def initialization_new_listpage_url_cache(database_config):
# 初始化 new_listpage_url_cache
# 读取 website, 删掉 new_listpage_url_cache 中目标website_no的记录，然后再重新插入主页url
# 清空 new_listpage_url_cache 已采的数据
delete_cache_sql = f"delete from new_listpage_url_cache where website_no in (select website_no from website);"
# common.query_mysql(database_config, delete_cache_sql)
website_sql = 'select * from website;'
website_result = common.query_mysql(database_config, website_sql)
for website in website_result:
Website_No = website['Website_No']
Listpage_Title = website['Website_Name']
Listpage_URL = website['Website_Main_Page_URL']
Is_Need_VPN = website['Is_Need_VPN']
# 计算url MD5 先去掉http和末尾斜杆
md5_source = Listpage_URL
md5_source = md5_source.replace('http://', '')
md5_source = md5_source.replace('https://', '')
md5_source = md5_source.rstrip('/')
Listpage_URL_MD5 = common.get_token(md5_source)
Domain_Code = common.get_domain_code(Listpage_URL)
Host_Code = common.get_host_code(Listpage_URL)
Score_Detail = '{"status": True, "message": "root page"}'
# 插入主页到 new_listpage_url_cache
insert_url_to_cache = f"""
insert ignore into new_listpage_url_cache(Column_Extraction_Deep, Listpage_URL,
Listpage_Title, Domain_Code, Host_Code, Listpage_URL_MD5, Level_Score, Score_Detail, Website_No, Is_Need_VPN)
value(1, '{Listpage_URL}', '{Listpage_Title}', '{Domain_Code}', '{Host_Code}', '{Listpage_URL_MD5}',
100, '{Score_Detail}', '{Website_No}', {Is_Need_VPN});
"""
print(insert_url_to_cache)
common.query_mysql(database_config, insert_url_to_cache)
def main():
# 导入参数文件
host = conf.get("database", "host")
port = conf.get("database", "port")
user = conf.get("database", "user")
passwd = conf.get("database", "passwd")
db = conf.get("database", "db")
table = conf.get("database", "table")
reject_domain_file = conf.get("reject_domain", "file_name")
# 数据库配置
# database_config = {'host': '192.168.1.118', 'port': 3306, 'user': '******', 'passwd': 'poms@db', 'db': 'datasource'}
database_config = {
'host': host,
'port': int(port),
'user': user,
'passwd': passwd,
'db': db
}
# 读取reject_domain
reject_domain_list = get_reject_domain_list(reject_domain_file)
# 执行删除操作
for str_domain in reject_domain_list:
logger.info(str_domain)
domain = common.get_domain_code(str_domain)
sql = f'''delete from {table} where domain_code='{domain}';'''
result = query_mysql(database_config, sql)
if result > 0:
logger.info(sql)
logger.info('delete count: ' + str(result))
def main():
extractor_116 = {
'host': '192.168.1.116',
'port': 3306,
'user': '******',
'passwd': 'poms@db',
'db': 'mymonitor'
}
extractor_118 = {
'host': '192.168.1.118',
'port': 3306,
'user': '******',
'passwd': 'poms@db',
'db': 'datasource'
}
select_column = f"""select ad.Article_URL, ad.Domain_Code, ad.Article_Source from article_detail ad
where 1=1 and ad.Website_No in (select Website_No from website where Website_Name like'%百度新闻%')
and ad.Extracted_Time>'2020-10-01' and ad.Extracted_Time<'2020-11-25'
and Article_Source is not NULL and Article_Source !='' GROUP BY Domain_Code;"""
try:
results = common.query_mysql(extractor_116, select_column)
except Exception as e:
results = []
column_list = []
for i in results:
title = i['Article_Source']
listpage_url = i['Article_URL']
domain_code = common.get_domain_code(listpage_url)
host_code = common.get_host_code(listpage_url)
host_code_index = listpage_url.index(host_code) + len(host_code)
listpage_url = listpage_url[0:host_code_index] + '/'
# 计算url MD5 先去掉http和末尾斜杆
md5_source = listpage_url
md5_source = md5_source.replace('http://', '')
md5_source = md5_source.replace('https://', '')
md5_source = md5_source.rstrip('/')
record_md5_id = common.get_token(md5_source)
level_score = '100'
Score_Detail = '"{\"status\": True, \"message\": \"root page\"}"'
website_no = 'BAIDU_NEWS'
column_extraction_deep = 0
column = f"('{title}', '{listpage_url}', '{record_md5_id}', '{website_no}', {column_extraction_deep}, '{domain_code}', '{host_code}', '{level_score}', '{Score_Detail}')"
# column_list.append(column)
print(column)
# 批量插入
values = column
insert_column = f"replace into column_link(Title, URL, record_md5_id, website_no, column_extraction_deep, domain_code, host_code, level_score, Score_Detail) values{values};"
# print(insert_column)
try:
common.query_mysql(extractor_118, insert_column)
except Exception as e:
pass
def main(start, end):
extractor_116 = {
'host': '192.168.1.116',
'port': 3306,
'user': '******',
'passwd': 'poms@db',
'db': 'mymonitor'
}
extractor_118 = {
'host': '192.168.1.118',
'port': 3306,
'user': '******',
'passwd': 'poms@db',
'db': 'datasource'
}
select_column = f"select Website_No,ListPage_URL,ListPage_Title from listpage_url where ListPage_URL_ID IN " \
f"(select ListPage_URL_ID from cloud_listpage_url where cloud_listpage_url_id BETWEEN {start} AND {end});"
try:
results = common.query_mysql(extractor_116, select_column)
column_list = []
for item in results:
URL = item['ListPage_URL']
Title = item['ListPage_Title']
Website_No = item['Website_No']
Domain_Code = common.get_domain_code(URL)
Host_Code = common.get_host_code(URL)
# 计算url MD5 先去掉http和末尾斜杆
md5_source = URL
md5_source = md5_source.replace('http://', '')
md5_source = md5_source.replace('https://', '')
md5_source = md5_source.rstrip('/')
Record_MD5_ID = common.get_token(md5_source)
Level_Score, Score_Detail = common.is_need_filter(Title, URL)
Column_Extraction_Deep = 1
column = f"({Column_Extraction_Deep}, '{URL}', '{Title}', '{Domain_Code}', '{Host_Code}', '{Record_MD5_ID}', {Level_Score}, '{Score_Detail}', '{Website_No}')"
if Level_Score > 20:
column_list.append(column)
# 批量插入
values = ",".join(column_list)
insert_column = f"insert ignore into column_link(Column_Extraction_Deep, URL, Title, Domain_Code, Host_Code, Record_MD5_ID, Level_Score, Score_Detail, Website_No) values{values};"
# print(insert_column)
common.query_mysql(extractor_118, insert_column)
except Exception as e:
print(e)
def main():
extractor_118 = {
'host': '192.168.1.118',
'port': 3306,
'user': '******',
'passwd': 'poms@db',
'db': 'datasource'
}
select_column = f"select * from website where ID>1577;"
try:
results = common.query_mysql(extractor_118, select_column)
column_list = []
for item in results:
ID = item['ID']
Column_Extraction_Deep = '0'
URL = item['URL']
Title = item['Title']
Domain_Code = common.get_domain_code(URL)
Host_Code = common.get_host_code(URL)
# 计算url MD5 先去掉http和末尾斜杆
md5_source = URL
md5_source = md5_source.replace('http://', '')
md5_source = md5_source.replace('https://', '')
md5_source = md5_source.rstrip('/')
Record_MD5_ID = common.get_token(md5_source)
Level_Score = '100'
Score_Detail = '{"status": True, "message": "root page"}'
Website_No = 'OVERSEA'
column = f"({Column_Extraction_Deep}, '{URL}', '{Title}', '{Domain_Code}', '{Host_Code}', '{Record_MD5_ID}', {Level_Score}, '{Score_Detail}', '{Website_No}')"
column_list.append(column)
# # 更新md5
update_website = f"update website set record_md5_id='{Record_MD5_ID}' where ID={ID}"
common.query_mysql(extractor_118, update_website)
insert_column = f"insert ignore into column_link_oversea(Column_Extraction_Deep, URL, Title, Domain_Code, Host_Code, Record_MD5_ID, Level_Score, Score_Detail, Website_No) values{column};"
print(insert_column)
common.query_mysql(extractor_118, insert_column)
# # 批量插入
# values = ",".join(column_list)
# insert_column = f"insert ignore into column_link_oversea(Column_Extraction_Deep, URL, Title, Domain_Code, Host_Code, Record_MD5_ID, Level_Score, Score_Detail, Website_No) values{values};"
# print(insert_column)
# common.query_mysql(extractor_118, insert_column)
except Exception as e:
print(e)
def parse_html_to_database(database_config, url, column_extraction_deep, domain_code_source, website_no, Is_Need_VPN, text):
try:
root = etree.HTML(text, parser=etree.HTMLParser(encoding='utf-8'))
column_extraction_deep = 0
items = root.xpath('//a')
column_list = []
for num, item in enumerate(items):
title = "".join(item.xpath('.//text()'))
listpage_url = "".join(item.xpath('./@href'))
if (len(title) > 0) and ('go.php?' in listpage_url):
listpage_url = listpage_url.replace('/go.php?', '')
listpage_url = listpage_url.replace('http://www.01-114.com', '')
print(title, listpage_url)
# 去掉标点符号
# 计算url MD5 先去掉http和末尾斜杆
md5_source = listpage_url
md5_source = md5_source.replace('http://', '')
md5_source = md5_source.replace('https://', '')
md5_source = md5_source.rstrip('/')
record_md5_id = common.get_token(md5_source)
domain_code = common.get_domain_code(listpage_url)
host_code = common.get_host_code(listpage_url)
host_code_index = listpage_url.index(host_code) + len(host_code)
listpage_url = listpage_url[0:host_code_index] + '/'
# 垃圾词、垃圾域名过滤
level_score = 100
score_detail = '{"status": True, "message": "root page"}'
if level_score > 20:
column = f"('{title}', '{listpage_url}', '{record_md5_id}', '{website_no}', {column_extraction_deep}, '{domain_code}', '{host_code}', {level_score}, '{score_detail}')"
column_list.append(column)
# 批量插入
print(len(column_list))
values = ",".join(column_list)
insert_column = f"replace into column_link(Title, URL, record_md5_id, website_no, column_extraction_deep, domain_code, host_code, level_score, score_detail) values{values};"
# print(insert_column)
query_mysql(database_config, insert_column)
return True
except Exception as e:
pass
def main():
extractor_118 = {
'host': '192.168.1.133',
'port': 3306,
'user': '******',
'passwd': 'poms@db',
'db': 'mymonitor'
}
# select_column = f"select Article_Current_Node_HTML, Website_URL from column_root_source where 1=1 " \
# f"and Source='baidu_news';"
select_column = f"select Website_URL,Website_Title,Website_Description,Website_Keywords from column_root_source " \
f"where 1=1 and Source='baidu_web' " \
f"and Website_Title not like '%新闻%' and Website_Title not like'%资讯%' and Website_Title not like'%论坛%' and Website_Title not like'%社区%'" \
f"and Website_Keywords like '%新闻%' and Website_Keywords like'%资讯%' and Website_Keywords like'%论坛%' and Website_Keywords like'%社区%';"
try:
results = common.query_mysql(extractor_118, select_column)
except Exception as e:
results = []
column_list = []
for i in results:
title = i['Website_Title']
listpage_url = i['Website_URL']
domain_code = common.get_domain_code(listpage_url)
host_code = common.get_host_code(listpage_url)
host_code_index = listpage_url.index(host_code) + len(host_code)
listpage_url = listpage_url[0:host_code_index] + '/'
# 计算url MD5 先去掉http和末尾斜杆
md5_source = listpage_url
md5_source = md5_source.replace('http://', '')
md5_source = md5_source.replace('https://', '')
md5_source = md5_source.rstrip('/')
record_md5_id = common.get_token(md5_source)
try:
title = title.split('-')[0].strip()
except Exception as e:
title = title
try:
title = title.split('_')[0].strip()
except Exception as e:
title = title
try:
title = title.split(',')[0].strip()
except Exception as e:
title = title
try:
title = title.split('，')[0].strip()
except Exception as e:
title = title
try:
title = title.split('|')[0].strip()
except Exception as e:
title = title
try:
title = title.split(' ')[0].strip()
except Exception as e:
title = title
level_score = '100'
Score_Detail = '"{\"status\": True, \"message\": \"root page\"}"'
# website_no = 'BAIDU_NEWS'
website_no = 'BAIDU_WEB'
column_extraction_deep = 0
column = f"('{title}', '{listpage_url}', '{record_md5_id}', '{website_no}', {column_extraction_deep}, '{domain_code}', '{host_code}', '{level_score}', '{Score_Detail}')"
# column_list.append(column)
print(column)
# 批量插入
values = column
# insert_column = f"replace into column_link(Title, URL, record_md5_id, website_no, column_extraction_deep, domain_code, host_code, level_score, Score_Detail) values{values};"
insert_column = f"insert ignore into column_link(Title, URL, record_md5_id, website_no, column_extraction_deep, domain_code, host_code, level_score, Score_Detail) values{values};"
# print(insert_column)
try:
common.query_mysql(extractor_118, insert_column)
except Exception as e:
pass
def main():
fn = open('listpage.txt', 'r', encoding='UTF-8') # 打开文件
column_list = []
for i in fn:
try:
title = i.split('=')[0].strip()
listpage_url = i.split('=')[1].strip()
domain_code = common.get_domain_code(listpage_url)
host_code = common.get_host_code(listpage_url)
host_code_index = listpage_url.index(host_code) + len(host_code)
listpage_url = listpage_url[0:host_code_index] + '/'
# 计算url MD5 先去掉http和末尾斜杆
md5_source = listpage_url
md5_source = md5_source.replace('http://', '')
md5_source = md5_source.replace('https://', '')
md5_source = md5_source.rstrip('/')
record_md5_id = common.get_token(md5_source)
if len(title) < 1:
try:
response = requests.get(listpage_url, timeout=5)
# print(response.status_code)
if response.status_code == 200:
encoding = get_encoding(response)
# print(encoding)
response.encoding = encoding
soup = BeautifulSoup(response.text, 'lxml')
# print(soup.title.text)
title = soup.title.text
else:
continue
except:
pass
try:
title = title.split('－')[0].strip()
except Exception as e:
title = title
try:
title = title.split('_')[0].strip()
except Exception as e:
title = title
try:
title = title.split('-')[0].strip()
except Exception as e:
title = title
level_score, Score_Detail = common.is_need_filter(
title, listpage_url, True)
# print(level_score, Score_Detail, title, listpage_url)
if level_score > -100:
level_score = '100'
Score_Detail = '"{\"status\": True, \"message\": \"root page\"}"'
website_no = 'AD_SELECTED'
column_extraction_deep = 0
column = f"('{title}', '{listpage_url}', '{record_md5_id}', '{website_no}', {column_extraction_deep}, '{domain_code}', '{host_code}', '{level_score}', '{Score_Detail}')"
# column_list.append(column)
print(column)
# 批量插入
extractor_118 = {
'host': '192.168.1.118',
'port': 3306,
'user': '******',
'passwd': 'poms@db',
'db': 'mymonitor'
}
values = column
insert_column = f"insert ignore into column_link(Title, URL, record_md5_id, website_no, column_extraction_deep, domain_code, host_code, level_score, Score_Detail) values{values};"
# print(insert_column)
try:
common.query_mysql(extractor_118, insert_column)
except Exception as e:
pass
except Exception as e:
pass
fn.close() # 关闭文件
def parse_html_to_database(database_config, url, column_extraction_deep,
domain_code_source, website_no, Is_Need_VPN, text):
try:
root = etree.HTML(text, parser=etree.HTMLParser(encoding='utf-8'))
column_extraction_deep = int(column_extraction_deep) + 1
items = root.xpath('//a')
column_list = []
for num, item in enumerate(items):
title = "".join(item.xpath('.//text()'))
listpage_url = "".join(item.xpath('./@href'))
listpage_url = urljoin(url, listpage_url)
# 去掉标点符号
title = common.filter_punctuation(title)
listpage_url = common.match_url(listpage_url)
# 计算url MD5 先去掉http和末尾斜杆
md5_source = listpage_url
md5_source = md5_source.replace('http://', '')
md5_source = md5_source.replace('https://', '')
md5_source = md5_source.rstrip('/')
record_md5_id = common.get_token(md5_source)
domain_code = common.get_domain_code(listpage_url)
host_code = common.get_host_code(listpage_url)
# domain 要与源域名一致
if domain_code_source != domain_code:
continue
# 计算a节点占全部a节点百分比，如果总节点小于50就去前50%的节点；如果在50和200之前，就取前30%；大于200，就取前20%
len_items = len(items)
node_percent = num / len(items)
# print(num, 'percent:{:.0%}'.format(node_percent), title)
logger.debug(
str(num) + 'percent:{:.0%}'.format(node_percent) + title)
if len_items < 50:
if node_percent > 0.5:
continue
if (len_items >= 50) and (len_items <= 200):
if node_percent > 0.3:
continue
if len_items > 200:
if node_percent > 0.2:
continue
# 垃圾词、垃圾域名过滤
level_score, score_detail = common.is_need_filter(
title, listpage_url, True)
# print(level_score, score_detail)
logger.debug(str(level_score) + '=' + score_detail)
# 入库分值，新闻要大于等于20，论坛要大于等于10
valid_score = 20
media_type = common.get_media_type(listpage_url)
if media_type == 'forum':
valid_score = 10
if level_score >= valid_score:
column = f"('{title}', '{listpage_url}', '{record_md5_id}', '{website_no}', {column_extraction_deep}, '{domain_code}', '{host_code}', '{level_score}', '{score_detail}')"
column_list.append(column)
# 批量插入
values = ",".join(column_list)
insert_column = f"insert ignore into column_link(Title, URL, record_md5_id, website_no, column_extraction_deep, domain_code, host_code, level_score, score_detail) values{values};"
# print(insert_column)
logger.info('column count: ' + str(len(column_list)))
count = common.query_mysql(database_config, insert_column)
# logger.info('insert count: ' + str(count))
except Exception as e:
logger.error(str(e))
async def get_response(semaphore,
url,
column_extraction_deep=1,
domain_code_source=None,
website_no=None):
try:
async with semaphore:
timeout = aiohttp.ClientTimeout(total=20)
# ValueError: too many file descriptoersin select()报错问题
# 一般是并发请求数太大导致的，通常通过减少并发数解决。
#
# 我遇到的情况：并发量设置的不高，运行一段时间后报该错误。通过搜索、调试，最后看aiohttp文档时发现是因为请求的https站点的服务器没有正确完成ssl连接，需要指定一个叫enable_cleanup_closed的参数为True：
#
# session = aiohttp.ClientSession(connector=aiohttp.TCPConnector(enable_cleanup_closed=True)
# 官方对enable_cleanup_closed参数的解释：
#
# Some ssl servers do not properly complete SSL shutdown process, in that case asyncio leaks SSL connections.
# If this parameter is set to True, aiohttp additionally aborts underlining transport after 2 seconds. It is off by default.
#
# 作者：Ovie
# 链接：https://www.jianshu.com/p/f7af4466f346
# 来源：简书
# 著作权归作者所有。商业转载请联系作者获得授权，非商业转载请注明出处。
connector = aiohttp.TCPConnector(
limit=60, verify_ssl=False,
enable_cleanup_closed=True) # 60小于64。也可以改成其他数
async with aiohttp.ClientSession(timeout=timeout,
connector=connector) as session:
headers = {
"user-agent":
"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/81.0.4044.113 Safari/537.36",
}
async with session.get(url, headers=headers) as response:
text = await response.text()
root = etree.HTML(
text, parser=etree.HTMLParser(encoding='utf-8'))
# print(url)
logging.info(url)
column_extraction_deep = int(column_extraction_deep) + 1
items = root.xpath('//a')
column_list = []
# print(len(items))
for item in items:
title = "".join(item.xpath('.//text()'))
listpage_url = "".join(item.xpath('./@href'))
listpage_url = urljoin(url, listpage_url)
# 去掉标点符号
title = common.filter_punctuation(title)
listpage_url = common.match_url(listpage_url)
# 计算url MD5 先去掉http和末尾斜杆
md5_source = listpage_url
md5_source = md5_source.replace('http://', '')
md5_source = md5_source.replace('https://', '')
md5_source = md5_source.rstrip('/')
record_md5_id = common.get_token(md5_source)
domain_code = common.get_domain_code(listpage_url)
host_code = common.get_host_code(listpage_url)
# domain 要与源域名一致
if domain_code_source != domain_code:
continue
# 垃圾词、垃圾域名过滤
level_score, score_detail = common.is_need_filter(
title, listpage_url, False)
# print(level_score, score_detail)
logging.info(str(level_score) + '=' + score_detail)
if level_score > 20:
column = f"('{title}', '{listpage_url}', '{record_md5_id}', '{website_no}', {column_extraction_deep}, '{domain_code}', '{host_code}', '{level_score}', '{score_detail}')"
column_list.append(column)
# 批量插入
values = ",".join(column_list)
insert_column = f"insert ignore into column_link(Title, URL, record_md5_id, website_no, column_extraction_deep, domain_code, host_code, level_score, score_detail) values{values};"
# print(insert_column)
common.query_mysql(extractor_118, insert_column)
return True
except Exception as e:
if len(str(e)) > 0:
logging.error(str(e))
return False
