rdblue commented on code in PR #61: URL: https://github.com/apache/iceberg-python/pull/61#discussion_r1416456616
########## pyiceberg/table/snapshots.py: ########## @@ -116,3 +168,193 @@ class MetadataLogEntry(IcebergBaseModel): class SnapshotLogEntry(IcebergBaseModel): snapshot_id: int = Field(alias="snapshot-id") timestamp_ms: int = Field(alias="timestamp-ms") + + +class SnapshotSummaryCollector: + added_size: int + removed_size: int + added_files: int + removed_files: int + added_eq_delete_files: int + removed_eq_delete_files: int + added_pos_delete_files: int + removed_pos_delete_files: int + added_delete_files: int + removed_delete_files: int + added_records: int + deleted_records: int + added_pos_deletes: int + removed_pos_deletes: int + added_eq_deletes: int + removed_eq_deletes: int + + def __init__(self) -> None: + self.added_size = 0 + self.removed_size = 0 + self.added_files = 0 + self.removed_files = 0 + self.added_eq_delete_files = 0 + self.removed_eq_delete_files = 0 + self.added_pos_delete_files = 0 + self.removed_pos_delete_files = 0 + self.added_delete_files = 0 + self.removed_delete_files = 0 + self.added_records = 0 + self.deleted_records = 0 + self.added_pos_deletes = 0 + self.removed_pos_deletes = 0 + self.added_eq_deletes = 0 + self.removed_eq_deletes = 0 + + def add_file(self, data_file: DataFile) -> None: + self.added_size += data_file.file_size_in_bytes + + if data_file.content == DataFileContent.DATA: + self.added_files += 1 + self.added_records += data_file.record_count + elif data_file.content == DataFileContent.POSITION_DELETES: + self.added_delete_files += 1 + self.added_pos_delete_files += 1 + self.added_pos_deletes += data_file.record_count + elif data_file.content == DataFileContent.EQUALITY_DELETES: + self.added_delete_files += 1 + self.added_eq_delete_files += 1 + self.added_eq_deletes += data_file.record_count + else: + raise ValueError(f"Unknown data file content: {data_file.content}") + + def remove_file(self, data_file: DataFile) -> None: + self.removed_size += data_file.file_size_in_bytes + + if data_file.content == DataFileContent.DATA: + self.removed_files += 1 + self.deleted_records += data_file.record_count + elif data_file.content == DataFileContent.POSITION_DELETES: + self.removed_delete_files += 1 + self.removed_pos_delete_files += 1 + self.removed_pos_deletes += data_file.record_count + elif data_file.content == DataFileContent.EQUALITY_DELETES: + self.removed_delete_files += 1 + self.removed_eq_delete_files += 1 + self.removed_eq_deletes += data_file.record_count + else: + raise ValueError(f"Unknown data file content: {data_file.content}") + + def build(self) -> Dict[str, str]: + def set_when_positive(properties: Dict[str, str], num: int, property_name: str) -> None: + if num > 0: + properties[property_name] = str(num) + + properties: Dict[str, str] = {} + set_when_positive(properties, self.added_size, ADDED_FILE_SIZE) + set_when_positive(properties, self.removed_size, REMOVED_FILE_SIZE) + set_when_positive(properties, self.added_files, ADDED_DATA_FILES) Review Comment: Minor: it would be better to use `self.added_data_files` and `self.removed_data_files` since those are the properties that we're tracking. -- 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