Seidel <eric at webkit.org>  2011-01-11 23:10:07 PST ---
(From update of attachment 78645)
View in context: https://bugs.webkit.org/attachment.cgi?id=78645&action=review

> Source/WebCore/html/HTMLElement.cpp:381
> +    ec = 0;

I think this is generally the callers responsibility in WebCore.

> Source/WebCore/html/HTMLElement.cpp:390
> +                fragment->appendChild(Text::create(document(), text.substring(lineStart, i - lineStart)), ec);

If this can run arbitrary javascript, "this" could get deleted, no?  Do we need to suspend mutation events during this?

> Source/WebCore/html/HTMLElement.cpp:399
> +            lineStart = i + 1;

I find it difficult to read this loop and understand what its doing.  I can't tell if that's a variable naming problem, the way the blocks are split up, or just my own thick-headedness at this hour.

> Source/WebCore/html/HTMLElement.cpp:498
> +        textPrev->appendData(textNode->data(), ec);

Does this cause JS to run?  If so, our pointers could go invalid.

> Source/WebCore/html/HTMLElement.cpp:511
> +        RefPtr<Text> textNext = static_cast<Text*>(next.get());
> +        RefPtr<Text> textNode = static_cast<Text*>(node);
> +        textNode->appendData(textNext->data(), ec);

Seems we just did this above.  Maybe there is code to share here?

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.