dbaron%dbaron.org
9a2f5780e9
Stop rounding widths to pixel boundaries to fix reftest failures on Mac, which does subpixel positioning and sizing of text. b=369318 r=bernd sr=roc
2007-02-08 19:21:58 +00:00
dbaron%dbaron.org
7a03b97f01
Stop rounding final column widths (up) to pixel boundaries. b=349297 r=bernd sr=roc
2007-02-08 19:20:51 +00:00
sharparrow1%yahoo.com
76a64e5c22
Bug 177805: Fix the use of units in Gecko. r+sr=roc
2007-02-07 07:46:44 +00:00
dbaron%dbaron.org
1e0db05c93
Make default table and hr border colors be gray rather than special values that depend on the background color, and remove the special values that depend on the background. b=84307 Patch by fantasai <fantasai.bugs@inkedblade.net>. r=bzbarsky sr=dbaron
2007-02-01 02:45:55 +00:00
cvshook%sicking.cc
a39ff30284
Kill remainders of ns(XBL|HTML|XUL|Layout)Atoms. r/sr=jst b=368128
2007-01-30 00:06:41 +00:00
bzbarsky%mit.edu
f29ad74749
Fix crash bug 368166. r=bernd, sr=roc
2007-01-28 16:49:26 +00:00
dbaron%dbaron.org
d8805141a6
Remove nsTableFrame::GetBorderPadding and related code. b=367458 r=bernd sr=roc
2007-01-27 22:22:24 +00:00
dbaron%dbaron.org
a74c1106d1
Address review comment on bug 9458 that depends on bug 367247 by making nsFrame::IsContainingBlock not check display type, but making it purely frame-class-based.
2007-01-27 18:45:35 +00:00
bzbarsky%mit.edu
baec9bd166
Make mComputedWidth private, with a getter/setter, and make the setter update
...
the resize flags. Bug 367442, r+sr=dbaron
2007-01-26 00:05:12 +00:00
bzbarsky%mit.edu
7413db90f4
Address review comment I missed. Bug 367749.
2007-01-24 20:23:08 +00:00
bzbarsky%mit.edu
27ca5847eb
Fix various bugs in the column iterator's keeping track of the current row.
...
Bug 367749, r=bernd, sr=roc
2007-01-24 20:18:04 +00:00
dbaron%dbaron.org
f6245f0ed9
Change the bulk of ascent (baseline) computation to happen on a virtual method on nsIFrame, and remove the redundant descent member from nsHTMLReflowMetrics, primarily to support having both first-line and last-line baselines. b=367332 r+sr=roc
2007-01-23 04:06:56 +00:00
bzbarsky%mit.edu
ed9ad7b7bf
Make it possible for callers who want to iterate all the row groups and
...
GetMapFor() for all of them do so in about O(N) time instead of O(N^2) (in
number of row groups). Bug 366892, r=bernd, sr=roc
2007-01-22 04:35:25 +00:00
bzbarsky%mit.edu
21d73573a8
Have a faster way of iterating down the columns of a table. Bug 352461,
...
r=bernd, sr=roc
2007-01-22 04:32:33 +00:00
bmlk%gmx.de
97b5794242
Stronger band-aid to prevent table splitting inside columns bug 362275 r/sr=roc
2007-01-21 17:51:33 +00:00
dbaron%dbaron.org
e47180b9d5
Make PushStackMemory never fail to simplify callers, and deCOMtaminate the stack memory API. b=366866 r+sr=bzbarsky
2007-01-17 23:53:47 +00:00
dbaron%dbaron.org
df3f524c0f
When possible, distribute min width of colspanning cells so it fits within the pref width of the spanned columns. b=363144 r=bernd sr=bzbarsky
2007-01-17 19:19:50 +00:00
dbaron%dbaron.org
52d0e2d098
Fix incorrectly placed PR_MAX(), since the span* values are additions to the base values. b=365173 r=bernd sr=bzbarsky
2007-01-13 01:08:18 +00:00
dbaron%dbaron.org
b32b8039f8
Accumulate column-spanning cells into column intrinsic widths separately for each number of columns spanned, in increasing order. b=363329 r=bernd sr=bzbarsky
2007-01-13 01:04:56 +00:00
bmlk%gmx.de
13abe1b209
Zerorowspans are expanded during cellmap entry creation. There is no need to adapt the rowspan value later, bug 364318 r/sr=bzbarsky
2007-01-01 18:18:53 +00:00
bmlk%gmx.de
27757bb08d
initialize variables before use bug 365075 r/sr=bzbarsky
2006-12-27 08:46:30 +00:00
timeless%mozdev.org
4e651390ef
Bug 355300 There should be only one nsGkAtoms creature
...
r=dbaron
2006-12-26 17:47:52 +00:00
bmlk%gmx.de
5f3f2bb2d3
be carefull with row indices when deleting cells bug=364318, patch by bz, r=me sr=roc
2006-12-26 14:24:15 +00:00
bmlk%gmx.de
ea94fabefd
rename variable rename mRowCount to mContentRowCount so that it becomes obvious what it contains. Adjust the cellmap grow mechanism to the new nsTArray world bug=363370 r/sr=bzbarsky
2006-12-26 14:04:20 +00:00
bmlk%gmx.de
0f8c9c56cf
add again bound checks that the BC code relies on. bug 364150 r/sr=bzbarsky
2006-12-26 13:55:32 +00:00
dbaron%dbaron.org
b80d119b57
Untabify.
2006-12-25 19:44:34 +00:00
dbaron%dbaron.org
146c1af6e5
Fix shadowing variable name. No bug.
2006-12-18 22:35:01 +00:00
dbaron%dbaron.org
1e894edf9f
Fix handling of columns in which no cell originates (i.e., unneeded extra colspan). b=363329 Patch from Bernd Mielke, bmlk%gmx.de, with review comments addressed by me. r+sr=dbaron
2006-12-18 22:34:30 +00:00
bzbarsky%mit.edu
6705ad44ff
Remove some unused table code. Bug 359139, r=bernd, sr=roc
2006-12-18 03:04:25 +00:00
mats.palmgren%bredband.net
6ee857d0b7
Make the CSS 'ch' length unit work in most cases. b=363573 r+sr=dbaron
2006-12-17 14:06:42 +00:00
dbaron%dbaron.org
f10ce086e7
Use unsigned value with 1-bit field width, per sicking's advice. No bug.
2006-12-16 02:11:38 +00:00
dbaron%dbaron.org
d5af630c22
Better API for used padding/border/margin: remove deprecated Calc*For on nsStyleStructs and add GetUsed* to nsIFrame. b=332922 r+sr=roc
2006-12-13 23:04:57 +00:00
dbaron%dbaron.org
824d3cb3bf
Fix bustage. b=363524
2006-12-13 04:21:05 +00:00
dbaron%dbaron.org
79275357a7
Reflow all of the table structure (although not its descendants) when frame structure within the table is changed. b=363524 r=bernd sr=bzbarsky
2006-12-13 03:45:28 +00:00
bmlk%gmx.de
52d6f885ce
report frames currently on the overflow list as frames on a additional childlist to get their views reparented if a frame higher in the tree needs to reparent the views bug 362724 r/sr=roc
2006-12-10 06:46:17 +00:00
dbaron%dbaron.org
f05e8007d1
Fix bad interaction of colspan handling with AddMinCoord's magic to deal with specified coord widths. Substantive changes by bzbarsky, comments by me. r+sr=dbaron/bzbarsky b=363150
2006-12-09 05:38:19 +00:00
dbaron%dbaron.org
31f1898810
Bug 300030: Move intrinsic width computation out of nsIFrame::Reflow and into its own methods on nsIFrame. Replace reflow reasons, types, and commands with dirty bits/notifications. Thanks to bzbarsky for almost all of the HTML form controls (mozilla/layout/forms) changes, and many others for help testing and patching. For detailed commit logs, see REFLOW_YYYYMMDD_BRANCH, where YYYYMMDD is one of 20061031, 20060830, 20060603, 20060302, 20060119, 20051011, 20050804, 20050429, 20050315, 20050111, and 20041213.
2006-12-08 05:38:33 +00:00
bzbarsky%mit.edu
99ed82e864
Fix build orange. Bug 356335, reviews hopefully coming up.
2006-12-07 03:43:52 +00:00
bzbarsky%mit.edu
66d9ef86aa
Fix build bustage. See bug 356335 comment 18 for what I forgot to do in the
...
initial checkin for that bug... I need a longer-term memory, clearly. :(
2006-12-07 02:53:35 +00:00
bzbarsky%mit.edu
132cdda426
Allocate the celldata structs from the presshell arena instead of allocating
...
them from the malloc heap with new. Use nsTArray to store them and an nsTArray
to store those arrays instead of using nsVoidArray. Bug 356335, r=bernd,
sr=sicking
2006-12-07 02:32:57 +00:00
bmlk%gmx.de
5bbcbbdb17
colframes are referenced via the firstinflow bug 362348 r/sr=roc
2006-12-02 06:52:55 +00:00
bmlk%gmx.de
d3be21f08f
don't rely on framconstructor to deliver a valid sibling for insert of colgroups and rowgroups as they are siblings contentwise but live on different childlists, bug 337124 r/sr=roc
2006-11-11 13:38:09 +00:00
bmlk%gmx.de
dff951d9ff
mark only cells where the colspan/rowspan offset is != 0 as zero colspans/rowspans if those properties apply bug 358871 r/sr=bzbarsky
2006-11-01 18:45:41 +00:00
bmlk%gmx.de
95696f562c
call the correct GetCellFrame function bug 358679 r/sr=bzbarsky
2006-10-30 19:21:05 +00:00
bmlk%gmx.de
489591b510
make zero rowspans dependent on the content row count and not how many rows are currently in the cellmap, this cause otherwise problems during rebuild of the cellmap bug 339129 r/sr=bzbarsky
2006-10-28 10:14:43 +00:00
bmlk%gmx.de
2722eb51e3
Revert a old design decision how to expand zero colspans in the cellmap. Once a cellmap hole was hit via the GetDataAt function the old code tried to repair this cellmap position by walking up and left till a possible origin of a zero span was hit. The new code uses a mark and sweep approach, which should remove the burden that ordinary tables without zero spans but large cellmap holes have to pay. bug 351942 r/sr=bzbarsky
2006-10-20 07:37:24 +00:00
roc+%cs.cmu.edu
516ce211dd
Bug 241796. Fix comments and handle OOM in SetProperty. Followup to the real fix. r+sr=bzbarsky
2006-09-22 01:56:04 +00:00
bmlk%gmx.de
56f7eba7ce
mark for all reflows the rowgroup as incomplete if there are still nextinflows bug 347367 r/sr=roc
2006-09-21 09:55:07 +00:00
bzbarsky%mit.edu
eb996179db
When painting backgrounds, don't loop over all rows. Just do the ones that
...
might intersect the dirty rect. Use the row cursor to optimize this. Bug
353455, r+sr=roc
2006-09-20 16:42:13 +00:00
roc+%cs.cmu.edu
c5f6e4e0b1
Bug 241796. Use a row cursor to speed up painting and event handling in large tables, similar to what we do for large blocks already. r+sr=bzbarsky
2006-09-19 21:39:33 +00:00
