pip install pypdf2 pdfplumber

import PyPDF2

def extract_text_with_pypdf2(pdf_path):
    """使用PyPDF2提取PDF文本"""
    with open(pdf_path, 'rb') as file:
        # 創(chuàng)建PDF閱讀器對(duì)象
        reader = PyPDF2.PdfReader(file)
        
        # 獲取總頁數(shù)
        num_pages = len(reader.pages)
        print(f"總頁數(shù): {num_pages}")
        
        # 提取每一頁文本
        full_text = ""
        for page_num in range(num_pages):
            page = reader.pages[page_num]
            text = page.extract_text()
            full_text += f"\n--- 第 {page_num + 1} 頁 ---\n{text}"
        
        return full_text

# 使用示例
text = extract_text_with_pypdf2("report.pdf")
print(text[:500])  # 打印前500個(gè)字符

import pdfplumber

def extract_text_with_pdfplumber(pdf_path):
    """使用pdfplumber提取PDF文本，保留布局"""
    full_text = ""
    with pdfplumber.open(pdf_path) as pdf:
        print(f"總頁數(shù): {len(pdf.pages)}")
        
        for i, page in enumerate(pdf.pages):
            # 提取文本（自動(dòng)處理布局）
            text = page.extract_text()
            full_text += f"\n--- 第 {i+1} 頁 ---\n{text}"
            
            # 可選：提取單詞及其位置信息
            words = page.extract_words()
            if words:
                print(f"第{i+1}頁包含{len(words)}個(gè)單詞")
                # 第一個(gè)單詞的坐標(biāo)信息示例
                first_word = words[0]
                print(f"首個(gè)單詞: '{first_word['text']}' 位置: ({first_word['x0']}, {first_word['top']})")
    
    return full_text

# 使用示例
text = extract_text_with_pdfplumber("復(fù)雜報(bào)表.pdf")

import pdfplumber
import pandas as pd

def extract_tables_to_dataframe(pdf_path):
    """提取PDF中的表格并轉(zhuǎn)為DataFrame"""
    all_tables = []
    
    with pdfplumber.open(pdf_path) as pdf:
        for page_num, page in enumerate(pdf.pages):
            # 提取當(dāng)前頁的所有表格
            tables = page.extract_tables()
            
            for table_num, table in enumerate(tables):
                print(f"第{page_num+1}頁，表格{table_num+1}，共{len(table)}行")
                
                # 轉(zhuǎn)為DataFrame（假設(shè)第一行為表頭）
                if len(table) > 0:
                    df = pd.DataFrame(table[1:], columns=table[0])
                    all_tables.append({
                        'page': page_num + 1,
                        'table_num': table_num + 1,
                        'dataframe': df
                    })
    
    return all_tables

# 優(yōu)化表格提取的參數(shù)設(shè)置
def extract_tables_with_settings(pdf_path):
    """使用自定義參數(shù)優(yōu)化表格提取"""
    results = []
    
    with pdfplumber.open(pdf_path) as pdf:
        for page in pdf.pages:
            # 自定義表格檢測策略
            table_settings = {
                "vertical_strategy": "lines",  # 基于線條檢測列
                "horizontal_strategy": "lines",  # 基于線條檢測行
                "snap_tolerance": 5,  # 線條對(duì)齊容差
                "edge_min_length": 3,  # 最小線段長度
            }
            
            # 或者使用"text"策略（當(dāng)表格沒有明顯邊框時(shí)）
            # table_settings = {
            #     "vertical_strategy": "text",
            #     "horizontal_strategy": "text",
            # }
            
            table = page.extract_table(table_settings)
            if table:
                results.append(table)
    
    return results

# 使用示例
tables = extract_tables_to_dataframe("財(cái)務(wù)報(bào)表.pdf")
for item in tables:
    print(item['dataframe'].head())

import pdfplumber
import pytesseract
from PIL import Image

def ocr_scanned_pdf(pdf_path):
    """對(duì)掃描件PDF進(jìn)行OCR識(shí)別"""
    text_results = []
    
    with pdfplumber.open(pdf_path) as pdf:
        for i, page in enumerate(pdf.pages):
            # 將PDF頁面轉(zhuǎn)為圖像
            im = page.to_image(resolution=300)
            
            # 使用pytesseract進(jìn)行OCR
            text = pytesseract.image_to_string(im.original, lang='chi_sim+eng')  # 中英文
            text_results.append({
                'page': i + 1,
                'text': text
            })
    
    return text_results

import os
from PyPDF2 import PdfMerger

def merge_pdfs(input_folder, output_filename, pattern=None):
    """
    合并指定文件夾中的所有PDF文件
    
    Args:
        input_folder: 包含PDF文件的文件夾路徑
        output_filename: 輸出文件名
        pattern: 文件名過濾模式，如"report_"開頭的文件
    """
    merger = PdfMerger()
    merged_count = 0
    
    # 獲取文件夾下所有PDF文件
    for filename in os.listdir(input_folder):
        if filename.lower().endswith('.pdf'):
            if pattern and pattern not in filename:
                continue
                
        file_path = os.path.join(input_folder, filename)
        print(f"正在添加: {filename}")
        
        try:
            merger.append(file_path)
            merged_count += 1
        except Exception as e:
            print(f"添加文件 {filename} 時(shí)出錯(cuò): {e}")
    
    # 寫入合并后的文件
    if merged_count > 0:
        merger.write(output_filename)
        merger.close()
        print(f"合并完成！共合并 {merged_count} 個(gè)文件，保存為: {output_filename}")
    else:
        print("未找到可合并的PDF文件")

# 按指定順序合并（指定文件列表）
def merge_pdfs_by_list(file_list, output_filename):
    """按指定文件列表順序合并PDF"""
    merger = PdfMerger()
    
    for file_path in file_list:
        if os.path.exists(file_path):
            print(f"添加: {os.path.basename(file_path)}")
            merger.append(file_path)
        else:
            print(f"文件不存在: {file_path}")
    
    merger.write(output_filename)
    merger.close()
    print(f"合并完成！保存為: {output_filename}")

# 使用示例
merge_pdfs('./月度報(bào)告', '2025年一季度合并報(bào)告.pdf')
merge_pdfs_by_list(['封面.pdf', '目錄.pdf', '正文.pdf', '附錄.pdf'], '完整報(bào)告.pdf')

from PyPDF2 import PdfReader, PdfWriter

def split_pdf_by_pages(input_pdf, output_pattern, pages_to_extract):
    """
    提取指定頁面并保存為新文件
    
    Args:
        input_pdf: 輸入PDF路徑
        output_pattern: 輸出文件名模式，如"提取_第{page}頁.pdf"
        pages_to_extract: 要提取的頁碼列表（從1開始）
    """
    reader = PdfReader(input_pdf)
    total_pages = len(reader.pages)
    
    for page_num in pages_to_extract:
        if page_num < 1 or page_num > total_pages:
            print(f"頁碼 {page_num} 超出范圍（1-{total_pages}）")
            continue
        
        writer = PdfWriter()
        # 注意：頁碼從0開始索引
        writer.add_page(reader.pages[page_num - 1])
        
        output_file = output_pattern.format(page=page_num)
        with open(output_file, 'wb') as output_pdf:
            writer.write(output_pdf)
        
        print(f"已提取第 {page_num} 頁，保存為: {output_file}")

def split_pdf_every_n_pages(input_pdf, n):
    """每n頁拆分成一個(gè)文件（如每10頁一個(gè)文件）"""
    reader = PdfReader(input_pdf)
    total_pages = len(reader.pages)
    
    for start in range(0, total_pages, n):
        end = min(start + n, total_pages)
        writer = PdfWriter()
        
        for page_num in range(start, end):
            writer.add_page(reader.pages[page_num])
        
        output_file = f"拆分_{start+1}-{end}.pdf"
        with open(output_file, 'wb') as output_pdf:
            writer.write(output_pdf)
        
        print(f"已生成: {output_file} (第{start+1}-{end}頁)")

# 使用示例
split_pdf_by_pages('年度報(bào)告.pdf', '第{page}頁_摘要.pdf', [1, 5, 10])
split_pdf_every_n_pages('超大文檔.pdf', 20)

import pdfplumber
from PyPDF2 import PdfReader, PdfWriter

def split_pdf_by_keyword(input_pdf, keyword, output_prefix):
    """
    根據(jù)關(guān)鍵字出現(xiàn)的頁碼拆分PDF
    找到關(guān)鍵字出現(xiàn)的所有頁碼，每個(gè)關(guān)鍵頁及其后續(xù)頁直到下一個(gè)關(guān)鍵頁之前，作為一個(gè)獨(dú)立文件
    """
    # 第一步：找出所有包含關(guān)鍵字的頁碼
    keyword_pages = []
    
    with pdfplumber.open(input_pdf) as pdf:
        for i, page in enumerate(pdf.pages):
            text = page.extract_text()
            if keyword in text:
                keyword_pages.append(i)  # 保存0-based索引
                print(f"第{i+1}頁包含關(guān)鍵字: {keyword}")
    
    if not keyword_pages:
        print(f"未找到包含關(guān)鍵字'{keyword}'的頁面")
        return
    
    # 第二步：根據(jù)關(guān)鍵頁拆分
    reader = PdfReader(input_pdf)
    total_pages = len(reader.pages)
    
    for idx, start_page in enumerate(keyword_pages):
        writer = PdfWriter()
        
        # 確定結(jié)束頁：下一個(gè)關(guān)鍵頁的前一頁，或最后一頁
        end_page = keyword_pages[idx + 1] - 1 if idx + 1 < len(keyword_pages) else total_pages - 1
        
        # 添加從start_page到end_page的所有頁
        for page_num in range(start_page, end_page + 1):
            writer.add_page(reader.pages[page_num])
        
        # 保存文件
        output_file = f"{output_prefix}_第{idx+1}段_頁{start_page+1}-{end_page+1}.pdf"
        with open(output_file, 'wb') as output_pdf:
            writer.write(output_pdf)
        
        print(f"已生成: {output_file}")

# 使用示例：按"第一章"、"第二章"拆分書籍
split_pdf_by_keyword('Python教程.pdf', '第', '章節(jié)')

from reportlab.pdfgen import canvas
from reportlab.lib.pagesizes import letter
from PyPDF2 import PdfReader, PdfWriter

def create_watermark_pdf(watermark_text, output_path, page_size=letter):
    """
    創(chuàng)建水印PDF文件
    """
    c = canvas.Canvas(output_path, pagesize=page_size)
    width, height = page_size
    
    # 設(shè)置透明度和旋轉(zhuǎn)
    c.setFillAlpha(0.3)  # 透明度30%
    c.setFont("Helvetica", 60)
    c.setFillColorRGB(0.5, 0.5, 0.5)  # 灰色
    
    # 旋轉(zhuǎn)45度，居中顯示
    c.saveState()
    c.translate(width/2, height/2)
    c.rotate(45)
    c.drawCentredString(0, 0, watermark_text)
    c.restore()
    
    c.save()
    print(f"水印文件已創(chuàng)建: {output_path}")

def add_watermark_to_pdf(input_pdf, watermark_pdf, output_pdf):
    """
    為PDF文件添加水印
    """
    # 讀取水印PDF
    watermark_reader = PdfReader(watermark_pdf)
    watermark_page = watermark_reader.pages[0]
    
    # 讀取原PDF
    reader = PdfReader(input_pdf)
    writer = PdfWriter()
    
    # 為每一頁添加水印
    for page_num in range(len(reader.pages)):
        page = reader.pages[page_num]
        page.merge_page(watermark_page)  # 合并水印
        writer.add_page(page)
    
    # 保存結(jié)果
    with open(output_pdf, 'wb') as output_file:
        writer.write(output_file)
    
    print(f"水印添加完成！保存為: {output_pdf}")

# 支持中文的水印
def create_chinese_watermark(watermark_text, output_path, font_path='simsun.ttc'):
    """
    創(chuàng)建支持中文的水印PDF
    """
    from reportlab.pdfbase import pdfmetrics
    from reportlab.pdfbase.ttfonts import TTFont
    
    # 注冊(cè)中文字體
    pdfmetrics.registerFont(TTFont('SimSun', font_path))
    
    c = canvas.Canvas(output_path, pagesize=letter)
    width, height = letter
    
    c.setFillAlpha(0.2)
    c.setFont('SimSun', 50)
    c.setFillColorRGB(0.7, 0.7, 0.7)
    
    # 居中旋轉(zhuǎn)
    c.saveState()
    c.translate(width/2, height/2)
    c.rotate(45)
    c.drawCentredString(0, 0, watermark_text)
    c.restore()
    
    c.save()

# 使用示例
create_watermark_pdf("CONFIDENTIAL", "watermark.pdf")
add_watermark_to_pdf("report.pdf", "watermark.pdf", "report_confidential.pdf")

create_chinese_watermark("絕密", "chinese_watermark.pdf")
add_watermark_to_pdf("合同.pdf", "chinese_watermark.pdf", "合同_帶水印.pdf")

from PyPDF2 import PdfReader, PdfWriter

def rotate_pdf_pages(input_pdf, output_pdf, rotation_degree=90, pages=None):
    """
    旋轉(zhuǎn)PDF頁面
    pages: 要旋轉(zhuǎn)的頁碼列表，None表示所有頁
    """
    reader = PdfReader(input_pdf)
    writer = PdfWriter()
    
    for page_num in range(len(reader.pages)):
        page = reader.pages[page_num]
        
        if pages is None or (page_num + 1) in pages:
            page.rotate(rotation_degree)
        
        writer.add_page(page)
    
    with open(output_pdf, 'wb') as output_file:
        writer.write(output_file)
    
    print(f"旋轉(zhuǎn)完成！保存為: {output_pdf}")

# 使用示例：將第2頁旋轉(zhuǎn)90度
rotate_pdf_pages("掃描件.pdf", "掃描件_校正.pdf", rotation_degree=90, pages=[2])

from PyPDF2 import PdfReader, PdfWriter

def encrypt_pdf(input_pdf, output_pdf, user_password, owner_password=None):
    """添加密碼保護(hù)"""
    reader = PdfReader(input_pdf)
    writer = PdfWriter()
    
    # 復(fù)制所有頁面
    for page in reader.pages:
        writer.add_page(page)
    
    # 加密
    writer.encrypt(user_password, owner_password)
    
    with open(output_pdf, 'wb') as output_file:
        writer.write(output_file)
    
    print(f"加密完成！保存為: {output_pdf}")

def decrypt_pdf(input_pdf, output_pdf, password):
    """解密PDF（需要知道密碼）"""
    reader = PdfReader(input_pdf)
    
    if reader.is_encrypted:
        reader.decrypt(password)
    
    writer = PdfWriter()
    for page in reader.pages:
        writer.add_page(page)
    
    with open(output_pdf, 'wb') as output_file:
        writer.write(output_file)
    
    print(f"解密完成！保存為: {output_pdf}")

# 使用示例
encrypt_pdf("財(cái)務(wù)報(bào)告.pdf", "財(cái)務(wù)報(bào)告_加密.pdf", "123456")
decrypt_pdf("財(cái)務(wù)報(bào)告_加密.pdf", "財(cái)務(wù)報(bào)告_解密.pdf", "123456")

import os
import pandas as pd
import pdfplumber
from PyPDF2 import PdfMerger, PdfReader, PdfWriter
from reportlab.pdfgen import canvas

class PDFAutomationPipeline:
    """PDF自動(dòng)化處理流水線"""
    
    def __init__(self, input_folder, output_folder):
        self.input_folder = input_folder
        self.output_folder = output_folder
        os.makedirs(output_folder, exist_ok=True)
    
    def extract_all_tables(self):
        """從所有PDF中提取表格"""
        all_data = []
        
        for filename in os.listdir(self.input_folder):
            if filename.endswith('.pdf'):
                file_path = os.path.join(self.input_folder, filename)
                print(f"處理文件: {filename}")
                
                with pdfplumber.open(file_path) as pdf:
                    for page_num, page in enumerate(pdf.pages):
                        tables = page.extract_tables()
                        for table_num, table in enumerate(tables):
                            # 假設(shè)第一行是表頭
                            if len(table) > 0:
                                df = pd.DataFrame(table[1:], columns=table[0])
                                all_data.append({
                                    'source': filename,
                                    'page': page_num + 1,
                                    'table': table_num + 1,
                                    'data': df
                                })
        
        return all_data
    
    def create_summary_pdf(self, all_tables, summary_pdf):
        """創(chuàng)建匯總PDF（使用reportlab繪制簡單匯總）"""
        from reportlab.lib.pagesizes import A4
        from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, Table as RLTable
        from reportlab.lib.styles import getSampleStyleSheet
        
        doc = SimpleDocTemplate(summary_pdf, pagesize=A4)
        styles = getSampleStyleSheet()
        story = []
        
        # 添加標(biāo)題
        story.append(Paragraph("數(shù)據(jù)匯總報(bào)告", styles['Title']))
        story.append(Spacer(1, 12))
        
        # 匯總每個(gè)表格的基本信息
        for idx, item in enumerate(all_tables[:5]):  # 只顯示前5個(gè)表格
            story.append(Paragraph(f"表格 {idx+1}: 來源 {item['source']} 第{item['page']}頁", styles['Heading2']))
            
            # 將DataFrame轉(zhuǎn)為列表
            data = item['data'].values.tolist()
            if data:
                # 添加表頭
                table_data = [item['data'].columns.tolist()] + data[:5]  # 只顯示前5行
                t = RLTable(table_data)
                story.append(t)
                story.append(Spacer(1, 12))
        
        doc.build(story)
        print(f"匯總PDF已創(chuàng)建: {summary_pdf}")
    
    def run_pipeline(self):
        """執(zhí)行完整流程"""
        print("=== 步驟1: 提取所有表格 ===")
        tables = self.extract_all_tables()
        print(f"共提取 {len(tables)} 個(gè)表格")
        
        print("\n=== 步驟2: 生成匯總PDF ===")
        summary_path = os.path.join(self.output_folder, "匯總報(bào)告.pdf")
        self.create_summary_pdf(tables, summary_path)
        
        print("\n=== 步驟3: 添加水印 ===")
        # 創(chuàng)建水印
        watermark_path = os.path.join(self.output_folder, "watermark.pdf")
        create_watermark_pdf("內(nèi)部資料", watermark_path)
        
        # 添加水印
        watermarked_path = os.path.join(self.output_folder, "匯總報(bào)告_帶水印.pdf")
        add_watermark_to_pdf(summary_path, watermark_path, watermarked_path)
        
        print("\n=== 步驟4: 加密保護(hù) ===")
        final_path = os.path.join(self.output_folder, "最終報(bào)告_加密.pdf")
        encrypt_pdf(watermarked_path, final_path, "secure123")
        
        print(f"\n? 所有處理完成！最終文件: {final_path}")

# 使用流水線
pipeline = PDFAutomationPipeline("./輸入文件夾", "./輸出文件夾")
pipeline.run_pipeline()

reader = PdfReader(open('large.pdf', 'rb'))
# 處理完成后
del reader

try:
    reader = PdfReader(file)
except PyPDF2.errors.PdfReadError as e:
    print(f"PDF讀取錯(cuò)誤: {e}")