170538 - When going to the URL, html is parially rendered.

Status: RESOLVED WORKSFORME

(Core :: DOM: HTML Parser, defect)

Description

[Igor Belayev]

User-Agent:       Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2a) Gecko/20020910
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2a) Gecko/20020910

After visiting this page:
http://bioethics.net/bioethics_intro.php?task=view&articleID=92

In right panel; bellow Introduction to Bioethics
there should be some text rendered. But there is nothing appearing.
Konqueror and IE render page ok.

The same problem persists on Mozilla 1.1 on windows XP.

Is it just a bad code on server ?
If so, how come other browsers show the page ok ?

Reproducible: Always

Steps to Reproduce:
1.
2.
3.

Comment 1

[Boris Zbarsky [:bzbarsky]]

worksforme as far as I can tell with linux build 2002-09-21-04 (identical
rendering to NS4).  Could you post a screenshot of what you see?

Comment 2

[Igor Belayev]

Attachment: A screen of what I see.

The bottom right panel is blank.
There should be some text there.

Could it be a missing font problem ?

Comment 3

[Boris Zbarsky [:bzbarsky]]

Unlikely... are there any errors in the JS console?

Comment 4

[Mats Palmgren (inactive)]

Attachment: Screenshot, 2002-09-21-04 Linux

Looks like a bad comment somewhere on the page. Odd that this did not occur for

you Boris, with the same build.

Comment 5

[Mats Palmgren (inactive)]

Attachment: Testcase

This looks like a regression in parsing of SGML markup declaration w. comments?

I don't recall stuff like this being visible before, even in Standards mode.

Comment 6

[Mats Palmgren (inactive)]

--> Parser

Comment 7

[Boris Zbarsky [:bzbarsky]]

Oh, I see the comment thing.  This looks like the "completely unterminated
comment in strict mode gets not parsed as a comment" thing which we do on
purpose, iirc.  Did we also do that in 1.2a?

Comment 8

[Heikki Toivonen (remove -bugzilla when emailing directly)]

This is on purpose - a malformed comment in strict mode is not a comment, but
becomes a text node. There is a bug that we eat the opening "<!", see bug 144902.

Comment 9

[Mats Palmgren (inactive)]

OK, I just saw that is was different from what we did earlier,
(see bug 102127 comment 16)
Seems like a good change though because I can see directly what the problem is
and the whole page doesn't disappear.

Comment 10

[Igor Belayev]

It's weird.

When I open a new window and go the url, it does not show the
portion of the page. Now, but when I refresh the page. It shows the complete
html page. I can reproduce it with both tabs and new windows.

No errors in java script console.
Another weird thing.
When I have java script console up. The page never gets rendered. Even after I
refresh. After I close java script console, and open a new tab and ask for the
link. It does not get rendered, but now after refreshing the page gets rendered
completely.

Weird.

Anybody else sees the same ?

Comment 11

[Mats Palmgren (inactive)]

I was able to reproduce the missing content problem as shown in
attachment 100430 [details] with 2002-09-23-08 trunk Linux.

Igor is right, there is something weird here - I get different results when I
click Reload, Shift-Reload and Stop in different combinations.

Comment 12

[Igor Belayev]

Webpage is too old.

Comment 13

[Mats Palmgren (inactive)]

The site has been redesigned so the problem is not reproducible anymore, that
does not say anything about wether the bug in Mozilla has been FIXED though.

Comment 14

[Mats Palmgren (inactive)]

If anyone can reproduce the Reload/Shift-Reload problem described in comment 11
please file a new bug report, thanks.

-> WORKSFORME