In Viewer: when automatically changing the background color from a function in JavaScript, only a part of the background region changes color and not the whole window (please see provided URL as an example.) http://www.chalcedony.com/javascript/scripts/chap03/script15.html

V -- Since I'm out this week, can you please take a quick look at this? It's likely a bug in the view system, but since JS is involved I wanted to make it was well formed.

This looks like a paint/invalidate problem. If you force a localized repaint (drag a window over the affected area) the correct colors show up as the localized area repaints. Tested on 19990809 build on Win98.

Moving this to layout. This looks like repaint problems relating to the BODY background fixup rule (to make it the whole canvas). It's only repainting within the border edge of the BODY element (note carried out margin of H1). EXPECTED: Entire canvas should rotate through colors. OBSERVED: Only the header (out to near the side edges) is rotating through colors, except when a repaint is forced by dragging a window over it or something. TESTED ON: * Linux viewer 1999-09-03-11-M11 * (Win98 1999-08-09 as mentioned above) Marking All/All.

Peter, I guess we're not invalidating the HTML window as well when the background color is changed?

Moving non-beta 1 items to M15

Reassigning peterl's bugs to myself.

Accepting peterl's bugs that have a Target Milestone

Resetting QA contact from leger.

Pushing my M15 bugs to M16

Bulk moving [testcase] code to new testcase keyword. Sorry for the spam!

Another good candidate for the bug contest this week... Could someone verify whether it's not a dup of bug 1886 and bug 24684? Thanks.

Replacing "need help" in the Status Whiteboard with the 'helpwanted' keyword.

This is a major ugliness for Composer -- When we set the background color on a page, which sets BGCOLOR on the BODY tag, only the frames behind the lines of selected text are colored. Resizing the window triggers the proper repaint. Attached is a test file that might help.

adding myself to Cc: list..

reassign to attinasi. Please look at this per rickg's request.

*** Bug 28494 has been marked as a duplicate of this bug. ***

*** Bug 1886 has been marked as a duplicate of this bug. ***

I have a fix for this: essentially I just update all of the views if the BODY is restyled. Not the most elegant solution, but all I could think of. I'm requesting input from troy on a potentially better approach.

*** Bug 30429 has been marked as a duplicate of this bug. ***

*** Bug 33251 has been marked as a duplicate of this bug. ***

Bug 8524, bug 33710 and this bug are related in the way that they rely on the BodyFixupRule that moves the background from the BODY to the HTML element.

*** Bug 36527 has been marked as a duplicate of this bug. ***

Hmm, that last dup is a restricted bug... weird. A similar problem exists for background images. Bug 34560.

*** Bug 34560 has been marked as a duplicate of this bug. ***

Fix is finally ready: in CSSFrameConstructor::AttributeChanged we check it the attribute being changed is the background. If it is, we find the canvas frame and call its (new) AttributeChanged method. That method then checks if the attribute being changed is the bgcolor or the background and invalidates the frame if it is. This way we get the correct region invaildated, and it happens whenever the bgcolor or image is changed, through the BODY tag or through script. I'm getting a checkin ready now...

When you mark this bug fixed, could check whether it has any effect on bug 33387? Please put the result under 33387. Thanks.

Finally, fixed.

Fixed in the July 6th build.