Spaces:
Sleeping
Sleeping
File size: 3,957 Bytes
7d134e4 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 |
# coding=utf-8
# Copyright 2019-present, the HuggingFace Inc. team.
#
# Licensed 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.
"""Git LFS related utilities"""
import io
import os
from contextlib import AbstractContextManager
from typing import BinaryIO
class SliceFileObj(AbstractContextManager):
"""
Utility context manager to read a *slice* of a seekable file-like object as a seekable, file-like object.
This is NOT thread safe
Inspired by stackoverflow.com/a/29838711/593036
Credits to @julien-c
Args:
fileobj (`BinaryIO`):
A file-like object to slice. MUST implement `tell()` and `seek()` (and `read()` of course).
`fileobj` will be reset to its original position when exiting the context manager.
seek_from (`int`):
The start of the slice (offset from position 0 in bytes).
read_limit (`int`):
The maximum number of bytes to read from the slice.
Attributes:
previous_position (`int`):
The previous position
Examples:
Reading 200 bytes with an offset of 128 bytes from a file (ie bytes 128 to 327):
```python
>>> with open("path/to/file", "rb") as file:
... with SliceFileObj(file, seek_from=128, read_limit=200) as fslice:
... fslice.read(...)
```
Reading a file in chunks of 512 bytes
```python
>>> import os
>>> chunk_size = 512
>>> file_size = os.getsize("path/to/file")
>>> with open("path/to/file", "rb") as file:
... for chunk_idx in range(ceil(file_size / chunk_size)):
... with SliceFileObj(file, seek_from=chunk_idx * chunk_size, read_limit=chunk_size) as fslice:
... chunk = fslice.read(...)
```
"""
def __init__(self, fileobj: BinaryIO, seek_from: int, read_limit: int):
self.fileobj = fileobj
self.seek_from = seek_from
self.read_limit = read_limit
def __enter__(self):
self._previous_position = self.fileobj.tell()
end_of_stream = self.fileobj.seek(0, os.SEEK_END)
self._len = min(self.read_limit, end_of_stream - self.seek_from)
# ^^ The actual number of bytes that can be read from the slice
self.fileobj.seek(self.seek_from, io.SEEK_SET)
return self
def __exit__(self, exc_type, exc_value, traceback):
self.fileobj.seek(self._previous_position, io.SEEK_SET)
def read(self, n: int = -1):
pos = self.tell()
if pos >= self._len:
return b""
remaining_amount = self._len - pos
data = self.fileobj.read(remaining_amount if n < 0 else min(n, remaining_amount))
return data
def tell(self) -> int:
return self.fileobj.tell() - self.seek_from
def seek(self, offset: int, whence: int = os.SEEK_SET) -> int:
start = self.seek_from
end = start + self._len
if whence in (os.SEEK_SET, os.SEEK_END):
offset = start + offset if whence == os.SEEK_SET else end + offset
offset = max(start, min(offset, end))
whence = os.SEEK_SET
elif whence == os.SEEK_CUR:
cur_pos = self.fileobj.tell()
offset = max(start - cur_pos, min(offset, end - cur_pos))
else:
raise ValueError(f"whence value {whence} is not supported")
return self.fileobj.seek(offset, whence) - self.seek_from
def __iter__(self):
yield self.read(n=4 * 1024 * 1024)
|