def get_user_info(uid):
user_info = {}
result = user_search.search_by_id(int(uid), fields=user_fields_list)
if result:
user_info['name'] = result['name']
user_info['location'] = result['location']
#user_info['gender'] = result['gender']
user_info['friends_count'] = result['friends_count']
user_info['followers_count'] = result['followers_count']
user_info['profile_image_url'] = result['profile_image_url']
user_info['friends_count'] = result['friends_count']
user_info['followers_count'] = result['followers_count']
user_info['created_at'] = result['created_at']
user_info['statuses_count'] = result['statuses_count']
else:
user_info['name'] = u'未知'
user_info['location'] = u'未知'
user_info['friends_count'] = u'未知'
user_info['followers_count'] = u'未知'
user_info['profile_image_url'] = 'no'
user_info['friends_count'] = u'未知'
user_info['followers_count'] = u'未知'
user_info['created_at'] = u'未知'
user_info['statuses_count'] = u'未知'
return user_info
def acquire_user_by_id(uid):
user_result = user_search.search_by_id(uid, fields=['name', 'profile_image_url'])
user = {}
if user_result:
user['name'] = user_result['name']
user['image'] = user_result['profile_image_url']
return user
def get_user_info(uid):
user_info = {}
result = user_search.search_by_id(int(uid), fields=user_fields_list)
if result:
user_info['name'] = result['name']
user_info['location'] = result['location']
#user_info['gender'] = result['gender']
user_info['friends_count'] = result['friends_count']
user_info['followers_count'] = result['followers_count']
user_info['profile_image_url'] = result['profile_image_url']
user_info['friends_count'] = result['friends_count']
user_info['followers_count'] = result['followers_count']
user_info['created_at'] = result['created_at']
user_info['statuses_count'] = result['statuses_count']
else:
user_info['name'] = u'未知'
user_info['location'] = u'未知'
user_info['friends_count'] = u'未知'
user_info['followers_count'] = u'未知'
user_info['profile_image_url'] = 'no'
user_info['friends_count'] = u'未知'
user_info['followers_count'] = u'未知'
user_info['created_at'] = u'未知'
user_info['statuses_count'] = u'未知'
return user_info
def get_province(uid):
results = user_search.search_by_id(uid, fields=['province'])
if results:
province = results['province']
else:
province = None
return province
def getUsernameByUid(uid):
if not uid:
return None
user = xapian_search_user.search_by_id(int(uid), fields=['name'])
if user:
name = user['name']
return name
return None
def acquire_user_by_id_v2(uid):
result = user_search.search_by_id(
int(uid),
fields=['name', 'location', 'followers_count', 'friends_count'])
user = {}
if result:
user['name'] = result['name']
user['location'] = result['location']
user['count1'] = result['followers_count']
user['count2'] = result['friends_count']
return user
def read_uid_weibos(topic, date, windowsize, uid):
# change
end_ts = datetime2ts(date)
start_ts = end_ts - Day * windowsize
xapian_search_weibo = getXapianWeiboByTopic(topic, start_ts, end_ts)
query_dict = {'user': uid}
count, results = xapian_search_weibo.search(query=query_dict,
fields=weibo_fields_list)
if count == 0:
weibo_list = []
else:
weibo_list = []
for weibo in results():
wid = weibo['_id']
uid = weibo['user']
result = user_search.search_by_id(uid, fields=user_fields_list)
if result:
name = result['name']
location = result['location']
friends_count = result['friends_count']
followers_count = result['followers_count']
created_at = result['created_at']
statuses_count = result['statuses_count']
profile_image_url = result['profile_image_url']
else:
name = u'未知'
location = u'未知'
friends_count = u'未知'
followers_count = u'未知'
created_at = u'未知'
statuses_count = u'未知'
profile_image_url = u'no'
text = weibo['text']
geo = weibo['geo']
source = weibo['source']
timestamp = weibo['timestamp']
date = ts2date(timestamp)
reposts_count = weibo['reposts_count']
comments_count = weibo['comments_count']
weibo_link = weiboinfo2url(uid, wid)
domain = uid2domain(uid)
row = [
wid, uid, name, location, friends_count, followers_count,
created_at, statuses_count, profile_image_url, date, text, geo,
source, reposts_count, comments_count, weibo_link
]
weibo_list.append(row)
sort_weibo_list = sorted(weibo_list, key=lambda x: x[9])
return sort_weibo_list
def get_impor_media():
reader = csv.reader(file('media.csv', 'rb'))
media_dict = {} # media_dict = {uid1:followers_count1}
for line in reader:
uid = line[0]
results = user_search.search_by_id(int(uid), fields=['followers_count'])
if results:
followers_count = results['followers_count']
media_dict[uid] = followers_count
else:
continue
sort_media = sorted(media_dict.iteritems(), key=lambda a:a[1], reverse=False)
topmedia = sort_media[len(sort_media)-500:] # topmedia = [(uid1, followers1),(uid2, followers2)]
write_impor_media(topmedia)
def get_impor_media():
reader = csv.reader(file('media.csv', 'rb'))
media_dict = {} # media_dict = {uid1:followers_count1}
for line in reader:
uid = line[0]
results = user_search.search_by_id(int(uid),
fields=['followers_count'])
if results:
followers_count = results['followers_count']
media_dict[uid] = followers_count
else:
continue
sort_media = sorted(media_dict.iteritems(),
key=lambda a: a[1],
reverse=False)
topmedia = sort_media[
len(sort_media) -
500:] # topmedia = [(uid1, followers1),(uid2, followers2)]
write_impor_media(topmedia)
def get_u_info(uid_list):
user_info_list = []
row = []
for uid in uid_list:
user = user_search.search_by_id(
uid, fields=['_id', 'name', 'profile_image_url', 'friends_count'])
if user:
name = user['name']
profile_image_user = user['profile_image_url']
friends_count = user['friends_count']
else:
name = u'未知'
profile_image_user = u'no'
friends_count = -1
row.append([uid, name, profile_image_user, friends_count])
sort_row = sorted(row, key=lambda x: x[3], reverse=True)
user_info_list = [user[:3] for user in sort_row]
return user_info_list
def community_result(community_user_list, topic, date, windowsize):
#change
end_ts = datetime2ts(date)
start_ts = end_ts - windowsize * Day
xapian_search_weibo = getXapianWeiboByTopic(topic, start_ts, end_ts)
query_dict = {'$or': []}
for uid in community_user_list:
query_dict['$or'].append({'user': int(uid)})
community_info = []
count, weibo_results = xapian_search_weibo.search(query=query_dict,
fields=weibo_fields_list)
if count == 0:
return None, None, None
sentiment_count = {}
for weibo in weibo_results():
uid = weibo['user']
_id = weibo['_id']
result = user_search.search_by_id(uid, fields=user_fields_list)
if result:
name = result['name']
location = result['location']
friends_count = result['friends_count']
followers_count = result['followers_count']
created_at = result['created_at']
statuses_count = result['statuses_count']
profile_image_url = result['profile_image_url']
else:
name = u'未知'
location = u'未知'
friends_count = u'未知'
followers_count = u'未知'
created_at = u'未知'
statuses_count = u'未知'
profile_image_url = u'no'
text = weibo['text']
timestamp = weibo['timestamp']
date = ts2date(timestamp)
reposts_count = weibo['reposts_count']
source = weibo['source']
geo = weibo['geo']
comments_count = weibo['comments_count']
sentiment = weibo['sentiment']
sentiment_name = emotions_kv[sentiment]
weibo_link = weiboinfo2url(uid, _id)
domain = uid2domain(uid)
try:
sentiment_count[sentiment] += 1
except KeyError:
sentiment_count[sentiment] = 1
community_info.append([
_id, name, location, friends_count, followers_count, created_at,
statuses_count, profile_image_url, text, date, reposts_count,
source, geo, comments_count, sentiment_name, weibo_link, domain
])
sort_community_info = sorted(community_info,
key=lambda x: x[10],
reverse=True) #以转发量排序
mset = xapian_search_weibo.search(query=query_dict,
max_offset=50,
mset_direct=True)
top_keyword = top_keywords(gen_mset_iter(xapian_search_weibo,
mset,
fields=['terms']),
top=50)
sort_top_keyword = sorted(top_keyword, key=lambda x: x[1], reverse=True)
new_sentiment_list = []
for sentiment in sentiment_count:
sentiment_ch = emotions_kv[int(sentiment)]
num = sentiment_count[sentiment]
ratio = float(num) / float(count)
new_sentiment_list.append([sentiment_ch, num, ratio])
return sort_community_info, sort_top_keyword, new_sentiment_list, query_dict
def c_weibo_by_ts(topic, date, windowsize, uid, network_type, cid, rank_method):
real_topic_id = acquire_real_topic_id(topic, date, windowsize)
if not real_topic_id:
return None, None, None
# 该话题存在进行下面的计算
key_pre = str(real_topic_id) + '_' + str(date) + '_' + str(windowsize)
# 选择有向图进行社区信息的计算
if network_type=='source_graph':
key = str(GRAPH_PATH)+key_pre + '_gg_graph.gexf'
elif network_type=='direct_superior_graph':
key = str(GRAPH_PATH)+key_pre + '_ds_udg_graph.gexf'
g = nx.read_gexf(key)
# 获取图结构中节点uid对应的社区包括的节点list
community_user_list = get_community_user(g, uid, cid)
# 考虑节点社区属性存放的位置
# change
end_ts = datetime2ts(date)
start_ts = end_ts - Day * windowsize
xapian_search_weibo = getXapianWeiboByTopic(topic, start_ts, end_ts)
query_dict = {
'$or' : []
}
for uid in community_user_list:
query_dict['$or'].append({'user': int(uid)})
community_info = []
count, weibo_results = xapian_search_weibo.search(query=query_dict, fields= weibo_fields_list)
if count==0:
return None
for weibo in weibo_results():
uid = weibo['user']
_id = weibo['_id']
result = user_search.search_by_id(uid, fields=user_fields_list)
if result:
name = result['name']
location = result['location']
friends_count = result['friends_count']
followers_count = result['followers_count']
created_at = result['created_at']
statuses_count = result['statuses_count']
profile_image_url = result['profile_image_url']
else:
name = u'未知'
location = u'未知'
friends_count = u'未知'
followers_count = u'未知'
created_at = u'未知'
statuses_count = u'未知'
profile_image_url = u'no'
text = weibo['text']
timestamp = weibo['timestamp']
date = ts2date(timestamp)
reposts_count = weibo['reposts_count']
source = weibo['source']
geo = weibo['geo']
comments_count = weibo['comments_count']
sentiment = weibo['sentiment']
sentiment_name = emotions_kv[sentiment]
weibo_link = weiboinfo2url(uid, _id)
community_info.append([_id, name, location, friends_count, followers_count, created_at, statuses_count, profile_image_url, text, date, reposts_count, source, geo, comments_count, sentiment_name,weibo_link, uid, timestamp])
if rank_method == 'reposts_count':
sort_community_info = sorted(community_info, key=lambda x:x[10], reverse=True) #以转发量排序
else:
sort_community_info = sorted(community_info, key=lambda x:x[17]) # 以时间戳排序
return sort_community_info