Fokko commented on code in PR #5298: URL: https://github.com/apache/iceberg/pull/5298#discussion_r927652488
########## python/pyiceberg/manifest.py: ########## @@ -0,0 +1,155 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. +from enum import Enum +from functools import singledispatch +from typing import ( + Any, + Dict, + Iterator, + List, + Optional, + Union, +) + +from pyiceberg.avro.file import AvroFile +from pyiceberg.avro.reader import AvroStruct +from pyiceberg.io.base import InputFile +from pyiceberg.schema import Schema +from pyiceberg.types import ( + IcebergType, + ListType, + MapType, + PrimitiveType, + StructType, +) +from pyiceberg.utils.iceberg_base_model import IcebergBaseModel + + +class FileFormat(str, Enum): + AVRO = "AVRO" + PARQUET = "PARQUET" + ORC = "ORC" + + +class DataFile(IcebergBaseModel): + file_path: str + file_format: FileFormat + partition: Dict[str, Any] + record_count: int + file_size_in_bytes: int + block_size_in_bytes: int + column_sizes: Optional[Dict[int, int]] + value_counts: Optional[Dict[int, int]] + null_value_counts: Optional[Dict[int, int]] + nan_value_counts: Optional[Dict[int, int]] + lower_bounds: Optional[Dict[int, bytes]] + upper_bounds: Optional[Dict[int, bytes]] + key_metadata: Optional[bytes] + split_offsets: Optional[List[int]] + sort_order_id: Optional[int] + + +class ManifestEntry(IcebergBaseModel): + status: int + snapshot_id: Optional[int] + data_file: DataFile + + +class Partition(IcebergBaseModel): + contains_null: bool + contains_nan: Optional[bool] + lower_bound: Optional[bytes] + upper_bound: Optional[bytes] + + +class ManifestFile(IcebergBaseModel): + manifest_path: str + manifest_length: int + partition_spec_id: int + added_snapshot_id: Optional[int] + added_data_files_count: Optional[int] + existing_data_files_count: Optional[int] + deleted_data_files_count: Optional[int] + partitions: Optional[List[Partition]] + added_rows_count: Optional[int] + existing_rows_counts: Optional[int] + deleted_rows_count: Optional[int] + + +class Manifest: + @staticmethod + def read_manifest_entry(input_file: InputFile) -> Iterator[ManifestEntry]: + with AvroFile(input_file) as reader: + schema = reader.schema + for record in reader: + dict_repr = convert_pos_to_dict(schema, record) Review Comment: Yes, there is some overhead involved. I've parsed 10k manifests (the same ones :), and it looks like 32% of the total time is spent constructing the object, including reading the actual data. Also, converting the schema from Avro to iceberg is even costlier (35% of total time). This is mostly because the singleton is quite slow with the lookup that it has to make.  Code for the profiling: ```python # Licensed to the Apache Software Foundation (ASF) under one # or more contributor license agreements. See the NOTICE file # distributed with this work for additional information # regarding copyright ownership. The ASF licenses this file # to you under the Apache License, Version 2.0 (the # "License"); you may not use this file except in compliance # with the License. You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, # software distributed under the License is distributed on an # "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY # KIND, either express or implied. See the License for the # specific language governing permissions and limitations # under the License. import os from tempfile import TemporaryDirectory from urllib.parse import ParseResult, urlparse from fastavro import parse_schema, writer from pyiceberg.io.base import InputFile, InputStream from pyiceberg.manifest import ( read_manifest_entry, ) class LocalInputFile(InputFile): """An InputFile implementation for local files (for test use only)""" def __init__(self, location: str): parsed_location = urlparse(location) # Create a ParseResult from the uri if parsed_location.scheme and parsed_location.scheme != "file": # Validate that a uri is provided with a scheme of `file` raise ValueError("LocalInputFile location must have a scheme of `file`") elif parsed_location.netloc: raise ValueError(f"Network location is not allowed for LocalInputFile: {parsed_location.netloc}") super().__init__(location=location) self._parsed_location = parsed_location @property def parsed_location(self) -> ParseResult: """The parsed location Returns: ParseResult: The parsed results which has attributes `scheme`, `netloc`, `path`, `params`, `query`, and `fragments`. """ return self._parsed_location def __len__(self): return os.path.getsize(self.parsed_location.path) def exists(self): return os.path.exists(self.parsed_location.path) def open(self) -> InputStream: input_file = open(self.parsed_location.path, "rb") if not isinstance(input_file, InputStream): raise TypeError("Object returned from LocalInputFile.open() does not match the OutputStream protocol.") return input_file avro_schema_manifest_entry = { "type": "record", "name": "manifest_entry", "fields": [ {"name": "status", "type": "int", "field-id": 0}, {"name": "snapshot_id", "type": ["null", "long"], "default": "null", "field-id": 1}, { "name": "data_file", "type": { "type": "record", "name": "r2", "fields": [ {"name": "file_path", "type": "string", "doc": "Location URI with FS scheme", "field-id": 100}, { "name": "file_format", "type": "string", "doc": "File format name: avro, orc, or parquet", "field-id": 101, }, { "name": "partition", "type": { "type": "record", "name": "r102", "fields": [ {"name": "VendorID", "type": ["null", "int"], "default": "null", "field-id": 1000}], }, "field-id": 102, }, {"name": "record_count", "type": "long", "doc": "Number of records in the file", "field-id": 103}, {"name": "file_size_in_bytes", "type": "long", "doc": "Total file size in bytes", "field-id": 104}, {"name": "block_size_in_bytes", "type": "long", "field-id": 105}, { "name": "column_sizes", "type": [ "null", { "type": "array", "items": { "type": "record", "name": "k117_v118", "fields": [ {"name": "key", "type": "int", "field-id": 117}, {"name": "value", "type": "long", "field-id": 118}, ], }, "logicalType": "map", }, ], "doc": "Map of column id to total size on disk", "default": "null", "field-id": 108, }, { "name": "value_counts", "type": [ "null", { "type": "array", "items": { "type": "record", "name": "k119_v120", "fields": [ {"name": "key", "type": "int", "field-id": 119}, {"name": "value", "type": "long", "field-id": 120}, ], }, "logicalType": "map", }, ], "doc": "Map of column id to total count, including null and NaN", "default": "null", "field-id": 109, }, { "name": "null_value_counts", "type": [ "null", { "type": "array", "items": { "type": "record", "name": "k121_v122", "fields": [ {"name": "key", "type": "int", "field-id": 121}, {"name": "value", "type": "long", "field-id": 122}, ], }, "logicalType": "map", }, ], "doc": "Map of column id to null value count", "default": "null", "field-id": 110, }, { "name": "nan_value_counts", "type": [ "null", { "type": "array", "items": { "type": "record", "name": "k138_v139", "fields": [ {"name": "key", "type": "int", "field-id": 138}, {"name": "value", "type": "long", "field-id": 139}, ], }, "logicalType": "map", }, ], "doc": "Map of column id to number of NaN values in the column", "default": "null", "field-id": 137, }, { "name": "lower_bounds", "type": [ "null", { "type": "array", "items": { "type": "record", "name": "k126_v127", "fields": [ {"name": "key", "type": "int", "field-id": 126}, {"name": "value", "type": "bytes", "field-id": 127}, ], }, "logicalType": "map", }, ], "doc": "Map of column id to lower bound", "default": "null", "field-id": 125, }, { "name": "upper_bounds", "type": [ "null", { "type": "array", "items": { "type": "record", "name": "k129_v130", "fields": [ {"name": "key", "type": "int", "field-id": 129}, {"name": "value", "type": "bytes", "field-id": 130}, ], }, "logicalType": "map", }, ], "doc": "Map of column id to upper bound", "default": "null", "field-id": 128, }, { "name": "key_metadata", "type": ["null", "bytes"], "doc": "Encryption key metadata blob", "default": "null", "field-id": 131, }, { "name": "split_offsets", "type": ["null", {"type": "array", "items": "long", "element-id": 133}], "doc": "Splittable offsets", "default": "null", "field-id": 132, }, { "name": "sort_order_id", "type": ["null", "int"], "doc": "Sort order ID", "default": "null", "field-id": 140, }, ], }, "field-id": 2, }, ], } manifest_entry_records = [ { "status": 1, "snapshot_id": 8744736658442914487, "data_file": { "file_path": "/home/iceberg/warehouse/nyc/taxis_partitioned/data/VendorID=null/00000-633-d8a4223e-dc97-45a1-86e1-adaba6e8abd7-00001.parquet", "file_format": "PARQUET", "partition": {"VendorID": None}, "record_count": 19513, "file_size_in_bytes": 388872, "block_size_in_bytes": 67108864, "column_sizes": [ {"key": 1, "value": 53}, {"key": 2, "value": 98153}, {"key": 3, "value": 98693}, {"key": 4, "value": 53}, {"key": 5, "value": 53}, {"key": 6, "value": 53}, {"key": 7, "value": 17425}, {"key": 8, "value": 18528}, {"key": 9, "value": 53}, {"key": 10, "value": 44788}, {"key": 11, "value": 35571}, {"key": 12, "value": 53}, {"key": 13, "value": 1243}, {"key": 14, "value": 2355}, {"key": 15, "value": 12750}, {"key": 16, "value": 4029}, {"key": 17, "value": 110}, {"key": 18, "value": 47194}, {"key": 19, "value": 2948}, ], "value_counts": [ {"key": 1, "value": 19513}, {"key": 2, "value": 19513}, {"key": 3, "value": 19513}, {"key": 4, "value": 19513}, {"key": 5, "value": 19513}, {"key": 6, "value": 19513}, {"key": 7, "value": 19513}, {"key": 8, "value": 19513}, {"key": 9, "value": 19513}, {"key": 10, "value": 19513}, {"key": 11, "value": 19513}, {"key": 12, "value": 19513}, {"key": 13, "value": 19513}, {"key": 14, "value": 19513}, {"key": 15, "value": 19513}, {"key": 16, "value": 19513}, {"key": 17, "value": 19513}, {"key": 18, "value": 19513}, {"key": 19, "value": 19513}, ], "null_value_counts": [ {"key": 1, "value": 19513}, {"key": 2, "value": 0}, {"key": 3, "value": 0}, {"key": 4, "value": 19513}, {"key": 5, "value": 19513}, {"key": 6, "value": 19513}, {"key": 7, "value": 0}, {"key": 8, "value": 0}, {"key": 9, "value": 19513}, {"key": 10, "value": 0}, {"key": 11, "value": 0}, {"key": 12, "value": 19513}, {"key": 13, "value": 0}, {"key": 14, "value": 0}, {"key": 15, "value": 0}, {"key": 16, "value": 0}, {"key": 17, "value": 0}, {"key": 18, "value": 0}, {"key": 19, "value": 0}, ], "nan_value_counts": [ {"key": 16, "value": 0}, {"key": 17, "value": 0}, {"key": 18, "value": 0}, {"key": 19, "value": 0}, {"key": 10, "value": 0}, {"key": 11, "value": 0}, {"key": 12, "value": 0}, {"key": 13, "value": 0}, {"key": 14, "value": 0}, {"key": 15, "value": 0}, ], "lower_bounds": [ {"key": 2, "value": b"2020-04-01 00:00"}, {"key": 3, "value": b"2020-04-01 00:12"}, {"key": 7, "value": b"\x03\x00\x00\x00"}, {"key": 8, "value": b"\x01\x00\x00\x00"}, {"key": 10, "value": b"\xf6(\\\x8f\xc2\x05S\xc0"}, {"key": 11, "value": b"\x00\x00\x00\x00\x00\x00\x00\x00"}, {"key": 13, "value": b"\x00\x00\x00\x00\x00\x00\x00\x00"}, {"key": 14, "value": b"\x00\x00\x00\x00\x00\x00\xe0\xbf"}, {"key": 15, "value": b")\\\x8f\xc2\xf5(\x08\xc0"}, {"key": 16, "value": b"\x00\x00\x00\x00\x00\x00\x00\x00"}, {"key": 17, "value": b"\x00\x00\x00\x00\x00\x00\x00\x00"}, {"key": 18, "value": b"\xf6(\\\x8f\xc2\xc5S\xc0"}, {"key": 19, "value": b"\x00\x00\x00\x00\x00\x00\x04\xc0"}, ], "upper_bounds": [ {"key": 2, "value": b"2020-04-30 23:5:"}, {"key": 3, "value": b"2020-05-01 00:41"}, {"key": 7, "value": b"\t\x01\x00\x00"}, {"key": 8, "value": b"\t\x01\x00\x00"}, {"key": 10, "value": b"\xcd\xcc\xcc\xcc\xcc,_@"}, {"key": 11, "value": b"\x1f\x85\xebQ\\\xe2\xfe@"}, {"key": 13, "value": b"\x00\x00\x00\x00\x00\x00\x12@"}, {"key": 14, "value": b"\x00\x00\x00\x00\x00\x00\xe0?"}, {"key": 15, "value": b"q=\n\xd7\xa3\xf01@"}, {"key": 16, "value": b"\x00\x00\x00\x00\x00`B@"}, {"key": 17, "value": b"333333\xd3?"}, {"key": 18, "value": b"\x00\x00\x00\x00\x00\x18b@"}, {"key": 19, "value": b"\x00\x00\x00\x00\x00\x00\x04@"}, ], "key_metadata": None, "split_offsets": [4], "sort_order_id": 0, }, }, { "status": 1, "snapshot_id": 8744736658442914487, "data_file": { "file_path": "/home/iceberg/warehouse/nyc/taxis_partitioned/data/VendorID=1/00000-633-d8a4223e-dc97-45a1-86e1-adaba6e8abd7-00002.parquet", "file_format": "PARQUET", "partition": {"VendorID": 1}, "record_count": 95050, "file_size_in_bytes": 1265950, "block_size_in_bytes": 67108864, "column_sizes": [ {"key": 1, "value": 318}, {"key": 2, "value": 329806}, {"key": 3, "value": 331632}, {"key": 4, "value": 15343}, {"key": 5, "value": 2351}, {"key": 6, "value": 3389}, {"key": 7, "value": 71269}, {"key": 8, "value": 76429}, {"key": 9, "value": 16383}, {"key": 10, "value": 86992}, {"key": 11, "value": 89608}, {"key": 12, "value": 265}, {"key": 13, "value": 19377}, {"key": 14, "value": 1692}, {"key": 15, "value": 76162}, {"key": 16, "value": 4354}, {"key": 17, "value": 759}, {"key": 18, "value": 120650}, {"key": 19, "value": 11804}, ], "value_counts": [ {"key": 1, "value": 95050}, {"key": 2, "value": 95050}, {"key": 3, "value": 95050}, {"key": 4, "value": 95050}, {"key": 5, "value": 95050}, {"key": 6, "value": 95050}, {"key": 7, "value": 95050}, {"key": 8, "value": 95050}, {"key": 9, "value": 95050}, {"key": 10, "value": 95050}, {"key": 11, "value": 95050}, {"key": 12, "value": 95050}, {"key": 13, "value": 95050}, {"key": 14, "value": 95050}, {"key": 15, "value": 95050}, {"key": 16, "value": 95050}, {"key": 17, "value": 95050}, {"key": 18, "value": 95050}, {"key": 19, "value": 95050}, ], "null_value_counts": [ {"key": 1, "value": 0}, {"key": 2, "value": 0}, {"key": 3, "value": 0}, {"key": 4, "value": 0}, {"key": 5, "value": 0}, {"key": 6, "value": 0}, {"key": 7, "value": 0}, {"key": 8, "value": 0}, {"key": 9, "value": 0}, {"key": 10, "value": 0}, {"key": 11, "value": 0}, {"key": 12, "value": 95050}, {"key": 13, "value": 0}, {"key": 14, "value": 0}, {"key": 15, "value": 0}, {"key": 16, "value": 0}, {"key": 17, "value": 0}, {"key": 18, "value": 0}, {"key": 19, "value": 0}, ], "nan_value_counts": [ {"key": 16, "value": 0}, {"key": 17, "value": 0}, {"key": 18, "value": 0}, {"key": 19, "value": 0}, {"key": 10, "value": 0}, {"key": 11, "value": 0}, {"key": 12, "value": 0}, {"key": 13, "value": 0}, {"key": 14, "value": 0}, {"key": 15, "value": 0}, ], "lower_bounds": [ {"key": 1, "value": b"\x01\x00\x00\x00"}, {"key": 2, "value": b"2020-04-01 00:00"}, {"key": 3, "value": b"2020-04-01 00:03"}, {"key": 4, "value": b"\x00\x00\x00\x00"}, {"key": 5, "value": b"\x01\x00\x00\x00"}, {"key": 6, "value": b"N"}, {"key": 7, "value": b"\x01\x00\x00\x00"}, {"key": 8, "value": b"\x01\x00\x00\x00"}, {"key": 9, "value": b"\x01\x00\x00\x00"}, {"key": 10, "value": b"\x00\x00\x00\x00\x00\x00\x00\x00"}, {"key": 11, "value": b"\x00\x00\x00\x00\x00\x00\x00\x00"}, {"key": 13, "value": b"\x00\x00\x00\x00\x00\x00\x00\x00"}, {"key": 14, "value": b"\x00\x00\x00\x00\x00\x00\x00\x00"}, {"key": 15, "value": b"\x00\x00\x00\x00\x00\x00\x00\x00"}, {"key": 16, "value": b"\x00\x00\x00\x00\x00\x00\x00\x00"}, {"key": 17, "value": b"\x00\x00\x00\x00\x00\x00\x00\x00"}, {"key": 18, "value": b"\x00\x00\x00\x00\x00\x00\x00\x00"}, {"key": 19, "value": b"\x00\x00\x00\x00\x00\x00\x00\x00"}, ], "upper_bounds": [ {"key": 1, "value": b"\x01\x00\x00\x00"}, {"key": 2, "value": b"2020-04-30 23:5:"}, {"key": 3, "value": b"2020-05-01 00:1:"}, {"key": 4, "value": b"\x06\x00\x00\x00"}, {"key": 5, "value": b"c\x00\x00\x00"}, {"key": 6, "value": b"Y"}, {"key": 7, "value": b"\t\x01\x00\x00"}, {"key": 8, "value": b"\t\x01\x00\x00"}, {"key": 9, "value": b"\x04\x00\x00\x00"}, {"key": 10, "value": b"\\\x8f\xc2\xf5(8\x8c@"}, {"key": 11, "value": b"\xcd\xcc\xcc\xcc\xcc,f@"}, {"key": 13, "value": b"\x00\x00\x00\x00\x00\x00\x1c@"}, {"key": 14, "value": b"\x9a\x99\x99\x99\x99\x99\xf1?"}, {"key": 15, "value": b"\x00\x00\x00\x00\x00\x00Y@"}, {"key": 16, "value": b"\x00\x00\x00\x00\x00\xb0X@"}, {"key": 17, "value": b"333333\xd3?"}, {"key": 18, "value": b"\xc3\xf5(\\\x8f:\x8c@"}, {"key": 19, "value": b"\x00\x00\x00\x00\x00\x00\x04@"}, ], "key_metadata": None, "split_offsets": [4], "sort_order_id": 0, }, }, ] parsed_schema = parse_schema(avro_schema_manifest_entry) with TemporaryDirectory() as tmpdir: generated_manifest_entry_file = tmpdir + "/manifest.avro" with open(generated_manifest_entry_file, "wb") as out: writer(out, parsed_schema, manifest_entry_records) input_file = LocalInputFile(generated_manifest_entry_file) for _ in range(10000): list(read_manifest_entry(input_file)) ``` -- 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
