geruh commented on code in PR #3131:
URL: https://github.com/apache/iceberg-python/pull/3131#discussion_r3095873141
##########
pyiceberg/table/update/snapshot.py:
##########
@@ -667,6 +667,96 @@ def _get_entries(manifest: ManifestFile) ->
list[ManifestEntry]:
return []
+class _RewriteFiles(_SnapshotProducer["_RewriteFiles"]):
+ """A snapshot producer that rewrites data files."""
+
+ def __init__(self, operation: Operation, transaction: Transaction, io:
FileIO, snapshot_properties: dict[str, str]):
+ super().__init__(operation, transaction, io,
snapshot_properties=snapshot_properties)
+
+ def _commit(self) -> UpdatesAndRequirements:
+ # Only produce a commit when there is something to rewrite
+ if self._deleted_data_files or self._added_data_files:
+ # Grab the entries that we actually found in the table's manifests
+ deleted_entries = self._deleted_entries()
+ found_deleted_files = {entry.data_file for entry in
deleted_entries}
+
+ # If the user asked to delete files that aren't in the table,
abort.
+ if len(found_deleted_files) != len(self._deleted_data_files):
+ raise ValueError("Cannot delete files that are not present in
the table")
+
+ added_records = sum(f.record_count for f in self._added_data_files)
+ deleted_records = sum(entry.data_file.record_count for entry in
deleted_entries)
+
+ if added_records > deleted_records:
Review Comment:
Okay let's keep the check I took a deeper look into the snapshot producer on
the java side so let's align closer to that:
https://github.com/apache/iceberg/blob/dde712ec9ed6c9d28183ee4615d50f97b246af5d/core/src/main/java/org/apache/iceberg/SnapshotProducer.java#L322-L334
--
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]
