Closed Bug 3854 Opened 26 years ago Closed 26 years ago

Pages not displayed correctly

Categories

(Core :: Internationalization, defect, P3)

Product:
Core
Component:
Internationalization
Platform:
x86
Windows NT
Type:
defect

Tracking

()

Status:
VERIFIED FIXED

People

(Reporter: marina, Assigned: karnaze)

References

(
URL
)

Details

Steps to reproduce:
go to the above URL and see that the page is not displayed.
(displays fine in 4.51)
observed with 3/15/99 on the FR Windows NT 4.0
QA Contact: 3853 → 4648
Component: Apprunner → Internationalization
Target Milestone: M4
Assignee: ftang → vidur
javascript issue, reassign to vidur
Assignee: vidur → troy
Summary: Pages written with JavaScript are not displayed correctly → Pages not displayed correctly
Sorry, the incorrect display of this page has nothing to do with JavaScript. The
fact that there is JavaScript on the page doesn't influence layout in this case.
Try Troy for this one.
Assignee: troy → karnaze
The problem is with the fieldset and how it behaves when content doesn't fit
within the fieldset.

Chris, I suspect the fieldset code isn't setting the flag that says that it has
child content that extends outside its frame, and it isn't correctly computing
the "combined area" member of nsHTMLReflowMetrics.

Talk with Kipp about this, it's stuff he added
Status: NEW → ASSIGNED
Comments from Kipp:

Okay, I looked at the test and there are several problems:

1. For marina: <span style="float:right; padding-top: 70px;"> will not work
because floating elements need a "width" value in
order to be rendered (see CSS2 spec, section 10.3.5 - auto widths are mapped to
zero).

2. For chris. Ignoring the form elements on top of each other, the only thing
else I see is that the legend width seems to constrain
itself instead of handling more content than fits. Is this correct? I dunno...

What troy is talking about is how we manage the NS_FRAME_OUTSIDE_CHILDREN bit in
the nsFrame::mState variable.
The way it works is that IF a frame sets that bit then it will also return in
the nsHTMLReflowMetrics.mCombinedArea the total
area that needs rendering as opposed to just its bounds. For example:

<body>
<img src=... align=left width=100 height=1000>Hello World
</body>

In the above example the floating image extends beyond the bottom of the body
(but not beyond the html container). Therefore
while the area of the body is small (one line that contains the "Hello World")
the combined-area of the body is large and includes
all of the floater.

What this does is allow the block elements to know how to render properly in the
face of children that stick outside their bounds.
Whiteboard: (04/07) 3jrgm@qlink.queensu.ca -- review complete (Kipp notes)
Target Milestone: M4 → M5
Moving to M5. I need to fix bug 4534 before this.
Target Milestone: M5 → M6
Moving to M6.
Target Milestone: M6 → M8
Moving to M8
Assignee: karnaze → evaughan
Status: ASSIGNED → NEW
Reassigning to Eric.
Assignee: evaughan → karnaze
bug 4534 has been fixed. Does this page now layout properly?
This page was modified from the URL mentioned in the bug 4534,
http://blues/users/ekrock/publish/traveldemo/0404/reservations.html.

I see the http://blues/users/ekrock/publish/traveldemo/0404/reservations.html
is displayed correctly. However, above URL's layout is not displayed correctly.
It is OK to move to M9.
QA Contact: marina → teruko
I forgot to mention that I tested 7-07 and 7-08 build.
I changed QA contact to teruko@netscape.com.
Whiteboard: (04/07) 3jrgm@qlink.queensu.ca -- review complete (Kipp notes)
teruko,  can you narrow down to a smaller test case of what is not working,
and also check to see if this has regressed recently, or if it is just
something that has not come on line yet.   moving to m9 unless karnaze
or evaughn have fixes.
Target Milestone: M8 → M9
teruko,  can you narrow down to a smaller test case of what is not working,
and also check to see if this has regressed recently, or if it is just
something that has not come on line yet.   moving to m9 unless karnaze
or evaughn have fixes.
Status: NEW → RESOLVED
Closed: 26 years ago
Resolution: --- → FIXED
I looked at this test cases.  This is not same as http://blues/users/ekrock/publish/traveldemo/0404/reservations.html.
I modified the test cases.  Now, it displayes fine.
Status: RESOLVED → VERIFIED
You need to log in before you can comment on or make changes to this bug.