Re: safari rendering bug: site display not correct
- From: Eric Lindsay <NOSPAmar2005@xxxxxxxxxxxxxxx>
- Date: Fri, 30 Dec 2005 13:04:06 +1000
In article <41j4dkF1ejpqlU1@xxxxxxxxxxxxxx>,
Markus Dehmann <markus.dehmann@xxxxxxxxx> wrote:
> I found a safari bug:
>
> Safari cannot display this site correctly:
> http://wahnzeit.libsyn.com/index.php?post_id=38396&comments=on
Given that the page in question is not valid HTML, I am not sure there
is any correct way to display it.
> The last four comments on the page (after the "test\ntest\ntest\n?\n?")
> should not be in bold font anymore. Firefox and Opera do it right, but
> Safari makes everything bold.
Also works that way in Safari for italics, underline and emphasis.
Probably the same for any inline element (although I didn't test them
all).
> One poster in the middle of the page had included a "<b><b><b>" in his
> posting, but did not close the tags. So, everything that follows is
> bold. A later posting tried to fix it, using "</b></b></b>" tags, but
> Safari ignores them.
Safari is obeying a delayed close to inline elements UNTIL it encounters
a new div tag (just closing the existing div does not stop Safari from
returning to normal text display, if it encounters appropriate close
tags).
Looks like it may be a general parsing problem in the tag soup parser.
I tried it with the inline font style elements bold, underline and the
inline phrase element emphasis, none of which are actually supposed to
be used outside the block element they occur within.
Personally I would have thought all browsers should stop displaying them
once the end of the enclosing block element (such as a paragraph) was
reached. However I know that many web pages have done things like put a
single bold tag around multiple paragraphs. So I can see why Safari
(and other browsers) don't just stop doing bold when they get to a new
paragraph. Looks like Safari loses track of non-closed inline elements
when it reaches a new div.
I did try throwing Safari into Standards mode with a HTML 4.01 Strict
Doctype, to see whether the problem was specific to Quirks mode or not.
Seems to be there in both modes.
You probably should report it as a problem. It should be easy to
generate a very short test page demonstrating all forms of the problem.
--
http://www.ericlindsay.com
.
- Follow-Ups:
- Re: safari rendering bug: site display not correct
- From: Warren Oates
- Re: safari rendering bug: site display not correct
- From: Markus Dehmann
- Re: safari rendering bug: site display not correct
- References:
- safari rendering bug: site display not correct
- From: Markus Dehmann
- safari rendering bug: site display not correct
- Prev by Date: remote control
- Next by Date: Re: safari rendering bug: site display not correct
- Previous by thread: safari rendering bug: site display not correct
- Next by thread: Re: safari rendering bug: site display not correct
- Index(es):
Relevant Pages
|