fix: prevent timeout in file encoding detection for large files (#21453)
Co-authored-by: crazywoola <427733928@qq.com>
This commit is contained in:
parent
ed54bd5121
commit
1c7404099d
2 changed files with 12 additions and 4 deletions
|
|
@ -1,7 +1,6 @@
|
||||||
"""Document loader helpers."""
|
"""Document loader helpers."""
|
||||||
|
|
||||||
import concurrent.futures
|
import concurrent.futures
|
||||||
from pathlib import Path
|
|
||||||
from typing import NamedTuple, Optional, cast
|
from typing import NamedTuple, Optional, cast
|
||||||
|
|
||||||
|
|
||||||
|
|
@ -16,7 +15,7 @@ class FileEncoding(NamedTuple):
|
||||||
"""The language of the file."""
|
"""The language of the file."""
|
||||||
|
|
||||||
|
|
||||||
def detect_file_encodings(file_path: str, timeout: int = 5) -> list[FileEncoding]:
|
def detect_file_encodings(file_path: str, timeout: int = 5, sample_size: int = 1024 * 1024) -> list[FileEncoding]:
|
||||||
"""Try to detect the file encoding.
|
"""Try to detect the file encoding.
|
||||||
|
|
||||||
Returns a list of `FileEncoding` tuples with the detected encodings ordered
|
Returns a list of `FileEncoding` tuples with the detected encodings ordered
|
||||||
|
|
@ -25,11 +24,16 @@ def detect_file_encodings(file_path: str, timeout: int = 5) -> list[FileEncoding
|
||||||
Args:
|
Args:
|
||||||
file_path: The path to the file to detect the encoding for.
|
file_path: The path to the file to detect the encoding for.
|
||||||
timeout: The timeout in seconds for the encoding detection.
|
timeout: The timeout in seconds for the encoding detection.
|
||||||
|
sample_size: The number of bytes to read for encoding detection. Default is 1MB.
|
||||||
|
For large files, reading only a sample is sufficient and prevents timeout.
|
||||||
"""
|
"""
|
||||||
import chardet
|
import chardet
|
||||||
|
|
||||||
def read_and_detect(file_path: str) -> list[dict]:
|
def read_and_detect(file_path: str) -> list[dict]:
|
||||||
rawdata = Path(file_path).read_bytes()
|
with open(file_path, "rb") as f:
|
||||||
|
# Read only a sample of the file for encoding detection
|
||||||
|
# This prevents timeout on large files while still providing accurate encoding detection
|
||||||
|
rawdata = f.read(sample_size)
|
||||||
return cast(list[dict], chardet.detect_all(rawdata))
|
return cast(list[dict], chardet.detect_all(rawdata))
|
||||||
|
|
||||||
with concurrent.futures.ThreadPoolExecutor() as executor:
|
with concurrent.futures.ThreadPoolExecutor() as executor:
|
||||||
|
|
|
||||||
|
|
@ -36,8 +36,12 @@ class TextExtractor(BaseExtractor):
|
||||||
break
|
break
|
||||||
except UnicodeDecodeError:
|
except UnicodeDecodeError:
|
||||||
continue
|
continue
|
||||||
|
else:
|
||||||
|
raise RuntimeError(
|
||||||
|
f"Decode failed: {self._file_path}, all detected encodings failed. Original error: {e}"
|
||||||
|
)
|
||||||
else:
|
else:
|
||||||
raise RuntimeError(f"Error loading {self._file_path}") from e
|
raise RuntimeError(f"Decode failed: {self._file_path}, specified encoding failed. Original error: {e}")
|
||||||
except Exception as e:
|
except Exception as e:
|
||||||
raise RuntimeError(f"Error loading {self._file_path}") from e
|
raise RuntimeError(f"Error loading {self._file_path}") from e
|
||||||
|
|
||||||
|
|
|
||||||
Loading…
Add table
Add a link
Reference in a new issue