Merge remote-tracking branch 'origin/master'

choozmo/click_commerce.py

@@ -0,0 +1,108 @@
+import time
+import json
+from selenium import webdriver
+from selenium.webdriver.common.desired_capabilities import DesiredCapabilities
+import time
+import os
+import urllib.parse
+from selenium.webdriver.support.ui import WebDriverWait
+from selenium.webdriver.common.by import By
+from selenium.webdriver.support import expected_conditions as EC
+import codecs
+import random
+import requests
+import dataset
+import time
+import traceback
+import sys
+
+driver=None
+headers = {
+        "Authorization": "Bearer " + "t35vhZtWNgvDNWHc3DJh0OKll3mcB9GvC8K2EAkBug2",
+        "Content-Type": "application/x-www-form-urlencoded"
+}
+
+
+
+def send_msg(kw):
+    params = {"message": "處理關鍵字: "+kw}  
+    r = requests.post("https://notify-api.line.me/api/notify",headers=headers, params=params)
+
+
+def empty_query(q):
+    global driver
+    googleurl='https://www.google.com/search?q='+urllib.parse.quote(q)
+    driver.get(googleurl)
+    time.sleep(3)
+
+
+def process_query(qs):
+    q=qs[0]
+    domain=qs[1]
+    global driver
+    googleurl = 'https://www.google.com/search?q={}&num={}&hl={}'.format(urllib.parse.quote(q), 100,'zh-TW')
+    print(googleurl)
+    driver.get(googleurl)
+    time.sleep(6)
+
+    elmts=driver.find_elements_by_xpath("//div[@class='yuRUbf']/a")
+
+    idx=1
+    ranking=-1
+    print(len(elmts))
+#    driver.save_screenshot('c:/tmp/test.png')
+
+    for elmt in elmts:
+
+        href=elmt.get_attribute('href')
+        txt=elmt.text
+        if len(txt)>10:
+            if domain in href:
+                print('clicked....')
+                print(href)
+                print(txt)
+                webdriver.ActionChains(driver).move_to_element(elmt).perform()
+                webdriver.ActionChains(driver).move_to_element(elmt).click().perform()
+                break
+
+
+def run_once(q):
+    global driver
+    result=[]
+    options = webdriver.ChromeOptions()
+    options.add_argument('--headless')
+#    options.add_argument("--user-agent=" +user_agent)
+    options.add_argument("--incognito")
+
+    driver = webdriver.Chrome(
+    options=options)
+
+    driver.delete_all_cookies()
+    driver.set_window_size(1400,1000)
+
+    print(q)
+    process_query(q)
+    time.sleep(3)
+    driver.quit()
+
+
+#lst=[{'kw':'幸福空間','domain':'hhh.com.tw','page':0}]
+lst=[]
+db = dataset.connect('mysql://choozmo:pAssw0rd@db.ptt.cx:3306/seo?charset=utf8mb4')
+
+cursor=db.query('SELECT kw as term,domain FROM seo.seo_clickjobs where category="hhh-faq"  order by rand()')
+
+for c in cursor:
+    lst.append(c)
+
+
+#for c in lst:
+while True:
+    try:
+        c=random.choice(lst)
+        run_once( (c['term'],c['domain'])   )
+    except:
+        traceback.print_exc()
+    sleepint=random.randint(380,520)
+    time.sleep(sleepint)
+

choozmo/list_clickjob.py

@@ -0,0 +1,108 @@
+import time
+import json
+from selenium import webdriver
+from selenium.webdriver.common.desired_capabilities import DesiredCapabilities
+import time
+import os
+import urllib.parse
+from selenium.webdriver.support.ui import WebDriverWait
+from selenium.webdriver.common.by import By
+from selenium.webdriver.support import expected_conditions as EC
+import codecs
+import random
+import requests
+import dataset
+import time
+import traceback
+import sys
+
+driver=None
+headers = {
+        "Authorization": "Bearer " + "t35vhZtWNgvDNWHc3DJh0OKll3mcB9GvC8K2EAkBug2",
+        "Content-Type": "application/x-www-form-urlencoded"
+}
+
+
+
+def send_msg(kw):
+    params = {"message": "處理關鍵字: "+kw}  
+    r = requests.post("https://notify-api.line.me/api/notify",headers=headers, params=params)
+
+
+def empty_query(q):
+    global driver
+    googleurl='https://www.google.com/search?q='+urllib.parse.quote(q)
+    driver.get(googleurl)
+    time.sleep(3)
+
+
+def process_query(qs):
+    q=qs[0]
+    domain=qs[1]
+    global driver
+    googleurl = 'https://www.google.com/search?q={}&num={}&hl={}'.format(urllib.parse.quote(q), 100,'zh-TW')
+    print(googleurl)
+    driver.get(googleurl)
+    time.sleep(6)
+
+    elmts=driver.find_elements_by_xpath("//div[@class='yuRUbf']/a")
+
+    idx=1
+    ranking=-1
+    print(len(elmts))
+#    driver.save_screenshot('c:/tmp/test.png')
+
+    for elmt in elmts:
+
+        href=elmt.get_attribute('href')
+        txt=elmt.text
+        if len(txt)>10:
+            if domain in href:
+                print('clicked....')
+                print(href)
+                print(txt)
+                webdriver.ActionChains(driver).move_to_element(elmt).perform()
+                webdriver.ActionChains(driver).move_to_element(elmt).click().perform()
+                break
+
+
+def run_once(q):
+    global driver
+    result=[]
+    options = webdriver.ChromeOptions()
+    options.add_argument('--headless')
+#    options.add_argument("--user-agent=" +user_agent)
+    options.add_argument("--incognito")
+
+    driver = webdriver.Chrome(
+    options=options)
+
+    driver.delete_all_cookies()
+    driver.set_window_size(1400,1000)
+
+    print(q)
+    process_query(q)
+    time.sleep(3)
+    driver.quit()
+
+
+#lst=[{'kw':'幸福空間','domain':'hhh.com.tw','page':0}]
+lst=[]
+db = dataset.connect('mysql://choozmo:pAssw0rd@db.ptt.cx:3306/seo?charset=utf8mb4')
+
+cursor=db.query('select term,domain from selected_kw  order by rand()')
+
+for c in cursor:
+    lst.append(c)
+
+
+#for c in lst:
+while True:
+    try:
+        c=random.choice(lst)
+        run_once( (c['term'],c['domain'])   )
+    except:
+        traceback.print_exc()
+    sleepint=random.randint(320,520)
+    time.sleep(sleepint)
+

hhh/SEO/fast_1999.py

@@ -10,7 +10,8 @@ import urllib.parse
 from selenium.webdriver.support.ui import WebDriverWait
 from selenium.webdriver.common.by import By
 from selenium.webdriver.support import expected_conditions as EC
-
+from userAgentRandomizer import userAgents
+from selenium.webdriver.chrome.options import Options
 import codecs
 import random
 import requests
@@ -88,17 +89,31 @@ def re_get_webdriver():
 #        if 'p17777' in l.name:
 #            ls[0].restart()
 
-#    time.sleep(4)
+    time.sleep(4)
     if driver is not None:
         print('closing....')
 #        driver.quit()
         os.system('docker container restart p'+portnum)
-        time.sleep(11)
+        time.sleep(10)
         
 #    options = webdriver.EdgeOptions()
     try:
+        ua = userAgents()
+
+        user_agent = ua.random()        
+#        options = webdriver.ChromeOptions()
+        options = Options()
+        # options.add_argument("--headless") # 視窗背後執行
+        print(user_agent)
+        options.add_argument("--user-agent=" +user_agent)
+        options.add_argument("--incognito")
+
         driver = webdriver.Remote(
-            command_executor='http://127.0.0.1:'+str(portnum)+'/wd/hub')
+            command_executor='http://127.0.0.1:'+str(portnum)+'/wd/hub',
+            options=options
+#                        desired_capabilities=options.to_capabilities(),
+            )
+        driver.delete_all_cookies()
         driver.set_window_size(1400,1000)
         return
     except:
@@ -110,7 +125,7 @@ def re_get_webdriver():
 
 def run_once(url):
     global driver
-    i=random.randint(0,25)
+    i=random.randint(0,5)
     if i==0 or driver is None:
 #    if True:
         re_get_webdriver()
@@ -145,7 +160,7 @@ db = dataset.connect('mysql://choozmo:pAssw0rd@db.ptt.cx:3306/seo?charset=utf8mb
 
 cursor=db.query('SELECT * FROM columnids order by rand()')
 for c in cursor:
-    lst.append('https://www.hhh.com.tw/columns/detail/'+str(c['cid'])+'/index.php')
+    lst.append('https://www.hhh.com.tw/columns/detail/'+str(c['cid'])+'/')
 
 
 #lst=['https://www.hhh.com.tw/columns/detail/3427/index.php']