- From: Maciej Stachowiak <mjs@apple.com>
- Date: Sat, 14 Apr 2007 15:00:37 -0700
- To: Matt Freels <freels@gmail.com>
- Cc: matt@builtfromsource.com, "public-html@w3.org" <public-html@w3.org>
On Apr 14, 2007, at 2:55 AM, Matt Freels wrote:
>
> At least as far as safari goes, Macej's quirksmode test:
>
> <!DOCTYPE html5>
> <style>
> .foo {
> width: 400px;
> height: 400px;
> background-color: lime;
> background-color: ff0000;
> }
> </style>
> <div class="foo"></div>
>
> ...shows a red square for either <!DOCTYPE html> or <!DOCTYPE
> html5>. I don't know if it's actually bailing out to quirksmode, or
> just incorrectly parsing the background-color hex incorrectly
> anyways. FWIW, the test works correctly in recent WebKit builds.
This is a bug in Safari 2's standards mode, here's a test based on a
different quirk that shows <!DOCTYPE html> triggering standards mode.
It depends on case-insensitive class selector matching in quirks mode.
<!DOCTYPE html>
<style>
.foo {
width: 400px;
height: 400px;
background-color: lime;
}
.FOO {
background-color: red;
}
</style>
<div class="foo"></div>
>
> Cheers!
>
> Matt (Freels)
>
> Il giorno 14/apr/07, alle ore 00:02, Matthew Ratzloff ha scritto:
>
>>
>> On Fri, April 13, 2007 2:55 pm, Arthur Jennings wrote:
>>> Perhaps this is a dumb question, but why can't "<!DOCTYPE html>"
>>> be that
>>> clearly identifiable identifier? I doubt it currently occurs in
>>> the wild
>>> outside of authors who are anticipating HTML 5.
>>
>> Safari and Firefox don't recognize "<!DOCTYPE html>" as a valid
>> doctype,
>> instead bailing out to quirks mode.
>>
>> -Matt
>>
>>
>
>
Received on Saturday, 14 April 2007 22:00:53 UTC