import os, signal, sys, pymysql,time, logging
import urllib.parse as urlparse
from datetime import datetime
import subprocess

def startNumber() :
    hour = int(time.strftime("%H"))
    day = int(time.strftime("%d"))
    if day in range(1,5) : vari = 1
    if day in range(6,10) : vari = 5
    if day in range(11,15) : vari = 10
    if day in range(16,20) : vari = 15
    if day in range(21,25) : vari = 20
    if day in range(26,30) : vari = 25
    else : vari = 30
    startnum = (hour*1000)+((23000*day-vari)-1)+((day-vari)-1)*1000
    return startnum

conn = pymysql.connect(host='localhost', user='root', password='dlsxjvkzmdkdlakzpt!',db='crawler', unix_socket='/var/run/mysqld/mysqld.sock', charset='utf8')

start = startNumber()
#print(int(time.strftime("%H")))

if start <= 100000 :
    curs = conn.cursor(pymysql.cursors.DictCursor)
    sql = "SELECT term FROM `naverpowerlinkterms` ORDER BY upTime ASC LIMIT " + str(start) + ",1000"
    curs.execute(sql)
    rows = curs.fetchall()
else :
    curs = conn.cursor(pymysql.cursors.DictCursor)
    sql = "SELECT extTerm AS term FROM relatedkeywords WHERE imarketRC > 0 AND adcheck = \"N\" ORDER BY upTime DESC LIMIT 0,1000"
    curs.execute(sql)
    rows = curs.fetchall()

for row in rows:
    shellcmd = "/home/maddiekorea/anaconda3/bin/python3 /home/maddiekorea/py/relatedTermCollector.py \"" + row['term'] + "\""
    executeCrawl = subprocess.Popen(shellcmd,stdout=subprocess.PIPE,shell=True)
    try: executeCrawl.communicate(timeout=80)
    except subprocess.TimeoutExpired:
        executeCrawl.terminate()
        continue
conn.close()