zhangbutao commented on issue #8951:
URL: https://github.com/apache/iceberg/issues/8951#issuecomment-1785506472
> This is why I asked if you checked the "files" metadata table. If the
Files metadata table has no tuples of {data: a} then there is no partition of
data = a.
After truncate(`DeleteFiles::deleteFromRowFilter`) the partition `data=a`,
I did a query to check the "files" metadata table using 1.3.1&1.41, and both
didn't find the `data = a` file path. And The partitions metadata table will
show the empty partition `data=a` in 1.4.1, but 1.3.1 didn't show this empty
partition.
> Now you can have a partition with no rows, this would happen if you have a
file for {data:a} but also have delete files which negate all rows in that
partition.
I extracted the key code snippet from the unit test:
```
// here delete a partition data
Expression finalExp = Expressions.in("data", "a");
DeleteFiles delete = table.newDelete();
delete.deleteFromRowFilter(finalExp);
delete.commit();
```
From the code snippet, I think we acutually did a metadata level delete and
no delete files generation. I didnt find the delete files. Am i right?
--
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]
