摘要

大多數APP裡面返回的是json格式資料，或者一堆加密過的資料 。這裡以超級課程表APP為例，抓取超級課程表裡使用者發的話題。

1

抓取APP資料包

方法詳細可以參考這篇博文：http://my.oschina.net/jhao104/blog/605963

得到超級課程表登入的地址：http://120.55.151.61/V2/StudentSkip/loginCheckV4.action

表單：

表單中包括了使用者名稱和密碼，當然都是加密過了的，還有一個裝置資訊，直接post過去就是。

另外必須加header,一開始我沒有加header得到的是登入錯誤，所以要帶上header資訊。

2

登入

登入程式碼：

import urllib2 
from cookielib import CookieJar 
loginUrl = 'http://120.55.151.61/V2/StudentSkip/loginCheckV4.action' headers = { 
 'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8', 
 'User-Agent': 'Dalvik/1.6.0 (Linux; U; Android 4.1.1; M040 Build/JRO03H)', 
 'Host': '120.55.151.61', 
 'Connection': 'Keep-Alive', 
 'Accept-Encoding': 'gzip', 
 'Content-Length': '207', 
 } 
loginData = 'phoneBrand=Meizu&platform=1&deviceCode=868033014919494&account=FCF030E1F2F6341C1C93BE5BBC422A3D&phoneVersion=16&password=A55B48BB75C79200379D82A18C5F47D6&channel=MXMarket&phoneModel=M040&versionNumber=7.2.1&' 
cookieJar = CookieJar() 
opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cookieJar)) 
req = urllib2.Request(loginUrl, loginData, headers) 
loginResult = opener.open(req).read() 
print loginResult

登入成功 會返回一串賬號資訊的json資料

和抓包時返回資料一樣，證明登入成功

3

抓取資料

用同樣方法得到話題的url和post引數

做法就和模擬登入網站一樣。詳見：http://my.oschina.net/jhao104/blog/547311

下見最終程式碼，有主頁獲取和下拉載入更新。可以無限載入話題內容。

#!/usr/local/bin/python2.7 # -*- coding: utf8 -*- """ 
 超級課程表話題抓取 
""" import urllib2 
from cookielib import CookieJar 
import json 
''' 讀Json資料 ''' def fetch_data(json_data): 
 data = json_data['data'] 
 timestampLong = data['timestampLong'] 
 messageBO = data['messageBOs'] 
 topicList = [] 
 for each in messageBO: 
 topicDict = {} 
 if each.get('content', False): 
 topicDict['content'] = each['content'] 
 topicDict['schoolName'] = each['schoolName'] 
 topicDict['messageId'] = each['messageId'] 
 topicDict['gender'] = each['studentBO']['gender'] 
 topicDict['time'] = each['issueTime'] 
 print each['schoolName'],each['content'] 
 topicList.append(topicDict) 
 return timestampLong, topicList 
''' 載入更多 ''' def load(timestamp, headers, url): 
 headers['Content-Length'] = '159' 
 loadData = 'timestamp=%s&phoneBrand=Meizu&platform=1&genderType=-1&topicId=19&phoneVersion=16&selectType=3&channel=MXMarket&phoneModel=M040&versionNumber=7.2.1&' % timestamp 
 req = urllib2.Request(url, loadData, headers) 
 loadResult = opener.open(req).read() 
 loginStatus = json.loads(loadResult).get('status', False) 
 if loginStatus == 1: 
 print 'load successful!' 
 timestamp, topicList = fetch_data(json.loads(loadResult)) 
 load(timestamp, headers, url) 
 else: 
 print 'load fail' 
 print loadResult 
 return False loginUrl = 'http://120.55.151.61/V2/StudentSkip/loginCheckV4.action' topicUrl = 'http://120.55.151.61/V2/Treehole/Message/getMessageByTopicIdV3.action' headers = { 
 'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8', 
 'User-Agent': 'Dalvik/1.6.0 (Linux; U; Android 4.1.1; M040 Build/JRO03H)', 
 'Host': '120.55.151.61', 
 'Connection': 'Keep-Alive', 
 'Accept-Encoding': 'gzip', 
 'Content-Length': '207', 
 } 
''' ---登入部分--- ''' 
loginData = 'phoneBrand=Meizu&platform=1&deviceCode=868033014919494&account=FCF030E1F2F6341C1C93BE5BBC422A3D&phoneVersion=16&password=A55B48BB75C79200379D82A18C5F47D6&channel=MXMarket&phoneModel=M040&versionNumber=7.2.1&' 
cookieJar = CookieJar() 
opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cookieJar)) 
req = urllib2.Request(loginUrl, loginData, headers) 
loginResult = opener.open(req).read() 
loginStatus = json.loads(loginResult).get('data', False) 
if loginResult: 
 print 'login successful!' else: 
 print 'login fail' 
 print loginResult 
''' ---獲取話題--- ''' 
topicData = 'timestamp=0&phoneBrand=Meizu&platform=1&genderType=-1&topicId=19&phoneVersion=16&selectType=3&channel=MXMarket&phoneModel=M040&versionNumber=7.2.1&' 
headers['Content-Length'] = '147' 
topicRequest = urllib2.Request(topicUrl, topicData, headers) 
topicHtml = opener.open(topicRequest).read() 
topicJson = json.loads(topicHtml) 
topicStatus = topicJson.get('status', False) 
print topicJson 
if topicStatus == 1: 
 print 'fetch topic success!' 
 timestamp, topicList = fetch_data(topicJson) 
 load(timestamp, headers, topicUrl)

結果：