PG1204 commented on code in PR #8617: URL: https://github.com/apache/texera/pull/8617#discussion_r4138963197
########## common/workflow-operator/src/main/scala/org/apache/texera/amber/operator/huggingFace/codegen/TextGenCodegen.scala: ########## @@ -28,7 +28,9 @@ package org.apache.texera.amber.operator.huggingFace.codegen * (Cerebras, Groq, Sambanova, Together, …) accepts. * * The parse step pulls `body["choices"][0]["message"]["content"]` out of - * the response. + * the response, degrading to the raw JSON body when a provider returns a + * shape that does not carry it — parsing runs per row, so raising here + * would abort the whole run over one malformed response. Review Comment: Same wrong claim, and I've taken your wording as-is. "Keeps the result column string-valued for malformed or non-string chat content" is more precise than what I'd written, and it covers the non-string cases (a numeric or list-valued content) that my phrasing missed. -- 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]
