def process_item(self, item, spider):
try:
if isinstance(item, XueqiuItem):
if item.get('article_content'):
item['article_content'] = self.parse_content(item['article_content'])
if item['article_content'] == None or item['article_content']=="":
item['article_content'] = item['article_title'].strip()
else:
item['article_content'] = item['article_title'].strip()
if item.get('article_title'):
item['article_title'] = item['article_title'].strip()[0:200]
if item.get('release_time'):
item['release_time'] = item['release_time'].strip()
item['release_time'] = self.parse_time(item['release_time'])
if item['author_province'] == None or item['author_province'] == '':
item['author_province'] = u"未知"
if item['author_city'] == None or item['author_city'] == "":
item['author_city'] = u"未知"
if item['author_description'] == None or item['author_description'] == "":
item['author_description'] = u'暂无'
return item
except Exception as ee:
log_error("XueqiuPipeline.process_item Error:{}".format(ee))
log_error(traceback.format_exc())
def parse_article_url(self, response):
try:
response_data = json.loads(response.text)
article_lists = response_data['statuses']
for article_dict in article_lists:
widget_info = {}
widget_info['like_count'] = article_dict['like_count']
widget_info['comment_count'] = article_dict['reply_count']
widget_info['forward_count'] = article_dict['retweet_count']
widget_info['article_id'] = article_dict['id']
widget_info['user_id'] = article_dict['user_id']
id = article_dict['id']
user_id = article_dict['user_id']
#获取作者新信息
author_info = {}
user_list = article_dict['user']
author_info['author_name'] = user_list['screen_name']
author_info['author_province'] = user_list['province']
author_info['author_city'] = user_list['city']
author_info['author_description'] = user_list['description']
author_info['friends_count'] = user_list['friends_count']
author_info['followers_count'] = user_list['followers_count']
author_info['author_id'] = user_list['id']
author_info['author_type'] = 1
request_url = 'https://xueqiu.com/{0}/{1}'.format(user_id, id)
yield Request(url=request_url,
callback=self.parse_article_page,
meta={
'widget_info': widget_info,
'author_info': author_info
},
dont_filter=settings.TYPE1_FILTER)
except Exception as ee:
log_error("Spider xueqiu.parse_article_url Error:{}".format(ee))
log_error(traceback.format_exc())
def process_item(self,item,spider):
try:
self.r.hset(self.redis_data_dict,item['article_url'],0)
return item
except Exception as ee:
log_error("RedisPipeline.process_item Error:{}".format(ee))
log_error(traceback.format_exc())
def open_spider(self,spider):
try:
self.r = redis.StrictRedis(host=self.redis_host,port=self.redis_port,db=self.redis_db)
self.redis_data_dict = 'url_hash'
except Exception as ee:
log_error("RedisPipeline.open_spider Error:{}".format(ee))
log_error(traceback.format_exc())
def deal_time(self,original_time):
try:
if re.match(u'今天.*', original_time):
date_time = re.match(u'今天(.*)', original_time).group(1).strip()
today = datetime.date.today()
date_time = today.strftime('%Y-%m-%d') + ' ' + date_time
elif re.match(u'昨天.*', original_time):
date_time = re.match(u'昨天(.*)', original_time).group(1).strip()
today = datetime.date.today()
yesterday = today - datetime.timedelta(days=1)
date_time = yesterday.strftime('%Y-%m-%d') + ' ' + date_time
elif re.match(u'.*小时前', original_time):
date_time = re.match(u'(.*)小时前', original_time).group(1).strip()
current_time = datetime.datetime.now()
real_time = current_time - datetime.timedelta(hours=int(date_time))
date_time = real_time.strftime('%Y-%m-%d %H:%M')
elif re.match(u'.*分钟前', original_time):
date_time = re.match(u'(.*)分钟前', original_time).group(1).strip()
current_time = datetime.datetime.now()
real_time = current_time - datetime.timedelta(minutes=int(date_time))
date_time = real_time.strftime('%Y-%m-%d %H:%M')
elif len(original_time) < 12:
date_time = time.strftime("%Y", time.localtime()) + '-' + original_time
else:
date_time = original_time
return date_time
except Exception as ee:
log_error("XueqiuPipeline.deal_time Error:{}".format(ee))
log_error(traceback.format_exc())
def open_spider(self,spider):
try:
self.connect = pymysql.connect(**self.dbparams)
self.cursor = self.connect.cursor()
except Exception as ee:
log_error("MysqlPipeline.open_spider Error:{}".format(ee))
log_error(traceback.format_exc())
def start_requests(self):
try:
random_id = ''.join(
str(random.choice(range(10))) for _ in range(13))
request_url = 'https://xueqiu.com/recommend/user/industry.json?detail=1&_={0}'.format(
random_id)
yield Request(url=request_url,
callback=self.parse_people_list,
dont_filter=settings.TYPE1_FILTER)
except Exception as ee:
log_error("Spider xueqiu.start_request Error:{}".format(ee))
log_error(traceback.format_exc())
def parse_content(self,article_content):
try:
if isinstance(article_content, list):
for i, item in enumerate(article_content):
article_content[i] = item.replace(u'\xa0', '').replace('\t', '').replace('\n', '').strip()
content = ''.join(article_content)
else:
content = article_content
return content
except Exception as ee:
log_error("XueqiuPipeline.parse_content Error:{}".format(ee))
log_error(traceback.format_exc())
def parse_time(self, date_time):
try:
if re.match(u'发布于.*',date_time):
re_data_time = re.match(u'发布于(.*)',date_time).group(1)
date_time = self.deal_time(re_data_time)
if re.match(u'修改于.*',date_time):
re_data_time = re.match(u'修改于(.*)',date_time).group(1)
date_time = self.deal_time(re_data_time)
return date_time
except Exception as ee:
log_error("XueqiuPipeline.parse_time Error:{}".format(ee))
log_error(traceback.format_exc())
def parse_people_page(self, response):
try:
response_data = json.loads(response.text)
response_url = response.url
max_page = response_data['maxPage']
for page in range(int(max_page)):
random_id = ''.join(
str(random.choice(range(10))) for _ in range(13))
based_url = ''.join(list(response_url)[:-13]) + random_id
request_url = based_url + '&page=' + str(page + 1)
yield Request(url=request_url,
callback=self.parse_article_url,
dont_filter=settings.TYPE1_FILTER)
except Exception as ee:
log_error("Spider xueqiu.parse_people_page Error:{}".format(ee))
log_error(traceback.format_exc())
def parse_people_url(self, response):
try:
response_data = json.loads(response.text)
user_lists = response_data['industries'][0]['users']
for user_list in user_lists:
#构建每个作者的url
id = user_list['id']
random_id = ''.join(
str(random.choice(range(10))) for _ in range(13))
request_url = 'https://xueqiu.com/v4/statuses/user_timeline.json?&user_id={0}&type=2&_={1}'.format(
id, random_id)
yield Request(url=request_url,
callback=self.parse_people_page,
dont_filter=settings.TYPE1_FILTER)
except Exception as ee:
log_error("Spider xueqiu.parse_people_url Error:{}".format(ee))
log_error(traceback.format_exc())
def parse_people_list(self, response):
try:
response_data = json.loads(response.text)
recommend_lists = response_data["list"]
for recommend_dict in recommend_lists:
id = recommend_dict['id']
random_id = ''.join(
str(random.choice(range(10))) for _ in range(13))
request_url = "https://xueqiu.com/recommend/user/industry.json?id={0}&_={1}".format(
id, random_id)
#logging.debug(resquest_url)
yield Request(url=request_url,
callback=self.parse_people_url,
dont_filter=settings.TYPE1_FILTER)
except Exception as ee:
log_error("Spider xueqiu.parse_people_list Error:{}".format(ee))
log_error(traceback.format_exc())
def process_item(self,item,spider):
author_info = {
'author_id': item['author_id'],
'author_name':item['author_name'],
'author_province':item['author_province'],
'author_city':item['author_city'],
'author_description':item['author_description'],
'author_friends_count':item['friends_count'],
'author_followers_count':item['followers_count'],
'author_type': item['author_type']
}
author_info_keys = ','.join(author_info.keys())
author_info_values = ','.join(['%s'] * len(author_info))
article_info = {
"index_id":item['index_id'],
'author_id':item['author_id'],
'article_id':item['article_id'],
'article_url':item['article_url'],
'article_title':item['article_title'],
'article_content':item['article_content'],
'release_time':item['release_time'],
'content_source':item['content_source'],
'forward_count':item['forward_count'],
'comment_count':item['comment_count'],
'like_count':item['like_count'],
}
article_info_keys = ','.join(article_info.keys())
article_info_values = ','.join(['%s'] * len(article_info))
try:
#将作者信息存入作者信息的数据表中
insert_author_sql = 'insert into author_info(%s) values (%s) ON duplicate KEY UPDATE author_id=author_id' %(author_info_keys,author_info_values)
self.cursor.execute(insert_author_sql,tuple(author_info.values()))
#文章信息存入文章信息的数据表中
insert_article_sql = 'insert into article_info(%s) values (%s) ON duplicate KEY UPDATE index_id=index_id' %(article_info_keys,article_info_values)
self.cursor.execute(insert_article_sql, tuple(article_info.values()))
self.connect.commit()
#self.logger.info('Success save to MySQL')
log_info("Success save to MySQL")
return item
except Exception as ee:
log_error("MysqlPipeline.process_item Error:{}".format(ee))
log_error(traceback.format_exc())
def parse_article_page(self, response):
try:
#获取文章信息
widget_info = response.meta['widget_info']
#获取文章的发布时间
release_time = response.xpath(
"//div[@class='avatar__subtitle']/a/text()").extract_first()
#获取文章的来源
isinstance_source = response.xpath(
"//article[@class='article__bd']/div[@class='article__bd__from']"
)
if isinstance_source:
content_source = isinstance_source.xpath(
'a/text()').extract_first()
else:
content_source = response.xpath(
"//div[@class='avatar__name']/a/text()").extract_first()
#文章的点赞数
like_count = widget_info['like_count']
#文章的评论数
comment_count = widget_info['comment_count']
#文章的转发数
forward_count = widget_info['forward_count']
#用户id
user_id = widget_info['user_id']
#文章id
article_id = widget_info['article_id']
#该文章的索引id
index_id = str(user_id) + '-' + str(article_id)
isinstance_title = response.xpath(
"//article[@class='article__bd']/h1")
isinstance_content = response.xpath(
"//div[@class='article__bd__detail']/p")
#获取文章的标题和内容
if isinstance_title:
article_title = ''.join(
isinstance_title.xpath('text()').extract())
if isinstance_content:
article_content = response.xpath(
"//div[@class='article__bd__detail']/p/text()"
).extract()
else:
article_content = response.xpath(
"//div[@class='article__bd__detail']/text()").extract(
)
else:
is_title_exist = response.xpath(
"//div[@class='article__bd__detail']/text()").extract()
if is_title_exist:
try:
article_title = ''.join(is_title_exist[0])
article_content = is_title_exist[1:]
except IndexError:
article_title = 'Wrong article {0}'.format(
response.url)
article_content = 'Wrong article {0}'.format(
response.url)
self.logger.error('Failed to get content {0}'.format(
response.url))
else:
article_title = 'None'
article_content = response.xpath(
"//div[@class='article__bd__detail']/p/text()"
).extract()
article_url = response.url
#获取作者信息
author_info = response.meta['author_info']
author_name = author_info['author_name']
author_province = author_info['author_province']
author_city = author_info['author_city']
author_description = author_info['author_description']
friends_count = author_info['friends_count']
followers_count = author_info['followers_count']
author_id = author_info['author_id']
author_type = author_info['author_type']
xueqiu_item = XueqiuItem()
for field in xueqiu_item.fields:
try:
xueqiu_item[field] = eval(field)
except NameError:
self.logger.error('Field is Not Defined')
yield xueqiu_item
except Exception as ee:
log_error("Spider xueqiu.parse_article_page Error:{}".format(ee))
log_error(traceback.format_exc())
def process_response(self, request, response, spider):
if response.status == 400:
try:
log_error('Cookie Invalidation url:{0}'.format(response.url))
self.logger.warning('Cookie Invalidation url:{0}'.format(
response.url))
except Exception as ee:
log_error('Cookie Invalidation ERROR {}'.format(ee))
log_error(traceback.format_exc())
if response.status == 404:
try:
log_error('ERROR 404 url:{0}'.format(response.url))
self.logger.warning('ERROR 404 url:{0}'.format(response.url))
except Exception as ee:
log_error('404 Error {}'.format(ee))
log_error(traceback.format_exc())
if response.status in [501, 502, 503, 504]:
#self.logger.info('IP use too match to enter waiting')
log_info("IP use too match to enter waiting url:{}".format(
response.url))
time.sleep(300)
return request
if response.status in [302]:
self.logger.warning("302 Error IP Invalidation")
log_error("302 Error IP Invalidation")
return request
else:
return response
