mCompositorVsyncDispatcher is created and destroyed on main thread. But it is
accessed from Compositor thread or VSyncBridge thread. To avoid race condition
between main thread and Compositor thread/VSyncBridge thread,
mCompositorVsyncDispatcher needs to be protected by lock.
Mostly just convert nsInt{Rect,Region} to LayoutDeviceInt{Rect,Region}.
One exception is to change the parameter of nsWindow::OnResize from
nsIntRect to LayoutDeviceIntSize, because it really only needs that.
MozReview-Commit-ID: 963Mzd5Wed6
--HG--
extra : rebase_source : efae039d26feb98a3591a7cd46c30ce4f052d70b
I accidentally broke the ability to retrieve a big string from the
clipboard, and there was no test that failed. So this provides a new
test that does the following:
1. Store a big string in a nsTransferable.
2. Copy it to the clipboard.
3. Create a new nsTransferable, initialize with small data.
4. Populate the nsTransferable with (big) data from the clipboard.
5. Populate the nsTransferable with small data.
After each step, the test checks whether the transferable holds the
expected data and length, and (on non-Windows) checks that the big
data is backed by a file, and small data is not.
MozReview-Commit-ID: 9yuXZxVqD6R
--HG--
extra : rebase_source : 6cec638c59e8ce5a18adbb642779c1dcd457cc5b
DataStruct cannot safely be copied if its mCacheFD member is set.
Currently there is no code for this case, but to avoid problems later,
mark the copy and assignment constructors private and delete them.
A move-constructor was added to compensate for the deleted copy
constructor. nsTransferable::AddDataFlavor uses this new constructor
instead of the previous implicit default copy constructor.
MozReview-Commit-ID: 3N5xjFXOUKB
--HG--
extra : rebase_source : dc609f20c7048b3630fa99fe2deef3f5be155334
- Count open file descriptors instead of testing the existence of a file
(because the clipboard is now only reachable through file descriptors,
and not through a file path).
- Use a fixed string instead of a random string. The previous way of
generating a string was non-deterministic, and there was a very small
chance that the generated string was not large enough to trigger the
cache-to-disk-mode.
- Use "text/unicode" instead of "text/plain", because JavaScript strings
use two bytes, not one bytes each.
- The cache file is already created when the Transferable is created, so
check the cache file after assigning data to the nsITransferable, but
before copying it to the clipboard.
MozReview-Commit-ID: KOkYOm280Oh
--HG--
extra : rebase_source : 3515ff55a316720b87f2473a1450e5d8304728e8
Ensure that only DataStruct::mData + mDataLen, XOR
DataStruct::mCacheFD is used.
(Previously it was possible that all of these members were populated,
which is a waste of memory.)
The effect of this change is visible when SetTransferData is called
multiple times with the same flavor, but with one below the threshold
for storing in-memory, and the other above (=store in a file).
MozReview-Commit-ID: 4UlkKAYsjf
--HG--
extra : rebase_source : 75398f84f039bdf2ff0195342f3db8efbdce6d3d
The cache file is never directly exposed to consumers of DataStruct,
so it does not make sense to keep the clipboardcache file around
forever.
The only change in this commit is to switch from using a filename to
using a file descriptor. In the destructor, the FD is explicitly closed
(which releases the file data). nsAnonymousTemporaryFile takes care
of removing the file when the destructor is not called (e.g. crashes).
Previously, the clipboard cache was stored in a file called:
TmpD/clipboardcache-N
As of this commit, the clipboard cache is stored at:
TmpD/mozilla-temp-randomN (macOS and Linux)
TmpD/mozilla-temp-files/mozilla-temp-randomN (Windows)
(see xpcom/io/nsAnonymousTemporaryFile.{h,cpp} for more details)
To verify that these files are really gone:
1. Create a document with 500k+ characters, open it in Firefox.
2. Copy its content - this will trigger the clipboard cache.
3. Look for the open file descriptor of the deleted file:
( macOS and Linux: )
lsof +L1 | grep mozilla-temp
4. Copy anything (under the 500k threshold), or quit/kill Firefox.
5. Repeat step 3 and observe that the number of file descriptors
has decreased.
MozReview-Commit-ID: 85GlKQrNUl5
--HG--
extra : rebase_source : 6937143639d6a6280ffe8b53b4c2fa4b1e7ef55d
We have stopped dispatching "keypress" events for non-printable keys
and key combinations for conforming to UI Events and following the
other browsers.
However, this change hits a serious bugs of Google Docs, Google
Spreadsheets and Gmail. Until they will fix their bugs, we should
take back the traditional behavior for keeping Nightly usable for
any Nightly testers.
MozReview-Commit-ID: 9CyEbsFit1S
--HG--
extra : rebase_source : 837288b1fb53121badff4e65094a87cebfe3cfee
Some themes (Ambiance for instance) uses first-child/last-child css selectors
to style titlebar buttons. Ubuntu Ambiance theme places titlebar buttons closer
by negative margin applied to them.
We put titlebar buttons to GtkBox as well as Gtk+ does and also keep
the button order here to match first-child/last-child selectors. It also means
we must have maximize/restore as one button to keep the correct order.
MozReview-Commit-ID: 9mqljOa4Vu7
--HG--
extra : rebase_source : 9c31a2073d1bb247ce9d0240333143661b8ae4b8
The titlebar rendering on Linux/Gtk+ is recently enabled at Beta59 but with many bugs fixed at Nightly.
Let's disable this feature for Beta / Release 59 and ship it at Firefox 60 where majority of the issues are fixed.
MozReview-Commit-ID: FQL7tNhcvUG
--HG--
extra : rebase_source : 03a495ebc81b5ae5b4093406bd40daf83251343a
This a remaining issue of bug 1435180. We need to skip keypress event check of
non-printable key operation in the test.
MozReview-Commit-ID: InobZCbrzWL
--HG--
extra : rebase_source : 9f17cd04500f4ef31c3674087188ec6ee157173b
UI Events declares that keypress event should be fired only when the keydown
sequence produces some characters. For conforming to UI Events and
compatibility with the other browsers, we should stop dispatching keypress
events for non-printable keys.
For getting regression reports, we should enable this new behavior only
on Nightly and early Beta.
MozReview-Commit-ID: 5IIL9huejXH
--HG--
extra : rebase_source : 0abdbe84a50d6fd1b4d52521b92e7513483b197c
https://bugzilla.mozilla.org/show_bug.cgi?id=1418749
This adds support for download progress reporting via the XApp
method currently used in the Cinnamon desktop, by establishing a new
X11 window property to be supported/read by the window manager.
See https://github.com/linuxmint/xapps/blob/master/libxapp/xapp-gtk-window.c,
as well as https://github.com/linuxmint/muffin/commit/39045da0ea06f
for more details.
The property-setting code lives in nsWindow - it's a small and stable
enough chunk that it made more sense to do this than actually depend on
another external library. As nsWindow is already using x11 calls, this
seemed the safest place for it, without affecting the build.
The TaskbarProgress instance is initialized via the DownloadsTaskbar
js module, and is handed a pointer to the current main window to call
SetProgress on. Most of the javascript side of this is in line with
how the other platforms are handled.
Without a supporting window manager/desktop environment (currently just
Cinnamon/Muffin 3.6,) the simplest way to observe working behavior is
by calling 'xprop -spy' on the browser window being testing and watching
for updates during a download.
--HG--
extra : rebase_source : 92cbe495ce097b080a9f94ec5312169da9a8648d
We still get occasional messages for the internal OLE Main Thread window.
Also, the PeekMessage call allows internal windows messages to be processed for
things like GDI.
GtkHeaderBar has property "spacing" which defines space between buttons at titlebar.
Get this property and apply as margin to titlebar buttons when there's more than one.
Also cache this value for furter use at titlebar metrics cache.
MozReview-Commit-ID: J7qAIWEnK4Y
--HG--
extra : rebase_source : 73f0f605444cf4e4ada3da335ff003e9df05c9c4
The DispatchEvent can manipulate with the mRefPoint we're later using to check if the
double click happened on the titlebar. We need to save it for later check to avoid
unwanted restore/maximize event when mouse event occurs near top border of any widget.
Also don't handle doubleclick on titlebar when CSD is not enabled.
MozReview-Commit-ID: KjxM1EsT4Lg
--HG--
extra : rebase_source : b880e4d89ebe3546b7ef70e3d94926a628f98598
A lot of "WARNING: Invalid monitor scale: -1" appering for the puppet widget since
fix for bug 1439881 landed. We don't need to print the warning, fallback to 1 is
sufficient enough.
MozReview-Commit-ID: 73BGc8neUmu
--HG--
extra : rebase_source : 74b2a818a29514095b5c56b05966cff5e4e9ba60
This is particularly useful for knowing when it's safe to query for style and
layout information for a window without causing a synchronous style or layout
flush.
Note that promiseDocumentFlushed was chosen over promiseDidRefresh or promiseRefreshed
to avoid potential confusion with the actual network-level refresh of browsers or
documents.
MozReview-Commit-ID: Am3G9yvSgdN
--HG--
extra : rebase_source : 20bdd2d6f624767d919d95a6601fc1c890aadf10
This is particularly useful for knowing when it's safe to query for style and
layout information for a window without causing a synchronous style or layout
flush.
Note that promiseDocumentFlushed was chosen over promiseDidRefresh or promiseRefreshed
to avoid potential confusion with the actual network-level refresh of browsers or
documents.
MozReview-Commit-ID: Am3G9yvSgdN
--HG--
extra : rebase_source : 20bdd2d6f624767d919d95a6601fc1c890aadf10
This gtest assumed the current working directory ($OBJDIR/_tests/gtest/) was in the DLL LoadLibrary path. That is no longer true with this bug's patches, so the gtest must specify the absolute path to the PDFium DLL to load it.
MozReview-Commit-ID: 5TXj6A9Tb9w
--HG--
extra : rebase_source : b097ae0c3430767e9ff16632ee96fd8739a902bf
extra : histedit_source : 1f13fa04c29039f8aeae05028d0a20ede5ee794b
Because of rounding errors there's a change that returned monitor scale
is 0. That would lead to SIGFPE because it's later used in division.
MozReview-Commit-ID: 4d7nHaBm4XG
--HG--
extra : rebase_source : d7fb5c64698d1e99afc223d401b30e0eb168fdb4
Windows default behavior will trigger the System menu bar when F10 is released.
Among other things, this causes the System menu bar to appear when a web page overrides the contextmenu event.
We *never* want this default behavior, so eat this key (never pass it to Windows).
MozReview-Commit-ID: 4fWOuj4mWvW
--HG--
extra : rebase_source : 45d38ba19b2f23aca74a60819c7d93af2ffa167d
Although we haven't any bug reports caused by this, this is a really old bug.
When we implement TSFTextStore, we decided to use queue of dispatching
events and flush it when document lock is unlocked. When we implement the
queue, we got this regression.
When TSFTextStore::SetText() is called with different range from current
selection range, TSFTextStore::SetSelectionInternal() add
PendingAction::SET_SELECTION into the queue first for replacing existing
text or inserting text into different position if there is no composition.
Then, TSFTextStore::InsertTextAtSelectionInternal() inserts text at the new
selection range.
When TSFTextStore::FlushPendingActions() is called after that, eSetSelection
should be dispatched and then, new text is inserted wit a set of composition
events. However, we forgot to dispatch creating eSetSelection event.
So, this patch just dispatches the event.
MozReview-Commit-ID: Hw8FTB1R5kR
--HG--
extra : rebase_source : 8a119f1f48b167d9423fc89ce0efc722313c3732
