kw_tools/choozmo/save_search_result.py

import gsearch_general
import sys
import dataset
import datetime
import codecs
import time
db = dataset.connect('mysql://choozmo:pAssw0rd@db.ptt.cx:3306/seo?charset=utf8mb4')
table=db['save_result']

driver=gsearch_general.restart_browser()
fr=codecs.open('c:/tmp/food.csv','r','utf-8')
lines=fr.readlines()
fr.close()

namedict={}
cursor=db.query('select distinct(keyword) as kw from save_result')
for c in cursor:
    kw=c['kw'].replace('\n','')
    kw=c['kw'].replace('\r','')
    namedict[kw]=1

print(namedict)
print(lines)
#time.sleep(9999)

#keyword='檸檬原汁'
#lines=['芋頭 外帶']
#lines=['布丁 牛奶 冰']
#lines=['芋圓 加盟']

for l in lines:
    l=l.replace('\r','')
    l=l.replace('\n','')
    if namedict.get(l) is not None:
        continue
    keyword=l
    print(keyword)
    idx=gsearch_general.process_query(driver,keyword,number_results=100,language_code='zh-TW',enable_next=False)
    print(idx)
    cnt=1
    for x in idx:
        x['keyword']=keyword
        x['dt']=datetime.datetime.now()
        x['num']=str(cnt)
        cnt+=1
        table.insert(x)
#    if idx==None:
#        print(driver.page_source)
    if '我們的系統偵測到您的電腦網路送出的流量有異常情況' in driver.page_source:
        print('baned.....')
        sys.exit()