[flutter_tts] Refactor the C++ code

[JSUYA]

Part of the code refactoring project.

• Extract method.
• Hiding tizen api in plugin.cc
• Minimize use of output parameters
• Change method name(Init-> TtsCreate, DeInit -> TtsDestroy, ...)

[HakkyuKim]

I'm not sure changing to exception throws is a good idea when the original code worked fine without them. If you are going to use exceptions, they must all be catched (it's easy to miss in callbacks) as @bbrto21 mentioned.

nit: remove "[WIP]" in commit message.

[JSUYA]

@bbrto21 @HakkyuKim Thank you for review. I reverted the code related to the exception.

[JSUYA]

@swift-kim Any opinions?

[swift-kim]

Some nitpicks. Otherwise looks good to me.

[swift-kim]

Calling getSpeechRateValidRange fails with a StateError on Tizen.

[ERROR:flutter/lib/ui/ui_dart_state.cc(198)] Unhandled Exception: Bad state: No element
#0      ListMixin.firstWhere (dart:collection/list.dart:167)
#1      FlutterTts.getSpeechRateValidRange (package:flutter_tts/flutter_tts.dart:320)
<asynchronous suspension>
#2      _MyAppState._speak (package:flutter_tts_tizen_example/main.dart:90)
<asynchronous suspension>

Maybe we can

• set the value of platform to "android" (instead of "tizen"), or
• just unsupport the method and automatically set tts_speed_ based on the platform's speed range (proportional to 0.0~1.0). For example, if the max speed is 10 and the given speech rate is 0.5, set the tts_speed_ to 5.

How do you think? The implementation can be done in a separate PR.

[JSUYA]

Calling getSpeechRateValidRange fails with a StateError on Tizen.

[ERROR:flutter/lib/ui/ui_dart_state.cc(198)] Unhandled Exception: Bad state: No element
#0      ListMixin.firstWhere (dart:collection/list.dart:167)
#1      FlutterTts.getSpeechRateValidRange (package:flutter_tts/flutter_tts.dart:320)
<asynchronous suspension>
#2      _MyAppState._speak (package:flutter_tts_tizen_example/main.dart:90)
<asynchronous suspension>

Maybe we can

• set the value of platform to "android" (instead of "tizen"), or
• just unsupport the method and automatically set tts_speed_ based on the platform's speed range (proportional to 0.0~1.0). For example, if the max speed is 10 and the given speech rate is 0.5, set the tts_speed_ to 5.

How do you think? The implementation can be done in a separate PR.

Could you please explain more how to reproduce this log?
As far as I know, the example in flutter_tts doesn't have the code to call getSpeechRateValidRange.
Do I need to test it separately? If the problem is not caused by this refactoring, I think it is better to handle it in another PR.

[swift-kim]

@JSUYA You can add

await FlutterTts.getSpeechRateValidRange;

to anywhere in the example app (like initState or _speak). And yes, the error has not been caused by this PR, but I mentioned here because the first solution in the above requires only a single line of change. If you prefer the second solution, we can work on it in another PR.

[JSUYA]

@JSUYA You can add

await FlutterTts.getSpeechRateValidRange;

to anywhere in the example app (like initState or _speak). And yes, the error has not been caused by this PR, but I mentioned here because the first solution in the above requires only a single line of change. If you prefer the second solution, we can work on it in another PR.

Thank you for your advice. I updated PR

[swift-kim]

Thank you!