Here is the patched build's log:
[Main Thread]: I/nsGtkIMModuleWidgets GTKIM: 7fab5a60a2c0 CreateTextRangeArray(aContext=7fab5a7bbbf0, aCompositionString="÷" (Length()=1))
[Main Thread]: W/nsGtkIMModuleWidgets GTKIM: 7fab5a60a2c0 SetTextRange(), FAILED, due to no attr, aTextRange= { mStartOffset=0, mEndOffset=1 }
[Main Thread]: W/nsGtkIMModuleWidgets GTKIM: 7fab5a60a2c0 SetTextRange(), FAILED, due to current clause length is 0
[Main Thread]: E/nsGtkIMModuleWidgets GTKIM: 7fab5a60a2c0 SetTextRange(), FAILED, due to g_utf8_to_utf16() failure (retrieving current clause)
[Main Thread]: W/nsGtkIMModuleWidgets GTKIM: 7fab5a60a2c0 CreateTextRangeArray(), inserting a dummy clause at the beginning of the composition string mStartOffset=0, mEndOffset=1, mRangeType=TextRangeType::eRawClause
iBus Chewing IME has two clauses when user presses Shift+p, one doesn't have pango_attr, the other is empty. These clauses are not useful in Gecko. Additionally, TextRangeArray assumes that there is a clause at beginning of the composition when there is one or more clauses. Therefore, this patch tries to insert dummy clause at the beggining of composition in such case.
MozReview-Commit-ID: 3hVGVmvFrhA
--HG--
extra : rebase_source : edbd3a6a1139cffb0d5bfbe0c92bf6870c9a2608
I think that we can drop nsIMEUpdatePreference::DontNotifyChangesCausedByComposition(), i.e., nsIMEUpdatePreference::NOTIFY_CHANGES_CAUSED_BY_COMPOSITION because it's now used only by TSFTextStore but TSFTextStore ignores if SelectionChangeDataBase::mCausedByComposition or TextChangeDataBase::mCausedOnlyByComposition is true (for supporting async changes in e10s mode). So, only issue is, dropping the flag might cause increasing computing TextChangeData cost during composition in TSF mode. However, now, it's already enough fast and even if it'd cause performance regression, we could add a hack with TextComposition's offset information. Therefore, we don't need to worry about the performance regression so seriously.
MozReview-Commit-ID: HNT3G4isONj
--HG--
extra : rebase_source : 164231023aa2a17ceab94d92fb49ba0a00dab429
Currently, all widgets request selection change notifications to IMEContentObserver. Additionally, IMEContentObserver needs to listen selection changes for caching latest selection for eQuerySelectedText. Therefore, it doesn't make sense to keep defining nsIMEUpdatePreference::NOTIFY_SELECTION_CHANGE.
If widgets didn't need selection change notifications, they could just ignore the unnecessary notifications.
Note that all widgets don't need selection change notifications if a plugin has focus and IMEContentObserver cannot observe selection changes in the plugin. Therefore, if IMEContentObserver is initialized with a plugin, it shouldn't listen selection changes (and doesn't need to notify widgets of selection changes).
MozReview-Commit-ID: FOVFFgA2nOz
--HG--
extra : rebase_source : 3e16d5023835f99f82934e754d2e7db70474f9ee
To optimize copy of text rect array, we should use mozilla::Move. Also, after using it, we should mark is invalid result into SetEventResult
MozReview-Commit-ID: HH9H7DhK12
--HG--
extra : rebase_source : 4e02ece73583306c386597bc92e203fa147cfcbc
extra : histedit_source : 726621cdaf262b9d173ae19d505575f10563cc36
Selection cache might be updated on commit string, so it should be updated correctly. Also, IMContextWrapper::OnSelectionChange() already has the check for composing using IsComposing().
MozReview-Commit-ID: 2n3f3I3aAjg
--HG--
extra : rebase_source : 5bd43d723a3170864dd7ac05c9101aaedd7dfac6
Use new event to update cotent process's cache.
MozReview-Commit-ID: CexTXW4knMQ
--HG--
extra : rebase_source : 525a8b0a1fe6ef533e06f76ac5156e264b158c70
extra : histedit_source : 2a8c46a64eb5c6a7a05adc28917422c488b2b1c4
It will use on ContentCache. Also, SetRangeFromFlatTextOffset issue will hanle on another bug.
MozReview-Commit-ID: 9Yu8bLlcZS5
--HG--
extra : rebase_source : c8eba70bdedf303b0fa649c1a609fe6120983e65
extra : histedit_source : 4189807b57ede13c1546e95052fcc298c8581d06
IMContextWrapper::OnDeleteSurroundingNative should output aOutput and aNChar correctly.
MozReview-Commit-ID: 46UrnGajnTG
--HG--
extra : rebase_source : 7ed7d3f389d492117e835c071bbea00f7ce4ff8d
TextInputHandler may dispatch keypress events after InsertText() is called if there was composition and it's committed by "current" keydown event. In that case, [NSEvent characters] may have the committing string. For example, when Opt+e of US keyboard layout started composition, Cmd+v causes committing the "`" character and pasting the clipboard text. Then, the "v" key's keydown event's |characters| is "`v". So, after InsertText() is called with "`", TextInputHandler shouldn't dispatch keypress event for "`" again. I.e., the KeyboardEvent.key value should be "v" rather than "`v".
For solving this issue, TextInputHandlerBase::AutoInsertStringClearer which is created at every InsertText() call should store the inserted string to TextInputHandlerBase::KeyEventState. However, for making the implemntation simpler, it should recode only when the inserting string is actually a part of [mKeyEvent characters]. Then, TextInputHandlerBase::KeyEventState can compute unhandled insert string at initializing WidgetKeyboardEvent.
So, finally, TextInputHandlerBase::InitKeyEvent() should be called via TextInputHandlerBase::KeyEventState::InitKeyEvent(). This ensures that all key events which may cause InsertText() calls are always initialized with unhandled string.
MozReview-Commit-ID: A9o8o9pV2XV
--HG--
extra : rebase_source : d99e9ab7633f45069d0a3940cc2b2b5ad859ea05
(Based on original patch by Chris Coulson.)
Replace session management through libgnome with direct use of libSM and
libICE. This allows xsmp session restore on gtk2 and gtk3 builds.
Changes in behaviour:
- It now only sends the "session-save" notification when the save style
is SmSaveLocal or SmSaveBoth. Saving internal state with a save style of
SmSaveGlobal is actually incorrect. This means that Firefox now
distinguishes between a normal session exit and a session exit with
session saving enabled.
- As "quit-application-requested" might pop up a dialog, it only does
this if the interact style is not SmInteractStyleNone
- "quit-application-requested" is only sent after sending
SmcInteractRequest and receiving an interact message.
- It defers closing the connection to the session manager until
the destructor, i.e after the "web-workers-shutdown" async shutdown
phase completes. This means that firefox shouldn't be killed too early
and lose data (*).
* It still might be killed prematurely if it takes too long to quit and
the session manager decides to timeout, but that's another story.
TextComposition should update its composition start offset after every DOM event dispatch with first clause's selection type if there is composition string.
MozReview-Commit-ID: HFkePci1PhU
--HG--
extra : rebase_source : c7af075115505b52e48d8bbd5f2dd1d9001fe36e
TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText.
First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor.
Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal.
Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases.
MozReview-Commit-ID: 9WzUx8b5piw
--HG--
extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
In an expression such as:
const auto& x = cond() ? AClass(...) : AClass();
the C++ standard specifies that the copy constructor of AClass is
invoked on the result of the conditional expression ([expr.cond]p6).
GCC does not honor this part of the specification, whereas clang does;
clang therefore complains about instances of code such as:
const auto& jstrVal = type == widget::PrefsHelper::PREF_STRING ?
jni::StringParam(strVal, aPrefName.Env()) :
jni::StringParam(nullptr);
as jni::StringParam is not copy-constructable.
The simplest solution that does not introduce unnecessary allocation
uses mozilla::Maybe to hold the temporary objects and to hide some of
the details of constructing objects in-place. The compiler may even be
able to optimize away some of the unnnecessary checks that Maybe
introduces (e.g. checking for whether the Maybe is a Some or None at
certain points).
