Skip to content

Commit 5092092

Browse files
authored
Integrate file_lock inside the lib for better logging control (#859)
* integrate file_lock in the lib for better logging control * remove file_lock from setup.py * style * minor
1 parent 3d1564e commit 5092092

File tree

6 files changed

+451
-8
lines changed

6 files changed

+451
-8
lines changed

setup.py

Lines changed: 0 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -78,8 +78,6 @@
7878
"tqdm>=4.27,<4.50.0",
7979
# dataclasses for Python versions that don't have it
8080
"dataclasses;python_version<'3.7'",
81-
# filesystem locks e.g. to prevent parallel downloads
82-
"filelock",
8381
# for fast hashing
8482
"xxhash",
8583
# for better multiprocessing

src/datasets/builder.py

Lines changed: 1 addition & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -26,8 +26,6 @@
2626
from functools import partial
2727
from typing import Dict, List, Optional, Union
2828

29-
from filelock import FileLock
30-
3129
from . import utils
3230
from .arrow_dataset import Dataset
3331
from .arrow_reader import HF_GCP_BASE_URL, ArrowReader, DatasetNotOnHfGcs, MissingFilesOnHfGcs
@@ -46,6 +44,7 @@
4644
from .splits import Split, SplitDict, SplitGenerator
4745
from .utils.download_manager import DownloadManager, GenerateMode
4846
from .utils.file_utils import HF_DATASETS_CACHE, DownloadConfig, is_remote_url
47+
from .utils.filelock import FileLock
4948
from .utils.info_utils import get_size_checksum_dict, verify_checksums, verify_splits
5049
from .utils.logging import WARNING, get_logger
5150

src/datasets/load.py

Lines changed: 1 addition & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -27,8 +27,6 @@
2727
from typing import Dict, List, Optional, Tuple, Union
2828
from urllib.parse import urlparse
2929

30-
from filelock import FileLock
31-
3230
from .arrow_dataset import Dataset
3331
from .builder import DatasetBuilder
3432
from .dataset_dict import DatasetDict
@@ -38,6 +36,7 @@
3836
from .splits import Split
3937
from .utils.download_manager import GenerateMode
4038
from .utils.file_utils import HF_MODULES_CACHE, DownloadConfig, cached_path, head_hf_s3, hf_bucket_url, hf_github_url
39+
from .utils.filelock import FileLock
4140
from .utils.logging import get_logger
4241
from .utils.version import Version
4342

src/datasets/metric.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -22,7 +22,6 @@
2222

2323
import numpy as np
2424
import pyarrow as pa
25-
from filelock import BaseFileLock, FileLock, Timeout
2625

2726
from .arrow_dataset import Dataset
2827
from .arrow_reader import ArrowReader
@@ -33,6 +32,7 @@
3332
from .utils import HF_METRICS_CACHE, copyfunc, temp_seed
3433
from .utils.download_manager import DownloadManager
3534
from .utils.file_utils import DownloadConfig
35+
from .utils.filelock import BaseFileLock, FileLock, Timeout
3636
from .utils.logging import get_logger
3737

3838

src/datasets/utils/file_utils.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -22,10 +22,10 @@
2222

2323
import numpy as np
2424
import requests
25-
from filelock import FileLock
2625
from tqdm.auto import tqdm
2726

2827
from .. import __version__
28+
from .filelock import FileLock
2929
from .logging import WARNING, get_logger
3030

3131

0 commit comments

Comments
 (0)