def compute_group_inner(task_name, task_user, start_ts):
#step1: get task_user in-monitor task user retweet relation from monitor_inner_r
#step2: get task_user in-task user retweet relation
#step3: compute every inner user be-retweet ratio in task
#step4: save top5 to es--monitor_result, doc_type=task_name, _id='inner_'+date e:'inner_2013-09-01'
group_status = 0
time_segment = 3600 * 24
iter_time_segment = 900
iter_ts = start_ts - time_segment
inner_group_dict = {}
user_count_dict = {}
print 'group inner ask_user:'******'''
if iter_ts >= start_ts:
break
'''
key = 'inner_' + str(iter_ts)
print 'iter_ts:', ts2date(iter_ts)
inner_retweet_string = monitor_inner_r.hget(root_uid, key)
print 'root_uid, key, inner_retweet_string:', root_uid, key, inner_retweet_string
if inner_retweet_string:
print 'yes'
inner_retweet_dict = json.loads(inner_retweet_string)
else:
inner_retweet_dict = None
if inner_retweet_dict:
inner_group_dict[root_uid] = merge_dict(
inner_group_dict[root_uid], inner_retweet_dict)
iter_ts += iter_time_segment
user_inner_retweet_count = sum(inner_group_dict[root_uid].values())
user_count_dict[root_uid] = user_inner_retweet_count
all_be_retweet_count = sum(user_count_dict.values())
if all_be_retweet_count == 0:
group_status = 1
return group_status
sort_user_inner_retweet_count = sorted(user_count_dict.items(),
key=lambda x: x[1],
reverse=True)
top5_user = sort_user_inner_retweet_count[:5]
# timestamp: '2013-09-01'
date = ts2datetime(start_ts - 24 * 3600)
index_body = {'date': date}
for rank in range(1, 6):
key = 'top' + str(rank)
index_body[key] = json.dumps(top5_user[rank - 1])
key = 'inner_' + date
# save inner-retweet graph by dict {root_uid1:{uid1:count1, uid2:count2}, ...}
index_body['inner_graph'] = json.dumps(inner_group_dict)
es.index(index=monitor_index_name,
doc_type=task_name,
id=key,
body=index_body)
group_status = 1
return group_status
def inner_group_retweet(item):
root_uid = str(item['root_uid'])
uid = str(item['uid'])
timestamp = item['timestamp']
date = ts2datetime(timestamp)
date_ts = datetime2ts(date)
time_segment = int((timestamp - date_ts) / 900)
start_ts = date_ts + time_segment * 900
key = 'inner_' + str(start_ts)
inner_retweet_exist = monitor_inner_r.hget(root_uid, key)
if not inner_retweet_exist:
monitor_inner_r.hset(root_uid, key, json.dumps({uid: 1}))
else:
inner_retweet_dict = json.loads(inner_retweet_exist)
if uid in inner_retweet_dict:
inner_retweet_dict[uid] += 1
else:
inner_retweet_dict[uid] = 1
monitor_inner_r.hset(root_uid, key, json.dumps(inner_retweet_dict))
def inner_group_retweet(item):
root_uid = str(item['root_uid'])
uid = str(item['uid'])
timestamp = item['timestamp']
date = ts2datetime(timestamp)
date_ts = datetime2ts(date)
time_segment = int((timestamp - date_ts) / 900)
start_ts = date_ts + time_segment * 900
key = 'inner_' + str(start_ts)
inner_retweet_exist = monitor_inner_r.hget(root_uid, key)
if not inner_retweet_exist:
monitor_inner_r.hset(root_uid, key, json.dumps({uid: 1}))
else:
inner_retweet_dict = json.loads(inner_retweet_exist)
if uid in inner_retweet_dict:
inner_retweet_dict[uid] += 1
else:
inner_retweet_dict[uid] = 1
monitor_inner_r.hset(root_uid, key, json.dumps(inner_retweet_dict))
def compute_group_inner(task_name, task_user, start_ts):
#step1: get task_user in-monitor task user retweet relation from monitor_inner_r
#step2: get task_user in-task user retweet relation
#step3: compute every inner user be-retweet ratio in task
#step4: save top5 to es--monitor_result, doc_type=task_name, _id='inner_'+date e:'inner_2013-09-01'
group_status = 0
time_segment = 3600*24
iter_time_segment = 900
iter_ts = start_ts - time_segment
inner_group_dict = {}
user_count_dict = {}
print 'group inner ask_user:'******'''
if iter_ts >= start_ts:
break
'''
key = 'inner_' + str(iter_ts)
print 'iter_ts:', ts2date(iter_ts)
inner_retweet_string = monitor_inner_r.hget(root_uid, key)
print 'root_uid, key, inner_retweet_string:', root_uid, key, inner_retweet_string
if inner_retweet_string:
print 'yes'
inner_retweet_dict = json.loads(inner_retweet_string)
else:
inner_retweet_dict = None
if inner_retweet_dict:
inner_group_dict[root_uid] = merge_dict(inner_group_dict[root_uid], inner_retweet_dict)
iter_ts += iter_time_segment
user_inner_retweet_count = sum(inner_group_dict[root_uid].values())
user_count_dict[root_uid] = user_inner_retweet_count
all_be_retweet_count = sum(user_count_dict.values())
if all_be_retweet_count==0:
group_status = 1
return group_status
sort_user_inner_retweet_count = sorted(user_count_dict.items(), key=lambda x:x[1], reverse=True)
top5_user = sort_user_inner_retweet_count[:5]
# timestamp: '2013-09-01'
date = ts2datetime(start_ts - 24*3600)
index_body = {'date': date}
for rank in range(1,6):
key = 'top' + str(rank)
index_body[key] = json.dumps(top5_user[rank-1])
key = 'inner_' + date
# save inner-retweet graph by dict {root_uid1:{uid1:count1, uid2:count2}, ...}
index_body['inner_graph'] = json.dumps(inner_group_dict)
es.index(index=monitor_index_name, doc_type=task_name, id=key, body=index_body)
group_status = 1
return group_status
