#-*-coding:utf-8-*- #编码声明，不要忘记！
import requests  #这里使用requests，小脚本用它最合适！
from lxml import html    #这里我们用lxml，也就是xpath的方法
import itertools
import re
import codecs

def formatstr(str1):
    pattern1 = re.compile(r'(([1-9]\d*(\.\d*[1-9])?)|(0\.\d*[1-9]))平')
    pattern2 = re.compile(r'(([1-9]\d*(\.\d*[1-9])?)|(0\.\d*[1-9]))万')

    match1_mianji = pattern1.search(str1)
    match2_jiage = pattern2.search(str1)
    if (match1_mianji and match2_jiage):
        mianji=match1_mianji.group(1)
        jiage=match2_jiage.group(1)
        danjia=float(jiage)/float(mianji)*10000
       # str_formated=str1 +'\t'+'面积：'+mianji+'\t'+'总价'+jiage+'\t'+str(int(danjia))
        str_formated =  '面积' +'\t'+ mianji + '\t' + '总价'  +'\t'+  jiage + '\t' + str(int(danjia)) + '\t' + str1
        print str_formated
        return str_formated
    else:
        return None

def  export_title(url,f):
    page = requests.get(url)
    if page is None:
        return False
    #对获取到的page格式化操作，方便后面用XPath来解析
    tree = html.fromstring(page.text)

    #XPath解析，获得你要的文字段落！#
    #intro_raw = tree.xpath('//th[contains(text(),"出售")]/a/text()')

    intro_raw = tree.xpath('//th[contains(@class,"new")]/a/text()')
    #intro_raw = tree.xpath('//th/a/text()')
    #简单的转码工作，这步根据需要可以省略
    #print intro_raw

    for i in intro_raw:
        intro = i.encode('utf-8').replace('New','')
        if intro!='':
            #print intro
            try:
                intro.replace('\t', '').replace('\n', '').replace(' ', '')
                intro_formated=formatstr(intro)
                if intro_formated is not None:
                    print intro_formated

                    f.write(intro_formated+'\n')
            except Exception as e:
                print e
                return False
    return True



if __name__ == "__main__":

    with open('C:/tobacco/11/dyfj.csv', 'w') as f:
        f.write(codecs.BOM_UTF8)
        for page in itertools.count(1):
            URL='http://bbs.212300.com/forum-56-%d.html' % page
            print URL
            flag =export_title(URL,f)
            if(not flag ):
                print 'finished!!'
                break
            else:
                pass
        # URL = 'http://bbs.212300.com/forum-56-1.html'
        # export_title(URL, f)
    f.close()