buzz2weibo.py

#!/usr/bin/python
# vim: set fileencoding=utf-8 :

# buzz2weibo
# Copyright 2011 Sun Zhigang
# See LICENSE for details.

from config import *
from urllib2 import urlopen, URLError
from simplejson import load
from activity import *
from weibopy.auth import OAuthHandler
from weibopy.api import API
from weibopy.error import WeibopError
from time import sleep
import os, errno, sys
from formaturl import changeurl
import re

WEIBO_APP_KEY = '3127127763'
WEIBO_APP_SECRET = '21cc35f55fc8fe73b73162964c0bb415'
BUZZ_API_KEY = 'AIzaSyAJ1zTsHUb12l1LCJbB20Fxsh8dz3zWt1o'

# 运行一次最多同步几条。缺省3。连续同步太多会被休息的
WEIBO_MAX_SYNC_COUNT = 3

def formaturl(content):

    urlreg = "([/w-]+/.)+[/w-]+.([^a-z])(/[/w- ./?%&=]*)?|[a-zA-Z0-9/-/.][/w-]+.([^a-z])(/[/w- ./?%&=]*)?" 
    
    urlfinders = [
            re.compile("([0-9]{1,3}\\.[0-9]{1,3}\\.[0-9]{1,3}\\.[0-9]{1,3}|(((news|telnet|nttp|file|http|ftp|https)://)|(www|ftp)[-A-Za-z0-9]*\\.)[-A-Za-z0-9\\.]+)(:[0-9]*)?/[-A-Za-z0-9_\\$\\.\\+\\!\\*\\(\\),;:@&=\\?/~\\#\\%]*[^]'\\.}>\\),\\\"]"),
            re.compile("([0-9]{1,3}\\.[0-9]{1,3}\\.[0-9]{1,3}\\.[0-9]{1,3}|(((news|telnet|nttp|file|http|ftp|https)://)|(www|ftp)[-A-Za-z0-9]*\\.)[-A-Za-z0-9\\.]+)(:[0-9]*)?"),
            re.compile("~/(~/|/|\\./)([-A-Za-z0-9_\\$\\.\\+\\!\\*\\(\\),;:@&=\\?/~\\#\\%]|\\\\)+"),
            re.compile("'\\<((mailto:)|)[-A-Za-z0-9\\.]+@[-A-Za-z0-9\\.]+"),
                            ]
    for urlfinder in urlfinders:
        a=urlfinder.finditer(content)
        for i in a:
            u1 = i.group()
            u1 = u1.strip()
            u2 = changeurl(u1) 
            content = content.replace(u1,u2)
    return content







def post2weibo(api, act):
    
    message = act.content + ' ' + act.link
    message = formaturl(message)
    if APPEND_SHARE_FROM_BUZZ_LINK:
        message += u' //转发自%s'.encode('utf-8') % act.origin_link

    if act.geo != '':
        geo = act.geo.split(' ')
    else:
        geo = [None, None]

    include_image = False
    if act.image != '':

        # 下载图像文件
        try:
            u = urlopen(act.image);
            data = u.read()
            u.close()
        except URLError:
            # 如果下载不下来，表示……，就别试了，当普通消息发吧
            pass
        else:
            filename = IMAGES_PATH + '/' + act.image_filename
            f = open(filename, 'w')
            f.write(data)
            f.close()
            include_image = True

    while (True):
        try:
            if include_image:
                api.upload(filename, status=message, lat=geo[0], long=geo[1])
            else:
                api.update_status(status=message, lat=geo[0], long=geo[1])
        except WeibopError, e:
            if e.reason.find('error_code:400,40013:Error:') == 0:
                # 微博太长，剪裁且留原始链接。原始链接不会太长，所以不会死循环
                message = unicode(message, 'utf-8')[0:80] + u'....更多:'
                message = message.encode('utf-8') + act.origin_link
                print '内容过长，截断发表:'
                print message
            else:
                raise
        else:
            break

    return True

# 测试config.py文件是否存在

if not os.path.exists(sys.path[0] + os.sep + 'config.py'):
    print '找不到配置文件。请先运行setup.py'
    sys.exit(1)

# 建图片目录
try:
    os.makedirs(IMAGES_PATH)
except OSError, e:
    if e.errno == errno.EEXIST:
        pass
    else:
        raise

if USE_HTTPS:
    prefix = 'https://'
else:
    prefix = 'http://'
buzz_url=prefix + 'www.googleapis.com/buzz/v1/activities/' + BUZZ_USERID + '/@public?alt=json&key=' + BUZZ_API_KEY

# 读buzz
fp = urlopen(buzz_url)
#fp = open('buzz.json')
buzz = load(fp)
fp.close()

# 微博认证
auth = OAuthHandler(WEIBO_APP_KEY, WEIBO_APP_SECRET)
auth.setToken(WEIBO_TOKEN_KEY, WEIBO_TOKEN_SECRET)
api = API(auth)


# 读已经同步过的activity id
synced_ids = set()
try:
    fp = open(HISTORY_FILE, 'r')
    for line in fp:
        synced_ids.add(line.strip())
    fp.close()
except IOError, e:
    # 如果文件不存在，就继续；否则，触发异常
    if e.errno != errno.ENOENT:
        raise

# 开始同步
count = 0
items = buzz['data']['items']
items.reverse()  # Buzz是后发的在前，所以翻转一下。感谢王瑞珩的建议
for item in items:

    # 解析buzz
    try:
        # 如果来源名是“Source Name”，就用SourceNameActivity处理
        classname = filter(unicode.isalpha, item['source']['title']) + 'Activity'
        act = eval(classname + '(item)')
    except (NameError, SyntaxError):
        # SourceNameActivity没有，或者Source Name是中文，用WildcardActivity做缺省处理
        act = WildcardActivity(item);

    # 同步未同步过的
    if act.id not in synced_ids:
        print '-----------------------'
        print 'syncing ' + act.id
        print item['source']['title']
        print act.origin_link
        if act.content != '':
            print act.content
        if act.link != '':
            print act.link
        if act.image != '':
            print act.image
        if act.image_filename != '':
            print act.image_filename
        if act.geo != '':
            print act.geo

        if DEBUG:
            continue

        if post2weibo(api, act):
            synced_ids.add(act.id)

            # 将同步过的activity id写入历史文件
            fp = open(HISTORY_FILE, 'w')
            for id in synced_ids:
                fp.write(id + '\n')
            fp.close()

            count = count + 1
            if count >= WEIBO_MAX_SYNC_COUNT:
                break

            sleep(1)  # 延时才能让新浪微博按正确顺序显示