amogh-jahagirdar commented on code in PR #245:
URL: https://github.com/apache/iceberg-python/pull/245#discussion_r1471453813
##########
pyiceberg/table/__init__.py:
##########
@@ -2271,3 +2325,243 @@ def commit(self) -> Snapshot:
)
return snapshot
+
+
+class UpdateSpec:
+ _table: Table
+ _name_to_field: Dict[str, PartitionField] = {}
+ _name_to_added_field: Dict[str, PartitionField] = {}
+ _transform_to_field: Dict[Tuple[int, str], PartitionField] = {}
+ _transform_to_added_field: Dict[Tuple[int, str], PartitionField] = {}
+ _renames: Dict[str, str] = {}
+ _added_time_fields: Dict[int, PartitionField] = {}
+ _case_sensitive: bool
+ _adds: List[PartitionField]
+ _deletes: Set[int]
+ _last_assigned_partition_id: int
+ _transaction: Optional[Transaction]
+ _unassigned_field_name = 'unassigned_field_name'
+
+ def __init__(self, table: Table, transaction: Optional[Transaction] =
None, case_sensitive: bool = True) -> None:
+ self._table = table
+ self._name_to_field = {field.name: field for field in
table.spec().fields}
+ self._name_to_added_field = {}
+ self._transform_to_field = {(field.source_id, repr(field.transform)):
field for field in table.spec().fields}
+ self._transform_to_added_field = {}
+ self._adds = []
+ self._deletes = set()
+ if len(table.specs()) == 1:
+ self._last_assigned_partition_id = PARTITION_FIELD_ID_START - 1
+ else:
+ self._last_assigned_partition_id =
table.spec().last_assigned_field_id
+ self._renames = {}
+ self._transaction = transaction
+ self._case_sensitive = case_sensitive
+ self._added_time_fields = {}
+
+ def add_field(
+ self,
+ source_column_name: str,
+ transform: Transform[Any, Any],
+ partition_field_name: Optional[str] = _unassigned_field_name,
+ ) -> UpdateSpec:
+ ref = Reference(source_column_name)
+ bound_ref = ref.bind(self._table.schema(), self._case_sensitive)
+ # verify transform can actually bind it
+ output_type = bound_ref.field.field_type
+ if not transform.can_transform(output_type):
+ raise ValueError(f"{transform} cannot transform {output_type}
values from {bound_ref.field.name}")
+
+ transform_key = (bound_ref.field.field_id, repr(transform))
+ existing_partition_field = self._transform_to_field.get(transform_key)
+ if existing_partition_field and
self._is_duplicate_partition(transform, existing_partition_field):
+ raise ValueError(f"Duplicate partition field for
${ref.name}=${ref}, ${existing_partition_field} already exists")
+
+ added = self._transform_to_added_field.get(transform_key)
+ if added:
+ raise ValueError(f"Already added partition: {added.name}")
+
+ new_field = self._partition_field((bound_ref.field.field_id,
transform), partition_field_name)
+ if new_field.name == self._unassigned_field_name:
Review Comment:
Ah yeah good point :) the unassigned name is really only temporarily needed.
I'll update to do this
--
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: issues-unsubscr...@iceberg.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org
For additional commands, e-mail: issues-h...@iceberg.apache.org
