data.crawler.r9.py

# -*- coding:utf-8 -*-

"""
Date 16-Jan-3q   lastest progres
"""

#from __future__ import print_function
import gevent
from gevent import monkey

# patches stdlib (including socket and ssl modules) to cooperate with other greenlets
monkey.patch_all()


import time
import sys
import re
import random


from bs4 import BeautifulSoup
import urlparse
#import urllib2
#import grequests
import requests
#from urllib2 import URLError
import codecs
import simplejson as json




#from cache_requests import requests

def find_page(html_page):

    question_url = "http://www.leleketang.com/lib/"

    soup = BeautifulSoup(html_page, "html5lib")

    #html = soup.find_all('a')
    # May need a list
    url_list = []
    for link in soup.find_all('a', {"class": "to_view"}):
        var = urlparse.urljoin(question_url, link.get('href'))
        url_list.append(var)

        #print url_list
    print (url_list)
    return url_list

def get_url_response(url):
    #try:
    request = urllib2.Request(url)
    request.add_header('Referer', "https://www.google.com.hk/#newwindow=1&safe=strict&q=%E4%B9%90%E4%B9%90%E8%AF%BE%E5%A0%82" , 'User-Agent', 'Mozilla/5.0 (Linux; Android 5.1.1; Nexus 5 Build/LMY48B; wv) AppleWebKit/537.36 (KHTML, like Gecko) Version/4.0 Chrome/43.0.2357.65 Mobile Safari/537.36 ' )
    resp = urllib2.urlopen(request)
    #except urllib2.HTTPError as e:
    #    print "get url func http status code wrong", e.code              # http status code
    #    print e.reason
    #    return

    #print "HTTP code is ", resp.code
    if resp.code != 200:
        return "END: %s" % url # change unknown

    #return resp

def get_requests(url):

    headers = {'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8) AppleWebKit/536.25 (KHTML, like Gecko) Version/6.0 Safari/536.25'}

    r = requests.get(url, headers=headers)
    r.encoding = 'utf-8'

    if r.status_code != 200:
        return " +++++++++++++ http status error END %s" % url

    return r.text


def crawler_one_page(dig_url):

    response = get_requests(dig_url)

    print
    print
    print ('dig_url')
    # print "Question, http code ",response.code

    #if response.code != 200:
    #    return "END: %s" % dig_url # change unknown

    # add exception handle logic
    if response:  page_content = response
    else: return "http status error END "

    question = parse_question(page_content)
    answer = parse_question_answer(page_content)
    # print ("html answer", answer)

    source = parse_question_otherinfo(page_content)
    question_info = parse_question_explained(page_content)

    ''' parse question over and then produce dict '''
    """ Somethimes something wrong here with empty list """
    try:
        question_text = question[0].text
    except:
        print "  question body not found, must be something wrong  "
        return 

    #if answer:
    #    answer_text = answer[0].text

    source_text = source[0].text

    if not answer:

        ''' [ Exception handle ] yet case , such as explain case '''

        dict_to_json = {"Question": question_text,
               "dianpin": question_info,
               "source": source_text,

               }

    else:
        answer_text = answer[0].text

        dict_to_json = {"Question": question_text,
               "dianpin": question_info,
               "source": source_text,
               "answer": answer_text,
               }


    return dict_to_json

def dict_write_fo_json_file(question_url, newdict):
    try:
        with codecs.open("%s.json" % question_url, "wb",'utf-8') as f:
            json.dump(newdict, f, indent=2, ensure_ascii=False)
    except IOError:
        print("Oops, file error...")


def parse_question(html):
    """ func work """
    soup = BeautifulSoup(html, "html5lib")


    #question = soup.findAll('div', {'class':"uc_q"})

    """ for fill question class """
    question = soup.findAll('div', {'class':"uc_q_caption"})

    #question = soup.title

    #print (question)

    #if not question:
    #    gevent.sleep(30)
    #    parse_question(html)
    #    return " Recursion call exit"

    print ("parse_ques done")
    #print  question
    return question

def parse_question_answer(html):

    soup = BeautifulSoup(html, "html5lib")

    print ("HTML with answer ")
    #print soup
    #print

    find_answer = soup.findAll('li', {'class':" ucqo_g_solution"}) # must have had spaces
    ''' for yet another case '''
    if not find_answer:
        find_answer = soup.findAll('span', {'class':"uc_q_object ucqo_text ucqo_g_blank ucqo_g_solution"})

    #print ("FIND ANSWER ", find_answer)
    return find_answer

def parse_question_otherinfo(r):

    soup = BeautifulSoup(r, "html5lib")
    source = soup.findAll('span', {'class':"ellipsis"})

    print ("parse_question_otherinfo done!")
    #print source

    return source

def parse_question_explained(html):

    soup = BeautifulSoup(html, "html5lib")

    htmlfind = soup.findAll('div', {'class':"question_part_content"})
    # print "HTML question explained \n", htmlfind

    if htmlfind:
        html_section = htmlfind[1] # must be [1]
    else: return

    question_info = {}

    index = 0
    tag_div_list = html_section.findAll('div', attrs={"class": "clearfix"})
    print
    print ("func html explain len", len(tag_div_list))
    print ("write json and then next")

    # 1 traceback at here
    tags = tag_div_list[0].findAll('div', attrs={'style':"float:left;width:45px;background: #4cd1d4;color: #fff;text-align:center;margin-right:10px;border-radius: 3px;"})
    if tags[0]:
        question_info['fenxi_tag'] = tags[0].text
    else: pass

    try:
        question_info['dianping_neirong'] = tag_div_list[0].findAll('span', attrs={'class': 'uc_q_object'})[0].text
    except IndexError:
        pass
    # 2
    tags = tag_div_list[1].findAll('div', attrs={'style':"float:left;width:45px;background: #4cd1d4;color: #fff;text-align:center;margin-right:10px;border-radius: 3px;"})
    if tags:
        question_info['jieda_tag'] = tags[0].text
    else: pass

    try: # not empty
        question_info['jieda_neirong'] = tag_div_list[1].findAll('span', attrs={'class': 'uc_q_object'})[0].text
    except:
        pass
    # 3
    tags = tag_div_list[2].findAll('div', attrs={'style':"float:left;width:45px;background: #4cd1d4;color: #fff;text-align:center;margin-right:10px;border-radius: 3px;"})

    if tags:
        question_info['dianping_tag'] = tags[0].text
    else: pass
    try:
        question_info['fenxi_neirong'] = tag_div_list[2].findAll('span', attrs={'class': 'uc_q_object'})[0].text
    except:
        pass

    return question_info




def dict_write_fo_json_file(question_url, newdict):

    a = question_url
    url = a.split('/')[-1].split('.')[0]
    try:

        with codecs.open("%s.json" % url, "wb",'utf-8') as f:
            json.dump(newdict, f, indent=2, ensure_ascii=False)
    except IOError:
        print(" +++++++++++++++++++ Oops, file error...")


def main():


    """ second level may need twice loop """

    """ list html iter """
    for visit_page in range(700, 801): #xz 550 done
        ' url = "http://www.leleketang.com/lib/list17-101-2-16-10129-113711-1-%s.shtml" % visit_page '
        ' url = "http://www.leleketang.com/lib/list17-101-2-16-10129-113711-4-%s.shtml" % visit_page '
        url = "http://www.leleketang.com/lib/list17-101-2-16-10129-113711-2-%s.shtml" % visit_page

        """ url 1 tiankong 2:xuanze 4:jieda """

        #time.sleep(5.10)
        response = get_requests(url)

        print ("working on page ", visit_page)

        #if response.status_code != 200:   # add status code case
        #    return "END: %s" % url # change unknown

        if response: page = response
        else:  return
        #print page

        found_list_url = find_page(page)

        if found_list_url:
        
          for question_url in found_list_url:

                #time.sleep(6.0)

                time.sleep(random.randint(7, 13))
                print (question_url)
                try:
                    dict_variable = crawler_one_page(question_url)
                    if dict_variable:
                        var = dict_write_fo_json_file(question_url, dict_variable)
                    else: return "http status error END"

                except:
                    return "Main loop question digging something wrong but continue "
        else:
            return #page list empty

if __name__ == '__main__':
    main()

    '''
    gevent.joinall([
        gevent.spawn(main),

    ])

    '''