def go_yahoo():
'''抓取雅虎的所有公司'''
baseCrawl = BaseCrawl()
baseExtract = BaseExtract()
cxn_db = None
try:
cxn_db = get_cxn_db()
cur_db = cxn_db.cursor()
in_sql = """ insert into yahoo_comps (code_name,crawl_time,return_type,add_time,count)
values (%s,now(),%s,now(),1)
on duplicate key update crawl_time = now(),return_type = %s,count=count+1 """
code_names = get_comps()
logger.info("number of companys is %d"%len(code_names))
for code in code_names:
logger.debug("crawl %s, start............ "%code)
status,headlines_links = go_crawl_headlines(code,baseCrawl,baseExtract)
cur_db.execute(in_sql,(code,status,status))
if status == 0 and len(headlines_links) > 0:
extract_headlines_news(code,headlines_links,cur_db)
logger.debug("crawl %s, end.............. "%code)
except:
logger.error("crawl_page failed ,Error:%s" % traceback.format_exc())
finally:
if cxn_db:
cxn_db.close()
def go_yahoo():
'''抓取雅虎的所有公司'''
baseCrawl = BaseCrawl()
baseExtract = BaseExtract()
cxn_db = None
try:
cxn_db = get_cxn_db()
cur_db = cxn_db.cursor()
in_sql = """ insert into yahoo_comps (code_name,crawl_time,return_type,add_time,count)
values (%s,now(),%s,now(),1)
on duplicate key update crawl_time = now(),return_type = %s,count=count+1 """
code_names = get_comps()
logger.info("number of companys is %d" % len(code_names))
for code in code_names:
logger.debug("crawl %s, start............ " % code)
status, headlines_links = go_crawl_headlines(
code, baseCrawl, baseExtract)
cur_db.execute(in_sql, (code, status, status))
if status == 0 and len(headlines_links) > 0:
extract_headlines_news(code, headlines_links, cur_db)
logger.debug("crawl %s, end.............. " % code)
except:
logger.error("crawl_page failed ,Error:%s" % traceback.format_exc())
finally:
if cxn_db:
cxn_db.close()
def go_newsextrac(source_type, code_names_file, url_prefix, req_referer,
finance_homes, extracteNewsurls):
'''抓取所有的新闻链接、抽取并存储
@param source_type: 抓取来源：比如yahoo、google等
@param code_names_file: 公司上市代码列表
@param url_prefix: 拼接上市公司新闻url的前缀
@param req_referer: 抓取前连，防封禁
@param finance_homes: 财经频道首页地址
@param extracteNewsurls: 个性化的抽取对象
'''
baseCrawl = BaseCrawl()
baseExtract = BaseExtract()
cxn_db = None
try:
cxn_db = get_cxn_db()
cur_db = cxn_db.cursor()
#构造一个boolm filter
initUrlsBloomFilter(cur_db, source_type)
in_sql = """ insert into com_news_extract_state (url,code_name,crawl_time,return_type,add_time,crawl_count,source_type)
values (%s,%s,now(),%s,now(),1,%s)
on duplicate key update crawl_time = now(),return_type = %s,crawl_count=crawl_count+1,source_type = %s """
news_chanel_urls = get_compnewsurls(code_names_file, url_prefix,
finance_homes)
logger.info("number of companys is %d" % len(news_chanel_urls))
for code, news_chanel_url in news_chanel_urls:
logger.debug("crawl %s, start............ " % news_chanel_url)
is_finance_home = False
if code == FINANCE_HOME:
is_finance_home = True
status, news_links = extracteNewsurls.extracte(news_chanel_url,
is_finance_home,
baseCrawl,
baseExtract,
req_referer,
try_times=1)
cur_db.execute(in_sql, (news_chanel_url, code, status, source_type,
status, source_type))
if status == 0 and len(news_links) > 0:
extract_news(code, news_links, source_type, cur_db)
logger.debug("crawl %s, end.............. " % code)
logger.info("number of news url is %s" % NEWS_URL_EXTRACTE)
except:
logger.error("crawl_page failed ,Error:%s" % traceback.format_exc())
finally:
if cxn_db:
cxn_db.close()
def go_newsextrac(source_type,code_names_file,url_prefix,req_referer,finance_homes,extracteNewsurls):
'''抓取所有的新闻链接、抽取并存储
@param source_type: 抓取来源：比如yahoo、google等
@param code_names_file: 公司上市代码列表
@param url_prefix: 拼接上市公司新闻url的前缀
@param req_referer: 抓取前连，防封禁
@param finance_homes: 财经频道首页地址
@param extracteNewsurls: 个性化的抽取对象
'''
baseCrawl = BaseCrawl()
baseExtract = BaseExtract()
cxn_db = None
try:
cxn_db = get_cxn_db()
cur_db = cxn_db.cursor()
#构造一个boolm filter
initUrlsBloomFilter(cur_db,source_type)
in_sql = """ insert into com_news_extract_state (url,code_name,crawl_time,return_type,add_time,crawl_count,source_type)
values (%s,%s,now(),%s,now(),1,%s)
on duplicate key update crawl_time = now(),return_type = %s,crawl_count=crawl_count+1,source_type = %s """
news_chanel_urls = get_compnewsurls(code_names_file,url_prefix,finance_homes)
logger.info("number of companys is %d"%len(news_chanel_urls))
for code,news_chanel_url in news_chanel_urls:
logger.debug("crawl %s, start............ "%news_chanel_url)
is_finance_home = False
if code == FINANCE_HOME:
is_finance_home = True
status,news_links = extracteNewsurls.extracte(news_chanel_url,is_finance_home,baseCrawl,baseExtract,req_referer,try_times = 1)
cur_db.execute(in_sql,(news_chanel_url,code,status,source_type,status,source_type))
if status == 0 and len(news_links) > 0:
extract_news(code,news_links,source_type,cur_db)
logger.debug("crawl %s, end.............. "%code)
logger.info("number of news url is %s"%NEWS_URL_EXTRACTE)
except:
logger.error("crawl_page failed ,Error:%s" % traceback.format_exc())
finally:
if cxn_db:
cxn_db.close()
if status == 0 and len(news_links) > 0:
extract_news(code,news_links,source_type,cur_db)
logger.debug("crawl %s, end.............. "%code)
logger.info("number of news url is %s"%NEWS_URL_EXTRACTE)
except:
logger.error("crawl_page failed ,Error:%s" % traceback.format_exc())
finally:
if cxn_db:
cxn_db.close()
if __name__ == '__main__':
cxn_db = None
try:
cxn_db = get_cxn_db()
cur_db = cxn_db.cursor()
#构造一个boolm filter
initUrlsBloomFilter(cur_db,1)
# print bf.exists('https://mail.yahoo.com/?.intl=us&.lang=en-US&.src=ym')
# print bf.exists('http://www.latimes.com/la-fi-hy-helmet-safety-20150409-story.html')
# print bf.exists('http://www.baidu.com')
except:
logger.error("crawl_page failed ,Error:%s" % traceback.format_exc())
finally:
if cxn_db:
cxn_db.close()
if status == 0 and len(news_links) > 0:
extract_news(code, news_links, source_type, cur_db)
logger.debug("crawl %s, end.............. " % code)
logger.info("number of news url is %s" % NEWS_URL_EXTRACTE)
except:
logger.error("crawl_page failed ,Error:%s" % traceback.format_exc())
finally:
if cxn_db:
cxn_db.close()
if __name__ == '__main__':
cxn_db = None
try:
cxn_db = get_cxn_db()
cur_db = cxn_db.cursor()
#构造一个boolm filter
initUrlsBloomFilter(cur_db, 1)
# print bf.exists('https://mail.yahoo.com/?.intl=us&.lang=en-US&.src=ym')
# print bf.exists('http://www.latimes.com/la-fi-hy-helmet-safety-20150409-story.html')
# print bf.exists('http://www.baidu.com')
except:
logger.error("crawl_page failed ,Error:%s" % traceback.format_exc())
finally:
if cxn_db:
cxn_db.close()