kevinjqliu commented on code in PR #2918: URL: https://github.com/apache/iceberg-python/pull/2918#discussion_r2717947622
########## pyiceberg/table/__init__.py: ########## @@ -33,7 +33,6 @@ ) from pydantic import Field -from sortedcontainers import SortedList Review Comment: https://github.com/apache/iceberg-python/blob/26ecfe77ffc9525e43a180c52b4ab49e76e66cb4/pyiceberg/table/update/snapshot.py#L798 last occurrence ########## pyiceberg/table/delete_file_index.py: ########## @@ -0,0 +1,143 @@ +# 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 __future__ import annotations + +from bisect import bisect_left +from typing import Any + +from pyiceberg.expressions import EqualTo +from pyiceberg.expressions.visitors import _InclusiveMetricsEvaluator +from pyiceberg.manifest import INITIAL_SEQUENCE_NUMBER, POSITIONAL_DELETE_SCHEMA, DataFile, ManifestEntry +from pyiceberg.typedef import Record + +PATH_FIELD_ID = 2147483546 + + +class PositionDeletes: + """Collects position delete files and indexes them by sequence number.""" + + __slots__ = ("_buffer", "_seqs", "_files") + + def __init__(self) -> None: + self._buffer: list[tuple[DataFile, int]] | None = [] + self._seqs: list[int] = [] + self._files: list[tuple[DataFile, int]] = [] + + def add(self, delete_file: DataFile, seq_num: int) -> None: + if self._buffer is None: + raise ValueError("Cannot add files after indexing") + self._buffer.append((delete_file, seq_num)) + + def _ensure_indexed(self) -> None: + if self._buffer is not None: + self._files = sorted(self._buffer, key=lambda file: file[1]) + self._seqs = [seq for _, seq in self._files] + self._buffer = None + + def filter_by_seq(self, seq: int) -> list[DataFile]: + self._ensure_indexed() + if not self._files: + return [] + start_idx = bisect_left(self._seqs, seq) Review Comment: note the old solution uses `bisect_right`, which might have been a bug ########## pyiceberg/table/delete_file_index.py: ########## @@ -0,0 +1,143 @@ +# 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 __future__ import annotations + +from bisect import bisect_left +from typing import Any + +from pyiceberg.expressions import EqualTo +from pyiceberg.expressions.visitors import _InclusiveMetricsEvaluator +from pyiceberg.manifest import INITIAL_SEQUENCE_NUMBER, POSITIONAL_DELETE_SCHEMA, DataFile, ManifestEntry +from pyiceberg.typedef import Record + +PATH_FIELD_ID = 2147483546 + + +class PositionDeletes: + """Collects position delete files and indexes them by sequence number.""" + + __slots__ = ("_buffer", "_seqs", "_files") + + def __init__(self) -> None: + self._buffer: list[tuple[DataFile, int]] | None = [] + self._seqs: list[int] = [] + self._files: list[tuple[DataFile, int]] = [] + + def add(self, delete_file: DataFile, seq_num: int) -> None: + if self._buffer is None: + raise ValueError("Cannot add files after indexing") + self._buffer.append((delete_file, seq_num)) + + def _ensure_indexed(self) -> None: + if self._buffer is not None: + self._files = sorted(self._buffer, key=lambda file: file[1]) + self._seqs = [seq for _, seq in self._files] + self._buffer = None + + def filter_by_seq(self, seq: int) -> list[DataFile]: + self._ensure_indexed() + if not self._files: + return [] + start_idx = bisect_left(self._seqs, seq) Review Comment: might be worth it to add an integration test scenario for this ########## tests/table/test_delete_file_index.py: ########## @@ -0,0 +1,192 @@ +# 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 pytest + +from pyiceberg.manifest import DataFile, DataFileContent, FileFormat, ManifestEntry, ManifestEntryStatus +from pyiceberg.table.delete_file_index import PATH_FIELD_ID, DeleteFileIndex, PositionDeletes +from pyiceberg.typedef import Record + + +def _create_data_file(file_path: str = "s3://bucket/data.parquet", spec_id: int = 0) -> DataFile: + data_file = DataFile.from_args( + content=DataFileContent.DATA, + file_path=file_path, + file_format=FileFormat.PARQUET, + partition=Record(), + record_count=100, + file_size_in_bytes=1000, + ) + data_file._spec_id = spec_id + return data_file + + +def _create_positional_delete( + sequence_number: int = 1, file_path: str = "s3://bucket/data.parquet", spec_id: int = 0 +) -> ManifestEntry: + delete_file = DataFile.from_args( + content=DataFileContent.POSITION_DELETES, + file_path=f"s3://bucket/pos-delete-{sequence_number}.parquet", + file_format=FileFormat.PARQUET, + partition=Record(), + record_count=10, + file_size_in_bytes=100, + lower_bounds={PATH_FIELD_ID: file_path.encode()}, + upper_bounds={PATH_FIELD_ID: file_path.encode()}, + ) + delete_file._spec_id = spec_id + return ManifestEntry.from_args(status=ManifestEntryStatus.ADDED, sequence_number=sequence_number, data_file=delete_file) + + +def _create_partition_delete(sequence_number: int = 1, spec_id: int = 0, partition: Record | None = None) -> ManifestEntry: + delete_file = DataFile.from_args( + content=DataFileContent.POSITION_DELETES, + file_path=f"s3://bucket/pos-delete-{sequence_number}.parquet", + file_format=FileFormat.PARQUET, + partition=partition or Record(), + record_count=10, + file_size_in_bytes=100, + ) + delete_file._spec_id = spec_id + return ManifestEntry.from_args(status=ManifestEntryStatus.ADDED, sequence_number=sequence_number, data_file=delete_file) + + +def _create_deletion_vector( + sequence_number: int = 1, file_path: str = "s3://bucket/data.parquet", spec_id: int = 0 +) -> ManifestEntry: + delete_file = DataFile.from_args( + content=DataFileContent.POSITION_DELETES, + file_path=f"s3://bucket/deletion-vector-{sequence_number}.puffin", + file_format=FileFormat.PUFFIN, + partition=Record(), + record_count=10, + file_size_in_bytes=100, + lower_bounds={PATH_FIELD_ID: file_path.encode()}, + upper_bounds={PATH_FIELD_ID: file_path.encode()}, + ) + delete_file._spec_id = spec_id + return ManifestEntry.from_args(status=ManifestEntryStatus.ADDED, sequence_number=sequence_number, data_file=delete_file) Review Comment: ha, might as well add it to the source code 😄 we can follow up with this ########## pyiceberg/table/delete_file_index.py: ########## @@ -0,0 +1,143 @@ +# 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 __future__ import annotations + +from bisect import bisect_left +from typing import Any + +from pyiceberg.expressions import EqualTo +from pyiceberg.expressions.visitors import _InclusiveMetricsEvaluator +from pyiceberg.manifest import INITIAL_SEQUENCE_NUMBER, POSITIONAL_DELETE_SCHEMA, DataFile, ManifestEntry +from pyiceberg.typedef import Record + +PATH_FIELD_ID = 2147483546 + + +class PositionDeletes: + """Collects position delete files and indexes them by sequence number.""" + + __slots__ = ("_buffer", "_seqs", "_files") + + def __init__(self) -> None: + self._buffer: list[tuple[DataFile, int]] | None = [] + self._seqs: list[int] = [] + self._files: list[tuple[DataFile, int]] = [] + + def add(self, delete_file: DataFile, seq_num: int) -> None: + if self._buffer is None: + raise ValueError("Cannot add files after indexing") + self._buffer.append((delete_file, seq_num)) + + def _ensure_indexed(self) -> None: + if self._buffer is not None: + self._files = sorted(self._buffer, key=lambda file: file[1]) + self._seqs = [seq for _, seq in self._files] + self._buffer = None + + def filter_by_seq(self, seq: int) -> list[DataFile]: + self._ensure_indexed() + if not self._files: + return [] + start_idx = bisect_left(self._seqs, seq) + return [delete_file for delete_file, _ in self._files[start_idx:]] + + +def _has_path_bounds(delete_file: DataFile) -> bool: + lower = delete_file.lower_bounds + upper = delete_file.upper_bounds + if not lower or not upper: + return False + + return PATH_FIELD_ID in lower and PATH_FIELD_ID in upper + + +def _applies_to_data_file(delete_file: DataFile, data_file: DataFile) -> bool: + if not _has_path_bounds(delete_file): + return True + + evaluator = _InclusiveMetricsEvaluator(POSITIONAL_DELETE_SCHEMA, EqualTo("file_path", data_file.file_path)) + return evaluator.eval(delete_file) + + +def _referenced_data_file_path(delete_file: DataFile) -> str | None: Review Comment: a little awkward that `delete_file` is of type `DataFile`, we can refactor this later perhaps -- 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]
