单页文字图片爬取保存到word

```python
import requests,csv,docx,os,time,random
from bs4 import BeautifulSoup
from docx.shared import Inches
# 0703完美测试版，记得更新cookie
# 定义函数，判断段落代码里是否有img，若有，则要单独提取图片
def img_ok(x):
    try:
        img = x.find('img')['src']
        if len(img)>1:
            return 1
        else:
            return 0
    except TypeError:
        return 0

headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win32; x32) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/83.0.4103.106 Safari/535.1',
           'Cookie': '__root_domain_v=.ipmph.com; _qddaz=QD.ky39uq.kayj8f.k9m46lu7; UM_distinctid=171c9f3608f367-0db723f0697f73-7373667-e1000-171c9f36090528; Hm_lvt_ec31b23a3a54fb0e85df69fc93bd5de9=1595478459,1595497069,1595562737,1595819138; jeesite.session.id=5bb73af84fdb41888c956f6d0daf9df0; CNZZDATA1268665702=515724504-1588225736-%7C1595897449; Hm_lpvt_ec31b23a3a54fb0e85df69fc93bd5de9=1595901255',}
session = requests.Session()
logurl = 'http://sso.ipmph.com//oauth/authorize?client_id=8fe837c9b8b94cf0b4a87287ea4c7b7a&response_type=code&scope=user_info&redirect_uri=http://ccdas.ipmph.com/sso/login?rw_lczsPath=http%3A%2F%2Fccdas.ipmph.com%2F'
response = session.get(logurl, headers=headers)

#url='http://ccdas.ipmph.com/rwDisease/getRwDiseaseDetail?diseaseId=24472'

with open(r'mulu.csv','r') as f:
    xls = csv.reader(f)
    for i, row in enumerate(xls):
        if i in range(0,1000):
            url = row[0]
            time.sleep(random.randrange(4,81)/3)
            print(url)

# 以下无误的爬取单个页面内容
            msg = session.get(url, headers=headers)
            sop = BeautifulSoup(msg.text, 'html.parser')
            try:
                name = sop.find('div',class_='gu_det_left_top').text.replace(' ','').replace('\n','').replace('\t','').replace('\r','').replace('/','、').replace('?','？').replace('*','')
            except AttributeError:
                name = sop.find('div',class_='gu_det_left_top')
            div = sop.find('div',class_='gu_det_left_con')
            # 记得先打开docx
            doc = docx.Document()
            doc.add_paragraph(name+'\n'+url)
            vvv = sop.find_all('div',class_='gu_det_left_con_h1')
            kkk = sop.find_all('div',class_='gu_det_left_con_main')
            ls1 = []
            # 提取段落标题，封装到列表
            for v in vvv:
                title = v.text.replace(' ','').replace('\n','').replace('\t','').replace('\r','').replace('：','')
                ls1.append(title)
            # 提取段落文字，n为段落的顺序
            n = -1
            for k in kkk:
                n += 1
                if img_ok(k) == 0:
                    print('【'+ls1[n]+'】'+'\n'+k.text)
                    doc.add_paragraph('【'+ls1[n]+'】'+k.text)
                elif img_ok(k) == 1:
                    print('【'+ls1[n]+'】'+'\n')
                    doc.add_paragraph('【'+ls1[n]+'】'+'\n')
                    for j in k:
                        if img_ok(j) == 0:
                            try:
                                # print(j.text)
                                doc.add_paragraph(j.text)
                            except AttributeError:
                                doc.add_paragraph(j)
                        elif img_ok(j) == 1:
                            img_ul = 'http://ccdas.ipmph.com/'+j.find('img')['src']
                            pic = requests.get(img_ul)
                            with open('img_tmp.png', 'wb') as f:
                                f.write(pic.content)
                            print(img_ul)
                            doc.add_picture('img_tmp.png', width=Inches(6))
                            os.remove('img_tmp.png')    #若无需保存原图片，一般删除
            # 切记保存word，否则不成功
            doc.save('人卫/0728/{}.docx'.format(name))
        else:
            continue
```