下载python-chinese 的邮件归档

 年前，我刚学Python的那阵子，先后两次换不同邮箱申请订阅 python-chinese 邮件列表，但两次都徒劳了。每次都是除了系统自动发给我一封确认信外（我回了），再也没啥消息了。python-chinese的可供下载的版本是gZip形式的，下载后使用rar解压不了，看了看winrar的版本，太古老了，是3.3的，还没有支持这种格式。后来下载了最新的winrar3.7.1绿色版，也未能解压，提示是一样的，说压缩文件已损坏，没有找到压缩文件云云。

      也罢，不让我订阅，也不解压下载提供的gZip版本，还不能让我下载过去的邮件归档么？其实我也只是想学习一下而已。昨晚完善好下载newsmth上Myphoto版照片的程序后，又看了好一会newsmth Python版的帖子，大概是扫完了。今天又写了一个新的脚本，去下载python-chinese 的邮件归档，下载完毕后做成电子书学习一下。

      python-chinese 的邮件归档的html格式比较简单，很好抓取。源代码如下，就当是学习Python的习作了。代码考虑到了网络IO读取失败的可能性，下载时没有使用多线程，只是挨个下载，就让机器慢慢工作吧。

#!/usr/bin/python
# -*- coding: cp936 -*-
# Filename: gethtml.py

import urllib
import os
from sgmllib import SGMLParser

#to parse
class URLLister(SGMLParser):
def reset(self):
   SGMLParser.reset(self)
  
   self.urls = []
   self.titles = []       
   self.isBody = False

def start_tr(self,attrs):
   self.isBody = True

def start_a(self, attrs):
   if self.isBody:
    href = [v for k, v in attrs if k == 'href']
    if href:
     self.urls.extend(href)
     self.isBody = False

class MailLister(SGMLParser):
def reset(self):
   SGMLParser.reset(self)
  
   self.mailUrls = []      
   self.isBody = False
   self.times = 0
   self.inLi = False

def start_p(self,attrs):
   if self.times == 0:
    self.isBody = True
    self.times = 1

def start_li(self,attrs):
   if self.isBody:
    self.inLi = True
   else:
    self.inLi = False

def start_a(self, attrs):
   href = [v for k, v in attrs if k == 'name']
   if href:
    if href[0] == "end":
     self.isBody = False
     self.inLi = False
  
   if self.inLi:
    href = [v for k, v in attrs if k == 'href']
    if href:
     self.mailUrls.extend(href)
     self.inLi = False

def getAllPaperMail(baseUrl = ""):
sock = urllib.urlopen(baseUrl)
htmlSource = sock.read()
sock.close()

parser = URLLister()
parser.feed(htmlSource)

for url in parser.urls:
   indexurl = baseUrl + url
   myList = url.split('/')
   path = myList[0]
   path = path.replace(' ',"")
   filename = myList[1]
   newurl = baseUrl + path

   if not os.path.exists(path):
    os.mkdir(path)
   filename = path + '/' + filename
  
   #download index page
   urllib.urlretrieve(indexurl, filename)

   #get papermail
   print "parse url= ",newurl
   getPaperMail(newurl,path)
   winsound.Beep(783,200)
  

def getPaperMail(baseUrl,path):
isOK = False
try1 = 0
try2 = 0
htmlSource = ""
while try1 < 6 and isOK == False:
   try:
    sock = urllib.urlopen(baseUrl)
    htmlSource = sock.read()
    sock.close()
   except:
    try2 = try2+1
    pass
   if try2 > try1:
    try1 = try1 + 1
   else:
    isOK = True

if isOK == False:
   print "get pageurl fail---",pageurl
   winsound.Beep(783,200)
   return
   
parser = MailLister()
parser.feed(htmlSource)

for url in parser.mailUrls:
   newurl = baseUrl + '/' + url
   filename = path + '/' + url

   #download each mail
   print "get: ",newurl
  
   isOK = False
   try1 = 0
   try2 = 0
   while try1 < 6 and isOK == False:
    try:
     urllib.urlretrieve(newurl, filename)
    except:
     try2 = try2+1
    if try2 > try1:
     try1 = try1 + 1
    else:
     isOK = True

if __name__ == "__main__":
print "Start."
url = r''
getAllPaperMail(url)
print "The end."