前些时间，公司让做一个自动回复卖家信息的程序，现在总结下（用python实现的）
1.登陆雅虎拍卖后台手动获取cookie文件
#coding=utf-8
import sqlite3
import cookielib
from cStringIO import StringIO
def sqlite2cookieChrome(filename):
con = sqlite3.connect(filename)
con.text_factory = str
cur = con.cursor()
cur.execute("select host_key, path, secure, expires_utc, name, value from cookies")
ftstr = ["FALSE","TRUE"]
s = StringIO()
s.write(
"""# Netscape HTTP Cookie File
# http://www.netscape.com/newsref/std/cookie_spec.html
# This is a generated file! Do not edit.\n""")
for item in cur.fetchall():
try:
s.write("%s\t%s\t%s\t%s\t%s\t%s\t%s\n" % (
item[0], ftstr[item[0].startswith('.')], item[1],
ftstr[item[2]], item[3], item[4], item[5]))
except UnicodeError:
continue
s.seek(0)
return s.read()
cookie_jar = cookielib.MozillaCookieJar()
cookie_jar._really_load(s, '', True, True)
return cookie_jar
yahoo_id = raw_input("please input yahoo id: ").strip()
cookie_jar = sqlite2cookieChrome('C:\Documents and Settings\Administrator\Local Settings\Application Data\Google\Chrome\User Data\Default\cookies') #windows xp
open('./%s.txt' % yahoo_id, 'w').write(cookie_jar)
2.回复卖家信息
主要实现代码：
# auction_url: http://page10.auctions.yahoo.co.jp/jp/auction/m114547886
# subject: 评论类型
# comment_content：评论内容
# cookie_jar: cookie
def post_yahoo_auction_comment(auction_url, subject, comment_content, cookie_jar):
try:
opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cookie_jar))
opener.addheaders = [("User-Agent", "Mozilla/5.0 (Windows NT 5.1) AppleWebKit/537.11 (KHTML, like Gecko) Chrome/23.0.1271.95 Safari/537.11")]
#-------------------------------抓取取引的链接--------------------------------------#
html = opener.open(auction_url).read()
write_file('thefile1.html',html,'w');
soup = BeautifulSoup(html)
url = soup.find("a", text="取引ナビ")['href']
#新版的yahoo页面
if(url[0:15] == 'https://contact'):
#-------------------------------抓提交的标单隐藏域--------------------------------------#
html = opener.open(url).read()
write_file('thefile2.html',html,'w');
soup = BeautifulSoup(html)
aid = soup.find("input", attrs={"name":"aid"})['value']
bid = soup.find("input", attrs={"name":"bid"})['value']
syid = soup.find("input", attrs={"name":"syid"})['value']
oid = soup.find("input", attrs={"name":"oid"})['value']
crumb = soup.find("input", attrs={"name":".crumb"})['value']
#-------------------------------初次提交标单--------------------------------------#
comment_content = comment_content.encode("utf-8", "ignore")
post_param = urllib.urlencode({
"aid" :aid,
"syid":syid,
"bid":bid,
"oid":oid,
"body" :comment_content,
".crumb":crumb,
"subject" : subject
})
html = opener.open("https://contact.auctions.yahoo.co.jp/preview", post_param).read()
write_file('thefile3.html',html,'w');
soup = BeautifulSoup(html)
crumb = soup.find("input", attrs={"name":".crumb"})['value']
#-------------------------------确认提交标单--------------------------------------#
confirm_param = urllib.urlencode({
".crumb" : crumb,
"aid" : aid,
"syid" : syid,
"bid" : bid,
"oid" : oid,
"subject" : subject,
"body" : comment_content
})
html = opener.open("https://contact.auctions.yahoo.co.jp/send", confirm_param).read()
write_file('thefile4.html',html,'w');
return True
else:
#旧版的雅虎页面
comment_content = comment_content.encode("EUC_JP", "ignore")
page_number = auction_url.split('.')[0].replace("http://", "").replace('page', '')
item_id = auction_url.split('/')[-1]
post_param = urllib.urlencode({
"aID" : item_id,
"subject" : subject,
"comment" : comment_content
})
html = opener.open("http://pageinfo%s.auctions.yahoo.co.jp/jp/show/contact_preview#message"%page_number, post_param).read()
write_file('thefile2.html',html,'w');
soup = BeautifulSoup(html)
crumb = soup.find("input", attrs={"name":".crumb"})['value']
target = soup.find("input", attrs={"name":"target"})['value']
confirm_param = urllib.urlencode({
".crumb" : crumb,
"aID" : item_id,
"subject" : subject,
"comment" : comment_content,
"target" : target,
})
html = opener.open("http://edit%s.auctions.yahoo.co.jp/jp/config/contact_submit"%page_number, confirm_param).read()
return True
except Exception as e:
print e
") == -1:
return -1
else:
return -2
3.雅虎后台自动评论
主要代码：
def post_yahoo_auction_comment(auction_url,item_id, seller,cookie_jar):
try:
opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cookie_jar))
opener.addheaders = [('User-Agent', 'Mozilla/5.0 (Windows NT 5.1) AppleWebKit/537.11 (KHTML, like Gecko) Chrome/23.0.1271.95 Safari/537.11')]
page_number = auction_url.split('.')[0].replace('http://', '').replace('page', '')
url = 'http://edit'+page_number+'.auctions.yahoo.co.jp/jp/config/leavefeedback?t='+seller+'&nr=1&aID='+item_id
#抓取表单提交页面
html = opener.open(url).read()
write_file('thefile1.html',html,'w')
findstr = '前回、評価した内容'
findstr = findstr.encode("utf-8", "ignore")
if html.find(findstr) > -1:
return 2
rating = 5
comment = '今回ありがとうございます！早速の発送ので取引ができます！ 今後もよろしくお願いします！！こちらの評価もお願いします。^_^ '
comment = comment.encode("EUC_JP", "ignore")
soup= BeautifulSoup(html, from_encoding="EUC_JP") #要指定编码，不然会出现编码错误
aID = soup.find('input', attrs={'name':'aID'})['value']
author = soup.find('input', attrs={'name':'author'})['value']
c_comment = soup.find('input', attrs={'name':'c_comment'})['value']
c_date = soup.find('input', attrs={'name':'c_date'})['value']
cacheid = soup.find('input', attrs={'name':'cacheid'})['value']
catid = soup.find('input', attrs={'name':'catid'})['value']
curprice = soup.find('input', attrs={'name':'curprice'})['value']
endtime = soup.find('input', attrs={'name':'endtime'})['value']
isBidByDocomo = soup.find('input', attrs={'name':'isBidByDocomo'})['value']
isIntlAuction = soup.find('input', attrs={'name':'isIntlAuction'})['value']
isIntlTarget = soup.find('input', attrs={'name':'isIntlTarget'})['value']
isIntlUser = soup.find('input', attrs={'name':'isIntlUser'})['value']
isResponse = soup.find('input', attrs={'name':'isResponse'})['value']
isRevision = soup.find('input', attrs={'name':'isRevision'})['value']
logininwinnerslist = soup.find('input', attrs={'name':'logininwinnerslist'})['value']
preview = soup.find('input', attrs={'name':'preview'})['value']
r_comment = soup.find('input', attrs={'name':'r_comment'})['value']
r_date = soup.find('input', attrs={'name':'r_date'})['value']
r_id = soup.find('input', attrs={'name':'r_id'})['value']
r_rating = soup.find('input', attrs={'name':'r_rating'})['value']
role = soup.find('input', attrs={'name':'role'})['value']
starttime = soup.find('input', attrs={'name':'starttime'})['value']
t = soup.find('input', attrs={'name':'t'})['value']
target = soup.find('input', attrs={'name':'target'})['value']
targetinwinnerslist = soup.find('input', attrs={'name':'targetinwinnerslist'})['value']
u = soup.find('input', attrs={'name':'u'})['value']
UEtitle = soup.find('input', attrs={'name':'UEtitle'})['value']
wc = soup.find('input', attrs={'name':'wc'})['value']
write = soup.find('input', attrs={'name':'write'})['value']
post_param = urllib.urlencode({
'aID':aID,
'author':author,
'c_comment':c_comment,
'c_date':c_date,
'cacheid':cacheid,
'catid':catid,
'comment':comment,
'curprice':curprice,
'endtime':endtime,
'isBidByDocomo':isBidByDocomo,
'isIntlAuction':isIntlAuction,
'isIntlTarget':isIntlTarget,
'isIntlUser':isIntlUser,
'isResponse':isResponse,
'isRevision':isRevision,
'logininwinnerslist':logininwinnerslist,
'preview':preview,
'r_comment':r_comment,
'r_date':r_date,
'r_id':r_id,
'r_rating':r_rating,
'rating':rating,
'role':role,
'starttime':starttime,
't' :t,
'target':target,
'targetinwinnerslist':targetinwinnerslist,
'u':u,
'UEtitle':UEtitle,
'wc':wc,
'write' :write,
})
#抓取确认提交后的页面
html = opener.open('http://redit.auctions.yahoo.co.jp/jp/rconfig/leavefeedback', post_param).read()
write_file('thefile2.html',html,'w');
soup= BeautifulSoup(html, from_encoding="EUC_JP")
post_param = urllib.urlencode({
'aID':aID,
'author':author,
'c_comment':c_comment,
'c_date':c_date,
'cacheid':cacheid,
'catid':catid,
'comment':comment,
'curprice':curprice,
'endtime':endtime,
'isBuyer':soup.find('input', attrs={'name':'isBuyer'})['value'],
'isBidByDocomo':isBidByDocomo,
'isIntlAuction':isIntlAuction,
'isIntlTarget':isIntlTarget,
'isIntlUser':isIntlUser,
'isResponse':isResponse,
'isRevision':isRevision,
'isSeller':soup.find('input', attrs={'name':'isSeller'})['value'],
'logininwinnerslist':logininwinnerslist,
'own':soup.find('input', attrs={'name':'own'})['value'],
'preview':soup.find('input', attrs={'name':'preview'})['value'],
'r_comment':r_comment,
'r_date':r_date,
'r_id':r_id,
'r_rating':soup.find('input', attrs={'name':'r_rating'})['value'],
'rating':rating,
'role':role,
'starttime':starttime,
't' :t,
'target':target,
'targetinwinnerslist':targetinwinnerslist,
'u':u,
'UEtitle':UEtitle,
'wc':wc,
'write' :soup.find('input', attrs={'name':'write'})['value'],
})
html = opener.open('http://redit.auctions.yahoo.co.jp/jp/rconfig/leavefeedback', post_param).read()
write_file('thefile3.html',html,'w');
return True
except Exception as e:
print e
') == -1:
return -1
else:
return -2
代码说明：其实自动回复卖家信息并不难，主要是抓取页面表单数据，并将数据提交到另一个页面。函数的auction_url, comment_content, seller等参数都是从数据库读的，我在程序用到了ConfigParser库读取配置文件。
python实现如下：
import ConfigParser
def get_db_con():
config = ConfigParser.ConfigParser();
config.read('conf/db.cfg')
host = config.get("mysqllink1","host")
port = config.getint("mysqllink1","port")
user = config.get("mysqllink1","user")
password = config.get("mysqllink1","password")
dbname = config.get("mysqllink1","dbname")
charset = config.get("mysqllink1","charset")
return [host,port,user,password,dbname,charset]
def get_bid_item_from_db():
link = get_db_con()
conn = MySQLdb.connect(host=link[0], port=link[1], user=link[2], passwd=link[3], db=link[4], charset=link[5], cursorclass=MySQLdb.cursors.DictCursor)
cursor = conn.cursor()
config = ConfigParser.ConfigParser();
config.read('conf/db.cfg')
limit = config.get("selpar","limit")
limit = '' if limit=='All' else ' limit '+limit
query = "SELECT a.*, b.seller, b.is_send_mail FROM bid_item a JOIN bid_win b ON a.bid = b.bid JOIN biogg_auction_data bd ON bd.AuctionId = a.item_id WHERE ( to_days(now()) - to_days(b.win_date) = 1 OR to_days(now()) = to_days(b.win_date)) AND a.is_store<>1 AND b.is_send_mail <= 0"+limit
cursor.execute(query)
bid_items = cursor.fetchall()
cursor.close()
conn.close()
return bid_items
db.cfg:
[mysqllink1] host=localhost user=root password=pwd port=3306 dbname=test charset=utf8 [selpar] limit=All
其中还用到BeautifulSoup库，链接：http://www.crummy.com/software/BeautifulSoup/bs4/doc/