def func(*args, **kw):
data = method(*args, **kw)
method_name = method.__name__
rank_range = method_name.split('_')[0]
rank_field = method_name.split('_')[1]
save_rank_results(data, rank_range, rank_field, args[1], args[2])
return data
def func(*args, **kw):
data = method(*args, **kw)
method_name = method.__name__
rank_field = method_name.split('_')[0]
## try:
## print 'yuan'
save_rank_results(data, 'whole', rank_field, args[1], args[2])
## except:
## print '%s save results failed, it is ok if not in web environment.' % method.__name__
return data
def pagerank_rank(top_n, date, window_size, topicname):
data = []
#tmp_file, N, ds_tmp_file, ds_N = prepare_data_for_pr(topic_id, date, window_size, topicname, real_topic_id)
#print '888888888888888888888888888888'
print date, window_size, topicname
tmp_file, N= prepare_data_for_pr( date, window_size, topicname)
top_n = N
#ds_top_n = ds_N
print 'page_rank start'
if not tmp_file:
return data
input_tmp_path = tmp_file.name
#ds_input_tmp_path = ds_tmp_file.name
#print input_tmp_path
iter_count = PAGERANK_ITER_MAX
print 'pagerank_source_network'
sorted_uids, all_uid_pr = pagerank(iter_count, input_tmp_path, top_n) # 排序的uid的序列
#print 'pagerank_direct_superior_network'
#ds_sorted_uids, ds_all_uid_pr = pagerank(iter_count, ds_input_tmp_path, ds_top_n)
print 'top_n:', top_n
#print 'len(sorted_uid):', len(ds_sorted_uids)
#print 'len(ds_all_uid_pr):', len(ds_all_uid_pr)
#print 'ds_top_n:', ds_top_n
#topicname = acquire_topic_name(topic_id)
print 'topicname:', topicname
if not topicname:
return data
print 'save_rank_results'
data = save_rank_results(sorted_uids, 'topic', 'spark_pagerank', date, window_size, topicname, all_uid_pr)
#ds_data = save_ds_rank_results(ds_sorted_uids, 'topic', 'spark_pagerank', date, window_size, topicname, ds_all_uid_pr)
return all_uid_pr, data
def pagerank_rank(top_n, date, topic_id, window_size, topicname, real_topic_id):
data = []
tmp_file, N, ds_tmp_file, ds_N = prepare_data_for_pr(topic_id, date, window_size, topicname, real_topic_id)
top_n = N
ds_top_n = ds_N
print 'page_rank start'
if not tmp_file or not ds_tmp_file:
return data
input_tmp_path = tmp_file.name
ds_input_tmp_path = ds_tmp_file.name
print input_tmp_path, ds_input_tmp_path
iter_count = PAGERANK_ITER_MAX
print 'pagerank_source_network'
sorted_uids, all_uid_pr = pagerank(iter_count, input_tmp_path, top_n) # 排序的uid的序列
print 'pagerank_direct_superior_network'
ds_sorted_uids, ds_all_uid_pr = pagerank(iter_count, ds_input_tmp_path, ds_top_n)
print 'top_n:', top_n
print 'len(sorted_uid):', len(ds_sorted_uids)
print 'len(ds_all_uid_pr):', len(ds_all_uid_pr)
print 'ds_top_n:', ds_top_n
topicname = acquire_topic_name(topic_id)
print 'topicname:', topicname
if not topicname:
return data
print 'save_rank_results'
data = save_rank_results(sorted_uids, 'topic', 'spark_pagerank', date, window_size, topicname, all_uid_pr)
ds_data = save_ds_rank_results(ds_sorted_uids, 'topic', 'spark_pagerank', date, window_size, topicname, ds_all_uid_pr)
return all_uid_pr, ds_all_uid_pr, data, ds_data
def pagerank_rank(top_n, date, topic_id, window_size):
data = []
tmp_file = prepare_data_for_pr(topic_id, date, window_size)
if not tmp_file:
return data
input_tmp_path = tmp_file.name
job_id = generate_job_id(datetime2ts(date), window_size, topic_id)
iter_count = PAGERANK_ITER_MAX
sorted_uids = pagerank(job_id, iter_count, input_tmp_path, top_n)
data = save_rank_results(sorted_uids, 'area', 'pagerank', date, window_size, topic_id=topic_id)
return data
def degree_rank(top_n, date, topic_id, window_size):
data = []
degree = prepare_data_for_degree(topic_id, date, window_size)
if not degree:
return data
sorted_degree = sorted(degree.iteritems(), key=operator.itemgetter(1), reverse=True)
sorted_uids = []
count = 0
for uid, value in sorted_degree:
if count >= top_n:
break
sorted_uids.append(uid)
count += 1
data = save_rank_results(sorted_uids, 'area', 'degree', date, window_size, topic_id=topic_id)
return data
