28 lines
754 B
Python
28 lines
754 B
Python
"""Base protocol and data structures for text extraction."""
|
|
|
|
from dataclasses import dataclass, field
|
|
from pathlib import Path
|
|
from typing import Any, Protocol
|
|
|
|
|
|
@dataclass
|
|
class ExtractionResult:
|
|
"""Result of extracting text from a document."""
|
|
|
|
text: str
|
|
source_file: Path
|
|
extraction_method: str # "direct", "ocr", "mixed"
|
|
metadata: dict[str, Any] = field(default_factory=dict)
|
|
|
|
|
|
class TextExtractor(Protocol):
|
|
"""Protocol that all text extractors must implement."""
|
|
|
|
def can_handle(self, file_path: Path) -> bool:
|
|
"""Return True if this extractor can process the given file."""
|
|
...
|
|
|
|
def extract(self, file_path: Path) -> ExtractionResult:
|
|
"""Extract text content from the file."""
|
|
...
|