funasr support payload.input and bug fix in transcription#142
funasr support payload.input and bug fix in transcription#142songguocola wants to merge 2 commits into
Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request introduces a wait_timeout parameter to the asynchronous and synchronous wait methods in base_api.py and transcription.py to handle task timeouts, and updates _only_parameters in api_request_data.py to extract raw_input from parameters. The review feedback highlights that mutating self.parameters in-place introduces side effects and recommends copying it first. Additionally, it is suggested to use time.monotonic() instead of time.time() for robust elapsed time measurements, and to update BaseAsyncApi.call() to pass wait_timeout so that synchronous calls respect the timeout.
Important
The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.
| task: Union[str, DashScopeAPIResponse], | ||
| api_key: str = None, | ||
| workspace: str = None, | ||
| wait_timeout: int = -1, |
There was a problem hiding this comment.
While wait_timeout is now accepted by BaseAsyncApi.wait(), note that BaseAsyncApi.call() (at line 618) does not pass **kwargs or wait_timeout to cls.wait(). This means any wait_timeout passed to synchronous call() methods (such as Transcription.call(..., wait_timeout=60)) will be silently ignored. Please update BaseAsyncApi.call() to pass wait_timeout or **kwargs to cls.wait() so that the timeout is respected in synchronous calls.
| task: Union[str, TranscriptionResponse], # type: ignore[override] | ||
| api_key: str = None, | ||
| workspace: str = None, | ||
| wait_timeout: int = -1, |
There was a problem hiding this comment.
Note that Transcription.call() delegates to BaseAsyncApi.call(), which currently does not pass **kwargs or wait_timeout to cls.wait(). Consequently, passing wait_timeout to Transcription.call() will have no effect. To fix this, BaseAsyncApi.call() in dashscope/client/base_api.py needs to be updated to pass **kwargs or wait_timeout to cls.wait().
| obj = {"model": self.model, "parameters": self.parameters, "input": {}} | ||
| if "raw_input" in self.parameters: | ||
| obj["input"] = self.parameters.pop("raw_input") |
There was a problem hiding this comment.
Mutating self.parameters in-place inside _only_parameters() introduces side effects. If the request is retried, logged, or reused, raw_input will be permanently missing from self.parameters. Copying self.parameters before popping avoids this side effect.
| obj = {"model": self.model, "parameters": self.parameters, "input": {}} | |
| if "raw_input" in self.parameters: | |
| obj["input"] = self.parameters.pop("raw_input") | |
| parameters = self.parameters.copy() | |
| obj = {"model": self.model, "parameters": parameters, "input": {}} | |
| if "raw_input" in parameters: | |
| obj["input"] = parameters.pop("raw_input") |
| max_wait_seconds = 5 | ||
| increment_steps = 3 | ||
| step = 0 | ||
| start_time = time.time() |
There was a problem hiding this comment.
Use time.monotonic() instead of time.time() to measure elapsed time. time.monotonic() is unaffected by system clock updates or adjustments, making it much more reliable for timeout calculations.
| start_time = time.time() | |
| start_time = time.monotonic() |
| # (server side return immediately when ready) | ||
| if wait_seconds < max_wait_seconds and step % increment_steps == 0: | ||
| wait_seconds = min(wait_seconds * 2, max_wait_seconds) | ||
| if 0 < wait_timeout <= (time.time() - start_time): |
There was a problem hiding this comment.
Use time.monotonic() instead of time.time() to measure elapsed time consistently with start_time.
| if 0 < wait_timeout <= (time.time() - start_time): | |
| if 0 < wait_timeout <= (time.monotonic() - start_time): |
| max_wait_seconds = 5 | ||
| increment_steps = 3 | ||
| step = 0 | ||
| start_time = time.time() |
There was a problem hiding this comment.
Use time.monotonic() instead of time.time() to measure elapsed time. time.monotonic() is unaffected by system clock updates or adjustments, making it much more reliable for timeout calculations.
| start_time = time.time() | |
| start_time = time.monotonic() |
| # (server side return immediately when ready) | ||
| if wait_seconds < max_wait_seconds and step % increment_steps == 0: | ||
| wait_seconds = min(wait_seconds * 2, max_wait_seconds) | ||
| if 0 < wait_timeout <= (time.time() - start_time): |
There was a problem hiding this comment.
Use time.monotonic() instead of time.time() to measure elapsed time consistently with start_time.
| if 0 < wait_timeout <= (time.time() - start_time): | |
| if 0 < wait_timeout <= (time.monotonic() - start_time): |
1 participant
No description provided.