DanielLeens opened a new issue, #11277: URL: https://github.com/apache/seatunnel/issues/11277
## Background The transform docs describe individual transforms, but users often need scenario-level recipes that combine source parsing, transforms, and sink schema requirements. Examples of recurring questions include: - Read a deeply nested Kafka JSON message, split part of it into many rows, and write the result to MySQL. - Read a MySQL JSON field and convert it into a Milvus `FLOAT_VECTOR` field. - Read file content and use LLM-related processing to extract or chunk content before writing it downstream. These questions are difficult to answer from individual transform option tables alone. ## Goal Add practical transform recipes that show how to solve common nested JSON and vector conversion scenarios with SeaTunnel. ## Suggested scope Please add or update English and Chinese transform docs. Possible locations: - `docs/en/transforms/` - `docs/zh/transforms/` - a new transform recipes page if that fits the current docs structure The recipes should cover: 1. Nested JSON extraction and field mapping from Kafka or file source input. 2. Whether SeaTunnel supports one input row producing multiple output rows through built-in transforms today. 3. How to handle unsupported recursive JSON flattening or one-to-many split scenarios. 4. MySQL JSON to Milvus vector conversion, including the expected SeaTunnel field type and required transform/conversion step. 5. LLM/file-content extraction boundaries: what the LLM transform can process directly and what must be handled before data enters the transform. ## Acceptance criteria - The docs include at least two complete HOCON snippets. - The docs clearly separate supported built-in behavior from scenarios requiring custom transform code. - English and Chinese docs are consistent. - The Milvus vector example matches the current Milvus connector type mapping. -- 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]
