利用Python将PDF转html并提取文字和图片

pip install PyMuPDF
pip install tqdm
pip install Beautifulsoup4

import fitz
import os,re,io
from tqdm import tqdm
import base64
from bs4 import BeautifulSoup
from PIL import Image

def decode_image(src, dir, index):

    # 1、信息提取
    result = re.search("data:image/(?P<ext>.*?);base64,(?P<data>.*)", src, re.DOTALL)
    if result:
        ext = result.groupdict().get("ext")
        data = result.groupdict().get("data")
    else:
        raise Exception("Do not parse!")

    imgdata = base64.b64decode(data)
    im = Image.open(io.BytesIO(imgdata))
    width, height = im.size

    print("width:{},height:{}".format(width, height))

    # 2、base64解码
    img = base64.urlsafe_b64decode(data)

    path = dir.replace('.pdf', '')
    if not os.path.exists(path):
        os.makedirs(path, 777)
    # 3、二进制文件保存
    filename = "{}{}.{}".format(path+'/', index, ext)
    print(filename)
    with open(filename, "wb") as f:
        f.write(img)
        f.close()
        
    # 人脸识别使用
    if width % 4 != 0:
        print('resize_4_pic:{}'.format(filename))
        nw = (width % 4 == 0) and width or (width + (4 - (width % 4)))
        nh = (height % 4 == 0) and height or (height + (4 - (height % 4)))

        img = Image.open(filename)
        img = img.resize( (nw,nh), Image.ANTIALIAS)
        img.save(filename)

    imgs.append(filename)
    return filename

def read_pdf(srcfile):
    doc = fitz.open(srcfile)
    html_content = ''
    for page in tqdm(doc):
        html_content += page.get_text('html')

    html_content = html_content.replace('<br/>', ' \n').replace('<br>', ' \n').replace('<br >', ' \n').replace('<br />', ' \n')
    html_content = html_content.replace(' ', ' ').replace('&#160;', ' ')

    soup =  BeautifulSoup(html_content,'lxml')

    index = 0
    for imgtag in soup.find_all('img'):
        decode_image(imgtag['src'], srcfile, index)
        index += 1

    return soup.text

if __name__ == '__main__':
    path = "D:/test.pdf"
    imgs = []
    content = read_pdf(path)
    print(content)
    print(imgs)