内蒙古人民政府

REITs专题数据/policy-neimenggu.py

-import os
+import os
@@ -6,10 +6,17 @@ import numpy as np
 import pandas as pd
 import requests
 from bs4 import BeautifulSoup
-from base import BaseCore
+import BaseCore
 
 baseCore = BaseCore.BaseCore()
 log = baseCore.getLogger()
+
+from reits import Policy
+policy = Policy()
+
+
+topic = 'policy'
+webname = '内蒙古自治区人民政府'
 headers = {
     'Accept': 'application/json, text/plain, */*',
     'Accept-Encoding': 'gzip, deflate, br',
@@ -64,9 +71,9 @@ def getSoup(url):
 
 
 def getPageSize():
-    ip = baseCore.get_proxy()
+
     url = 'https://www.nmg.gov.cn/nmsearch/trssearch/searchAll.do?siteId=32&searchTag=zc&allKeywords=REITs&fullKeywords=&orKeywords=&notKeywords=&sort=&position=0&organization=&pageNum=1&pageSize=10&zcYear=&zcMonth=&docno=&cdesc=&publisher=&cityName=&isAlways=1&isSearchRmzfAndBgt=&isAccurate=1'
-    req = requests.get(url, headers=headers, proxies=ip)
+    req = requests.get(url, headers=headers)
     req.encoding = req.apparent_encoding
     total = int(req.json()['data']['total'])
     if total % 10 == 0:
@@ -77,23 +84,22 @@ def getPageSize():
 
 
 def getJson(page):
-    ip = baseCore.get_proxy()
+    # ip = baseCore.get_proxy()
     url = f'https://www.nmg.gov.cn/nmsearch/trssearch/searchAll.do?siteId=32&searchTag=zc&allKeywords=REITs&fullKeywords=&orKeywords=&notKeywords=&sort=&position=0&organization=&pageNum={page}&pageSize=10&zcYear=&zcMonth=&docno=&cdesc=&publisher=&cityName=&isAlways=1&isSearchRmzfAndBgt=&isAccurate=1'
-    req = requests.get(url, headers=headers, proxies=ip)
+    req = requests.get(url, headers=headers)
     req.encoding = req.apparent_encoding
     return req.json()['data']['data']
 
 
 def getFjContent(url):
-    ip = baseCore.get_proxy()
-    req = requests.get(url, headers=headers_, proxies=ip)
+    # ip = baseCore.get_proxy()
+    req = requests.get(url, headers=headers_)
     req.encoding = req.apparent_encoding
     return req.content
 
 
 def getContent(num, data):
-    fjhref_list = ''
-    fjtitle_list = ''
+    id_list = []
     title = data['title']
     pub_hao = data['docno']
     origin = data['sitedesc']
@@ -102,12 +108,16 @@ def getContent(num, data):
     try:
         writtenDate = data['scrq']
     except:
-        writtenDate = ''
+        writtenDate = None
     summary = BeautifulSoup(data['zc_doccontent'], 'html.parser').text.lstrip().strip()
     url = data['docpuburl']
+    # 根据链接判重
+    is_member = baseCore.r.sismember('REITs::' + webname, url)
+    if is_member:
+        return
     soup = getSoup(url)
     if soup == '':
-        return ''
+        return
     url_ = url.split('/')[-1]
     soup = paserUrl(soup, url.replace(url_, ''))
     contentWithTag = soup.find('div', attrs={'id': 'pare'})
@@ -119,7 +129,9 @@ def getContent(num, data):
                 contentWithTag = soup.find('div', attrs={'class': 'zoomCon'})
                 if not contentWithTag:
                     contentWithTag = soup.find('div', attrs={'id': 'pagecontent'})
-    if writtenDate == '':
+                if not contentWithTag:
+                    contentWithTag = soup.find('div', id="docContent")
+    if not writtenDate:
         try:
             tr_list = soup.find('table', class_='m-detailtb').find_all('tr')
             for tr in tr_list:
@@ -154,46 +166,69 @@ def getContent(num, data):
         a_list = contentWithTag.find_all('a')
         for a in a_list:
             href = a.get('href')
-            fjhref_list += href + '\n'
+
             category = os.path.splitext(href)[1]
             fj_title = f'{num}-{publishDate}-{a.text.lstrip().strip()}'
             if '<' in fj_title or '>' in fj_title:
                 fj_title = fj_title.replace('<', '').replace('>', '')
             if category not in fj_title:
                 fj_title = fj_title + category
-            fjtitle_list += fj_title + '\n'
-            fjcontent = getFjContent(href)
-            file = f'./相关政策/内蒙古自治区人民政府/政策文件/{fj_title}'
-            with open(file, 'wb') as f:
-                f.write(fjcontent)
-            log.info(f'{fj_title}===附件下载成功')
+            # 附件
+            att_id, full_path = policy.attuributefile(fj_title, href, num, publishDate)
+            if att_id:
+                id_list.append(att_id)
+                a['href'] = full_path
+
     except Exception as e:
         log.error(title, '=====', e)
-    content = contentWithTag.text.lstrip().strip()
-    data_ = [num, title, publishDate, origin, url, writtenDate, organ, pub_hao, summary, content, fjtitle_list,
-             fjhref_list]
-    return data_
+    try:
+        content = contentWithTag.text.lstrip().strip()
+    except:
+        log.info(url)
+        return
+    contentWithTag_str = str(contentWithTag)
+    time_now = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
+    dic_info = {
+        'attachmentIds': id_list,
+        'author': '',
+        'content': content,
+        'contentWithTag': contentWithTag_str,
+        'deleteFlag': 0,
+        'id': '',
+        'title': title,
+        'publishDate': publishDate,
+        'origin': origin,
+        'sourceAddress': url,
+        'writtenDate': writtenDate,
+        'organ': organ,
+        'topicClassification': '',
+        'issuedNumber': pub_hao,
+        'summary': summary,
+        'createDate': time_now,
+        'sid': '1729041959772860417',
+    }
+    try:
+        baseCore.sendkafka(dic_info, topic)
+        baseCore.r.sadd('REITs::' + webname, url)
+        log.info(f'采集成功--{title}--{url}')
+    except Exception as e:
+        for att_id in id_list:
+            baseCore.deliteATT(att_id)
+
+    return
 
 
 def doJob():
-    if not os.path.exists('./相关政策/内蒙古自治区人民政府/政策文件'):
-        os.makedirs('./相关政策/内蒙古自治区人民政府/政策文件')
-    data_list = []
+
+
     pageSize = getPageSize()
     num = 1
     for page in range(1, pageSize + 1):
         data_json = getJson(page)
         for data_ in data_json:
             if data_['chnldesc'] == '政策文件':
-                data = getContent(num, data_)
-                if data:
-                    data_list.append(data)
-                    num += 1
-                    log.info(f'{data[1]}===采集成功')
-    df = pd.DataFrame(np.array(data_list))
-    df.columns = ['序号', '标题', '发布时间', '来源', '原文链接', '发文时间', '发文机构', '发文字号', '摘要', '正文', '附件名称', '附件连接']
-    df.to_excel('./内蒙古自治区人民政府政策文件.xlsx', index=False)
-
+                getContent(num, data_)
+                num += 1
 
 if __name__ == '__main__':
     doJob()