This patch basically does:
* remove StyleSetHandle and its corresponding files
* revisit #includes of related header files and change correspondingly
* change nsIPresShell::mStyleSet to be UniquePtr<ServoStyleSet>
* change the creating path of ServoStyleSet to pass UniquePtr
* change other mentions of StyleSetHandle to ServoStyleSet*
* remove AsServo() calls on ServoStyleSet
Some unfortunate bits:
* some methods of (Servo)StyleSet only accepts ServoStyleSheet while
many places call into the methods with StyleSheet, so there are many
->AsServo() added to sheets
MozReview-Commit-ID: K4zYnuhOurA
--HG--
extra : rebase_source : 459e8efeb171adad089d94272e143e8c244bd279
extra : source : 65ba2f174fcf7dba4e59c00ee8908b1bd0820a48
Now that what we use to decide whether a document is styled by Servo are only
prefs and the doc principal, we don't need to inherit the style backend type,
since unless the pref has changed, the result will be the same.
MozReview-Commit-ID: KBmeBn1cRne
We need to get rid of BindingHolder to handle properly the case of an invalid
binding URL.
MozReview-Commit-ID: 3sIGtcVOt0r
--HG--
extra : rebase_source : a56cebc74ff5f3102ee8ef6d048ad05056fbef41
They're useless now, provided we remove the hack to not traverse XBL-bound
elements on initial styling.
This also allows us to get rid of the fallback case.
MozReview-Commit-ID: AvBVdyF1wb6
We not only need to care about children getting inserted in the flat tree, but
also about children moving _out_ of the flat tree.
In particular, as of right now we may leave stale data on elements when they
disappear from the flattened tree.
We're lucky enough that in 99% of the situations we enter in[1] and that clears
all the stuff, including servo data. But my assertions for bug 1414999 caught
the template / observes case.
Thus, just clear the whole bound element subtree data, and restyle it in the
end, no need for StyleNewChildren. This matches what we do for shadow DOM
(though in the shadow DOM case we do it async in DestroyFramesForAndRestyle).
[1]: https://searchfox.org/mozilla-central/rev/9bab9dc5a9472e3c163ab279847d2249322c206e/dom/xbl/nsXBLBinding.cpp#368
MozReview-Commit-ID: 69A0aR0AFfU
I'm drive-by removing the comment about the frame tree state because I looked
into it, and the answer is: we properly restore it.
The gotcha is that we retain it too much, indeed, we retain it enough that it
can leak. See bug 1397239.
MozReview-Commit-ID: LP6bXkduEZ4
--HG--
extra : rebase_source : f7e18fc35e48b75c07fcc84b939614d379926828
This is straightforward, with only two notable things.
- `#include "nsXPIDLString.h" is replaced with `#include "nsString.h"`
throughout, because all nsXPIDLString.h did was include nsString.h. The
exception is for files which already include nsString.h, in which case the
patch just removes the nsXPIDLString.h inclusion.
- The patch removes the |xpidl_string| gtest, but improves the |voided| test to
cover some of its ground, e.g. testing Adopt(nullptr).
--HG--
extra : rebase_source : 452cc4a08046a1adb1a8099a7e85a1917de5add8
There's only one case of sync frame construction from ContentRemoved now, and
it's not on the element being removed, but on the whitespace siblings if needed,
and _only_ when they don't support lazy frame construction.
Basically, this switches all the RecreateFramesForContent calls to use
`aAsyncInsert` (which I changed to an enum class for readability), except when
we're already reframing.
Also, it switches ReframeTextIfNeeded to opt-in into lazy frame construction,
since it's used only when aFlags == CONTENT_REMOVED.
This allows to simplify the DestroyFramesFor API (which I'm happy to rename to
something more meaningful, since now it's something like
DestroyFramesForAndRecreateThemAsync), and do some other consistency cleanups.
A bunch of the ContentRemoved callsites were pretty random at passing
aAsyncInsert, and that was some kind of a mess. This patch ensures consistency,
and makes it impossible to do O(n^2) work when removing DOM nodes, which is
nice.
The underlying reason for this is explained in the description of bug 1377848,
and basically allows us to remove a bunch of Servo hacks on the longer term (a
few of them are going away already, yay!).
MozReview-Commit-ID: 2DrUTxGV8RX
--HG--
extra : rebase_source : f428d839a5482477dea22c0fea600d54f3e8799c
The issue here is that the DestroyFramesFor call does a ClearServoDataFromSubtree,
and then we subsequently fail to load the bindings, leaving ourselves with an unstyled
subtree that isn't marked with descendant bits and isn't rooted at a display:none element,
which is forbidden. This can trigger crashes when we call the innerText getter on one of
the unstyled elements, since that will first flush layout (which won't find the unstyled
subtree), and then invoke IsOrHasAncestorWithDisplayNone, which passes LazyComputeBehavior::Assert.
MozReview-Commit-ID: 7roBkH9fTGZ
We'll need the bound element to get PresContext in a later patch.
MozReview-Commit-ID: 2Gcb0yKgbTH
--HG--
extra : rebase_source : 6549eb726c32e25bea6045027805492984521d48
The <content> in an XBL document could be injected into documents with
different style backend types. Therefore, we need to set the XBL document's
style backend to the same as the bound document's so that the style
attribute of the content can be processed by the correct backend. <marquee>
elements in xbl-marquee.xml is one such example.
MozReview-Commit-ID: 7M33zlbZqNF
--HG--
extra : rebase_source : 363fd7c2bc91e959fbcfb288bdc1b65cfd1bba97
We assume those XBL documents without a bound document use gecko style
backend. Their backend type will be explicitly set when loading in Part 6.
MozReview-Commit-ID: 5c5xtYFvgAD
--HG--
extra : rebase_source : 578b50953d98ffa5280dac3440979e2aba7519cb
This function is an infallible alternative to nsIURI::GetSpec(). It's useful
when it's appropriate to handle a GetSpec() failure with a failure string, e.g.
for log/warning/error messages. It allows code like this:
nsAutoCString spec;
uri->GetSpec(spec);
printf("uri: %s", spec.get());
to be changed to this:
printf("uri: %s", uri->GetSpecOrDefault().get());
This introduces a slight behavioural change. Previously, if GetSpec() failed,
an empty string would be used here. Now, "[nsIURI::GetSpec failed]" will be
produced instead. In most cases this failure string will make for a clearer
log/warning/error message than the empty string.
* * *
Bug 1297961 (part 1b) - More GetSpecOrDefault() additions. r=hurley.
I will fold this into part 1 before landing.
--HG--
extra : rebase_source : ddc19a5624354ac098be019ca13cc24b99b80ddc
This is kind of a long story, stay with me on this.
In bug 990290, a WeakMap was added to any JS scope that loaded an XBL
binding. That WeakMap stored the JS prototypes that are injected into
a bound node's prototype chain.
When a binding is removed, we search the prototype chain for the
JS prototype that we'd added, and remove it.
The XUL prototype cache caches numerous things, including nsXBLDocumentInfo,
which we use to get at the nsXBLPrototypeBinding for a particular binding,
which is then used to generate the class object that's put into the WeakMap.
When the XUL prototype cache is flushed, that means that when a binding
is bound, its definition needs to be reloaded off of the disk. If, however,
there was a pre-existing instance of the binding already being used in a
document, now we were in a funny case.
We were in a funny case, because when attempting to remove a binding, we
would look up the nsXBLPrototypeBinding via the nsXBLDocumentInfo that's
being held within the nsXULPrototypeCache, find (or load off the disk) a
_new_ nsXBLDocumentInfo and generate a _new_ nsXBLPrototypeBinding that
did not match to the one that we'd already stored in the WeakMap. This
would mean that removal would go wrong, and things would break horribly.
This patch makes it so that we prioritize checking the nsBindingManager
for a document for the nsXBLDocumentInfo before checking the
global nsXULPrototypeCache. That way, even if the cache gets cleared,
if the binding was ever used in this document, it'll be in the
nsBindingManager, and we'll get the same nsXULProtoypeBinding that
we'd been using before, and sanity will prevail.
MozReview-Commit-ID: G8iLDbCPRAC
--HG--
extra : rebase_source : 4322965c0b7150b22454651ad7a9461ee76d766b
Chris Peterson
Boris Zbarsky
Adrian Wielgosik
Emilio Cobos Álvarez
Xidorn Quan
Andrea Marchesini
Brindusan Cristian
Dave Townsend
Nicholas Nethercote
Jonathan Watt
Bobby Holley
Cameron McCormack
Ting-Yu Lin
Phil Ringnalda
Andrew McCreight
Mike Conley
Masayuki Nakano