Closed Bug 150625 Opened 22 years ago Closed 22 years ago

Tables background critical problem on win32

Categories

(Core :: Layout: Tables, defect, P4)

x86
Windows NT
defect

Tracking

()

RESOLVED WORKSFORME

People

(Reporter: deepmind, Assigned: karnaze)

References

()

Details

Attachments

(3 files)

See these two shots:
http://deepmind.brainlex.com/tmp/with_mozilla.jpg
and
http://deepmind.brainlex.com/tmp/screenshot%20brainlex%20beta.png

The 1st one is with Mozilla on Linux. The 2nd one is with Mozilla on Windows.
The page is HTML 4.01 compliant, and even Internet Explorer has no problem with it.

1st bug: the table's background image, which is simply inherited from the body,
does not properly match the body's one.

2nd bug: the background-image: and background-repeat: properties given to posts
titles' TD are not rendered (works fine on Linux, and on every gfx browser
around on Windows -except Mozilla/Win).
Sorry, there is nothing critical in this bug.

You did not say, which versions you use.

Please provide testcase.

pi
Severity: critical → normal
I can reproduce this bug in Mozilla 1.0 - 20020530 in Windows2000 on my page at
http://dirtyredcommie.com/mozbug.html .  If you scroll down the page,
Alpha-layered PNGs in table cell backgrounds that should be appearing will be
visible as they scroll behind other alpha-layered PNGs.  They will either not
appear or only be partially drawn.  

It is particularly apparent if you take a small window and drag it around on top
of the mozilla window.  

This does not occur in Mac OSX. In Linux, I have only been able to generate very
slight abnormalities in the drawing of the table background graphics.  Scrolling
up and down occasionally causes a 2-pixel-thick horizontal line to be missing
from the background.  

I can reproduce this both when the background is part of the table cell tag (as
in <td background="shade.png">) and when using CSS background (as in <td
style="background='shade.png'">)

I will attatch screenshots as I can generate them.
This is the effect I can create when dragging a smaller window around on top of
mozilla.  The table backgrounds are partially redrawn, but they are offset so
that the top left corner of the background graphic appears at the point where
the redraw starts.
Here you can see the effect caused by scrolling my page at
http://dirtyredcommie.com/mozbug.html .  The images that should be drawn as
backgrounds to tables are not drawn during scrolling, but parts of them can be
seen through the alpha-layered PNG that they scroll under.
I can reproduce this bug in Mozilla 1.0 - 20020530 in Windows2000 on my page at
http://dirtyredcommie.com/mozbug.html .  If you scroll down the page,
Alpha-layered PNGs in table cell backgrounds that should be appearing will be
visible as they scroll behind other alpha-layered PNGs.  They will either not
appear or only be partially drawn.  

It is particularly apparent if you take a small window and drag it around on top
of the mozilla window.  

This does not occur in Mac OSX. In Linux, I have only been able to generate very
slight abnormalities in the drawing of the table background graphics.  Scrolling
up and down occasionally causes a 2-pixel-thick horizontal line to be missing
from the background.  

I can reproduce this both when the background is part of the table cell tag (as
in <td background="shade.png">) and when using CSS background (as in <td
style="background='shade.png'">)

I will attatch screenshots as I can generate them.
Ben, I am not sure this is the problem the original reporter is talking about.

Also I am not really sure what I sould see. Maybe you can simplify the testcase
even more.

pi
This is the expected behavior of this page as seen from Mozilla 1.0 in Linux.
I think my bug is a manifestation of Julien's <b>second</b> bug involving the
behavior seen on
http://deepmind.brainlex.com/tmp/screenshot%20brainlex%20beta.png .  

Since I'm using alpha-layered PNG's and more complex layer stuff, he doesn't
have the missing graphics showing up only through other graphics, but I'm
confident that it's the same bug.  

I meant to supply the screenshot from linux last night, but I got busy.  Let me
know if you need any more clarification.
So this WFM under Linux. To get this clear. This is marked as a tables bug. What
you are seeing looks more like an image bug.

pi
It seems to be related to both tables and images, since the images only show
this behavior when used as backgrounds of tables.  

Do you think I should file a separate bug or see if there's another occurance of
this behavior filed as an image bug, then?
Priority: -- → P4
Is it possible to attach a reduced testcase for this problem? 
http://deepmind.brainlex.com/tmp/screenshot%20brainlex%20beta.png is almost
certainly an image error, because it is platform-dependent.  Does this problem
only occur with PNG images (with alpha-channel, perhaps?)  Currently this bug
does not have enough information to fix or confirm.
This may also be a result of ATI graphics card driver problems.  Please check
your graphics driver to see if it is ATI.   Marking dependent on Bug 104992
Depends on: 104992
No response to the last two comments, marking WFM.

pi
Status: UNCONFIRMED → RESOLVED
Closed: 22 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: