3行Python代码快速实现对比两份PDF文档差异

pip install Spire.Pdf.Free

from spire.pdf.common import *
from spire.pdf import *

from spire.pdf.common import *
from spire.pdf import *

def compare_pdf(base_path: str, revised_path: str, output_path: str) -> None:
    """
    对比两个 PDF 文档，生成带高亮标记的差异报告
    
    Args:
        base_path:  基准文档路径（作为参照版本）
        revised_path: 待比对文档路径（修订版本）
        output_path: 对比结果 PDF 的输出路径
    """
    # 加载两个 PDF 文档
    base_doc = PdfDocument(base_path)
    revised_doc = PdfDocument(revised_path)
    
    try:
        # 创建对比器，第一个参数为基准，第二个为待比较版本
        comparer = PdfComparer(base_doc, revised_doc)
        # 执行对比并保存结果
        comparer.Compare(output_path)
        print(f"对比完成，结果已保存至: {output_path}")
    finally:
        # 显式释放资源，避免内存泄漏
        base_doc.Dispose()
        revised_doc.Dispose()

# 使用示例
if __name__ == "__main__":
    compare_pdf("document1.pdf", "document2.pdf", "comparison_result.pdf")

comparer = PdfComparer(base_doc, revised_doc)
# 开启仅文本对比，忽略排版和图形差异
comparer.PdfCompareOptions.OnlyCompareText
comparer.Compare(output_path)

comparer = PdfComparer(base_doc, revised_doc)
# 参数依次为：基准文档起始页、结束页，待比对文档起始页、结束页（索引均从 0 开始）
comparer.PdfCompareOptions.SetPageRanges(0, 2, 0, 2)
comparer.Compare(output_path)

pip install PyPDF2

pip install pdfplumber

import pdfplumber
import difflib
import sys
def extract_text(pdf_path):
    """提取 PDF 所有页面的文本"""
    with pdfplumber.open(pdf_path) as pdf:
        text = ""
        for page in pdf.pages:
            text += page.extract_text() or ""
    return text
def compare_pdf_text(pdf1, pdf2, output_html="diff_report.html"):
    """对比两个 PDF 的文本差异，生成 HTML 报告"""
    text1 = extract_text(pdf1)
    text2 = extract_text(pdf2)
    # 按行分割
    lines1 = text1.splitlines()
    lines2 = text2.splitlines()
    # 生成差异 HTML
    diff = difflib.HtmlDiff(wrapcolumn=80)
    html = diff.make_file(lines1, lines2, fromdesc=pdf1, todesc=pdf2)
    with open(output_html, "w", encoding="utf-8") as f:
        f.write(html)
    print(f"差异报告已生成：{output_html}")
    # 也可简单打印差异摘要
    differ = difflib.Differ()
    diff_lines = list(differ.compare(lines1, lines2))
    added = sum(1 for line in diff_lines if line.startswith('+ '))
    removed = sum(1 for line in diff_lines if line.startswith('- '))
    print(f"新增行数：{added}，删除行数：{removed}")
if __name__ == "__main__":
    compare_pdf_text("old.pdf", "new.pdf")

pip install pdf2image pillow numpy opencv-python

import cv2
import numpy as np
from pdf2image import convert_from_path
from PIL import Image, ImageChops
def compare_pdf_images(pdf1, pdf2, output_diff="diff_image.png"):
    """
    逐页对比两个 PDF 的图像，输出第一处差异的标注图像
    """
    # 转换为图像列表（分辨率可调）
    images1 = convert_from_path(pdf1, dpi=150)
    images2 = convert_from_path(pdf2, dpi=150)
    # 确保页数相同
    if len(images1) != len(images2):
        print(f"页数不同：{len(images1)} vs {len(images2)}，无法逐页对比")
        return
    for i, (img1, img2) in enumerate(zip(images1, images2)):
        # 转换为 numpy 数组（OpenCV 格式）
        np_img1 = np.array(img1)
        np_img2 = np.array(img2)
        # 计算像素差异
        diff = cv2.absdiff(np_img1, np_img2)
        diff_gray = cv2.cvtColor(diff, cv2.COLOR_BGR2GRAY)
        _, thresh = cv2.threshold(diff_gray, 30, 255, cv2.THRESH_BINARY)
        # 如果有差异，标记并保存
        if np.sum(thresh) > 0:
            # 在差异区域画框
            contours, _ = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
            for cnt in contours:
                x, y, w, h = cv2.boundingRect(cnt)
                cv2.rectangle(np_img1, (x, y), (x+w, y+h), (0, 0, 255), 2)
            # 保存标注后的图像
            result_img = Image.fromarray(np_img1)
            result_img.save(f"diff_page_{i+1}.png")
            print(f"第 {i+1} 页存在差异，已保存标注图像 diff_page_{i+1}.png")
            break
    else:
        print("所有页面完全相同")
if __name__ == "__main__":
    compare_pdf_images("old.pdf", "new.pdf")

# 安装 comparepdf（Ubuntu）
sudo apt-get install comparepdf
comparepdf --diff=diff.pdf old.pdf new.pdf