ethanstanley3 opened a new issue, #2367: URL: https://github.com/apache/iceberg-go/issues/2367
### Apache Iceberg version main (development) ### Please describe the bug 🐞 # Partition paths included despite write.object-storage.partitioned-paths=false With `github.com/apache/iceberg-go v0.7.0`, writing to a partitioned Iceberg table with these properties produces object keys containing both hash directories and partition paths: ```text write.object-storage.enabled=true write.object-storage.partitioned-paths=false ``` Expected: data-file keys include the hash but omit partition directories. Actual: keys still include the partition directory, for example: ``` s3://warehouse/repro/partition_paths/data/category=example/0011/1010/1101/01110000-00000-0-085dc2e2-1424-4e29-8b16-286759ed332f-00001.parquet ``` To trigger this, create a table in Spark partitioned by `category` with the properties above, then load it through the REST catalog in Go and append a row with `category='example'`. The resulting key includes `category=example/` even though partition paths are disabled. [reproduction.zip](https://github.com/user-attachments/files/34153987/reproduction.zip) contains a minimal Docker Compose stack, Spark SQL table creation script, and Go writer. Extract the archive and follow its README to reproduce and verify the written file path. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
