Python代码实现将多个PDF合并为单个文档并添加页码

pip install pypdf reportlab

import os
from pathlib import Path
from pypdf import PdfReader, PdfWriter, PdfMerger

# ==================== 方案 1：简单合并 ====================
def merge_pdfs_simple(pdf_files, output_file="合并结果.pdf"):
    """
    将多个 PDF 文件按顺序合并为一个

    参数:
        pdf_files: PDF 文件路径列表
        output_file: 输出文件名
    """
    merger = PdfMerger()

    for pdf_file in pdf_files:
        if os.path.exists(pdf_file):
            merger.append(pdf_file)
            print(f"已添加: {os.path.basename(pdf_file)}")
        else:
            print(f"[跳过] 文件不存在: {pdf_file}")

    merger.write(output_file)
    merger.close()
    print(f"\n合并完成: {output_file}")


# ==================== 方案 2：带书签的合并 ====================
def merge_pdfs_with_bookmarks(pdf_files, output_file="合并结果_带书签.pdf"):
    """
    合并 PDF 并添加书签（每个文件一个书签）
    """
    merger = PdfMerger()

    for pdf_file in pdf_files:
        if not os.path.exists(pdf_file):
            continue

        bookmark_name = Path(pdf_file).stem
        merger.append(pdf_file)

        # 获取此文件在合并文件中的起始页码
        total_pages = sum(PdfReader(f).get_num_pages() for f in pdf_files[:pdf_files.index(pdf_file) + 1] if os.path.exists(f))
        start_page = total_pages - PdfReader(pdf_file).get_num_pages()

        # 添加书签
        merger.add_outline_item(bookmark_name, start_page)
        print(f"已添加书签: {bookmark_name} (第 {start_page + 1} 页)")

    merger.write(output_file)
    merger.close()
    print(f"\n合并完成: {output_file}")


# ==================== 方案 3：批量合并文件夹中的所有 PDF ====================
def merge_folder_pdfs(folder_path, output_file="合并结果.pdf", sorted_by='name'):
    """
    自动合并文件夹中的所有 PDF

    参数:
        folder_path: PDF 文件所在文件夹
        output_file: 输出文件名
        sorted_by: 排序方式 'name'(文件名) / 'date'(修改日期)
    """
    folder = Path(folder_path)
    pdf_files = sorted(
        folder.glob('*.pdf'),
        key=lambda f: f.stat().st_mtime if sorted_by == 'date' else f.name
    )

    if not pdf_files:
        print(f"文件夹 {folder_path} 中没有 PDF 文件！")
        return

    print(f"找到 {len(pdf_files)} 个 PDF 文件:")
    for f in pdf_files:
        print(f"  - {f.name}")

    merge_pdfs_with_bookmarks(
        [str(f) for f in pdf_files],
        output_file
    )


# ==================== 方案 4：添加页码 ====================
def add_page_numbers(input_pdf, output_file="带页码版本.pdf"):
    """
    为 PDF 添加页码（每页底部居中）
    """
    from reportlab.lib.pagesizes import A4
    from reportlab.pdfgen import canvas
    import io

    reader = PdfReader(input_pdf)
    writer = PdfWriter()

    total_pages = len(reader.pages)

    for i, page in enumerate(reader.pages):
        # 创建页码水印
        packet = io.BytesIO()
        c = canvas.Canvas(packet, pagesize=A4)
        width, height = A4

        # 绘制页码
        c.setFont("Helvetica", 9)
        page_number = f"- {i + 1} -"
        text_width = c.stringWidth(page_number, "Helvetica", 9)
        c.drawString((width - text_width) / 2, 30, page_number)

        c.save()
        packet.seek(0)

        # 叠加水印到原页面
        watermark = PdfReader(packet)
        page.merge_page(watermark.pages[0])
        writer.add_page(page)

    with open(output_file, 'wb') as f:
        writer.write(f)

    print(f"页码已添加: {output_file}")


# ==================== 使用示例 ====================
if __name__ == "__main__":
    # 方案 1：手动指定文件列表
    # pdf_files = [
    #     "第1章.pdf",
    #     "第2章.pdf",
    #     "第3章.pdf",
    # ]
    # merge_pdfs_simple(pdf_files, "完整手册.pdf")

    # 方案 2：批量合并文件夹
    merge_folder_pdfs(
        r"D:\部门报告",
        output_file="年度总报告.pdf",
        sorted_by='name'
    )

    # 方案 3：添加页码
    # add_page_numbers("年度总报告.pdf", "年度总报告_带页码.pdf")

from reportlab.pdfbase import pdfmetrics
from reportlab.pdfbase.ttfonts import TTFont

pdfmetrics.registerFont(TTFont('SimSun', 'C:/Windows/Fonts/simsun.ttc'))
c.setFont('SimSun', 9)

pip install pypdf

reader = PdfReader("encrypted.pdf")
if reader.is_encrypted:
    reader.decrypt("password")