def saveMessage(god_name, github_name, god_id, message):
'''
create by bigzhu at 15/07/16 09:44:39 为了抽取数据方便,合并数据到 content 里
'''
content = storage()
content.type = message.type
content.repo = message.repo
content.payload = message.payload
content = json.dumps(content)
m = public_bz.storage()
m.god_id = god_id
m.god_name = god_name
m.name = github_name
# m.avatar = message.actor['avatar_url']
m.id_str = message['id']
m.m_type = 'github'
m.created_at = time_bz.unicodeToDateTIme(message.created_at)
m.created_at += timedelta(hours=8)
m.content = content
m.text = None
m.href = None
id = pg.insertIfNotExist('message', m,
"id_str='%s' and m_type='github'" % m.id_str)
if id is not None:
print '%s new github %s' % (m.name, m.id_str)
return id
def sendTwitter(openid, tweet, screen_name, id):
'''
发送twitter的消息
'''
articles = []
if hasattr(tweet,
'extended_entities') and tweet.extended_entities['media']:
for media in tweet.extended_entities['media']:
article = storage()
article.picurl = "http://follow.center/ProxyHandler/%s" % media[
'media_url_https']
article.url = "http://follow.center/message?t=twitter&id=%s" % id
articles.append(article)
if len(articles) == 1:
articles[0].title = screen_name
else:
articles[0].title = screen_name + ': ' + tweet.text
articles[0].description = tweet.text
else:
article = storage()
article.title = screen_name
article.url = "http://follow.center/message?t=twitter&id=%s" % id
article.description = tweet.text
articles = [article]
sendArticle(openid, articles)
def saveMessage(message):
'''
create by bigzhu at 15/07/16 09:44:39 为了抽取数据方便,合并数据到 content 里
'''
message.id_str = message.pop('id')
content = storage()
content.type = message.type
content.repo = message.pop('repo')
content.payload = message.pop('payload')
message.content = json.dumps(content)
if message.get('org'):
message.org = json.dumps(message.org)
# return db_bz.insertIfNotExist(pg, 'github_message', message, "id_str='%s'" % message.id_str)
m = public_bz.storage()
m.id_str = message.id_str
m.m_type = 'github'
m.m_user_id = message.actor
m.created_at = message.created_at
m.content = message.content
m.text = None
m.href = None
return db_bz.insertIfNotExist(
pg, 'm', m, "id_str='%s' and m_type='github'" % message.id_str)
def saveMessage(god_name, facebook_name, god_id, message):
'''
'''
message = public_bz.storage(message)
m = public_bz.storage()
m.god_id = god_id
m.god_name = god_name
m.name = facebook_name
m.m_type = 'facebook'
m.id_str = message.id
m.created_at = message.created_time
m.content = json.dumps({'description': message.get('description')})
m.text = message.get('message')
m.extended_entities = json.dumps({
'pictrue': message.get('full_picture'),
'source': message.get('source')
})
m.type = message.get('type')
if m.type == 'video':
print('facebook 不再同步 video, 有时效, 无法一直看')
return
m.href = message.get('link')
id = pg.insertIfNotExist('message', m,
"id_str='%s' and m_type='facebook'" % m.id_str)
if id is not None:
print '%s new facebook message %s' % (m.name, m.id_str)
return id
def saveMessage(message):
'''
create by bigzhu at 15/07/16 09:44:39 为了抽取数据方便,合并数据到 content 里
'''
message.id_str = message.pop('id')
content = storage()
content.type = message.type
content.repo = message.pop('repo')
content.payload = message.pop('payload')
message.content = json.dumps(content)
if message.get('org'):
message.org = json.dumps(message.org)
# return db_bz.insertIfNotExist(pg, 'github_message', message, "id_str='%s'" % message.id_str)
m = public_bz.storage()
m.id_str = message.id_str
m.m_type = 'github'
m.m_user_id = message.actor
m.created_at = message.created_at
m.content = message.content
m.text = None
m.href = None
return db_bz.insertIfNotExist(pg, 'm', m, "id_str='%s' and m_type='github'" % message.id_str)
def getHtmlListByNameLike(search_name):
'''
根据名字查找对应html
返回 list
{name:'', time: ''}
'''
html_list = []
for html in os.listdir(HTML_PATH):
html_name = cutName(html)
if html_name == '':
continue
if isNameLike(html_name, search_name):
pass
else:
continue
try:
modify_time = getModifyTime(html_name)
except OSError as e: # 可能会有非html结尾的文件，忽略
print e
continue
the_html = storage()
the_html.name = html_name
the_html.time = modify_time
html_list.append(the_html)
# 按时间排序
html_list = sorted(html_list, key=lambda d: d.time, reverse=True)
return html_list
def saveMessage(god_name, twitter_name, god_id, tweet):
'''
create by bigzhu at 15/07/10 14:39:48
保存twitter
create by bigzhu at 16/03/26 06:05:12 重构，改很多
modify by bigzhu at 16/03/26 20:33:59 重构
user 本系统用户信息
tweet 消息本身
modify by bigzhu at 17/01/13 15:38:11 去了用不到的
'''
m = public_bz.storage()
m.god_id = god_id
m.god_name = god_name
m.name = twitter_name
m.id_str = tweet.id_str
m.m_type = 'twitter'
m.created_at = tweet.created_at
m.content = None
m.text = tweet.text
if hasattr(tweet, 'extended_entities'):
m.extended_entities = json.dumps(tweet.extended_entities)
m.type = tweet.extended_entities['media'][0]['type']
m.href = 'https://twitter.com/' + m.name + '/status/' + m.id_str
id = pg.insertIfNotExist('message', m, "id_str='%s' and m_type='twitter'" % tweet.id_str)
if id is not None:
print '%s new twitter %s' % (m.name, m.id_str)
return id
def post(self):
'''
新增文件
'''
global md5
self.set_header("Content-Type", "application/json")
results = []
if self.request.files:
for i in self.request.files:
fd = self.request.files[i]
for f in fd:
file_name = f.get("filename")
file_suffix = file_name[file_name.rfind("."):]
file_body = f["body"]
md5.update(file_body)
file_hash = md5.hexdigest()
file_path = "static/uploaded_files/%s_%s_%s" % (self.current_user, int(time.time()), file_name)
# hash
img = open(file_path, 'w')
img.write(file_body)
img.close()
new_file = storage(file_name=file_name, file_path="/" + file_path, file_hash=file_hash, file_type="file", suffix=file_suffix, seqname='uploaded_files_id_seq')
file_id = self.pg.db.insert("uploaded_files", **new_file)
r = {
'file_id': file_id,
'file_name': file_name,
'file_path': '/' + file_path,
'suffix': file_suffix
}
results.append(r)
self.write(json.dumps({'error': '0', 'results': results}))
def render(self, comments, key_type, key):
parm = storage()
parm.comments = comments
parm.key_type = key_type
parm.key = key
return self.render_string(self.html_name, parm=parm)
def getUser(user_name):
'''
modify by bigzhu at 15/07/31 16:41:16 api找用户时是按专注度来排序的,名字绝对匹配的未必是第一位, 从10个里面找
'''
users = list(pg.select('instagram_user', where="lower(username)=lower('%s')" % user_name))
if users:
return users[0]
else:
user = None
for this_user in api.user_search(user_name, 10):
if this_user.username.lower() == user_name.lower():
user = this_user
break
#如果没有这个用户
if user is None:
public_db.delNoName('instagram', user_name)
#user = api.user_search(user_name, 1)[0]
user = api.user(user.id)
db_user = storage()
db_user.id = user.id
db_user.username = user.username
db_user.full_name = user.full_name
db_user.profile_picture = user.profile_picture
db_user.bio = user.bio
db_user.website = user.website
db_user.counts = json.dumps(user.counts)
pg.insert('instagram_user', **db_user)
#db_bz.insertIfNotExist(pg, 'instagram_user', db_user, "id=%s" % db_user.id)
return getUser(user_name)
def main(god, wait):
'''
create by bigzhu at 15/07/15 17:54:08 取github
modify by bigzhu at 15/07/22 16:20:42 时间同样要加入8小时,否则不正确
'''
god_name = god.name
github_name = god.github['name']
god_id = god.id
etag = god.github.get('sync_key')
github_user = getGithubUser(github_name, god_name)
if github_user is None:
return
headers = {'If-None-Match': etag}
try:
r = requests.get('https://api.github.com/users/%s/events' %
github_name,
headers=headers,
params=params)
except requests.exceptions.ConnectionError:
print public_bz.getExpInfoAll()
return
if r.status_code == 200:
etag = r.headers['etag']
limit = r.headers['X-RateLimit-Remaining']
if limit == '0':
return
for message in r.json():
message = storage(message)
saveMessage(god_name, github_name, god_id, message)
# oper.noMessageTooLong('github', github_name)
saveUser(god_name, github_name, github_user, etag) # 为了update etag
if r.status_code == 404:
# public_db.sendDelApply('github', god_name, github_name, '404')
god_oper.delNoName('github', god_name)
def saveMessage(god_name, twitter_name, god_id, blog):
m = public_bz.storage()
m.god_id = god_id
m.god_name = god_name
m.name = twitter_name
m.id_str = blog['id']
m.m_type = 'tumblr'
m.created_at = time_bz.timestampToDateTime(blog['timestamp'])
type = blog.get('type')
m.href = blog.get('short_url')
m.type = type
if type == 'text':
m.title = blog.get('title')
m.text = blog.get('body')
elif type == 'photo':
m.text = blog.get('caption')
m.extended_entities = json.dumps(blog.get('photos'))
elif type == 'video':
m.extended_entities = json.dumps({'video_url': blog.get('video_url')})
m.content = None
id = pg.insertIfNotExist('message', m,
"id_str='%s' and m_type='tumblr' " % m.id_str)
if id is None:
pass
else:
print '%s new tumblr message %s' % (m.name, m.id_str)
def saveBlogs(user_name, blogs, offset):
if blogs:
pass
else:
return
for blog in blogs:
m = public_bz.storage()
m.id_str = blog['id']
m.m_type = 'tumblr'
m.m_user_id = user_name
m.created_at = time_bz.timestampToDateTime(blog['timestamp'])
m.extended_entities = json.dumps(blog.get('photos'))
m.content = None
m.text = blog.get('caption')
m.href = blog.get('short_url')
m.type = blog.get('type')
result = pg.insertIfNotExist(pg, 'm', m, "id_str='%s' and m_type='tumblr' " % m.id_str)
if result is None: # 有重复记录了,就不再继续了
print 'have same data'
return
else:
print 'new ', m.id_str, m.type, 'offset:', offset, 'name:', user_name
# 继续取
new_offset = offset + 20
new_blogs = callGetMeidaApi(user_name, offset=new_offset)['response']['posts']
saveBlogs(user_name, new_blogs, new_offset)
def saveGraphqlMessage(ins_name, user_name, god_id, message):
'''
用 Graphql 取到的数据
'''
message = storage(message)
m = public_bz.storage()
m.god_name = user_name
m.name = ins_name
m.m_type = 'instagram'
m.id_str = message.id
m.created_at = time_bz.timestampToDateTime(message.taken_at_timestamp)
if message.get('edge_media_to_caption').get('edges'):
m.text = message.get('edge_media_to_caption').get('edges')[0].get(
'node').get('text')
else:
m.text = None
m.extended_entities = json.dumps({'url': message.display_url})
m.href = 'https://www.instagram.com/p/%s/' % message.shortcode
if message.__typename == 'GraphSidecar': # mutiple image
edges = getMutipleImage(message.shortcode)
images = []
for edge in edges:
url = edge['node']['display_url']
images.append({'url': url})
m.extended_entities = json.dumps(images)
m.type = 'images'
elif message.is_video:
m.type = 'video'
video_url = getVideoUrl(m.href)
m.extended_entities = json.dumps({
'url': message.display_url,
'video_url': video_url
})
else:
m.type = 'image'
id = pg.insertIfNotExist('message', m,
"id_str='%s' and m_type='instagram'" % m.id_str)
if id is not None:
print '%s new instagram message %s' % (m.name, m.id_str)
# 肯定会有一条重复
# else:
# print '%s 重复记录 %s' % (m.user_name, m.id_str)
return id
def addTimeLine(self, oper, user_id, target_type, target_id, other_info=None):
timeline = storage()
timeline.oper = oper
timeline.user_id = user_id
timeline.target_type = target_type
timeline.target_id = target_id
if other_info:
timeline.other_info = json.dumps(other_info, cls=public_bz.ExtEncoder)
return self.pg.db.insert('timeline', **timeline)
def getTenContent(name):
ten_names = getList(name)[0][1][:SHOW_COUNT]
lists = []
for i in ten_names:
c = public_bz.storage()
c.name = i[0]
c.time = i[1]
c.content = getHtmlContent(c.name)
lists.append(c)
return lists
def sendGithub(openid, text, user_name, id):
'''
create by bigzhu at 15/07/22 15:05:01 发送github的消息
'''
articles = []
article = storage()
article.title = user_name
article.url = "http://follow.center/message?t=github&id=%s" % id
article.description = text
articles = [article]
sendArticle(openid, articles)
def post(self):
self.set_header("Content-Type", "application/json")
data = json.loads(self.request.body)
anki_info = storage()
anki_info.user_name = data['user_name']
anki_info.password = data['password']
anki_info.user_id = self.current_user
db_bz.insertOrUpdate(pg, 'anki', anki_info,
"user_id='%s'" % anki_info.user_id)
anki.getMidAndCsrfTokenHolder(anki_info.user_id, reset_cookie=True)
self.write(json.dumps(self.data))
def sendGithub(openid, text, user_name, id):
'''
create by bigzhu at 15/07/22 15:05:01 发送github的消息
'''
articles = []
article = storage()
article.title = user_name
article.url = "http://follow.center/message?t=github&id=%s" % id
article.description = text
articles = [article]
sendArticle(openid, articles)
def getUserEvent(user_name, etag):
'''
create by bigzhu at 15/07/15 17:54:08 取github
modify by bigzhu at 15/07/22 16:20:42 时间同样要加入8小时,否则不正确
'''
headers = {'If-None-Match': etag}
try:
r = requests.get('https://api.github.com/users/%s/events' % user_name,
headers=headers)
except requests.exceptions.ConnectionError:
print public_bz.getExpInfoAll()
return
if r.status_code == 200:
messages = r.json()
if not messages:
delGithubUser(user_name)
# 没有这个github用户,取消
return
actor = messages[0]['actor']
# actor不定是作者名字，有可能org才是
if actor['login'].lower() == user_name.lower():
the_user = actor
else:
org = messages[0]['org']
if org['login'].lower() == user_name.lower():
the_user = org
# 如果是org，那么url不同
the_user['url'] = "https://api.github.com/users/" + user_name
else:
raise "in this github can't find user_name=%s"
user_id = saveUser(the_user['id'], the_user['url'])
# 更新etag
etag = r.headers['etag']
updateEtag(user_name, etag)
for i in r.json():
i['actor'] = user_id
message = storage(i)
message.created_at = time_bz.unicodeToDateTIme(message.created_at)
message.created_at += timedelta(hours=8)
saveMessage(copy.deepcopy(message))
#id = saveMessage(copy.deepcopy(message))
# if id is not None:
# text = formatInfo(message)
# print text
# openids = public_db.getOpenidsByName('github', user_name)
# for data in openids:
# wechat_oper.sendGithub(data.openid, text, user_name, id)
if r.status_code == 404:
delGithubUser(user_name)
else:
print r.status_code
def sendInstagram(openid, text, img_url, user_name, id):
'''
create by bigzhu at 15/08/01 00:35:08
modify by bigzhu at 15/08/01 00:57:28 不用代理,没被屏蔽
'''
article = storage()
article.title = user_name
article.picurl = 'http://follow.center/sp/' + oper.encodeUrl(img_url)
article.url = "http://follow.center/message?t=instagram&id=%s" % id
article.description = text
articles = [article]
sendArticle(openid, articles)
def initialize(self):
'''
针对 oauth2 ,需要你重载的时候来设置为你自己的参数, 以下是 google twitter douban 的例子
modify by bigzhu at 15/04/26 21:56:09 对应的oauth登录的参数,应该在对应的oauth里面来设置,这里不用再设置了
'''
UserInfoHandler.initialize(self)
oauth2 = storage()
oauth2.google = storage(enabled=False, url='/google')
oauth2.twitter = storage(enabled=False, url='/twitter')
oauth2.douban = storage(enabled=False, url='/douban')
oauth2.github = storage(enabled=False, url='/github')
self.oauth2 = oauth2
# 用户操作相关的
self.user_oper = user_bz.UserOper(self.pg)
# 是否要验证
self.validate = False
#salt
self.salt = "hold is watching you"
def getUserEvent(user_name, etag):
'''
create by bigzhu at 15/07/15 17:54:08 取github
modify by bigzhu at 15/07/22 16:20:42 时间同样要加入8小时,否则不正确
'''
headers = {'If-None-Match': etag}
try:
r = requests.get('https://api.github.com/users/%s/events' % user_name, headers=headers)
except requests.exceptions.ConnectionError:
print public_bz.getExpInfoAll()
return
if r.status_code == 200:
messages = r.json()
if not messages:
delGithubUser(user_name)
# 没有这个github用户,取消
return
actor = messages[0]['actor']
# actor不定是作者名字，有可能org才是
if actor['login'].lower() == user_name.lower():
the_user = actor
else:
org = messages[0]['org']
if org['login'].lower() == user_name.lower():
the_user = org
# 如果是org，那么url不同
the_user['url'] = "https://api.github.com/users/" + user_name
else:
raise "in this github can't find user_name=%s"
user_id = saveUser(the_user['id'], the_user['url'])
# 更新etag
etag = r.headers['etag']
updateEtag(user_name, etag)
for i in r.json():
i['actor'] = user_id
message = storage(i)
message.created_at = time_bz.unicodeToDateTIme(message.created_at)
message.created_at += timedelta(hours=8)
saveMessage(copy.deepcopy(message))
#id = saveMessage(copy.deepcopy(message))
# if id is not None:
# text = formatInfo(message)
# print text
# openids = public_db.getOpenidsByName('github', user_name)
# for data in openids:
# wechat_oper.sendGithub(data.openid, text, user_name, id)
if r.status_code == 404:
delGithubUser(user_name)
else:
print r.status_code
def post(self):
self.set_header("Content-Type", "application/json")
data = json.loads(self.request.body)
god_id = data['god_id']
remark = data['remark']
where = "user_id='%s' and god_id=%s" % (self.current_user, god_id)
values = storage()
values.user_id = self.current_user
values.remark = remark
values.god_id = god_id
db_bz.insertOrUpdate(pg, 'remark', values, where)
self.write(json.dumps({'error': '0'}))
def sendInstagram(openid, text, img_url, user_name, id):
'''
create by bigzhu at 15/08/01 00:35:08
modify by bigzhu at 15/08/01 00:57:28 不用代理,没被屏蔽
'''
article = storage()
article.title = user_name
article.picurl = 'http://follow.center/sp/' + oper.encodeUrl(img_url)
article.url = "http://follow.center/message?t=instagram&id=%s" % id
article.description = text
articles = [article]
sendArticle(openid, articles)
def saveUser(god_name, tumblr_name, user):
social_user = public_bz.storage()
social_user.type = 'tumblr'
# social_user.name = user['name']
social_user.name = tumblr_name
social_user.count = user.get('likes', -1) # 有人会不分享likes数
avatar_url = 'https://api.tumblr.com/v2/blog/%s.tumblr.com/avatar/512' % user[
'name']
social_user.avatar = tumblrRealAvatar(avatar_url)
social_user.description = user['description']
social_user.sync_key = user['updated']
pg.update('god', where={'name': god_name}, tumblr=json.dumps(social_user))
return social_user
def saveUser(god_name, twitter_name, twitter_user):
social_user = public_bz.storage()
# 不要用返回的name, 大小写会发生变化
# social_user.name = twitter_user.screen_name
social_user.name = twitter_name
social_user.type = 'twitter'
social_user.count = twitter_user.followers_count
social_user.avatar = twitter_user.profile_image_url_https.replace('_normal', '_400x400')
social_user.description = twitter_user.description
# 没有找到
# social_user.sync_key = twitter_user.description
pg.update('god', where={'name': god_name}, twitter=json.dumps(social_user))
return social_user
def saveUser(god_name, ins_name, user, sync_key):
social_user = public_bz.storage()
social_user.type = 'instagram'
# social_user.name = user['username']
social_user.name = ins_name
social_user.count = user['followed_by']['count']
social_user.avatar = user['profile_pic_url']
social_user.description = user['biography']
social_user.id = user['id']
social_user.sync_key = sync_key
pg.update('god',
where={'name': god_name},
instagram=json.dumps(social_user))
return social_user
def getContent(name, count=15, lenght=None):
ten_names = getList(name)[0][1][:count]
lists = []
for i in ten_names:
c = public_bz.storage()
c.name = i[0]
c.time = i[1]
c.content = getHtmlContent(c.name)
if lenght:
print len(c.content)
if len(c.content) < lenght:
print 'continue ' + c.name
continue
lists.append(c)
return lists
def getContent(name, count=15, lenght=None):
ten_names = getList(name)[0][1][:count]
lists = []
for i in ten_names:
c = public_bz.storage()
c.name = i[0]
c.time = i[1]
c.content = getHtmlContent(c.name)
if lenght:
print len(c.content)
if len(c.content) < lenght:
print 'continue ' + c.name
continue
lists.append(c)
return lists
def sendTwitter(openid, tweet, screen_name, id):
'''
发送twitter的消息
'''
articles = []
if hasattr(tweet, 'extended_entities') and tweet.extended_entities['media']:
for media in tweet.extended_entities['media']:
article = storage()
article.picurl = "http://follow.center/ProxyHandler/%s" % media['media_url_https']
article.url = "http://follow.center/message?t=twitter&id=%s" % id
articles.append(article)
if len(articles) == 1:
articles[0].title = screen_name
else:
articles[0].title = screen_name + ': ' + tweet.text
articles[0].description = tweet.text
else:
article = storage()
article.title = screen_name
article.url = "http://follow.center/message?t=twitter&id=%s" % id
article.description = tweet.text
articles = [article]
sendArticle(openid, articles)
def saveMessage(message):
'''
create by bigzhu at 15/07/16 09:44:39 为了抽取数据方便,合并数据到 content 里
'''
message.id_str = message.pop('id')
content = storage()
content.type = message.type
content.repo = message.pop('repo')
content.payload = message.pop('payload')
message.content = json.dumps(content)
if message.get('org'):
message.org = json.dumps(message.org)
return db_bz.insertIfNotExist(pg, 'github_message', message, "id_str='%s'" % message.id_str)
def saveUser(god_name, facebook_name, user, etag):
social_user = public_bz.storage()
social_user.type = 'facebook'
# social_user.name = user['username']
social_user.name = facebook_name
# facebook 取不到 friend count or followed count
social_user.count = -1
social_user.avatar = user['picture']['data']['url']
social_user.description = user.get('bio') # bio 可能没有
if etag is not None:
social_user.sync_key = etag
social_user.out_id = user['id']
pg.update('god',
where={'name': god_name},
facebook=json.dumps(social_user))
return social_user
def groupByCreatedDateDay(self, timelines):
'''
create by bigzhu at 15/02/03 13:44:24 按照天的精度,归并timeline
modify by bigzhu at 15/02/06 13:29:58 修改为返回 list 而不是一个 dic, 以便于对时间按天的排序
'''
day_time_lines = []
group_time_line = {}
for timeline in timelines:
day = timeline.created_date.strftime('%Y年%m月%d日')
this_day_timeline = group_time_line.get(day)
if this_day_timeline:
#this_day_timeline.insert(0, timeline)
this_day_timeline.append(timeline)
else:
group_time_line[day] = [timeline]
day_time_lines.append(storage(day = day, timelines = group_time_line[day]))
return day_time_lines
def put(self):
self.set_header("Content-Type", "application/json")
data = json.loads(self.request.body)
last_time = int(data.get('last_time'))
last_time = time_bz.timestampToDateTime(last_time, True)
# last_message_id = data.get('message_id')
user_id = self.current_user
if user_id is None:
pass
else:
last_oper.saveLast(last_time, user_id)
data = storage()
data.error = OK
data.unread_message_count = oper.getUnreadCount(user_id)
self.write(json.dumps(data, cls=json_bz.ExtEncoder))
def getUser(user_name, always_check=False):
'''
modify by bigzhu at 15/07/31 16:41:16 api找用户时是按专注度来排序的,名字绝对匹配的未必是第一位, 从10个里面找
'''
users = list(
pg.select('instagram_user',
where="lower(username)=lower('%s')" % user_name))
if users and not always_check:
return users[0]
else:
user = None
try:
for this_user in api.user_search(user_name, 10):
if this_user.username.lower() == user_name.lower():
user = this_user
break
except instagram.bind.InstagramClientError:
print public_bz.getExpInfoAll()
return
# 如果没有这个用户
if user is None:
public_db.delNoName('instagram', user_name)
return
try:
user = api.user(user.id)
except (instagram.bind.InstagramAPIError,
instagram.bind.InstagramClientError):
# 通常是没有访问权限
print public_bz.getExpInfoAll()
public_db.delNoName('instagram', user_name)
return
db_user = storage()
db_user.id_str = user.id
db_user.username = user.username
db_user.full_name = user.full_name
db_user.profile_picture = user.profile_picture
db_user.bio = user.bio
db_user.website = user.website
db_user.counts = json.dumps(user.counts)
#pg.insert('instagram_user', **db_user)
pg.insertOrUpdate(pg, 'instagram_user', db_user,
"id_str='%s'" % db_user.id_str)
#db_bz.insertIfNotExist(pg, 'instagram_user', db_user, "id=%s" % db_user.id)
return getUser(user_name)
def getUser(user_name, always_check=False):
'''
modify by bigzhu at 15/07/31 16:41:16 api找用户时是按专注度来排序的,名字绝对匹配的未必是第一位, 从10个里面找
'''
users = list(pg.select('instagram_user', where="lower(username)=lower('%s')" % user_name))
if users and not always_check:
return users[0]
else:
user = None
try:
for this_user in api.user_search(user_name, 10):
if this_user.username.lower() == user_name.lower():
user = this_user
break
except instagram.bind.InstagramClientError:
print public_bz.getExpInfoAll()
return
# 如果没有这个用户
if user is None:
public_db.delNoName('instagram', user_name)
return
try:
user = api.user(user.id)
except (instagram.bind.InstagramAPIError, instagram.bind.InstagramClientError):
# 通常是没有访问权限
print public_bz.getExpInfoAll()
public_db.delNoName('instagram', user_name)
return
db_user = storage()
db_user.id_str = user.id
db_user.username = user.username
db_user.full_name = user.full_name
db_user.profile_picture = user.profile_picture
db_user.bio = user.bio
db_user.website = user.website
db_user.counts = json.dumps(user.counts)
#pg.insert('instagram_user', **db_user)
pg.insertOrUpdate(pg, 'instagram_user', db_user, "id_str='%s'" % db_user.id_str)
#db_bz.insertIfNotExist(pg, 'instagram_user', db_user, "id=%s" % db_user.id)
return getUser(user_name)
def saveUser(god_name, github_name, user, sync_key=None):
social_user = public_bz.storage()
try:
# social_user.name = user['login']
social_user.name = github_name
except Exception as e:
print e
print user
social_user.type = 'github'
if user.get('followers') is None:
social_user.count = -1
else:
social_user.count = user['followers']
social_user.avatar = user.get('avatar_url', '')
social_user.description = user.get('bio')
if sync_key is not None:
social_user.sync_key = sync_key
pg.update('god', where={'name': god_name}, github=json.dumps(social_user))
return social_user
def get(self, parm=None):
starttime = datetime.datetime.now()
self.set_header("Content-Type", "application/json")
after = None
limit = None
search_key = None
god_name = None
if parm:
parm = json.loads(parm)
after = parm.get('after') # 晚于这个时间的
limit = parm.get('limit')
search_key = parm.get('search_key')
god_name = parm.get('god_name') # 只查这个god
user_id = self.current_user
if after:
after = time_bz.timestampToDateTime(after, True)
elif search_key is None and god_name is None: # 这些条件查询不能卡上次看到那条的时间
after = last_oper.getLastTime(user_id)
messages = public_db.getNewMessages(user_id=user_id,
after=after,
limit=limit,
god_name=god_name,
search_key=search_key)
data = storage()
data.error = OK
data.messages = messages
data.unread_message_count = oper.getUnreadCount(user_id)
if (len(messages) == 0):
if (user_id):
data.followed_god_count = god_oper.getFollowedGodCount(user_id)
else:
data.followed_god_count = 0
endtime = datetime.datetime.now()
print((endtime - starttime).seconds)
self.write(json.dumps(data, cls=json_bz.ExtEncoder))
def saveUser(id, url):
'''
create by bigzhu at 15/07/15 21:27:19 保存github信息
create by bigzhu at 15/07/22 16:17:37 fix bug, not return id
'''
if list(pg.select('github_user', where='id=%s' % id)):
return id
else:
r = requests.get(url)
user = storage(r.json())
del user.url
del user.followers_url
del user.following_url
del user.gists_url
del user.starred_url
del user.subscriptions_url
del user.organizations_url
del user.repos_url
del user.events_url
del user.received_events_url
del user.type
return db_bz.insertIfNotExist(pg, 'github_user', user)
def saveUser(id, url):
'''
create by bigzhu at 15/07/15 21:27:19 保存github信息
create by bigzhu at 15/07/22 16:17:37 fix bug, not return id
'''
if list(pg.select('github_user', where='id=%s' % id)):
return id
else:
r = requests.get(url)
user = storage(r.json())
print user
if hasattr(user, 'url'):
del user.url
del user.followers_url
del user.following_url
del user.gists_url
del user.starred_url
del user.subscriptions_url
del user.organizations_url
del user.repos_url
del user.events_url
del user.received_events_url
del user.type
return db_bz.insertIfNotExist(pg, 'github_user', user)
def getHtmlByName(name):
the_html = storage()
the_html.name = name
the_html.time = getModifyTime(name)
the_html.content = getHtmlContent(name)
return the_html
def saveMedias(user, medias):
'''
create by bigzhu at 15/09/04 20:58:54 保存meedias
"attribution":null,
"tags":[ ],
"type":"image",
"location":{ },
"comments":{ },
"filter":"Normal",
"created_time":"1441362020",
"link":"https:\/\/instagram.com\/p\/7NIHiLJJs3\/",
"likes":{ },
"images":{ },
"users_in_photo":[ ],
"caption":{ },
"user_has_liked":false,
"id":"1066544388859271991_262341",
"user":{ }
'''
for media_d in medias['data']:
media = storage(media_d)
db_media = storage()
#db_media.attribution = media.attribution
#db_media.tags = json.dumps(media.tags, cls=public_bz.ExtEncoder)
db_media.type = media.type
#db_media.location = json.dumps(media.location, cls=public_bz.ExtEncoder)
db_media.comments = json.dumps(media.comments, cls=public_bz.ExtEncoder)
db_media.filter = media.filter
#db_media.created_time = time_bz.timestampToDateTime(media.created_time) + timedelta(hours=8)
db_media.created_time = time_bz.timestampToDateTime(media.created_time)
db_media.link = media.link
#db_media.likes = json.dumps(media.likes, cls=public_bz.ExtEncoder)
db_media.low_resolution = json.dumps(media.images['low_resolution'])
db_media.standard_resolution = json.dumps(media.images['standard_resolution'])
db_media.thumbnail = json.dumps(media.images['thumbnail'])
#db_media.users_in_photo = json.dumps(media.users_in_photo, cls=public_bz.ExtEncoder)
if media.caption:
caption = media.caption
caption['user_id'] = caption['from']['id']
del caption['from']
else:
caption = ''
db_media.caption = json.dumps(caption, cls=public_bz.ExtEncoder)
db_media.id_str = media.id
db_media.user_id = user.id
id = pg.insertIfNotExist(pg, 'instagram_media', db_media, "id_str='%s'" % db_media.id_str)
if id is None:
raise Exception('重复记录 id=%s, name=%s' % (media.id, user.username))
else:
print 'new=', media.id, user.username
if id is not None and len(medias) <= 2: # 新增加消息,微信通知只通知2条以内
openids = public_db.getOpenidsByName('instagram', user.username)
for data in openids:
if caption != '':
text = caption.get('text')
else:
text = ''
wechat_oper.sendInstagram(data.openid, text, media.images['low_resolution']['url'], user.username, id)
if medias['pagination']:
next_url = medias['pagination']['next_url']
medias = callGetMeidaApi(next_url=next_url)
saveMedias(user, medias)
def getMedia(user_name=None, with_next_url=None, user=None):
if user_name:
user = getUser(user_name)
if user is None:
return
# min_id 会查出大于等于这个id的
try:
medias, next_ = api.user_recent_media(user_id=user.id, min_id=user.last_id)
except instagram.bind.InstagramClientError:
print public_bz.getExpInfoAll()
public_db.delNoName('instagram', user_name)
return
if medias:
last_id = medias[0].id
pg.update('instagram_user', where="lower(username)=lower('%s')" % user_name, last_id=last_id)
else:
medias, next_ = api.user_recent_media(with_next_url=with_next_url)
for media in medias:
db_media = storage()
if media.caption:
caption = media.caption.__dict__
caption['user_id'] = caption['user'].id
del caption['user']
else:
caption = ''
db_media.caption = json.dumps(caption, cls=public_bz.ExtEncoder)
db_media.comment_count = media.comment_count
if media.comments:
media.comments = [d.__dict__ for d in media.comments]
for comment in media.comments:
comment['user'] = comment['user'].__dict__
db_media.comments = json.dumps(media.comments, cls=public_bz.ExtEncoder)
db_media.created_time = media.created_time
# 8小时的问题
db_media.created_time += timedelta(hours=8)
db_media.filter = media.filter
db_media.low_resolution = json.dumps(media.images['low_resolution'].__dict__)
db_media.standard_resolution = json.dumps(media.images['standard_resolution'].__dict__)
db_media.thumbnail = json.dumps(media.images['thumbnail'].__dict__)
db_media.id_str = media.id
db_media.like_count = media.like_count
# likes里有User对象,暂时不存了
#db_media.likes = json.dumps(media.likes)
db_media.link = media.link
db_media.type = media.type
db_media.user_id = user.id
id = db_bz.insertIfNotExist(pg, 'instagram_media', db_media, "id_str='%s'" % db_media.id_str)
print 'new=', media.id, user.username
if id is not None and len(medias) <= 2: # 新增加消息,微信通知只通知2条以内
openids = public_db.getOpenidsByName('instagram', user.username)
for data in openids:
if caption != '':
text = caption.get('text')
else:
text = ''
wechat_oper.sendInstagram(data.openid, text, media.images['standard_resolution'].url, user.username, id)
# 递归查出
if next_ != with_next_url:
getMedia(with_next_url=next_, user=user)
def getEmptyUserInfo(self):
user_info = public_bz.storage()
for p in model_oper_bz.getModelAttributes(model_bz.user_info):
user_info[p] = ""
return user_info
def saveMedias(user, medias):
'''
create by bigzhu at 15/09/04 20:58:54 保存meedias
"attribution":null,
"tags":[ ],
"type":"image",
"location":{ },
"comments":{ },
"filter":"Normal",
"created_time":"1441362020",
"link":"https:\/\/instagram.com\/p\/7NIHiLJJs3\/",
"likes":{ },
"images":{ },
"users_in_photo":[ ],
"caption":{ },
"user_has_liked":false,
"id":"1066544388859271991_262341",
"user":{ }
'''
for media_d in medias['data']:
media = storage(media_d)
db_media = storage()
#db_media.attribution = media.attribution
#db_media.tags = json.dumps(media.tags, cls=public_bz.ExtEncoder)
db_media.type = media.type
#db_media.location = json.dumps(media.location, cls=public_bz.ExtEncoder)
db_media.comments = json.dumps(media.comments,
cls=public_bz.ExtEncoder)
db_media.filter = media.filter
#db_media.created_time = time_bz.timestampToDateTime(media.created_time) + timedelta(hours=8)
db_media.created_time = time_bz.timestampToDateTime(media.created_time)
db_media.link = media.link
#db_media.likes = json.dumps(media.likes, cls=public_bz.ExtEncoder)
db_media.low_resolution = json.dumps(media.images['low_resolution'])
db_media.standard_resolution = json.dumps(
media.images['standard_resolution'])
db_media.thumbnail = json.dumps(media.images['thumbnail'])
#db_media.users_in_photo = json.dumps(media.users_in_photo, cls=public_bz.ExtEncoder)
if media.caption:
caption = media.caption
caption['user_id'] = caption['from']['id']
del caption['from']
else:
caption = ''
db_media.caption = json.dumps(caption, cls=public_bz.ExtEncoder)
db_media.id_str = media.id
db_media.user_id_str = user.id
#id = pg.insertIfNotExist(pg, 'instagram_media', db_media, "id_str='%s'" % db_media.id_str)
m = public_bz.storage()
m.id_str = db_media.id_str
m.m_type = 'instagram'
m.m_user_id = db_media.user_id_str
m.created_at = db_media.created_time
m.extended_entities = db_media.standard_resolution
m.content = db_media.comments
if media.caption:
m.text = media.caption['text']
else:
m.text = None
m.href = db_media.link
m.type = db_media.type
id = pg.insertIfNotExist(pg, 'm', m, "id_str='%s'" % db_media.id_str)
if id is None:
# 似乎就是会有重复的
error = '重复记录 id=%s, name=%s' % (media.id, user.username)
print error
#raise Exception(error)
else:
print 'new=', media.id, user.username
# if id is not None and len(medias) <= 2: # 新增加消息,微信通知只通知2条以内
# openids = public_db.getOpenidsByName('instagram', user.username)
# for data in openids:
# if caption != '':
# text = caption.get('text')
# else:
# text = ''
# wechat_oper.sendInstagram(data.openid, text, media.images['low_resolution']['url'], user.username, id)
if medias['pagination']:
next_url = medias['pagination']['next_url']
medias = callGetMeidaApi(next_url=next_url)
saveMedias(user, medias)
def saveTwitter(tweet):
'''
create by bigzhu at 15/07/10 14:39:48
保存twitter
'''
del tweet.user._json
#del tweet.user._api
tweet.user.entities = json.dumps(tweet.user.entities)
del tweet.user.id
pg.insertOrUpdate(pg, 'twitter_user', vars(tweet.user), "id_str='%s'" % tweet.user.id_str)
tweet.t_user_id = tweet.user.id_str
screen_name = tweet.user.screen_name
del tweet.user
if hasattr(tweet, 'author'):
#del tweet.author.id
pg.insertOrUpdate(pg, 'twitter_user', vars(tweet.author), "id_str='%s'" % tweet.author.id_str)
tweet.t_author_id = tweet.author.id_str
del tweet.author
if hasattr(tweet, '_api'):
del tweet._api
if hasattr(tweet, '_json'):
del tweet._json
# twitter id 太大了 "id": 618948810941673472 导致 psycopg2.DataError: integer out of range
if hasattr(tweet, 'id'):
del tweet.id
if hasattr(tweet, 'entities'):
tweet.entities = json.dumps(tweet.entities)
if hasattr(tweet, 'geo'):
tweet.geo = json.dumps(tweet.geo)
if hasattr(tweet, 'coordinates'):
tweet.coordinates = json.dumps(tweet.coordinates)
if hasattr(tweet, 'extended_entities'):
tweet.extended_entities = json.dumps(tweet.extended_entities)
if hasattr(tweet, 'scopes'):
tweet.scopes = json.dumps(tweet.scopes)
if hasattr(tweet, 'retweeted_status'):
saveTwitter(tweet.retweeted_status)
tweet.retweeted_status = tweet.retweeted_status.id_str
if hasattr(tweet, 'quoted_status'):
# print tweet.quoted_status
del tweet.quoted_status
# saveTwitter(tweet.quoted_status)
# tweet.quoted_status = tweet.quoted_status.id_str
# place 是一个对象(我不知道如何处理): Place(_api=<tweepy.api.API object at 0x1808050>
if hasattr(tweet, 'place'):
del tweet.place
# for k, v in vars(tweet).items():
# print '%s=%s' % (k, v)
#return pg.insertIfNotExist(pg, 'twitter_message', vars(tweet), "id_str='%s'" % tweet.id_str)
m = public_bz.storage()
m.id_str = tweet.id_str
m.m_type = 'twitter'
m.m_user_id = tweet.t_user_id
m.created_at = tweet.created_at
if hasattr(tweet, 'extended_entities'):
m.extended_entities = tweet.extended_entities
m.content = None
m.text = tweet.text
m.href = 'https://twitter.com/'+screen_name+'/status/'+tweet.id_str
return pg.insertIfNotExist(pg, 'm', m, "id_str='%s' and m_type='twitter'" % tweet.id_str)
def saveTwitter(tweet):
'''
create by bigzhu at 15/07/10 14:39:48
保存twitter
'''
del tweet.user._json
#del tweet.user._api
tweet.user.entities = json.dumps(tweet.user.entities)
del tweet.user.id
pg.insertOrUpdate(pg, 'twitter_user', vars(tweet.user),
"id_str='%s'" % tweet.user.id_str)
tweet.t_user_id = tweet.user.id_str
screen_name = tweet.user.screen_name
del tweet.user
if hasattr(tweet, 'author'):
#del tweet.author.id
pg.insertOrUpdate(pg, 'twitter_user', vars(tweet.author),
"id_str='%s'" % tweet.author.id_str)
tweet.t_author_id = tweet.author.id_str
del tweet.author
if hasattr(tweet, '_api'):
del tweet._api
if hasattr(tweet, '_json'):
del tweet._json
# twitter id 太大了 "id": 618948810941673472 导致 psycopg2.DataError: integer out of range
if hasattr(tweet, 'id'):
del tweet.id
if hasattr(tweet, 'entities'):
tweet.entities = json.dumps(tweet.entities)
if hasattr(tweet, 'geo'):
tweet.geo = json.dumps(tweet.geo)
if hasattr(tweet, 'coordinates'):
tweet.coordinates = json.dumps(tweet.coordinates)
if hasattr(tweet, 'extended_entities'):
tweet.extended_entities = json.dumps(tweet.extended_entities)
if hasattr(tweet, 'scopes'):
tweet.scopes = json.dumps(tweet.scopes)
if hasattr(tweet, 'retweeted_status'):
saveTwitter(tweet.retweeted_status)
tweet.retweeted_status = tweet.retweeted_status.id_str
if hasattr(tweet, 'quoted_status'):
# print tweet.quoted_status
del tweet.quoted_status
# saveTwitter(tweet.quoted_status)
# tweet.quoted_status = tweet.quoted_status.id_str
# place 是一个对象(我不知道如何处理): Place(_api=<tweepy.api.API object at 0x1808050>
if hasattr(tweet, 'place'):
del tweet.place
# for k, v in vars(tweet).items():
# print '%s=%s' % (k, v)
#return pg.insertIfNotExist(pg, 'twitter_message', vars(tweet), "id_str='%s'" % tweet.id_str)
m = public_bz.storage()
m.id_str = tweet.id_str
m.m_type = 'twitter'
m.m_user_id = tweet.t_user_id
m.created_at = tweet.created_at
if hasattr(tweet, 'extended_entities'):
m.extended_entities = tweet.extended_entities
m.content = None
m.text = tweet.text
m.href = 'https://twitter.com/' + screen_name + '/status/' + tweet.id_str
return pg.insertIfNotExist(
pg, 'm', m, "id_str='%s' and m_type='twitter'" % tweet.id_str)
