tanmayrauth commented on issue #1082:
URL: https://github.com/apache/iceberg-go/issues/1082#issuecomment-4468507472
Welcome to iceberg-go!
Your diagnosis is correct. PartitionType(schema) skips any partition field
whose source ID isn't found in the provided schema. since
PositionalDeleteSchema only has file_path and pos, all partition source columns
are missing, producing an empty partition type and an empty path for every
target file.
The fix you outlined (store the table schema separately, use it for
PartitionType/PartitionToPath, keep PositionalDeleteSchema for file contents)
is the right approach. The constructor already receives metadata, so you can
get the table schema from there.
For the regression test: two target data files in different partitions
asserting they produce distinct partition paths (and that the resulting
DataFile entries carry the correct partition metadata) would cover it well.
@zeroshade @laskoviymishka anything to add before @fallintoplace open 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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
