Bug 1318137 - Added charLength to speech synth boundary events. r?m_kato draft
authorEitan Isaacson <eitan@monotonous.org>
Wed, 16 Nov 2016 17:22:36 -0800
changeset 441632 3f7b7251755ca80834ebab24fd33f78ec80550a8
parent 439734 51750761f2c61c64cf0553f6cb5fefd4999d3bc0
child 537594 583948b149fa011fad378aec8c432c4d2a679f35
push id36471
push userbmo:eitan@monotonous.org
push dateSun, 20 Nov 2016 06:20:13 +0000
reviewersm_kato
bugs1318137
milestone53.0a1
Bug 1318137 - Added charLength to speech synth boundary events. r?m_kato MozReview-Commit-ID: 2B9gM3wmEVQ
dom/media/webspeech/synth/SpeechSynthesisUtterance.cpp
dom/media/webspeech/synth/SpeechSynthesisUtterance.h
dom/media/webspeech/synth/cocoa/OSXSpeechSynthesizerService.mm
dom/media/webspeech/synth/ipc/PSpeechSynthesisRequest.ipdl
dom/media/webspeech/synth/ipc/SpeechSynthesisChild.cpp
dom/media/webspeech/synth/ipc/SpeechSynthesisChild.h
dom/media/webspeech/synth/ipc/SpeechSynthesisParent.cpp
dom/media/webspeech/synth/ipc/SpeechSynthesisParent.h
dom/media/webspeech/synth/nsISpeechService.idl
dom/media/webspeech/synth/nsSpeechTask.cpp
dom/media/webspeech/synth/nsSpeechTask.h
dom/media/webspeech/synth/windows/SapiService.cpp
dom/webidl/SpeechSynthesisEvent.webidl
--- a/dom/media/webspeech/synth/SpeechSynthesisUtterance.cpp
+++ b/dom/media/webspeech/synth/SpeechSynthesisUtterance.cpp
@@ -153,24 +153,26 @@ void
 SpeechSynthesisUtterance::GetChosenVoiceURI(nsString& aResult) const
 {
   aResult = mChosenVoiceURI;
 }
 
 void
 SpeechSynthesisUtterance::DispatchSpeechSynthesisEvent(const nsAString& aEventType,
                                                        uint32_t aCharIndex,
+                                                       Nullable<uint32_t> aCharLength,
                                                        float aElapsedTime,
                                                        const nsAString& aName)
 {
   SpeechSynthesisEventInit init;
   init.mBubbles = false;
   init.mCancelable = false;
   init.mUtterance = this;
   init.mCharIndex = aCharIndex;
+  init.mCharLength = aCharLength;
   init.mElapsedTime = aElapsedTime;
   init.mName = aName;
 
   RefPtr<SpeechSynthesisEvent> event =
     SpeechSynthesisEvent::Constructor(this, aEventType, init);
   DispatchTrustedEvent(event);
 }
 
--- a/dom/media/webspeech/synth/SpeechSynthesisUtterance.h
+++ b/dom/media/webspeech/synth/SpeechSynthesisUtterance.h
@@ -92,16 +92,17 @@ public:
   IMPL_EVENT_HANDLER(mark)
   IMPL_EVENT_HANDLER(boundary)
 
 private:
   virtual ~SpeechSynthesisUtterance();
 
   void DispatchSpeechSynthesisEvent(const nsAString& aEventType,
                                     uint32_t aCharIndex,
+                                    Nullable<uint32_t> aCharLength,
                                     float aElapsedTime, const nsAString& aName);
 
   nsString mText;
 
   nsString mLang;
 
   float mVolume;
 
--- a/dom/media/webspeech/synth/cocoa/OSXSpeechSynthesizerService.mm
+++ b/dom/media/webspeech/synth/cocoa/OSXSpeechSynthesizerService.mm
@@ -38,17 +38,17 @@ public:
     mStartingTime = TimeStamp::Now();
   }
 
   NS_DECL_CYCLE_COLLECTING_ISUPPORTS
   NS_DECL_CYCLE_COLLECTION_CLASS_AMBIGUOUS(SpeechTaskCallback, nsISpeechTaskCallback)
 
   NS_DECL_NSISPEECHTASKCALLBACK
 
-  void OnWillSpeakWord(uint32_t aIndex);
+  void OnWillSpeakWord(uint32_t aIndex, uint32_t aLength);
   void OnError(uint32_t aIndex);
   void OnDidFinishSpeaking();
 
 private:
   virtual ~SpeechTaskCallback()
   {
     [mSpeechSynthesizer release];
   }
@@ -132,24 +132,25 @@ SpeechTaskCallback::OnVolumeChanged(floa
 float
 SpeechTaskCallback::GetTimeDurationFromStart()
 {
   TimeDuration duration = TimeStamp::Now() - mStartingTime;
   return duration.ToMilliseconds();
 }
 
 void
-SpeechTaskCallback::OnWillSpeakWord(uint32_t aIndex)
+SpeechTaskCallback::OnWillSpeakWord(uint32_t aIndex, uint32_t aLength)
 {
   mCurrentIndex = aIndex < mOffsets.Length() ? mOffsets[aIndex] : mCurrentIndex;
   if (!mTask) {
     return;
   }
   mTask->DispatchBoundary(NS_LITERAL_STRING("word"),
-                          GetTimeDurationFromStart(), mCurrentIndex);
+                          GetTimeDurationFromStart(),
+                          mCurrentIndex, aLength, 1);
 }
 
 void
 SpeechTaskCallback::OnError(uint32_t aIndex)
 {
   if (!mTask) {
     return;
   }
@@ -180,17 +181,17 @@ SpeechTaskCallback::OnDidFinishSpeaking(
   [super init];
   mCallback = aCallback;
   return self;
 }
 
 - (void)speechSynthesizer:(NSSpeechSynthesizer *)aSender
             willSpeakWord:(NSRange)aRange ofString:(NSString*)aString
 {
-  mCallback->OnWillSpeakWord(aRange.location);
+  mCallback->OnWillSpeakWord(aRange.location, aRange.length);
 }
 
 - (void)speechSynthesizer:(NSSpeechSynthesizer *)aSender
         didFinishSpeaking:(BOOL)aFinishedSpeaking
 {
   mCallback->OnDidFinishSpeaking();
 }
 
--- a/dom/media/webspeech/synth/ipc/PSpeechSynthesisRequest.ipdl
+++ b/dom/media/webspeech/synth/ipc/PSpeechSynthesisRequest.ipdl
@@ -32,15 +32,16 @@ async protocol PSpeechSynthesisRequest
   async OnEnd(bool aIsError, float aElapsedTime, uint32_t aCharIndex);
 
   async OnStart(nsString aUri);
 
   async OnPause(float aElapsedTime, uint32_t aCharIndex);
 
   async OnResume(float aElapsedTime, uint32_t aCharIndex);
 
-  async OnBoundary(nsString aName, float aElapsedTime, uint32_t aCharIndex);
+  async OnBoundary(nsString aName, float aElapsedTime, uint32_t aCharIndex,
+                   uint32_t aCharLength, uint8_t argc);
 
   async OnMark(nsString aName, float aElapsedTime, uint32_t aCharIndex);
 };
 
 } // namespace dom
 } // namespace mozilla
--- a/dom/media/webspeech/synth/ipc/SpeechSynthesisChild.cpp
+++ b/dom/media/webspeech/synth/ipc/SpeechSynthesisChild.cpp
@@ -126,19 +126,21 @@ SpeechSynthesisRequestChild::RecvOnResum
 {
   mTask->DispatchResumeImpl(aElapsedTime, aCharIndex);
   return IPC_OK();
 }
 
 mozilla::ipc::IPCResult
 SpeechSynthesisRequestChild::RecvOnBoundary(const nsString& aName,
                                             const float& aElapsedTime,
-                                            const uint32_t& aCharIndex)
+                                            const uint32_t& aCharIndex,
+                                            const uint32_t& aCharLength,
+                                            const uint8_t& argc)
 {
-  mTask->DispatchBoundaryImpl(aName, aElapsedTime, aCharIndex);
+  mTask->DispatchBoundaryImpl(aName, aElapsedTime, aCharIndex, aCharLength, argc);
   return IPC_OK();
 }
 
 mozilla::ipc::IPCResult
 SpeechSynthesisRequestChild::RecvOnMark(const nsString& aName,
                                         const float& aElapsedTime,
                                         const uint32_t& aCharIndex)
 {
--- a/dom/media/webspeech/synth/ipc/SpeechSynthesisChild.h
+++ b/dom/media/webspeech/synth/ipc/SpeechSynthesisChild.h
@@ -58,17 +58,19 @@ protected:
                                     const float& aElapsedTime,
                                     const uint32_t& aCharIndex) override;
 
   mozilla::ipc::IPCResult RecvOnPause(const float& aElapsedTime, const uint32_t& aCharIndex) override;
 
   mozilla::ipc::IPCResult RecvOnResume(const float& aElapsedTime, const uint32_t& aCharIndex) override;
 
   mozilla::ipc::IPCResult RecvOnBoundary(const nsString& aName, const float& aElapsedTime,
-                                         const uint32_t& aCharIndex) override;
+                                         const uint32_t& aCharIndex,
+                                         const uint32_t& aCharLength,
+                                         const uint8_t& argc) override;
 
   mozilla::ipc::IPCResult RecvOnMark(const nsString& aName, const float& aElapsedTime,
                                      const uint32_t& aCharIndex) override;
 
   RefPtr<SpeechTaskChild> mTask;
 };
 
 class SpeechTaskChild : public nsSpeechTask
--- a/dom/media/webspeech/synth/ipc/SpeechSynthesisParent.cpp
+++ b/dom/media/webspeech/synth/ipc/SpeechSynthesisParent.cpp
@@ -203,20 +203,22 @@ SpeechTaskParent::DispatchErrorImpl(floa
     return NS_ERROR_FAILURE;
   }
 
   return NS_OK;
 }
 
 nsresult
 SpeechTaskParent::DispatchBoundaryImpl(const nsAString& aName,
-                                       float aElapsedTime, uint32_t aCharIndex)
+                                       float aElapsedTime, uint32_t aCharIndex,
+                                       uint32_t aCharLength, uint8_t argc)
 {
   MOZ_ASSERT(mActor);
-  if(NS_WARN_IF(!(mActor->SendOnBoundary(nsString(aName), aElapsedTime, aCharIndex)))) {
+  if(NS_WARN_IF(!(mActor->SendOnBoundary(nsString(aName), aElapsedTime,
+                                         aCharIndex, aCharLength, argc)))) {
     return NS_ERROR_FAILURE;
   }
 
   return NS_OK;
 }
 
 nsresult
 SpeechTaskParent::DispatchMarkImpl(const nsAString& aName,
--- a/dom/media/webspeech/synth/ipc/SpeechSynthesisParent.h
+++ b/dom/media/webspeech/synth/ipc/SpeechSynthesisParent.h
@@ -88,17 +88,18 @@ public:
 
   nsresult DispatchPauseImpl(float aElapsedTime, uint32_t aCharIndex);
 
   nsresult DispatchResumeImpl(float aElapsedTime, uint32_t aCharIndex);
 
   nsresult DispatchErrorImpl(float aElapsedTime, uint32_t aCharIndex);
 
   nsresult DispatchBoundaryImpl(const nsAString& aName,
-                                float aElapsedTime, uint32_t aCharIndex);
+                                float aElapsedTime, uint32_t aCharIndex,
+                                uint32_t aCharLength, uint8_t argc);
 
   nsresult DispatchMarkImpl(const nsAString& aName,
                             float aElapsedTime, uint32_t aCharIndex);
 
 private:
   SpeechSynthesisRequestParent* mActor;
 };
 
--- a/dom/media/webspeech/synth/nsISpeechService.idl
+++ b/dom/media/webspeech/synth/nsISpeechService.idl
@@ -108,19 +108,21 @@ interface nsISpeechTask : nsISupports
   void dispatchError(in float aElapsedTime, in unsigned long aCharIndex);
 
   /**
    * Dispatch boundary event.
    *
    * @param aName        name of boundary, 'word' or 'sentence'
    * @param aElapsedTime time in seconds since speech has started.
    * @param aCharIndex   offset of spoken characters.
+   * @param aCharLength  length of text in boundary event to be spoken.
    */
-  void dispatchBoundary(in DOMString aName, in float aElapsedTime,
-                        in unsigned long aCharIndex);
+  [optional_argc] void dispatchBoundary(in DOMString aName, in float aElapsedTime,
+                                        in unsigned long aCharIndex,
+                                        [optional] in unsigned long aCharLength);
 
   /**
    * Dispatch mark event.
    *
    * @param aName        mark identifier.
    * @param aElapsedTime time in seconds since speech has started.
    * @param aCharIndex   offset of spoken characters.
    */
--- a/dom/media/webspeech/synth/nsSpeechTask.cpp
+++ b/dom/media/webspeech/synth/nsSpeechTask.cpp
@@ -373,18 +373,18 @@ nsSpeechTask::DispatchStartImpl(const ns
   if(NS_WARN_IF(!(mUtterance->mState == SpeechSynthesisUtterance::STATE_PENDING))) {
     return NS_ERROR_NOT_AVAILABLE;
   }
 
   CreateAudioChannelAgent();
 
   mUtterance->mState = SpeechSynthesisUtterance::STATE_SPEAKING;
   mUtterance->mChosenVoiceURI = aUri;
-  mUtterance->DispatchSpeechSynthesisEvent(NS_LITERAL_STRING("start"), 0, 0,
-                                           EmptyString());
+  mUtterance->DispatchSpeechSynthesisEvent(NS_LITERAL_STRING("start"), 0,
+                                           nullptr, 0, EmptyString());
 
   return NS_OK;
 }
 
 NS_IMETHODIMP
 nsSpeechTask::DispatchEnd(float aElapsedTime, uint32_t aCharIndex)
 {
   if (!mIndirectAudio) {
@@ -428,17 +428,17 @@ nsSpeechTask::DispatchEndImpl(float aEla
     mSpeechSynthesis->OnEnd(this);
   }
 
   if (utterance->mState == SpeechSynthesisUtterance::STATE_PENDING) {
     utterance->mState = SpeechSynthesisUtterance::STATE_NONE;
   } else {
     utterance->mState = SpeechSynthesisUtterance::STATE_ENDED;
     utterance->DispatchSpeechSynthesisEvent(NS_LITERAL_STRING("end"),
-                                            aCharIndex, aElapsedTime,
+                                            aCharIndex, nullptr, aElapsedTime,
                                             EmptyString());
   }
 
   return NS_OK;
 }
 
 NS_IMETHODIMP
 nsSpeechTask::DispatchPause(float aElapsedTime, uint32_t aCharIndex)
@@ -461,17 +461,17 @@ nsSpeechTask::DispatchPauseImpl(float aE
   }
   if(NS_WARN_IF(mUtterance->mState == SpeechSynthesisUtterance::STATE_ENDED)) {
     return NS_ERROR_NOT_AVAILABLE;
   }
 
   mUtterance->mPaused = true;
   if (mUtterance->mState == SpeechSynthesisUtterance::STATE_SPEAKING) {
     mUtterance->DispatchSpeechSynthesisEvent(NS_LITERAL_STRING("pause"),
-                                             aCharIndex, aElapsedTime,
+                                             aCharIndex, nullptr, aElapsedTime,
                                              EmptyString());
   }
   return NS_OK;
 }
 
 NS_IMETHODIMP
 nsSpeechTask::DispatchResume(float aElapsedTime, uint32_t aCharIndex)
 {
@@ -493,17 +493,17 @@ nsSpeechTask::DispatchResumeImpl(float a
   }
   if(NS_WARN_IF(mUtterance->mState == SpeechSynthesisUtterance::STATE_ENDED)) {
     return NS_ERROR_NOT_AVAILABLE;
   }
 
   mUtterance->mPaused = false;
   if (mUtterance->mState == SpeechSynthesisUtterance::STATE_SPEAKING) {
     mUtterance->DispatchSpeechSynthesisEvent(NS_LITERAL_STRING("resume"),
-                                             aCharIndex, aElapsedTime,
+                                             aCharIndex, nullptr, aElapsedTime,
                                              EmptyString());
   }
 
   return NS_OK;
 }
 
 NS_IMETHODIMP
 nsSpeechTask::DispatchError(float aElapsedTime, uint32_t aCharIndex)
@@ -531,45 +531,48 @@ nsSpeechTask::DispatchErrorImpl(float aE
   }
 
   if (mSpeechSynthesis) {
     mSpeechSynthesis->OnEnd(this);
   }
 
   mUtterance->mState = SpeechSynthesisUtterance::STATE_ENDED;
   mUtterance->DispatchSpeechSynthesisEvent(NS_LITERAL_STRING("error"),
-                                           aCharIndex, aElapsedTime,
+                                           aCharIndex, nullptr, aElapsedTime,
                                            EmptyString());
   return NS_OK;
 }
 
 NS_IMETHODIMP
 nsSpeechTask::DispatchBoundary(const nsAString& aName,
-                               float aElapsedTime, uint32_t aCharIndex)
+                               float aElapsedTime, uint32_t aCharIndex,
+                               uint32_t aCharLength, uint8_t argc)
 {
   if (!mIndirectAudio) {
     NS_WARNING("Can't call DispatchBoundary() from a direct audio speech service");
     return NS_ERROR_FAILURE;
   }
 
-  return DispatchBoundaryImpl(aName, aElapsedTime, aCharIndex);
+  return DispatchBoundaryImpl(aName, aElapsedTime, aCharIndex, aCharLength, argc);
 }
 
 nsresult
 nsSpeechTask::DispatchBoundaryImpl(const nsAString& aName,
-                                   float aElapsedTime, uint32_t aCharIndex)
+                                   float aElapsedTime, uint32_t aCharIndex,
+                                   uint32_t aCharLength, uint8_t argc)
 {
   MOZ_ASSERT(mUtterance);
   if(NS_WARN_IF(!(mUtterance->mState == SpeechSynthesisUtterance::STATE_SPEAKING))) {
     return NS_ERROR_NOT_AVAILABLE;
   }
+  mUtterance->DispatchSpeechSynthesisEvent(NS_LITERAL_STRING("boundary"),
+                                           aCharIndex,
+                                           argc ? static_cast<Nullable<uint32_t> >(aCharLength) : nullptr,
+                                           aElapsedTime, aName);
 
-  mUtterance->DispatchSpeechSynthesisEvent(NS_LITERAL_STRING("boundary"),
-                                           aCharIndex, aElapsedTime,
-                                           aName);
   return NS_OK;
 }
 
 NS_IMETHODIMP
 nsSpeechTask::DispatchMark(const nsAString& aName,
                            float aElapsedTime, uint32_t aCharIndex)
 {
   if (!mIndirectAudio) {
@@ -585,17 +588,17 @@ nsSpeechTask::DispatchMarkImpl(const nsA
                                float aElapsedTime, uint32_t aCharIndex)
 {
   MOZ_ASSERT(mUtterance);
   if(NS_WARN_IF(!(mUtterance->mState == SpeechSynthesisUtterance::STATE_SPEAKING))) {
     return NS_ERROR_NOT_AVAILABLE;
   }
 
   mUtterance->DispatchSpeechSynthesisEvent(NS_LITERAL_STRING("mark"),
-                                           aCharIndex, aElapsedTime,
+                                           aCharIndex, nullptr, aElapsedTime,
                                            aName);
   return NS_OK;
 }
 
 void
 nsSpeechTask::Pause()
 {
   MOZ_ASSERT(XRE_IsParentProcess());
--- a/dom/media/webspeech/synth/nsSpeechTask.h
+++ b/dom/media/webspeech/synth/nsSpeechTask.h
@@ -81,17 +81,19 @@ protected:
   virtual nsresult DispatchPauseImpl(float aElapsedTime, uint32_t aCharIndex);
 
   virtual nsresult DispatchResumeImpl(float aElapsedTime, uint32_t aCharIndex);
 
   virtual nsresult DispatchErrorImpl(float aElapsedTime, uint32_t aCharIndex);
 
   virtual nsresult DispatchBoundaryImpl(const nsAString& aName,
                                         float aElapsedTime,
-                                        uint32_t aCharIndex);
+                                        uint32_t aCharIndex,
+                                        uint32_t aCharLength,
+                                        uint8_t argc);
 
   virtual nsresult DispatchMarkImpl(const nsAString& aName,
                                     float aElapsedTime, uint32_t aCharIndex);
 
   RefPtr<SpeechSynthesisUtterance> mUtterance;
 
   float mVolume;
 
--- a/dom/media/webspeech/synth/windows/SapiService.cpp
+++ b/dom/media/webspeech/synth/windows/SapiService.cpp
@@ -134,27 +134,29 @@ SapiCallback::OnSpeechEvent(const SPEVEN
       mCurrentIndex = mSpeakTextLen;
     }
     mTask->DispatchEnd(GetTickCount() - mStartingTime, mCurrentIndex);
     mTask = nullptr;
     break;
   case SPEI_TTS_BOOKMARK:
     mCurrentIndex = static_cast<ULONG>(speechEvent.lParam) - mTextOffset;
     mTask->DispatchBoundary(NS_LITERAL_STRING("mark"),
-                            GetTickCount() - mStartingTime, mCurrentIndex);
+                            GetTickCount() - mStartingTime, mCurrentIndex, 0, 0);
     break;
   case SPEI_WORD_BOUNDARY:
     mCurrentIndex = static_cast<ULONG>(speechEvent.lParam) - mTextOffset;
     mTask->DispatchBoundary(NS_LITERAL_STRING("word"),
-                            GetTickCount() - mStartingTime, mCurrentIndex);
+                            GetTickCount() - mStartingTime, mCurrentIndex,
+                            static_cast<ULONG>(speechEvent.wParam), 1);
     break;
   case SPEI_SENTENCE_BOUNDARY:
     mCurrentIndex = static_cast<ULONG>(speechEvent.lParam) - mTextOffset;
     mTask->DispatchBoundary(NS_LITERAL_STRING("sentence"),
-                            GetTickCount() - mStartingTime, mCurrentIndex);
+                            GetTickCount() - mStartingTime, mCurrentIndex,
+                            static_cast<ULONG>(speechEvent.wParam), 1);
     break;
   default:
     break;
   }
 }
 
 // static
 void __stdcall
--- a/dom/webidl/SpeechSynthesisEvent.webidl
+++ b/dom/webidl/SpeechSynthesisEvent.webidl
@@ -11,19 +11,21 @@
  */
 
 [Constructor(DOMString type, SpeechSynthesisEventInit eventInitDict),
  Pref="media.webspeech.synth.enabled"]
 interface SpeechSynthesisEvent : Event
 {
   readonly attribute SpeechSynthesisUtterance utterance;
   readonly attribute unsigned long charIndex;
+  readonly attribute unsigned long? charLength;
   readonly attribute float elapsedTime;
   readonly attribute DOMString? name;
 };
 
 dictionary SpeechSynthesisEventInit : EventInit
 {
   required SpeechSynthesisUtterance utterance;
   unsigned long charIndex = 0;
+  unsigned long? charLength = null;
   float elapsedTime = 0;
   DOMString name = "";
 };