def click1(url, proxy):
try:
ip = utils.get_proxy(1, proxy)
utils.write_debug(utils.LINE(), 'request', ip)
proxies = {"http": "http://" + ip, "https": "https://" + ip}
UA = random.choice(userAgent.USERAGENT['pc'])
headers = {"User-Agent": UA}
requests.adapters.DEFAULT_RETRIES = 5
session = requests.session()
session.keep_alive = False
session.get(url, proxies=proxies, headers=headers)
time.sleep(30 + random.random() * 30)
except Exception:
utils.write_debug(utils.LINE(), 'request', traceback.print_exc())
def select_one(condition={}, collection='ip_pool', db=DB):
try:
db = get_db(db)
results = db[collection].find_one(condition)
return results
except:
utils.write_debug(utils.LINE(), "dealIp", traceback.print_exc())
def browser_click1(url, proxy):
try:
WIDTH = 1336
HEIGHT = 750
PIXEL_RATIO = 3.0
UA = random.choice(userAgent.USERAGENT['google'] +
userAgent.USERAGENT['ios'])
mobileEmulation = {
"deviceMetrics": {
"width": WIDTH,
"height": HEIGHT,
"pixelRatio": PIXEL_RATIO
},
"userAgent": UA
}
options = webdriver.ChromeOptions()
# proxy = int(proxy)
# ip = utils.get_proxy(1,proxy)
# utils.write_debug(utils.LINE(), 'browser', ip)
# options.add_argument('--proxy-server=%s' % ip)
options.add_argument("headless")
options.add_argument("disable-gpu")
options.add_experimental_option('mobileEmulation', mobileEmulation)
driver = webdriver.Chrome(chrome_options=options)
driver.set_page_load_timeout(20)
driver.implicitly_wait(10)
driver.get(url)
time.sleep(30 + random.random() * 30)
# el = WebDriverWait(driver,60,0.5).until(EC.presence_of_element_located((By.XPATH,'//*[@id="bs-example-navbar-collapse-1"]/ul/li[3]/a')))
# el.click()
# time.sleep(random.random() * 5)
driver.quit()
except Exception:
driver.quit()
utils.write_debug(utils.LINE(), 'browser', traceback.print_exc())
def insert(datas, collection='ip_pool', db=DB):
try:
db = get_db(db)
# for data in datas:
# data['create_at'] = time.strftime('%Y-%m-%d %H:%M:%S')
result = db[collection].insert(datas)
return result.inserted_ids
except:
utils.write_debug(utils.LINE(), "dealIp", traceback.print_exc())
def hour_task():
# 控量,每小时查找任务
print('start')
now = datetime.datetime.now()
hour = 'field_' + str(now.hour)
# 进程队列
ts = []
# 查找status不为2,且时间已到的任务
try:
query_results = ClickTask.filter(task_status__in=[0, 1],
task_time=datetime.date.today())
for query_result in query_results:
# url = query_result.url
md5 = query_result.url_md5
# proxy = query_result.task_area
time_task_obj = TimeTask.get(url_md5=md5)
# querySet序列化
time_task_dic = json.loads(
serializers.serialize('json', [time_task_obj]))[0]
count = time_task_dic['fields'][hour]
params = json.loads(serializers.serialize(
'json', [query_result]))[0]['fields']
params['count'] = count
# celery中起子进程会报错
# AssertionError: daemonic processes are not allowed to have children
process = multiprocessing.current_process()
# 进程启动前设为非守护进程启动
process.daemon = False
p = multiprocessing.Process(target=browser.browser_click, args=(params,)) \
if int(query_result.request) == 1 else multiprocessing.Process(target=request.click,
args=(params,))
p.start()
ts.append(p)
# 启动后再设为true
process.daemon = True
utils.write_debug(
utils.LINE(), 'tasks', 'start:' + md5 + ' time: ' +
time.strftime('%X') + '\ncount:' + str(count))
# status置位
query_results.update(task_status=1)
for t in ts:
t.join()
except Exception:
utils.write_debug(utils.LINE(), 'tasks', traceback.print_exc())
def update(data, condition, collection='ip_pool', db=DB):
try:
db = get_db(db)
# "$currentDate": {"lastModified": True}
data['update_at'] = time.strftime('%Y-%m-%d %H:%M:%S')
db[collection].update_many(condition, {
"$set": data,
})
except:
utils.write_debug(utils.LINE(), "dealIp", traceback.print_exc())
def dealIp():
m = 0
try:
datas = {}
ips = utils.get_proxy(10)
n = len(ips.split('\r\n'))
for data in ips.split('\r\n'):
datas['_id'] = data
if select_one({'_id': data}, 'ip_pool'):
pass
else:
insert([datas], 'ip_pool')
m += 1
datas = {}
utils.write_debug(utils.LINE(), "dealIp",
'total ip num is {}'.format(n))
utils.write_debug(utils.LINE(), "dealIp",
'success ip num is {}'.format(m))
except:
utils.write_debug(utils.LINE(), "dealIp", traceback.print_exc())
def click1(url):
try:
ip = utils.get_proxy(1)
utils.write_debug(utils.LINE(), 'request', ip)
proxies = {"http": "http://" + ip, "https": "https://" + ip}
UA = random.choice(userAgent.USERAGENT['pc'])
headers = {"User-Agent": UA,"Referer":"http://www.ebrun.com/retail/b2c/"}
requests.adapters.DEFAULT_RETRIES = 5
session = requests.session()
session.keep_alive = False
r = session.get(url, proxies=proxies, headers=headers)
time.sleep(random.random() * 3)
soup = BeautifulSoup(r.text,'lxml')
links = soup.find_all('a',class_='buyEm')
link = random.choice(links)
url = link.get('href')
print (url)
session.get(url, proxies=proxies, headers=headers)
time.sleep(random.random()*30)
except Exception:
utils.write_debug(utils.LINE(), 'request', traceback.print_exc())
def check_task():
try:
results = TaskCount.filter(task_status=0)
for result in results:
if result.finish_count >= result.task_count:
result.task_status = 1
md5 = result.url_md5
result.save(update_fields=['task_status'])
q = ClickTask.get(url_md5=md5)
q.task_status = 2
q.save(update_fields=['task_status'])
except Exception:
utils.write_debug(utils.LINE(), 'tasks', traceback.print_exc())
# @app.task
# def day_task():
# # 不控量
# today = datetime.date.today()
# # 进程队列
# ts = []
# # 查找status不为2,且时间已到的任务
# try:
# query_results = ClickTask.filter(task_status=0, task_detail=0, task_time=today)
# for query_result in query_results:
# url = query_result.url
# md5 = query_result.url_md5
# proxy = query_result.task_area
# task_count_obj = TaskCount.get(url_md5=md5)
# # querySet序列化
# time_task_dic = json.loads(serializers.serialize('json', [task_count_obj]))[0]
# count = time_task_dic['fields']['task_count']
# # celery中起子进程会报错
# # AssertionError: daemonic processes are not allowed to have children
# process = multiprocessing.current_process()
# # 进程启动前设为非守护进程启动
# process.daemon = False
# p = multiprocessing.Process(target=browser.browser_click, args=(url, count, proxy, md5)) \
# if int(query_result.request) == 1 else multiprocessing.Process(target=request.click,
# args=(url, count, proxy,md5))
# p.start()
# ts.append(p)
# # 启动后再设为true
# process.daemon = True
# utils.write_debug(utils.LINE(), 'tasks',
# 'start:' + md5 + ' time: ' + time.strftime('%X') + '\ncount:' + str(count))
# # status置位
# query_results.update(task_status=1)
# for t in ts:
# t.join()
# except Exception:
# utils.write_debug(utils.LINE(), 'tasks', traceback.print_exc())
def get_db(db=DB):
try:
# 建立连接
client = pymongo.MongoClient(IP, PORT)
db_auth = client.admin
db_auth.authenticate(USER, PASSWD)
# admin 数据库有帐号，连接-认证-切换库
db = client[db]
client['ad_click']['ip_pool'].create_index(
[("create_at", pymongo.ASCENDING)], expireAfterSeconds=30)
return db
except:
utils.write_debug(utils.LINE(), "dealIp", traceback.print_exc())
def dosql(request):
try:
url = request.GET.get('url')
task_detail = request.GET.get('task_detail')
task_time = request.GET.get('task_time')
if task_time:
task_time = datetime.datetime.strptime(task_time, '%Y-%mm-%d')
if task_time < datetime.datetime.now():
return HttpResponse('任务发布失败,请检查任务开始的时间!')
else:
task_time = datetime.date.today() + datetime.timedelta(days=1)
md5 = hashlib.md5()
md5.update((url + str(time.time())).encode('utf8'))
MD5 = md5.hexdigest()
app_clicktask = {
'url': url,
'url_md5': MD5,
'task_detail': task_detail,
'task_time': task_time
}
ClickTask.create(**app_clicktask)
ClickTask.objects.filter(id)
if int(task_detail) == 0:
task_count = request.GET.get('task_count')
app_taskcount = {'url_md5': MD5, 'task_count': task_count}
TaskCount.create(**app_taskcount)
else:
app_timetask = {'url_md5': MD5}
task_count = 0
for i in range(24):
app_timetask['field_' + str(i)] = request.GET.get(str(i))
task_count += int(request.GET.get(str(i)))
TimeTask.create(**app_timetask)
app_taskcount = {'url_md5': MD5, 'task_count': task_count}
TaskCount.create(**app_taskcount)
return HttpResponse('任务发布成功,您的任务编号为: ' + MD5)
except Exception as e:
utils.write_debug(utils.LINE(), 'dosql', traceback.print_exc())
return HttpResponse('任务发布失败\n' + str(e))
def CPM(request):
try:
# 获取数据
task_area = request.POST.get('task_area')
flow_demand = request.POST.get('flow_demand')
referer = request.POST.get('referer')
remaintime = request.POST.get('remaintime')
url = request.POST.get('url')
tfip = request.POST.get('tfip')
pvbs = request.POST.get('pvbs')
is_control = request.POST.get('is_control')
task_time = request.POST.get('task_time')
remark = request.POST.get('remark')
user_id = request.POST.get('user_id')
# 计算和处理
md5 = hashlib.md5()
md5.update((url + str(time.time())).encode('utf8'))
url_md5 = md5.hexdigest()
if tfip.isdigit():
tfip = int(tfip)
else:
raise Exception(u"投放ip请输入数字")
# click_task
if task_time:
task_time = datetime.datetime.strptime(task_time, '%Y-%m-%d')
if task_time < datetime.datetime.now():
raise Exception(u'任务日期小于当前时间,请检查!')
else:
task_time = datetime.date.today() + datetime.timedelta(days=1)
click_task = {
'url': url,
'url_md5': url_md5,
'task_time': task_time,
'task_area': task_area,
'pv': pvbs,
'user_id': user_id
}
if int(flow_demand) == 0 or int(flow_demand) == 1:
if int(flow_demand) == 1:
if remaintime.isdigit():
click_task['remaintime'] = remaintime
if referer:
click_task['referer'] = referer
click_task['task_detail'] = 0
else:
if int(flow_demand) == 3:
if remaintime:
click_task['remaintime'] = remaintime
if referer:
click_task['referer'] = referer
click_task['task_detail'] = 1
click_task['remark'] = remark if remark else u'无'
task_id = ClickTask.objects.create(**click_task).id
# task_count
task_count = {
'url_md5': url_md5,
'task_count': tfip,
'task_id_id': task_id
}
TaskCount.objects.create(**task_count)
# time_task
time_task = {'url_md5': url_md5, 'task_id_id': task_id}
if int(is_control) == 0:
count = int(tfip / 24)
for i in range(24):
time_task['field_' + str(i)] = count
elif int(is_control) == 1:
for i in range(24):
time_task['field_' + str(i)] = request.POST.get(str(i))
TimeTask.objects.create(**time_task)
# business_cpmwork
cpmwork = {
'user_name_id': user_id,
'task_id_id': task_id,
'url': url,
'click_nums': tfip,
'ip_nums': 0,
'pv_nums': 0,
'status': 0,
'is_control': is_control,
'remark': remark,
'task_time': task_time
}
CPMWork.objects.create(**cpmwork)
# cpmwork['task_time'] = time.strftime('%Y-%m-%d %X',time.localtime())
return render(request, 'CPMWork.html', {'cpmworks': [cpmwork]})
# return HttpResponse(click_task)
except Exception as e:
utils.write_debug(utils.LINE(), 'business/views',
traceback.print_exc())
return HttpResponse(e)
def browser_click1(url, proxy, referer, remaintime, task_detail, pv, width,
height, is_random, sectime, securl, element, task_type):
try:
WIDTH = width
HEIGHT = height
PIXEL_RATIO = 3.0
if task_detail == 0:
UA = random.choice(userAgent.USERAGENT['pc'])
elif task_detail == 1:
UA = random.choice(userAgent.USERAGENT['google'] +
userAgent.USERAGENT['ios'])
elif task_detail == 2:
UA = random.choice(userAgent.USERAGENT['ios'])
else:
UA = random.choice(userAgent.USERAGENT['google'])
mobileEmulation = {
"deviceMetrics": {
"width": WIDTH,
"height": HEIGHT,
"pixelRatio": PIXEL_RATIO
},
"userAgent": UA
}
if referer:
referer = random.choice(referer.split(';'))
pass
options = webdriver.ChromeOptions()
if int(proxy) == 9999:
pass
else:
proxy = int(proxy)
ip = utils.get_proxy(proxy)
utils.write_debug(utils.LINE(), 'browser', ip)
options.add_argument('--proxy-server=%s' % ip)
options.add_argument("headless")
options.add_argument("disable-gpu")
options.add_experimental_option('mobileEmulation', mobileEmulation)
driver = webdriver.Chrome(chrome_options=options)
driver.set_page_load_timeout(60)
driver.implicitly_wait(60)
driver.get(url)
for i in range(pv - 1):
js = 'window.open("{}");'.format(url)
driver.execute_script(js)
time.sleep(random.uniform(remaintime - 3, remaintime + 3))
'''
CPC
'''
if task_type == 1:
if is_random == 1:
urls = BeautifulSoup(driver.page_source, 'lxml').find_all('a')
url = random.choice(urls).get('href')
driver.get(url)
time.sleep(sectime)
else:
if securl:
# <a class="swf-top" href="http://tv.sohu.com/s2015/dsjwxfs" target="_blank"></a>
urls = securl.split(';')
url = random.choice(urls)
print(url)
driver.get(url)
time.sleep(sectime)
# soup.findAll('a',{'class':'lb'})
# Out[14]: [<a class="lb" href="http://www.baidu.com/bdorz/login.gif?login&tpl=mn&u=http%3A%2F%2Fwww.baidu.com%2f%3fbdorz_come%3d1" name="tj_login">ç»å½</a>]
elif element:
element_key = element.split(':')[0]
element_value = element.split(':')[1]
urls = BeautifulSoup(driver.page_source, 'lxml').findAll(
'a', {element_key: element_value})
url = random.choice(urls)
driver.get(url)
time.sleep(sectime)
# el = WebDriverWait(driver,60,0.5).until(EC.presence_of_element_located((By.XPATH,'//*[@id="bs-example-navbar-collapse-1"]/ul/li[3]/a')))
# el.click()
# time.sleep(random.random() * 5)
print(driver.title)
driver.get('about:blank')
driver.close()
driver.quit()
except Exception as e:
print(e)
driver.close()
driver.quit()
utils.write_debug(utils.LINE(), 'browser', traceback.print_exc())