python实现的一个火车票转让信息采集器，python采集器

#coding: utf-8
'''
春运查询火车票转让信息
Author: piglei2007@gmail.com
Date: 2011.01.25
'''
import re
import os
import time
import urlparse
import datetime
import traceback
import urllib2
import socket
socket.setdefaulttimeout(20)

BLANK_RE = re.compile(r"\s+")

opener = urllib2.build_opener(urllib2.HTTPCookieProcessor())
opener.addheaders = [
  ("User-agent", "Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.9.1) Gecko/20090704 Firefox/3.5"),
  ("Accept", "*/*"),
]
urllib2.install_opener(opener)

from BeautifulSoup import BeautifulSoup

SOURCE = {
  "58": "http://bj.58.com/huochepiao/?Num=%(train)s&StartTime=%(date)s00",
  "ganji": "http://bj.ganji.com/piao/cc_%(train)s/%(date)s/",
}
RECORD_FILE = "/tmp/ticket_records.txt"

def parse_record():
  try:
    return set([x.strip() for x in open(RECORD_FILE, "r").readlines()])
  except IOError:
    open(RECORD_FILE, "w")
    return set()

def flush_record(records):
  open(RECORD_FILE, "w").write("\n".join(records))

def main(config):
  """
  开始抓取
  """
  existed = parse_record()
  to_email = []

  for train in config["trains"]:
    for date in config["dates"]:
      for type, _url in SOURCE.items():
        url = _url % dict(train=train, date=date)
        content = urllib2.urlopen(url).read()
        soup = BeautifulSoup(content)
        result = parse_content(type, soup, train)
        for url, text in result:
          url = urlparse.urljoin(_url, url)
          # 只要卧铺！
          if url not in existed and u"卧" in text:
            to_email.append([text, url])
          existed.add(url)
  if to_email:
    content = "".join(
      [x for x in [" | ".join(y) for y in to_email]]
    ).encode("utf-8")
    simple_mail(config["people"], content)
  flush_record(existed)

def parse_content(type, soup, train):
  """
  获得车次信息
  """
  result = []
  if type == "58":
    info_table = soup.find("table", id="infolist")
    if info_table:
      for x in info_table.findAll("tr", text=re.compile(ur"%s(?!时刻表)" % train, re.I)):
        a = x.parent
        _text = BLANK_RE.sub("", a.text)
        result.append([a["href"], _text])
  if type == "ganji":
    for x in soup.findAll("dl", {"class": "list_piao"}):
      a = x.dt.a
      result.append([a["href"], a.text])
  return result

EMAIL_HOST = 'smtp.sohu.com'
EMAIL_HOST_USER = 'yourname@sohu.com'
EMAIL_HOST_PASSWORD = 'yourpassword'
EMAIL_PORT = 25

def simple_mail(to, content):
  """
  发送邮件
  """
  import smtplib
  from email.mime.text import MIMEText

  msgRoot = MIMEText(content, 'html', 'UTF-8')
  msgRoot['Subject'] = "[%s]有票来啦！！！！" % datetime.datetime.today().isoformat(" ")
  msgRoot['From'] = EMAIL_HOST_USER
  msgRoot['To'] = ", ".join(to)

  s = smtplib.SMTP(EMAIL_HOST, EMAIL_PORT)
  s.login(EMAIL_HOST_USER, EMAIL_HOST_PASSWORD)
  s.sendmail(EMAIL_HOST_USER, to, msgRoot.as_string())
  s.close()

def switch_time_zone():
  """
  切换时区
  """
  os.environ["TZ"] = "Asia/Shanghai"
  time.tzset()

switch_time_zone()

if __name__ == '__main__':
  config = {
    "trains": ("k471",),
    "dates": ("20110129",),
    "people": (
      "youremail@sohu.com",
    )
  }
  try:
    main(config)
    print "%s: ok" % datetime.datetime.today()
  except Exception, e:
    print traceback.format_exc()