#MySQL数据库建表
CREATE TABLE catetable(
cate_name VARCHAR(255),
cate_url VARCHAR(255),
total_num INT(19),
focus INT(19),
KEY cate_name(cate_name),
KEY cate_url(cate_url)
)ENGINE=INNODB DEFAULT CHARSET=utf8

#coding:utf-8
import sys
reload(sys)
sys.setdefaultencoding(‘utf-8‘)
import requests
from lxml import etree
import MySQLdb
from multiprocessing import Pool

‘‘‘   
获取所有所有专题所在页的url列表下一步提取每个专题的url
分析：http://www.jianshu.com/recommendations/collections页面为异步加载,查看热门和城市导航栏请求构造url列表
‘‘‘
def get_cateurls():
    urls=[]
    for i in range(1, 3):
        cityurl=‘http://www.jianshu.com/recommendations/collections?page=%s&order_by=city‘ % i
        urls.append(cityurl)
    for j in range(1,40):
        hoturl=‘http://www.jianshu.com/recommendations/collections?page=%s&order_by=hot‘% j
        urls.append(hoturl)
    return urls

‘‘‘ 解析页面 ‘‘‘ 
def get_response(url):
    html=requests.get(url).content
    selector=etree.HTML(html)
    return selector

‘‘‘ 获取专题数据 得到专题名称 专题url 收录文章数 关注人数 通过zip函数转置数据返回data对象‘‘‘ 
def get_catedata(url):
    selector=get_response(url)
    cate_url= map(lambda x: ‘http://www.jianshu.com‘+x, selector.xpath(‘//div[@id="list-container"]//div[contains(@class,"count")]/a/@href‘))
    cate_name=selector.xpath(‘//div/h4/a/text()‘)
    total_num=map(lambda x: int(x.strip("篇文章").strip()),selector.xpath(‘//div[@id="list-container"]//div[contains(@class,"count")]/a/text()‘))
    focus1=selector.xpath(‘//div[@id="list-container"]//div[contains(@class,"count")]/text()‘)
    focus=[]
    for i in focus1:
        focus_num=i.split("·")[1].rstrip("人关注")
        if "K" in focus_num:
            focus_num=int(float(focus_num[:-1])*1000)
        else:
            focus_num=int(focus_num)
        #print i,focus_num
        focus.append(focus_num)
    data=zip(cate_name,cate_url,total_num,focus)
    return data

‘‘‘ 写入数据库‘‘‘ 
def insert_into_Mysql(url):
    try:
        conn=MySQLdb.connect(host=‘127.0.0.1‘,user=‘root‘,passwd=‘你的密码‘,db=‘local_db‘,port=3306,charset=‘utf8‘)
        with conn:
            cursor = conn.cursor()
            print u‘正在加载 %s 页面‘ % url
            data=get_catedata(url)
            for i in data:
                #print i[0], i[1], i[2],i[3]
                cursor.execute("INSERT INTO  catetable (cate_name,cate_url,total_num,focus) values(%s,%s,%s,%s)", (i[0], i[1], i[2],i[3]))
                conn.commit()
            sql=‘select * from catetable‘
            count = cursor.execute(sql)
            print u‘总共输入了%s条数据‘ % count
    except MySQLdb.Error,e:
        print e


‘‘‘ 从数据库获取所有的专题url便于以后专题页面数据的进一步获取‘‘‘ 
def get_allcate_urls_from_mysql():
    try:
        conn = MySQLdb.connect(host=‘127.0.0.1‘, user=‘root‘, passwd=‘你的密码‘, db=‘local_db‘, port=3306, charset=‘utf8‘)
        with conn:
            cursor = conn.cursor()
            sql = ‘select cate_url from catetable‘
            count = cursor.execute(sql)
            print u‘共输入了%s条数据‘ % count
            print u‘正在获取专题url‘
            all_cate_urls=cursor.fetchall()
            return all_cate_urls
    except MySQLdb.Error, e:
        print url,e

‘‘‘ 多进程执行‘‘‘ 
def get_allcate_urls():
    urls=get_cateurls()
    pool = Pool(processes=4)
    pool.map(insert_into_Mysql,urls)
    allcate_urls=get_allcate_urls_from_mysql()
    return allcate_urls

‘‘‘ 先获取到所有专题的数据 下次获取每个专题的数据‘‘‘ 
if __name__ == ‘__main__‘:
    allcate_urls=get_allcate_urls()