beryllw opened a new issue, #1412: URL: https://github.com/apache/fluss/issues/1412
### Search before asking - [x] I searched in the [issues](https://github.com/alibaba/fluss/issues) and found nothing similar. ### Fluss version 0.7.0 (latest release) ### Please describe the bug 🐞 When using the Paimon lake format for Fluss primary key tables, keys are encoded using the Paimon format, which stores length and offset information for variable-length fields. While primary key point queries work correctly, prefix queries fail to return any data. This issue arises because the encoded fixed-length region of the stored key differs from that of the prefix query key. **Root Cause:** The PaimonBinaryRowWriter class (see com.alibaba.fluss.row.encode.paimon.PaimonBinaryRowWriter) encodes variable-length fields differently for primary key look up queries and prefix look up queries: 1. Primary Key Point Query:Encodes all key fields (e.g., two string fields) 2. Encodes only the prefix field (e.g., the first string field) ### Solution Do we need to use Paimon's encoding for storing keys in primary key tables? Is it possible to use Fluss's default KeyEncoder instead? ### Are you willing to submit a PR? - [x] I'm willing to submit a PR! -- 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]
