{list} [BLOCK] List numbering code is inefficient

NEW
Unassigned

Status

()

P3
normal
20 years ago
4 months ago

People

(Reporter: vidur, Unassigned)

Tracking

(Depends on: 1 bug, {hang, perf, testcase})

Trunk
All
Windows NT
hang, perf, testcase
Points:
---
Dependency tree / graph

Firefox Tracking Flags

(Not tracked)

Details

(Reporter)

Description

20 years ago
Since list renumbering happens for any block frame, this is potentially a O(n^2)
algorithm. I was looking at this when I hooked up attribute changes for lists
and list items. If you make modifications to list renumbering, look at the
AttributeChanged method for nsBlockFrame. You will notice that I renumber lists
when a list item's "value" attribute is changed. The code currently makes the
assumption that the renumbering must be done by the first block ancestor of the
list item. If you modify list renumbering to make different assumptions about
parenting, please modify this code as well. (test to come)

Updated

20 years ago
Summary: List numbering code is inefficient → {perf} List numbering code is inefficient

Updated

20 years ago
Priority: P2 → P5

Updated

20 years ago
QA Contact: 4110 → 3849

Updated

20 years ago
Whiteboard: [Perf]

Comment 1

20 years ago
putting on [Perf] radar
Summary: {perf} List numbering code is inefficient → {perf} {list} List numbering code is inefficient

Updated

19 years ago
Blocks: 8691

Updated

19 years ago
QA Contact: beppe → petersen

Comment 2

19 years ago
Updating to default Layout Assignee...kipp no longer with us :-(

Comment 3

19 years ago
Why are you re-reassing layout bugs? Do NOT touch layout bugs.

The bugs are assigned to Kipp so they can stay neatly organized until we have a
new owner for the block/inline code.

Updated

19 years ago
Summary: {perf} {list} List numbering code is inefficient → {perf} {list} [BLOCK] List numbering code is inefficient

Updated

19 years ago
Keywords: perf

Comment 4

19 years ago
Bulk add of "perf" to new keyword field.  This will replace the [PERF] we were
using in the Status Summary field.

Updated

19 years ago
Summary: {perf} {list} [BLOCK] List numbering code is inefficient → {list} [BLOCK] List numbering code is inefficient
Whiteboard: [Perf]

Comment 5

19 years ago
kipp is very unlikely to fix these, since he's not working ont he project any 
longer.  so I'll take a look.
Assignee: kipp → buster

Comment 6

19 years ago
another possibly "future" perf bug.

This bug has been marked "future" because we have determined that it is not 
critical for netscape6.0. 
If you feel this is an error, or if it blocks your work in some way -- please 
attach your concern to the bug for reconsideration.
Assignee: buster → waterson
Target Milestone: M17 → Future

Updated

18 years ago
Status: NEW → ASSIGNED
putting on the nsbeta1 radar for consideration by engineering
Keywords: nsbeta1

Updated

17 years ago
Blocks: 42138

Comment 8

17 years ago
Cc'ing myself. Is this still an issue?

Updated

17 years ago
Blocks: 71668

Updated

17 years ago
Blocks: 104166

Updated

16 years ago
Component: Layout → Layout: Block & Inline

Comment 9

15 years ago
Reassigning ...
Assignee: waterson → core.layout.block-and-inline
Status: ASSIGNED → NEW
Assignee: layout.block-and-inline → nobody
QA Contact: chrispetersen → layout.block-and-inline
The testcase in bug 881090 hangs in nsBlockFrame::RenumberListsInBlock / RenumberListsFor.
Keywords: hang, testcase
Target Milestone: Future → ---
You need to log in before you can comment on or make changes to this bug.