We need to pull outerFrame one level up to the outer scope because it's
needed when calling FinishBuildingColumns().
The patch is very similar to implementing column-span:all for <fieldset>
in bug 1535200.
Differential Revision: https://phabricator.services.mozilla.com/D24396
After supporting column-span, the top-level container for multicol
layout is no longer the ColumnSetFrame, so remove the "SET" suffix from
FCDATA_ALLOW_GRID_FLEX_COLUMNSET bit to make it sound more general.
Also, update the document and add an assertion.
Differential Revision: https://phabricator.services.mozilla.com/D24394
Per spec, "Note: Column boxes do not become the containing block for
elements with position: fixed or position: absolute. The containing
block is the multicol container, it being the principal box."
https://drafts.csswg.org/css-multicol-1/#the-multi-column-model
contentFrame and contentFrameTop are different only if fieldset has
multicol layout. In that case, contentFrameTop is nsColumnSetFrame (or
ColumnSetWrapperFrame after applying Part 2 with
layout.css.column-span.enabled=true).
Differential Revision: https://phabricator.services.mozilla.com/D23617
The test case triggers MOZ_ASSERT(!IsFramePartOfIBSplit(aParentFrame))
in MaybeRecreateForColumnSpan() because WipeContainingBlock() returns
early when the FrameConstructionItemList is empty. Thus, it doesn't wipe
the aParentFrame even if it's part of an IB split.
An empty FrameConstructionItemList constructs no frames. Therefore,
MaybeRecreateForColumnSpan() doesn't need to do anything if aFrameList is empty
since an empty frame list cannot contain any column-span.
Differential Revision: https://phabricator.services.mozilla.com/D23820
The assertion can be reproduced locally by running
"./mach test dom/canvas/test/chrome/test_drawWindow_widget_layers.html"
with layout.accessiblecaret.enabled=true.
When AccessibleCaret is enabled, caret elements will be inserted into
nsCanvasFrame::mCustomContentContainer, thus it recursively invokes
ConstructFramesFromItemList() to construct frames for carets before it had a
chance to construct popup group.
I feel it's too strict to assume that ConstructFramesFromItemList() cannot be
invoke recursively whenever there's a popup group item. I move the assertion to
the end of ConstructDocElementFrame() to ensure the popup group is processed by
then.
Differential Revision: https://phabricator.services.mozilla.com/D10298
aFrameList can contain placeholder frames. If we decide to nuke
aFrameList, we need to destroy the out-of-flow frames gracefully.
In this case, out-of-flow frames are still in nsFrameConstructorState's
absolute item lists. To rely on nsPlaceholderFrame::DestroyFrom() to
remove its out-of-flow frame properly, we manually flush all the frame
insertions for all the lists in aState before destroying aFrameList.
Differential Revision: https://phabricator.services.mozilla.com/D19460
This is more consistent with what the Rust bits of the style system do, and
removes a pointer from ComputedStyle which is always nice.
This also aligns the Rust bits with the C++ bits re. not treating xul pseudos as
anonymous boxes. See the comment in nsTreeStyleCache.cpp regarding those.
Can't wait for XUL trees to die.
Depends on D19001
Differential Revision: https://phabricator.services.mozilla.com/D19002
Many of the modifications are guarded by #ifdefs. I verify them locally
by manually define them in nsBlockDebugFlags.h and nsLinelayout.cpp.
Note that I replace "mFrame" with "frame" in lines guarded by
NOISY_BLOCK_DIR_MARGINS in nsBlockFrame.cpp because they were
incorrectly renamed in Bug 1277129 Part 6a.
https://hg.mozilla.org/mozilla-central/rev/a70b04f074fc
Differential Revision: https://phabricator.services.mozilla.com/D17733
The crash happens when we try to reparent the absolute/fixed
positioned children to the non-column-span wrapper's absolute list.
When constructing the multicol container, we want it to be the
absolute/fixed position container, not the moz-column-content anonymous
blocks. Hence the modification in AppendFramesToParent() and
ConstructBlock().
Delete AdjustAbsoluteContainingBlock() because we'd like to reparent
absolute/fixed children to non-first continuation of block descendant of
multicol. And it doesn't crash anymore today.
Differential Revision: https://phabricator.services.mozilla.com/D16728
Move CaptureStateForFramesOf into ContentRemoved, so we can traverse frames
which were under display: contents as well.
Differential Revision: https://phabricator.services.mozilla.com/D15319
This commit adds categories to all markers. This way the profiler's
marker categories and frame label categories agree. There are a few
duplicate category properties on some of the marker payloads, but
this could be cleaned up in a follow-up if needed.
Differential Revision: https://phabricator.services.mozilla.com/D16864
There's no real behavior change, since:
* aFlags right now is only used for whitespace suppression optimizations (which
we don't do when display: contents is involved anyway).
* The CouldBeDisplayContents() check is only an optimization: we're
unconditionally recursing into ContentRemoved now, which is not incorrect,
just inefficient.
I found this while reviewing https://phabricator.services.mozilla.com/D15319
(where this _would_ have an impact on behavior).
Differential Revision: https://phabricator.services.mozilla.com/D15373
The elements under <body> are treat as content append, so their frames
will be construct by nsCSSFrameConstructor::ContentAppended.
This patch fixed only the simple "append" case which is appending to the
last continuation of ::moz-column-content. For other more complex
appending or inserting cases, we might need to reframe (bug 1504053).
Differential Revision: https://phabricator.services.mozilla.com/D16076
To reproduce, execute
"GECKO_FRAMECTOR_DEBUG_FLAGS=really-noisy-content-updates ./mach run".
nsIFrame::List()'s second argument should be a const char*, not an
integer. We can fix the bug by omitting it because the default value of
the argument is an empty string.
Differential Revision: https://phabricator.services.mozilla.com/D16075
Summary: Really sorry for the size of the patch. It's mostly automatic
s/nsIDocument/Document/ but I had to fix up in a bunch of places manually to
add the right namespacing and such.
Overall it's not a very interesting patch I think.
nsDocument.cpp turns into Document.cpp, nsIDocument.h into Document.h and
nsIDocumentInlines.h into DocumentInlines.h.
I also changed a bunch of nsCOMPtr usage to RefPtr, but not all of it.
While fixing up some of the bits I also removed some unneeded OwnerDoc() null
checks and such, but I didn't do anything riskier than that.
GetParent() already returns nsContainerFrame*, and it's easier to use
with API like FinishBuildingColumns() demanding a nsContainerFrame*.
Differential Revision: https://phabricator.services.mozilla.com/D15549
This is to prevent Part 1 from breaking
testing/web-platform/tests/css/css-multicol/multicol-span-all-restyle-002.html
After introducing Part 1, we now use CreateContinuingFrame() to create
non-column-span wrapper frames, which use nsFrame::Init() rather than
InitAndRestoreFrame() to initialize them. Because the bits in
aState.mAdditionalStateBits are added to frames only when
nsCSSFrameConstructor::InitAndRestoreFrame() is called. We need a new
way to add the NS_FRAME_PART_OF_IBSPLIT to non-column-span wrapper
frames.
Note that we used to add NS_FRAME_PART_OF_IBSPLIT to both column-span
wrapper and non-column-span wrapper. After this patch, column-span
wrapper won't have NS_FRAME_PART_OF_IBSPLIT. That's fine because
column-span wrappers are not linked with other frames and they
shouldn't get this bit set.
Depends on D15134
Differential Revision: https://phabricator.services.mozilla.com/D15452
In CreateColumnSpanSiblings(), aInitialBlock can be a DetailsFrame.
Therefore, when creating a non-column-span wrapper, we cannot assume
it's always a block frame. We need to use CreateContinuingFrame to
ensure the correct type of continuation is created.
Differential Revision: https://phabricator.services.mozilla.com/D15134
