oscerd opened a new pull request, #25854: URL: https://github.com/apache/camel/pull/25854
## Issue [CAMEL-24528](https://issues.apache.org/jira/browse/CAMEL-24528) ## Problem Ten task predictors delegate to a Python inference script and read the result via `output.getAsString("data")`. Six of them guard the output before using it: ```java String resultJson = output.getAsString("data"); if (resultJson.contains("\"error\"")) { throw new RuntimeCamelException("Python inference failed: " + resultJson); } ``` Four do **not**: `TextGenerationPredictor`, `SummarizationPredictor`, `QuestionAnsweringPredictor` and `TextToImagePredictor`. When the Python script fails and returns a JSON error payload (`{"error": ...}`), these four set it as the successful result: - text generation / summarization / question answering return the raw error JSON as the body; - text-to-image serves the error JSON bytes to the caller as `image/png`. ## Fix Add the same error guard to the four predictors. For text-to-image (whose output is raw image bytes) the returned bytes are decoded to a string to check for the error marker before the image body is set. ## Testing - New `PredictorErrorHandlingTest` asserts each of the four predictors throws on an error payload, and that a successful text-generation result is still passed through. - `mvn -Psourcecheck validate` green. _Claude Code on behalf of oscerd_ -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
