import fitz
import os

BASE_DIR = r'c:\Users\zhang\Desktop\demo'
PDF_PATH = os.path.join(BASE_DIR, 'document.pdf')
OUTPUT_DIR = os.path.join(BASE_DIR, 'document_images')

# 創(chuàng)建輸出目錄
os.makedirs(OUTPUT_DIR, exist_ok=True)

print(f'=== PDF轉(zhuǎn)圖片 ===')
print(f'輸入文件: {PDF_PATH}')
print(f'輸出目錄: {OUTPUT_DIR}')
print()

# 打開PDF
doc = fitz.open(PDF_PATH)
print(f'PDF頁數(shù): {len(doc)}')
print()

# 逐頁轉(zhuǎn)換
for page_num in range(len(doc)):
    page = doc[page_num]
    
    # 渲染為圖片 (dpi=200 保證高質(zhì)量)
    pix = page.get_pixmap(dpi=200)
    
    # 生成文件名
    output_path = os.path.join(OUTPUT_DIR, f'page_{page_num + 1:03d}.png')
    
    # 保存圖片
    pix.save(output_path)
    
    print(f'  第 {page_num + 1} 頁 -> {os.path.basename(output_path)} ({pix.width}x{pix.height})')

page_count = len(doc)
doc.close()

print()
print('=== 完成 ===')
print(f'共轉(zhuǎn)換 {page_count} 頁，圖片保存在: {OUTPUT_DIR}')

# 驗證生成的文件
output_files = sorted([f for f in os.listdir(OUTPUT_DIR) if f.endswith('.png')])
print(f'生成的文件: {output_files}')

# -*- coding:utf-8 -*-
# Author :  yyzhang56
# 所有的圖片與PDF轉(zhuǎn)換的操作都在這里進行定義
from multiprocessing import Pool
# 安裝fitz需要安裝PyMuPDF才能使用
import fitz
import os
tmp = r'C:\Users\Downloads\'   #pdf路徑
export_file=r"D:\new_dates\數(shù)據(jù)"
save_path=r"D:\new_dates\ocr_result"
os.makedirs(export_file,exist_ok=True)
pdf_dir = [i for i in os.listdir(tmp) if os.path.splitext(i)[-1] == ".pdf"]
def pdf_to_jpg(name):
    # lock.acquire()
    #拼接pdf的文件路徑
    pwd_name=os.path.join(tmp,name)
    doc=fitz.open(pwd_name)
    # 將文件名同我們的保存路徑拼接起來（保存圖片的文件夾）
    dir_name=os.path.splitext(name)[0]
    pdf_name = os.path.join(export_file,  dir_name)
    # print(pdf_name)
    temp = 0
    #（保存圖片的文件夾）不存咋則生成
    # exsitsdir.judge(pdf_name)
    os.makedirs(pdf_name,exist_ok=True)
    for pg in range(doc.pageCount):
        page = doc[pg]
        temp += 1
        rotate = int(0)
        # 每個尺寸的縮放系數(shù)為2，這將為我們生成分辨率提高四倍的圖像。
        zoom_x = 2.0
        zoom_y = 2.0
        trans = fitz.Matrix(zoom_x, zoom_y).preRotate(rotate)
        pm = page.getPixmap(matrix=trans, alpha=False)
        pic_name = '{}.jpg'.format(temp)
        #拼接生成pdf的文件路徑
        pic_pwd = os.path.join(pdf_name, pic_name)
        # print(pic_pwd)
        pm.writePNG(pic_pwd)
def main():
    pool = Pool(10)
    for i in pdf_dir:
        res = pool.apply_async(pdf_to_jpg, (i,))
    pool.close()
    pool.join()
if __name__ == '__main__':
    main()       #需要pdf切圖就開啟

#Author：yyzhang56
from PIL import Image
import os
from PyPDF2 import PdfFileReader, PdfFileWriter
# 單張圖片轉(zhuǎn)pdf
# img = Image.open('優(yōu)化.png')
# img.save('pypdf01.pdf', 'PDF')  # 通過PIL庫保存為pdf格式
# 多張圖片轉(zhuǎn)pdf
# ilst = ['D:/docOfStu/pypdf2-mindmap-01.jpg', 'D:/docOfStu/pypdf2-mindmap-02.jpg']  # 圖片列表
path=r"C:\Users\Desktop\pdf\tif_pic"
ilst=os.listdir(path)
out_pdf = PdfFileWriter()
sdfs=[os.path.join(path,i) for i in ilst ]
print(sdfs)
for f in sdfs:
    img = Image.open(f)
    fw = f.replace('.png', '.pdf')
    img.save(fw)
    out_pdf.appendPagesFromReader(PdfFileReader(
        open(fw, 'rb')))  # 也可拆這句為 sc_pdf=PdfFileReader(open(fw,'rb')); out_pdf.addPage(sc_pdf.getPage(0))
out_pdf.write(open('./pypdf2.pdf', 'wb'))

import os
from pdf2docx import Converter
'''
author--yyzhang
'''
pdf_path = r"C:\Users\Desktop\pdf\files"  # pdf路徑
docx_path = r"C:\Users\Desktop\pdf\docx"  # 保存docx路徑
# 如果保存的路徑不存在可以自動生成
if os.path.exists(docx_path):
    pass
else:
    os.makedirs(docx_path)
pdf_files = os.listdir(pdf_path)
# 遍歷pdf路徑下的文件
for pdf_file in pdf_files:
    if pdf_file.endswith(".pdf") or pdf_file.endswith(".PDF"):
        head_name = os.path.splitext(pdf_file)[0]   #獲取文件名
        docx_name = head_name + ".docx"             #創(chuàng)建docx名稱
        file_path = os.path.join(pdf_path, pdf_file)    #獲取pdf絕對路徑
        docx_file_path = os.path.join(docx_path, docx_name)   #獲取docx絕對路徑
        cv = Converter(file_path)
        cv.convert(docx_file_path, start=0, end=None)
        cv.close()

#安裝庫
pip install pyautogui
#安裝庫
pip install  pillow

import os
import glob
from PyPDF2 import PdfReader
from pdf2image import convert_from_path
pdf_dir = "path/to/pdf_dir/"  #pdf目錄
# 遍歷目錄中的PDF文件
pdf_files = glob.glob(os.path.join(pdf_dir, "*.pdf"))
# 遍歷每個PDF文件，并將其轉(zhuǎn)換為圖片
for pdf_file in pdf_files:
    # 創(chuàng)建PdfReader對象
    pdf = open(pdf_file, 'rb')
    pdf_reader = PdfReader(pdf)
    # 遍歷PDF的頁面并將其轉(zhuǎn)換為圖片
    for page_num in range(len(pdf_reader.pages)):
        # 獲取頁面對象
        page = pdf_reader.pages[page_num]
        # 將PDF頁面轉(zhuǎn)換為圖像
        images = convert_from_path(pdf_file, first_page=page_num+1, last_page=page_num+1)
        # 定義圖像保存路徑
        filename = os.path.splitext(os.path.basename(pdf_file))[0]
        image_path = os.path.join(pdf_dir, f"{filename}_page_{page_num+1}.jpg")
        # 保存圖像
        for i, image in enumerate(images):
            if i == 0:
                image.save(image_path, "JPEG")
            else:
                image_path = os.path.join(pdf_dir, f"{filename}_page_{page_num+1}_{i+1}.jpg")
                image.save(image_path, "JPEG")
    # 關(guān)閉PDF文件
    pdf.close()

#手動選擇目錄下的pdf文件
import os
from tkinter import Tk
from tkinter.filedialog import askdirectory
from PyPDF2 import PdfReader
from pdf2image import convert_from_path
# 打開選擇目錄的對話框
Tk().withdraw()  # 隱藏Tkinter根窗口
pdf_dir = askdirectory(title="選擇PDF所在目錄")
# 遍歷目錄中的PDF文件
pdf_files = [f for f in os.listdir(pdf_dir) if f.endswith(".pdf")]
# 遍歷每個PDF文件，并將其轉(zhuǎn)換為圖片
for pdf_file in pdf_files:
    # 創(chuàng)建PdfReader對象
    pdf_path = os.path.join(pdf_dir, pdf_file)
    pdf = open(pdf_path, 'rb')
    pdf_reader = PdfReader(pdf)
    # 遍歷PDF的頁面并將其轉(zhuǎn)換為圖片
    for page_num in range(len(pdf_reader.pages)):
        # 獲取頁面對象
        page = pdf_reader.pages[page_num]
        # 將PDF頁面轉(zhuǎn)換為圖像
        images = convert_from_path(pdf_path, first_page=page_num+1, last_page=page_num+1)
        # 定義圖像保存路徑
        filename = os.path.splitext(pdf_file)[0]
        image_path = os.path.join(pdf_dir, f"{filename}_page_{page_num+1}.jpg")
        # 保存圖像
        for i, image in enumerate(images):
            if i == 0:
                image.save(image_path, "JPEG")
            else:
                image_path = os.path.join(pdf_dir, f"{filename}_page_{page_num+1}_{i+1}.jpg")
                image.save(image_path, "JPEG")
    # 關(guān)閉PDF文件
    pdf.close()