Closed
Bug 53460
Opened 24 years ago
Closed 24 years ago
Cellpadding destroys hover effect on table cells
Categories
(Core :: Layout, defect, P3)
Tracking
()
People
(Reporter: allen.sam, Assigned: joki)
Details
(Keywords: css2, perf, regression)
Attachments
(1 file)
5.54 KB,
text/html
|
Details |
BuildID: 2000092005
When the hover pseudo-class is activated on a table cell that has cellpadding,
a reflow is done even when one is not required, such as when only the
background-color of the table cell changes. The cellpadding is removed, and the
entire table shifts, potentially causing a performance degradation.
Reproducible: Always
Steps to Reproduce:
1. Open the attached testcase.
2. Mouse over the text "Here" in the left table cells for each part of the
testcase.
Actual Results: No reflows should have been triggered; only repaints should
have occurred.
Expected Results: For the first part of the testcase (the part that uses
cellpadding), reflows are consistantly triggered, resulting in a messy repaint
and an unacceptable user experience.
When testing this bug you will run into bug 48771, which means that it may be
tricky to activate the hover psuedo-class.
Admittedly, the "performance degradation" demonstration part of the testcase is
somewhat pathological, but any unneccessary reflow is actually a performance
performance degradtion.
Adding keywords.
I'm fairly certain this is a regression because I tested this stuff more than a
year ago and it wasn't this bad.
Updated•24 years ago
|
Assignee: clayton → joki
joki changed the hover code a little while ago so it could be him, reassigning.
Setting
pref("nglayout.events.showHierarchicalHover", false);
in all.js makes the problem go away, so I believe this is a duplicate of bug
5693.
*** This bug has been marked as a duplicate of 5693 ***
Status: NEW → RESOLVED
Closed: 24 years ago
Resolution: --- → DUPLICATE
You need to log in
before you can comment on or make changes to this bug.
Description
•