pdf2text.py

import os
import pytesseract
from PyPDF2 import PdfReader
from pdf2image import convert_from_path
from typing import Literal

def extract_text_simple(pdf_path: str) -> str:
    """
    使用 PyPDF2 解析 PDF 純文字
    """
    reader = PdfReader(pdf_path)
    all_text = []
    for page in reader.pages:
        text = page.extract_text()
        if text:
            all_text.append(text.strip())
    return "\n".join(all_text)

def extract_text_ocr(pdf_path: str, dpi: int = 300) -> str:
    """
    使用 Tesseract OCR 提取圖片形式的 PDF 內容
    """
    images = convert_from_path(pdf_path, dpi=dpi)
    all_text = []
    for img in images:
        text = pytesseract.image_to_string(img, lang="chi_tra+eng")
        if text:
            all_text.append(text.strip())
    return "\n".join(all_text)

def extract_text(pdf_path: str, mode: Literal["simple", "ocr"] = "simple") -> str:
    """
    根據模式選擇提取方法
    """
    if mode == "ocr":
        return extract_text_ocr(pdf_path)
    else:
        return extract_text_simple(pdf_path)

# 為 app.py 提供相容介面
def convert_PDF_to_Text(pdf_path: str, ocr_model=None, max_pages: int = 20) -> dict:
    """
    模擬 app.py 所需的 convert_PDF_to_Text 介面
    """
    text = extract_text(pdf_path, mode="ocr" if ocr_model else "simple")
    return {
        "converted_text": text,
        "source_path": pdf_path,
        "used_ocr": bool(ocr_model),
        "page_count": "N/A",
    }

convert_pdf_to_text = convert_PDF_to_Text