def download_page(self, urlmd5, url, proxy):
downloader = PageDownload(proxy=proxy)
page = downloader.simple_download(url=url)
if page is not None:
file_name = self.extract_field_from_page(
page=page, reg=r'<li class="fl ml_1 mt_08 c999">([^=]+)</li>')
# if file_name is None:
# file_name = self.extract_field_from_page(page=page, reg=r'<h1>([^<]+)下载?</h1>')
singer_name = self.extract_field_from_page(
page=page,
reg=r'<li><a class="fl c3b ml_1 mt_08" href="http:'
r'//www.51ape.com/[^=]+/" title="[^=]+">([^=]+)'
r'</a></li>')
baiduyun_url = self.extract_field_from_page(
page=page, reg=r'href="(https?://pan.baidu.com/s/[^=]+)"')
baiduyun_password = self.extract_field_from_page(
page=page, reg=r'提取<em class="dn"></em>密码：(\w+)</b>')
sql = "insert into " + self.page_table + " values (%s,%s,%s,%s,%s,%s,now())"
db = MysqlHandle()
db.insert(sql=sql,
value_list=[(urlmd5, url, file_name, singer_name,
baiduyun_url, baiduyun_password)])
db.close()
update_sql = "update " + self.url_table + " set status=200 where urlmd5='%s'" % (
urlmd5)
db = MysqlHandle()
db.update(sql=update_sql)
db.close()
return True
else:
return False
def split_boundary_outline(b_type, city_code, name):
boundary_table = "bdmap_api_" + name + "_" + str(
city_code) + "_boundary_table"
sql = 'select lng_min,lat_min,lng_max,lat_max from ' + boundary_table + " where type=" + str(
b_type) + " and total_count=400"
db = MysqlHandle()
query_res = db.query(sql)
for (lng_min, lat_min, lng_max, lat_max) in query_res:
boundarys = split_boundary(float(lat_max), float(lat_min),
float(lng_max), float(lng_min), 10, 0.1)
for _boundary in boundarys:
_lng_min = _boundary[1][0]
_lat_min = _boundary[0][0]
_lng_max = _boundary[1][1]
_lat_max = _boundary[0][1]
_boundary_st = str(_boundary[0][0]) + "," + str(
_boundary[1][0]) + "," + str(_boundary[0][1]) + "," + str(
_boundary[1][1])
md5 = to_md5(_boundary_st)
db = MysqlHandle()
sql = "insert into " + boundary_table + " values(%s,%s,2,%s,%s,%s,%s,0,0,now())"
db.insert(
sql,
[[md5, _boundary_st, _lng_min, _lat_min, _lng_max, _lat_max]])
db.close()
def download_page(self, urlmd5, url, proxy):
downloader = PageDownload(proxy=proxy)
page = downloader.simple_download(url=url)
if page is not None:
file_name = self.extract_field_from_page(
page=page, reg=r'<h1>([^<]+)\[FLAC格式\]下载</h1>')
if file_name is None:
file_name = self.extract_field_from_page(
page=page, reg=r'<h1>([^<]+)下载?</h1>')
baiduyun_url = self.extract_field_from_page(
page=page,
reg=r'<meta name="description" content="'
r'[^<]*(https?://pan.baidu.com/[^\s]+) [^<]+"/>')
baiduyun_password = self.extract_field_from_page(
page=page,
reg=r'<meta name="description" content'
r'="[^<]+密码[\W]+(\w+)..."/>')
sql = "insert into " + self.page_table + " values (%s,%s,%s,%s,%s,now())"
db = MysqlHandle()
db.insert(sql=sql,
value_list=[(urlmd5, url, file_name, baiduyun_url,
baiduyun_password)])
db.close()
update_sql = "update " + self.url_table + " set status=200 where urlmd5='%s'" % (
urlmd5)
db = MysqlHandle()
db.update(sql=update_sql)
db.close()
return True
else:
return False
def filter_avaliable_ips():
db = MysqlHandle()
is_success = db.delete('DELETE FROM TEMP_IPS_MANAGE')
if not is_success:
db.close()
return
db.close()
sql = 'SELECT PROXY FROM AI_PROXY_IPS'
db = MysqlHandle()
# 查询出所有代理ip
IP_LIST = db.query(sql)
db.close()
for ip in IP_LIST:
PROXY = {'http': 'http://' + ip[0]} # 代理
print 'filtering ip:' + ip[0]
downloader = PageDownload(hd=mobike_headers, proxy=PROXY, timeout=3)
try:
post_data = {
'longitude': '121.1883',
'latitude': '31.05147',
'citycode': '021',
'errMsg': 'getMapCenterLocation:ok'
}
page = downloader.download_with_post(url=TEST_URL,
post_data=post_data)
if page is not None:
AVALIABLE_IPS.append(ip)
print ip[0] + " is ok!"
else:
pass
except Exception, e:
print str(e)
pass
def download_page(self, urlmd5, url, proxy):
downloader = PageDownload(proxy=proxy, timeout=5)
page = downloader.simple_download(url=url)
if page is not None:
company_name = self.extract_field_from_page(
page=page,
reg=r"""<meta name="keywords" content="([^=]+)"/>""",
index=-1)
company_net = self.extract_field_from_page(
page=page,
reg=
r'"/companyurlimg.php\?url=([^<]+)" alt="myImage" style="border:none'
)
address = self.extract_field_from_page(page=page.decode("utf-8"),
reg="<p>([^=]+)</p>",
index=-1)
# if file_name is None:
# file_name = self.extract_field_from_page(page=page, reg=r'<h1>([^<]+)下载?</h1>')
sql = "insert into " + self.page_table + " values (%s,%s,%s,%s,%s,now())"
db = MysqlHandle()
db.insert(sql=sql,
value_list=[(urlmd5, url, company_name, company_net,
address)])
db.close()
update_sql = "update " + self.url_table + " set status=200 where urlmd5='%s'" % (
urlmd5)
db = MysqlHandle()
db.update(sql=update_sql)
db.close()
return True
else:
return False
def download_page():
db = MysqlHandle()
query_sql = "select uid,min(name),min(line_type), min(page_url) from baidu_busline_url_analyse where status=0 group by uid "
page_infs = db.query(query_sql)
db.close()
downloader = PageDownload()
for item in page_infs:
print(item[0])
page = downloader.simple_download(item[3])
# if is_json(page):
# json_page = json.loads(page)
# if json_page.has_key("content"):
# main_info = json_page["content"][0]
# name = main_info["name"]
# timeable = main_info["timeable"]
db = MysqlHandle()
is_success = False
if page is not None:
insert_sql = "insert into baidu_busline_page values(%s,%s,%s,%s,NULL )"
is_success = db.insert(insert_sql,
[(item[0], item[1], item[2], page)])
if is_success and page is not None:
update_sql = "update baidu_busline_url_analyse set status=200 where uid='%s'" % (
item[0])
db.update(update_sql)
db.close()
def download_page(self, urlmd5, url, proxy):
downloader = PageDownload(proxy=proxy)
page = downloader.simple_download(url=url)
if page is not None:
#print page.decode("utf-8")
file_name = self.extract_field_from_page(page=page, reg=r'专辑名称：([^<]+)')
if file_name is None:
file_name = self.extract_field_from_page(page=page, reg=r'<h1 class="title">([^<]+)</h1>')
music_type = self.extract_field_from_page(page=page, reg=r' <a href="/\w+/">([^<]+)</a>')
# if file_name is None:
# file_name = self.extract_field_from_page(page=page, reg=r'<h1>([^<]+)下载?</h1>')
singer_name = self.extract_field_from_page(page=page, reg=r'专辑艺人：([^<]+)')
baiduyun_url = self.extract_field_from_page(page=page, reg=r"""<a href="#ecms" onclick="window.open\('([^<]+)','','width=300,height=300,resizable=yes'\)""")
print baiduyun_url
if baiduyun_url is None:
return False
if baiduyun_url is not None:
baiduyun_url = self.domain+baiduyun_url
baiduyun_password = self.extract_field_from_page(page=page, reg=r'密码: (\w+)')
sql = "insert into " + self.page_table + " values (%s,%s,%s,%s,%s,%s,%s,now())"
db = MysqlHandle()
db.insert(sql=sql, value_list=[(urlmd5, url, file_name,music_type, singer_name, baiduyun_url,
baiduyun_password)])
db.close()
update_sql = "update " + self.url_table + " set status=200 where urlmd5='%s'" % (urlmd5)
db = MysqlHandle()
db.update(sql=update_sql)
db.close()
return True
else:
return False
def download_page(self, urlmd5, url, proxy, url_table, page_table):
downloader = PageDownload(proxy=proxy)
page = downloader.simple_download(url=url)
if page is not None:
file_name = self.extract_field_from_page(
page=page,
reg=r'<h1 class="yh mt_1 f_32">([^<]+\.[a-z]+)</h1>')
file_size = self.extract_field_from_page(
page=page,
reg=r'<h3 class="c999 fl mt_05 f_12 n yh">'
r'<em class="n ml_1 mr_1">·</em>(\d+\.?\d+M)</h3>')
baiduyun_url = self.extract_field_from_page(
page=page, reg=r'href="(https?://pan.baidu.com/[^\s]+)"')
baiduyun_password = self.extract_field_from_page(
page=page, reg=r'<em class="dn"></em>密码：(\w+)</b>')
sql = "insert into " + page_table + " values (%s,%s,%s,%s,%s,%s,now())"
db = MysqlHandle()
db.insert(sql=sql,
value_list=[(urlmd5, url, file_name, file_size,
baiduyun_url, baiduyun_password)])
db.close()
update_sql = "update " + url_table + " set status=200 where urlmd5='%s'" % (
urlmd5)
db = MysqlHandle()
db.update(sql=update_sql)
db.close()
return True
else:
return False
def download_list_page(self, urlmd5, url, proxy, boundary, domain=None):
downloader = PageDownload(proxy=proxy)
page = downloader.simple_download(url=url)
if is_json(page):
json_page = json.loads(page)
result = json_page["result"]
total_count = result["total"]
print ("total:"+str(total_count))
if int(total_count) <= 10 and int(total_count)>0:
content = json_page["content"]
for item in content:
uid = item["uid"]
primary_uid = item["primary_uid"]
new_url = self.page_url % (uid, primary_uid)
new_urlmd5 = to_md5(in_str=new_url)
url_type = 0
boundary = None
status = 0
sql = "select * from " + self.url_table + " where urlmd5='%s'" % (new_urlmd5)
db = MysqlHandle()
results = db.query(sql=sql)
db.close()
if not results:
db = MysqlHandle()
insert_sql = "insert into " + self.url_table + " values (%s,%s,%s,%s,%s,now())"
db.insert(sql=insert_sql, value_list=[(new_urlmd5, new_url, url_type, boundary,status)])
db.close()
else:
print "This url is already in the database!!"
elif int(total_count) <= 0:
pass
else:
min_interval = boundary.split(";")[0]
max_interval = boundary.split(";")[1]
lat_min = min_interval.split(",")[1]
lat_max = max_interval.split(",")[1]
lng_min = min_interval.split(",")[0]
lng_max = max_interval.split(",")[0]
boundarys = split_boundary(int(float(lat_max)), int(float(lat_min)), int(float(lng_max)), int(float(lng_min)), 4, 0.2)
for _boundary in boundarys:
_boundary_st = str(_boundary[1][0])+","+str(_boundary[0][0])+";"+str(_boundary[1][1])+","+str(_boundary[0][1])
new_url = self.list_url % (self.city_code,self.keyword, _boundary_st)
new_urlmd5 = to_md5(in_str=new_url)
url_type = 1
boundary = _boundary_st
status = 0
db = MysqlHandle()
insert_sql = "insert into " + self.url_table + " values (%s,%s,%s,%s,%s,now())"
db.insert(sql=insert_sql, value_list=[(new_urlmd5, new_url, url_type, boundary, status)])
db.close()
update_sql = "update "+self.url_table+" set status=200 where urlmd5='%s'" % (urlmd5)
db = MysqlHandle()
db.update(sql=update_sql)
db.close()
return True
else:
return False
def spider(city_code, name, keyword, key_token):
boundary_table = "bdmap_api_" + name + "_" + str(
city_code) + "_boundary_table"
page_table = "bdmap_api_" + name + "_" + str(city_code) + "_page_table"
base_url = "http://api.map.baidu.com/place/v2/search?query=%s&scope=2&bounds=%s&output=json&ak=%s&page_num=%d"
sql = "select md5, boundary from " + boundary_table + " where status=0"
db = MysqlHandle()
res_data = db.query(sql)
for (md5, boundary) in res_data:
url = base_url % (keyword, boundary, key_token, 0)
downloader = PageDownload()
page = downloader.simple_download(url)
if is_json(page):
json_data = json.loads(page)
status = json_data["status"]
total = json_data["total"]
print(boundary, url, total)
if status == 0 and int(total) > 0:
page_count = int(total) / 10
for x in range(0, page_count + 2):
_url = base_url % (keyword, boundary, key_token, x)
downloader = PageDownload()
_page = downloader.simple_download(_url)
if is_json(_page):
_json_data = json.loads(_page)
results = _json_data["results"]
for item in results:
name = item["name"]
address = item["address"]
province = item["province"]
city = item["city"]
area = item["area"]
uid = item["uid"]
_md5 = to_md5(uid)
lat = item["location"]["lat"]
lng = item["location"]["lng"]
try:
tag = item["detail_info"]["tag"]
except Exception, e:
tag = None
print(e.message)
sql = "insert into " + page_table + " values(%s,%s,%s,%s,null,%s,%s,%s,%s,%s,null,null,%s,null,now(),null)"
db = MysqlHandle()
db.insert(sql, [[
_md5, uid, name, address, province, city, area,
lng, lat, tag
]])
db.close()
sql = 'update ' + boundary_table + ' set status=200,total_count=' + str(
total) + ' where md5="' + md5 + '"'
db = MysqlHandle()
db.update(sql)
db.close()
def download_page(self, urlmd5, url, proxy):
downloader = PageDownload(proxy=proxy)
page = downloader.simple_download(url=url)
if is_json(page):
page_json = json.loads(page)
content = page_json["content"]
uid = content["uid"]
name = content["name"]
address = content["addr"]
if content.has_key("phone"):
phone = content["phone"]
else:
phone = None
x = content["navi_x"]
y = content["navi_y"]
# geo = content["geo"]
ext = content["ext"]
if type(ext) == type({"ee":""}):
detail_info = ext["detail_info"]
else:
detail_info = {"info":""}
if detail_info.has_key("tag"):
tag = detail_info["tag"]
else:
tag = None
if detail_info.has_key("image"):
image = detail_info["image"]
else:
image = None
if detail_info.has_key("display_info_redu"):
dispaly_redu = detail_info["display_info_redu"]
else:
dispaly_redu = None
if detail_info.has_key("price"):
price = detail_info["price"]
else:
price = None
sql = "insert into " + self.page_table + " values (%s,%s,%s,%s,%s,%s,%s,%s,null,null,null,%s,%s,%s,null,%s,now())"
db = MysqlHandle()
db.insert(sql=sql, value_list=[(urlmd5, url, uid, name, address, phone, x, y, tag, image, price, dispaly_redu)])
db.close()
update_sql = "update " + self.url_table + " set status=200 where urlmd5='%s'" % (urlmd5)
db = MysqlHandle()
db.update(sql=update_sql)
db.close()
return True
else:
return False
def district_table(table_name):
query_sql = 'select distinct proxy from '+table_name
db = MysqlHandle()
proxys = db.query(query_sql)
db.close()
delete_sql = 'delete from '+table_name
db = MysqlHandle()
db.delete(delete_sql)
db.close()
db = MysqlHandle()
insert_sql = 'insert into '+table_name+' values (%s,now(),0,100)'
is_success = db.insert(insert_sql, proxys)
if is_success:
print u'The filtering has finished!'
db.close()
def GET(self):
inputs = web.input()
if inputs.has_key("token"):
token = inputs["token"]
if token == "whanys":
if inputs.has_key("username") and inputs.has_key("password"):
username = inputs["username"]
password = inputs["password"]
sql = "select * from user_table_yixin where username='******' and password='******' and status=0"
db = MysqlHandle()
res = db.query(sql)
if res:
result = {"status": "0", "msg": "success"}
else:
result = {"status": "1", "msg": "failed!"}
else:
result = {
"status": "1",
"msg": "failed,parameters not enough!"
}
else:
result = {
"status": "1",
"msg": "failed,your token is not true!"
}
else:
result = {"status": "1", "msg": "failed,you need a token!"}
return result
def download(city_code, l_type, name, tag, city):
page_table = "bdmap_api_" + name + "_" + str(city_code) + "_page_table"
sql = 'select uid from ' + page_table + ' where city="' + city + '" and geo is null'
db = MysqlHandle()
query_data = db.query(sql)
for (uid, ) in query_data:
update_geo_data(uid, l_type, city_code, name)
def generate_samples():
base_url = "http://api.map.baidu.com/geoconv/v1/?coords=%s,%s&from=5&to=6&ak=XwpZGfXMn45W9Czd1UwmC6RwMMULD1Ue"
work_book = xlwt.Workbook()
sheet = work_book.add_sheet("sheet")
sql = "select x , y from bdmap_api_school_218_page_table LIMIT 1000"
db = MysqlHandle()
query_res = db.query(sql)
i = 0
for (x, y) in query_res:
url = base_url % (str(x), str(y))
downloader = PageDownload()
page = downloader.simple_download(url)
if is_json(page):
json_page = json.loads(page)
status = json_page["status"]
if status == 0:
new_x = json_page["result"][0]["x"]
new_y = json_page["result"][0]["y"]
print (x, y, new_x, new_y)
sheet.write(i, 0, x)
sheet.write(i, 1, y)
sheet.write(i, 2, new_x)
sheet.write(i, 3, new_y)
i = i+1
work_book.save("sample.xls")
def download():
db = MysqlHandle()
sql = "select s.division, s.parent from divisions t,divisions s where t.division=s.parent and t.parent='湖北省' "
query_res = db.query(sql)
webSelenium = WebSelenium()
for (division, parent) in query_res:
division = division.replace("+", "")
division = division.replace("☆", "")
search_word = parent + division
if arcpy.Exists(r'G:\xin.data\spiders_data\hbs' + "\\" + search_word +
".shp"):
continue
print(division)
try:
webdriver = webSelenium.simple_download(
"http://127.0.0.1/common/get_bmap_boundary?city=" + division,
"chrome")
# webdriver = webSelenium.login_with_cookies(login_url="http://pan.baidu.com/s/1c03zJGW", cookies_data=cookies, domain="pan.baidu.com")
button_path = webdriver.find_elements_by_xpath(
"/html/body/input[2]")[0]
button_path.click()
time.sleep(5)
button_path.click()
button_download = webdriver.find_element_by_xpath(
"/html/body/input[3]")
time.sleep(5)
button_download.click()
time.sleep(3)
webdriver.close()
except Exception, e:
print(e.message)
def download_info_page(self, e_id):
page_url = self.page_base_url % (e_id)
print page_url
downloader = PageDownload()
page = downloader.simple_download(page_url)
if page:
site = re.findall(r"<br />网址:([^<]+)<br />",page)
if site:
site = site[0]
else:
site = None
company_name = re.findall(r"<strong>([^<]+)</strong>",page)
if company_name:
company_name = company_name[0]
else:
company_name = None
zw_num = re.findall(r'<span class="glyphicon glyphicon-envelope"></span> 展位号: (\w+)',page)
if zw_num:
zw_num = zw_num[0]
else:
zw_num = None
mail = re.findall(r'</span> 邮箱：<a href="mailto:([^<]+@[^<]+)">[^<]+@[^<]+</a>',page)
if mail:
mail = mail[0]
else:
mail = None
db = MysqlHandle()
sql = "insert into diecast VALUES (%s,%s,%s,%s)"
db.insert(sql,[(site,company_name,zw_num,mail)])
db.close()
def get_info(urlmd5, url, driver):
driver.get(url)
page = driver.page_source.encode("utf-8")
# print page
contacts = re.findall(r"<strong>联系人</strong>：([^=]+)<a", page)
if contacts:
contacts = contacts[0]
else:
contacts = None
e_mail = re.findall(r'<a href="mailto:([^"]+@[^"]+)"', page)
if e_mail:
e_mail = e_mail[0]
else:
e_mail = None
phone_num = re.findall(r"<p><strong>电话</strong>： ([^/]+)</p>", page)
if phone_num:
phone_num = phone_num[0]
else:
phone_num = None
#print phone_num
db = MysqlHandle()
sql = "insert into cphi_info_table values(%s,%s,%s,%s,%s)"
print e_mail
db.insert(sql, [(urlmd5, url, contacts, e_mail, phone_num)])
db.close()
def add_init_url(self, url_table_name, filter_config, city_code, keyword):
list_url = 'https://map.baidu.com/?newmap=1&reqflag=pcmap&biz=1&from=webmap&da_par=direct&pcevaname=pc4.1&qt=spot&from=webmap&c=%d&wd=%s&wd2=&pn=0&nn=0&db=0&sug=0&addr=0&pl_data_type=life&pl_sort_type=data_type&pl_sort_rule=0&pl_business_type=cinema&pl_business_id=&da_src=pcmappg.poi.page&on_gel=1&src=7&gr=3&l=12&rn=10&tn=B_NORMAL_MAP&ie=utf-8&b=(%s)'
url = filter_config["url"]
# db = MysqlHandle()
# insert_sql = "insert into " + url_table_name + " values(%s,%s,%s,%s,%s,now())"
# db.insert(sql=insert_sql, value_list=[(url["urlmd5"], url["url"], url["type"], url["boundary"], url["status"])])
# db.close()
boundary = url["boundary"]
min_interval = boundary.split(";")[0]
max_interval = boundary.split(";")[1]
lat_min = min_interval.split(",")[1]
lat_max = max_interval.split(",")[1]
lng_min = min_interval.split(",")[0]
lng_max = max_interval.split(",")[0]
boundarys = split_boundary(int(float(lat_max)), int(float(lat_min)),
int(float(lng_max)), int(float(lng_min)),
20, 0.2)
for _boundary in boundarys:
_boundary_st = str(_boundary[1][0]) + "," + str(
_boundary[0][0]) + ";" + str(_boundary[1][1]) + "," + str(
_boundary[0][1])
new_url = list_url % (city_code, keyword, _boundary_st)
new_urlmd5 = to_md5(in_str=new_url)
url_type = 2
boundary = _boundary_st
status = 0
db = MysqlHandle()
insert_sql = "insert into " + self.url_table + " values (%s,%s,%s,%s,%s,now())"
db.insert(sql=insert_sql,
value_list=[(new_urlmd5, new_url, url_type, boundary,
status)])
db.close()
def get_page_url(line_name, line_type, city_code, coords):
base_url = 'http://map.baidu.com/?newmap=1&reqflag=pcmap&biz=1&from=webmap&da_par=direct&pcevaname=pc4.1&qt=bl&da_src=searchBox.button&wd=%s&c=%d&l=13&b=(%s)&from=webmap&sug_forward=&tn=B_NORMAL_MAP&nn=0'
downloader = PageDownload(timeout=5)
page = downloader.simple_download(base_url %
(line_name, city_code, coords))
if is_json(page):
json_data = json.loads(page)
if not json_data.has_key('content'):
print base_url
return
contents = json_data['content']
line_list = []
for item in contents:
name = item['name']
if not item.has_key("uid"):
print name, base_url
continue
uid = item['uid']
page_url = 'http://map.baidu.com/?qt=bsl&tps=&newmap=1&uid=' + uid + '&c=%d' % (
city_code)
line_list.append((name, uid, page_url, line_type))
db = MysqlHandle()
insert_sql = "insert into baidu_busline_url_analyse values(%s,%s,%s,%s,0)"
db.insert(insert_sql, line_list)
db.close()
def download_page(self, base_url, geo, proxy, city_code, sec, page_table):
downloader = PageDownload(proxy=proxy, hd=mobike_headers)
post_dic = {
'longitude': str(geo[0]),
'latitude': str(geo[1]),
'citycode': str(city_code),
'errMsg': 'getMapCenterLocation:ok'
}
page = downloader.download_with_post(url=base_url, post_data=post_dic)
if is_json(page):
json_page = json.loads(page)
if json_page.has_key("object"):
mobike_object = json_page["object"]
items = []
for mobike in mobike_object:
bike_id = mobike["distId"]
bike_type = mobike["biketype"]
b_type = mobike["type"]
lng = mobike["distX"]
lat = mobike["distY"]
dis_source = str(geo[0]) + "," + str(geo[1])
item = (bike_id, bike_type, b_type, lat, lng, dis_source,
sec)
items.append(item)
db = MysqlHandle()
sql = "insert into " + page_table + " values(%s,%s,%s,%s,%s,%s,%s,now())"
db.insert(sql=sql, value_list=items)
return True
else:
return False
def run(self):
download_count = 0
total_count = 0
fail_count_20 = 0
start_time = time.time()
while 1:
try:
self.tuple_from_queue = self.queue.get()
if self.tuple_from_queue is None:
print u'urls pool is empty,please wait for 90s'
time.sleep(90)
break
else:
print self.tuple_from_queue
start_time_this_round = time.time()
download_count += 1
self.lock.acquire()
if not self.need_proxy:
self.proxy = None
download_result = self.spider.process(tuple_from_queue=self.tuple_from_queue, proxy=self.proxy)
self.lock.release()
total_count += download_result['total']
if download_result['total'] != download_result['success']: # 如果爬取url没有全部成功，失败的次数相应增加
fail_count_20 += download_result['total'] - download_result['success']
for failed_data in download_result['failed_list']: # 将失败的url记录再加入持久队列
self.queue.put(failed_data)
print failed_data
batch = download_count % self.CHKTHD
if batch == 0:
self.activity = 1 - float(fail_count_20) / float(total_count)
print '[%s]COUNT: %d, FAIL-IN-this%d:%d , avail:%f:' % (
self.proxy_ip, self.CHKTHD, total_count, fail_count_20,
self.activity)
fail_count_20 = 0
total_count = 0
if self.activity < 0.3:
print '[%s]rate of download is %f，too low' % (self.proxy_ip, self.activity)
db = MysqlHandle()
sql = "update temp_ips_manage set availabity=%s where proxy='%s'" % (
self.activity, self.proxy_ip)
db.update(sql=sql)
db.close()
# self.change_proxy()
self.proxy = {'http': 'http://' + self.proxy_manger.change_proxy(self.proxy_ip)[1]}
spider_time = time.time() - start_time
if spider_time > 600:
print '[%s]timeout，we will quit' % (self.proxy_ip)
self.proxy = {'http': 'http://' + self.proxy_manger.change_proxy(self.proxy_ip)[1]}
start_time = time.time()
elaspsed = time.time() - start_time_this_round
interval = random.randint(self.interval_down, self.interval_upp)
if elaspsed < interval:
time.sleep(interval - elaspsed)
except Exception,e:
print e.message
print self.tuple_from_queue
if type(self.tuple_from_queue)==list and len(self.tuple_from_queue)==3:
self.queue.put(self.tuple_from_queue)
def download_list_page(self,
urlmd5,
url,
proxy,
url_table,
filter_table,
domain=None):
downloader = PageDownload(proxy=proxy)
page = downloader.simple_download(url=url)
if page is not None:
new_urls = re.findall(self.reg, page)
for _url in new_urls:
if domain is not None:
if _url.startswith("/"):
new_url = domain + _url
else:
new_url = _url
else:
new_url = _url
url_type = self.filter_url(url=new_url,
filter_table=filter_table)
if url_type is not None:
new_urlmd5 = to_md5(in_str=new_url)
sql = "select * from " + url_table + " where urlmd5='%s'" % (
new_urlmd5)
db = MysqlHandle()
results = db.query(sql=sql)
db.close()
if not results:
db = MysqlHandle()
insert_sql = "insert into " + url_table + " values (%s,%s,%s,%s,now())"
db.insert(sql=insert_sql,
value_list=[(new_urlmd5, new_url, url_type,
0)])
db.close()
else:
print "This url is already in the database!!"
else:
pass
update_sql = "update " + url_table + " set status=200 where urlmd5='%s'" % (
urlmd5)
db = MysqlHandle()
db.update(sql=update_sql)
db.close()
return True
else:
return False
def add_filter_urls(self, filter_table_name, filter_config):
filters = filter_config["filters"]
value_list = []
for filter in filters:
value_list.append((filter["type"], filter["filter"]))
db = MysqlHandle()
insert_sql = "insert into "+filter_table_name+" values(%s,%s)"
db.insert(sql=insert_sql, value_list=value_list)
db.close()
def start_feed(self):
round_num = 0
last_url_count = 0
this_url_count = 0
same_count = 0
while self.run_sign:
if len(self.queue) < 2500:
start_time = time.time()
round_index = round_num % 10
db = MysqlHandle()
url_list = db.query(self.sql)
db.close()
count = 0
for url_data in url_list:
self.queue.put(
(url_data[0], url_data[1], url_data[2], url_data[3]))
update_sql = self.update_sql_base % url_data[0]
db = MysqlHandle()
db.update(update_sql)
db.close()
count += 1
print 'FinishedQueue-' + self.url_table + ': %d TIME ELASPSED: %f ' % (
count, time.time() - start_time)
if round_index == 0:
if this_url_count == last_url_count:
same_count += 1
else:
last_url_count = this_url_count
same_count = 0
this_url_count = 0
else:
this_url_count += 1
round_num += 1
else:
print 'The Queue is full!'
if same_count == 100:
print 'THE SAME NUM %d appeared for 10 rounds, feeding frequency turn down.'
time.sleep(360)
self.set_stop()
else:
time.sleep(5)
print 'Existed Successfully!!'
def load_urls():
db = MysqlHandle()
sql = "select urlmd5,url from cphi_page_table where urlmd5 not in (select urlmd5 FROM cphi_info_table)"
results = db.query(sql)
queue = Queue()
for item in results:
_url = item[1].replace("/introduce/", "/")
#print _url
queue.put_nowait((item[0], _url + "contactinfo/"))
return queue
def filter_url(self, url):
sql = "select type, filter from "+self.filter_table
db = MysqlHandle()
filter_data = db.query(sql=sql)
for _filter in filter_data:
_type = _filter[0]
_url = _filter[1]
if re.match(_url, url) is not None:
return _type
return None
def spider(d_type):
db = MysqlHandle()
sql = "select division, abb_name from divisions where status=0 and type =" + str(
d_type)
query_res = db.query(sql)
for (division, abb_name) in query_res:
print division
try:
download(division, d_type, abb_name)
except Exception, e:
print(e.message)
def download_list_page(self, urlmd5, url, proxy, domain=None):
downloader = PageDownload(proxy=proxy,timeout=10)
page = downloader.simple_download(url=url)
if page is not None:
new_urls = re.findall(self.reg, page)
# singer_names = re.findall(self.js0_reg, page)
# for singer_name in singer_names:
# merge_url = "http://www.51ape.com/skin/ape/php/qx_2.php?qx=" + singer_name
# new_urls.append(merge_url)
for _url in new_urls:
if domain is not None:
if _url.startswith("/"):
new_url = domain + _url
else:
new_url = _url
else:
new_url = _url
url_type = self.filter_url(url=new_url)
if url_type is not None:
new_urlmd5 = to_md5(in_str=new_url)
sql = "select * from "+self.url_table+" where urlmd5='%s'" % (new_urlmd5)
db = MysqlHandle()
results = db.query(sql=sql)
db.close()
if not results:
db = MysqlHandle()
insert_sql = "insert into "+self.url_table+" values (%s,%s,%s,%s,now())"
db.insert(sql=insert_sql, value_list=[(new_urlmd5, new_url, url_type, 0)])
db.close()
else:
print "This url is already in the database!!"
else:
pass
update_sql = "update "+self.url_table+" set status=200 where urlmd5='%s'" % (urlmd5)
db = MysqlHandle()
db.update(sql=update_sql)
db.close()
return True
else:
return False
def download_page(self, urlmd5, url, proxy):
downloader = PageDownload(proxy=proxy)
page = downloader.simple_download(url=url)
if page is not None:
file_name = self.extract_field_from_page(
page=page,
reg=r'<h2>([^\]]+) <span class="c?b?">'
r'WAV</span>')
# if file_name is None:
# file_name = self.extract_field_from_page(page=page, reg=r'<h1>([^<]+)下载?</h1>')
singer_name = self.extract_field_from_page(
page=page,
reg=r'<p><a href="/detail/\d+.html" style="color:'
r'#217fbc">([^=]+)</a></p>')
result = self.extract_field_from_page(
page=page,
reg=
r'<p class="downurl">链接: (https?://pan.baidu.com/s/[^=]+) 密码: (\w+)</p>'
)
if result is not None:
[baiduyun_url, baiduyun_password] = result
else:
baiduyun_url = None
baiduyun_password = None
sql = "insert into " + self.page_table + " values (%s,%s,%s,%s,%s,%s,now())"
db = MysqlHandle()
db.insert(sql=sql,
value_list=[(urlmd5, url, file_name, singer_name,
baiduyun_url, baiduyun_password)])
db.close()
update_sql = "update " + self.url_table + " set status=200 where urlmd5='%s'" % (
urlmd5)
db = MysqlHandle()
db.update(sql=update_sql)
db.close()
return True
else:
return False
