Unify StreamError and OpenAIError #413
Open
+106
−51
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
timonv
reviewed
Aug 14, 2025
| @@ -20,13 +24,35 @@ pub enum OpenAIError { | |||
| FileReadError(String), | |||
| /// Error on SSE streaming | |||
| #[error("stream failed: {0}")] | |||
| StreamError(String), | |||
| StreamError(StreamError), | |||
There was a problem hiding this comment.
Nice! Since you always map the inner eventsource error anyway, It would simplify the code quite a bit I think to wrap StreamError(EventSourceError) instead with #[from].
If I'm not mistaken, that would remove map_stream_error instead as we can use Into, remove the intermediate error, and avoid reading out the bytes of the response early.
There was a problem hiding this comment.
We can't because we need to determine what OpenAIError is wrapped inside the EventSourceError::InvalidStatusCode and as a bonus we also map the EventSourceError::Transport to OpenAIError::Reqwest.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The current StreamError implementation wraps whatever happens in a string, throwing away important underlying information. Specifically this prevents us from catching ContextLengthExceeded errors when streaming is enabled.
This PR rewrites the error handling so that the streaming and non-streaming implementations share the same error parsing code and both return OpenAIError's of which StreamError is now a stream specific subset.
Unfortunately this is a breaking change, any users that explicitly match on StreamError will have to be modified to accommodate the extra information that's now returned in the error.