[나라장터] 크롤링

from selenium import webdriver
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.support.ui import Select
import pandas as pd 
import datetime
import time
import os
times=20
while True:
    if(str(datetime.datetime.now())[11:13]=='10'):
        
        times=20
        print(times)
        if(not(str(datetime.datetime.now())[0:10].replace('-','')+'.csv' in os.listdir('G:/내 드라이브'))):
            times=50
            print(times)
            driver=webdriver.Chrome('c:/selenium/chromedriver')
            results=[]
            querylist=['홈페이지','플랫폼','포털','시스템','구축','빅데이터','산학협력','산학포털','LINC','업무지원','평생교육','솔루션']
            querylist2=['중앙신체검사소','한국감정원','한국교육학술정보원','한국산업단지공단','한국사학진흥재단','한국산업기술평가관리원','한국가스공사','신용보증기금','중앙119구조본부','한국정보화진흥원','중앙교육연수원','한국장학재단','대구경북첨단의료산업진흥재단','의료기술시험훈련원','한국뇌연구원','한의기술응용센터','한국도로공사','한국건설관리공사','교통안전공단','국립농산물품질관리원','농림축산검역본부','국립종자원','한국전력기술','대한법률구조공단','우정사업조달사무소','기상통신소','조달품질원','한국법무보호복지공단','에너지경제연구원','근로복지공단','노동부고객상담센터','한국산업인력공단','한국산업안전보건공단','국립재난안전연구원','한국동서발전','한국석유공사','한국에너지공단','한국토지주택공사','주택관리공단','한국시설안전공단','중소기업진흥공단','한국산업기술시험원','한국세라믹기술원','한국남동발전','한국승강기안전관리원','국방기술품질원','중앙관세분석소','한국저작권위원회','대구광역시','대구광역시교육청','대구광역시교육연구정보원','경상북도교육청','경상북도교육연구원','경상북도교육정보센터','국립대구과학관']
            n=len(querylist)*len(querylist2)
            t=0
            try:
                for query in querylist:
                    for query2 in querylist2:
                        t=t+1
                        driver.get('http://www.g2b.go.kr:8101/ep/tbid/tbidFwd.do')
                        
                        # 업무 종류 체크
            #            task_dict = {'용역': 'taskClCds5', '민간': 'taskClCds20', '기타': 'taskClCds4'}
            #            for task in task_dict.values():
            #                checkbox = driver.find_element_by_id(task)
            #                checkbox.click()
            
                        instNm=driver.find_element_by_id('instNm')
                        instNm.clear()
                        instNm.send_keys(query2)
                        instNm.send_keys(Keys.RETURN)
                        # 입찰정보 검색 페이지로 이동
                        
                        # 검색어
                        # id값이 bidNm인 태그 가져오기
                        bidNm = driver.find_element_by_id('bidNm')
                        # 내용을 삭제 (버릇처럼 사용할 것!)
                        bidNm.clear()
                        # 검색어 입력후 엔터
                        bidNm.send_keys(query)
                        bidNm.send_keys(Keys.RETURN)
                        # 검색 조건 체크
                        option_dict = {'검색기간 1달': 'setMonth1_1', '입찰마감건 제외': 'exceptEnd'}
                        for option in option_dict.values():
                            checkbox = driver.find_element_by_id(option)
                            checkbox.click()
                        # 목록수 100건 선택 (드롭다운)
                        recordcountperpage = driver.find_element_by_name('recordCountPerPage')
                        selector = Select(recordcountperpage)
                        selector.select_by_value('100')
                        # 검색 버튼 클릭
                        search_button = driver.find_element_by_class_name('btn_mdl')
                        search_button.click()
                        # 검색 결과 확인
                        elem = driver.find_element_by_class_name('results')
                        div_list = elem.find_elements_by_tag_name('div')
                        # 검색 결과 모두 긁어서 리스트로 저장
                        #results.append(query)
                        #if len(div_list)!=0:
                        #    results.append(query)
                        for div in div_list:
                            results.append(div.text)
                            a_tags = div.find_elements_by_tag_name('a')
                            if a_tags:
                                for a_tag in a_tags:
                                    link = a_tag.get_attribute('href')
                                    results.append(link)
                        # 검색결과 모음 리스트를 12개씩 분할하여 새로운 리스트로 저장 
                        #result = [results[i * 13:(i + 1) * 13] for i in range((len(results) + 13 - 1) // 13 )]
                        result = [results[i * 12:(i + 1) * 12] for i in range((len(results) + 12 - 1) // 12 )]
                        # 결과 출력
                        print(round(t/n*100,2))
                        #print(result)
            except Exception as e:
                # 위 코드에서 에러가 발생한 경우 출력
                print(e)
            finally:
                # 에러와 관계없이 실행되고, 크롬 드라이버를 종료
                driver.quit()
            
            
            my_df=pd.DataFrame(result)
            my_df.to_csv('G:/내 드라이브/test.csv',index=False,header=False,encoding='cp949')
            my_df.to_csv('G:/내 드라이브/'+str(datetime.datetime.now())[0:10].replace('-','')+'.csv',index=False,header=False,encoding='cp949')
    time.sleep(60*times)