def do_statistics_member_quantity(cache_key, city_code_list, choice_time):
"""开始统计
:param cache_key:
:param city_code_list:
:param choice_time
:return:
"""
RedisCache.set(cache_key, KEY_CACHE_REPORT_DOING_NOW, 5 * 60)
stage_list = []
if city_code_list:
stage_list.append(MatchStage({'city_code': {'$in': city_code_list}}))
if not choice_time:
# 取前一天凌晨12点之前的数据
yesterday_time = get_yesterday()
time_match = MatchStage({'updated_dt': {'$lt': yesterday_time}})
else:
# 当天下一天凌晨的时候
max_choice_time = choice_time.replace(hour=23, minute=59, second=59, microsecond=999)
time_match = MatchStage({'updated_dt': {'$gte': choice_time, '$lt': max_choice_time}})
stage_list.append(time_match)
stage_list.append(MatchStage({'status': STATUS_USER_ACTIVE}))
group_stage = GroupStage('province_code', quantity={'$sum': 1})
lookup_stage = LookupStage(AdministrativeDivision, '_id', 'post_code', 'ad_list')
sort_stage = SortStage([('quantity', DESC)])
stage_list += [group_stage, lookup_stage, sort_stage]
province_cursor = Member.sync_aggregate(stage_list)
province_dict = {}
while True:
try:
province_stat = province_cursor.next()
if province_stat:
province_code = province_stat.id if province_stat.id else '000000'
quantity = province_stat.quantity
title = 'undefined'
ad_list = province_stat.ad_list
if ad_list:
ad: FacadeO = ad_list[0]
if ad:
title = ad.title.replace('省', '').replace('市', '')
province_dict[province_code] = {
'code': province_code,
'title': title,
'data': quantity
}
except StopIteration:
break
# 合并城市统计信息
do_merge_city_stat_member_quantity(province_dict, choice_time, city_code_list)
data = [v for v in province_dict.values()]
if not data:
early_warning_empty("start_statistics_member_quantity", cache_key, city_code_list, '学习近况中人数数据为空，请检查！')
RedisCache.set(cache_key, msgpack.packb(data))
def do_statistics_subject_parameter(cache_key, m_province_code_list,
m_city_code_list, s_province_code_list,
s_city_code_list, s_gender_list,
s_age_group_list, s_education_list):
"""
:param cache_key:
:param m_province_code_list:
:param m_city_code_list:
:param s_province_code_list:
:param s_city_code_list:
:param s_gender_list:
:param s_age_group_list:
:param s_education_list:
:return:
"""
RedisCache.set(cache_key, KEY_CACHE_REPORT_DOING_NOW, 5 * 60)
data = {}
max_q = None
max_q_list = SubjectChoiceRules.sync_aggregate(
[GroupStage('max', max={'$max': '$quantity'})]).to_list(1)
if max_q_list:
max_q = max_q_list[0]
if max_q and max_q.max > 0:
stage_list = do_create_query(max_q.max + 1, m_province_code_list,
m_city_code_list, s_province_code_list,
s_city_code_list, s_gender_list,
s_age_group_list, s_education_list)
if stage_list:
if stage_list:
stat_result = None
stat_result_list = MemberDailyStatistics.sync_aggregate(
stage_list).to_list(1)
if stat_result_list:
stat_result = stat_result_list[0]
if stat_result:
for i in range(max_q.max + 1):
attr = str(i)
if hasattr(stat_result, attr):
data[attr] = getattr(stat_result, attr, 0)
if not data:
early_warning_empty(
"start_statistics_subject_quantity", cache_key,
str(
dict(cache_key=cache_key,
m_province_code_list=m_province_code_list,
m_city_code_list=m_city_code_list,
s_province_code_list=s_province_code_list,
s_city_code_list=s_city_code_list,
s_gender_list=s_gender_list,
s_age_group_list=s_age_group_list,
s_education_list=s_education_list)), '学习趋势统计数据为空，请检查！')
RedisCache.set(cache_key, msgpack.packb(data))
def do_statistics_member_time(cache_key, city_code_list, choice_time):
"""开始统计
:param cache_key:
:param city_code_list:
:param choice_time:
:return:
"""
RedisCache.set(cache_key, KEY_CACHE_REPORT_DOING_NOW, 5 * 60)
ad_map = {}
game_data, ad_map = do_stat_in_history(MemberGameHistory, city_code_list, choice_time, ad_map)
ckpt_data, ad_map = do_stat_in_history(MemberCheckPointHistory, city_code_list, choice_time, ad_map)
# 对学习之旅和科协答题历史记录进行数据整合
for k, city_dict in game_data.items():
if k not in ckpt_data:
ckpt_data[k] = city_dict
continue
# loop city_list
for c_name, c_data in city_dict.items():
try:
# try to merge
ckpt_data[k][c_name] += c_data
except KeyError:
ckpt_data[k][c_name] = c_data
ret_data = []
for prov_code, city_data in ckpt_data.items():
prov = ad_map.get(prov_code)
if not prov:
prov = AdministrativeDivision.sync_find_one({'code': prov_code, 'parent_code': None})
city_list = [{'title': _k, 'data': _v} for _k, _v in city_data.items()]
_ds = [_.get('data') for _ in city_list]
city_list.sort(key=lambda x: -x.get('data'))
ret_data.append(
{'title': prov.title.replace('省', '').replace('市', ''), 'data': sum(_ds), 'city_list': city_list})
ret_data.sort(key=lambda x: -x.get('data'))
if not ret_data:
early_warning_empty("start_statistics_member_time", cache_key, city_code_list, '学习近况中次数数据为空，请检查！')
RedisCache.set(cache_key, msgpack.packb(ret_data))
def do_statistics_subject_radar(cache_key, root_dimension_code,
m_city_code_list, province_code_list,
city_code_list, gender_list, age_group_list,
education_list):
"""
:param cache_key:
:param root_dimension_code:
:param m_city_code_list:
:param province_code_list:
:param city_code_list:
:param gender_list:
:param age_group_list:
:param education_list:
:return:
"""
RedisCache.set(cache_key, KEY_CACHE_REPORT_DOING_NOW, 5 * 60)
data = []
dimension = SubjectDimension.sync_find_one(
dict(code=root_dimension_code, status=STATUS_SUBJECT_DIMENSION_ACTIVE))
if not dimension:
raise ValueError(
'can not find dimension by `root_dimension_code`(%s)' %
root_dimension_code)
stage_list = []
# 取前一天凌晨12点之前的数据
time_match = get_yesterday()
stage_list.append(MatchStage({'updated_dt': {'$lt': time_match}}))
if m_city_code_list:
stage_list.append(MatchStage({'city_code': {'$in': m_city_code_list}}))
query_dict = {}
if province_code_list:
query_dict['province_code'] = {'$in': province_code_list}
if city_code_list:
query_dict['city_code'] = {'$in': city_code_list}
if gender_list:
query_dict['gender'] = {
'$in': [int(s_gender) for s_gender in gender_list]
}
if age_group_list:
query_dict['age_group'] = {
'$in': [int(s_age_group) for s_age_group in age_group_list]
}
if education_list:
query_dict['education'] = {
'$in': [int(s_education) for s_education in education_list]
}
if query_dict:
stage_list.append(MatchStage(query_dict))
stage_list.append(
GroupStage('dimension.%s' % dimension.cid,
total={'$sum': '$total'},
correct={'$sum': '$correct'}))
stage_list.append(
LookupStage(SubjectDimension, '_id', 'cid', 'dimension_list'))
stat_result = MemberSubjectStatistics.sync_aggregate(stage_list)
while True:
try:
mds = stat_result.next()
if mds:
code, title, ordered = '', '', 0
if hasattr(mds, 'dimension_list') and mds.dimension_list:
dimension = mds.dimension_list[0]
if dimension:
code = dimension.code
title = dimension.title
ordered = dimension.ordered
data.append(
dict(code=code,
title=title,
ordered=ordered,
correct=mds.correct,
total=mds.total))
except StopIteration:
break
if not data:
early_warning_empty("start_statistics_subject_parameter_radar",
cache_key, locals(), '获取维度正确率雷达图统计数据为空，请检查！')
RedisCache.set(cache_key, msgpack.packb(data))
def do_statistics_accuracy(cache_key, city_code_list, choice_time):
"""
学习状况-正确率
:param cache_key:
:param city_code_list:
:param choice_time
:return:
"""
RedisCache.set(cache_key, KEY_CACHE_REPORT_DOING_NOW)
# 取前一天凌晨12点之前的数据
time_match = get_yesterday()
if not choice_time:
match_stage = MatchStage({'updated_dt': {'$lt': time_match}})
else:
# 当天下一天凌晨的时候
max_choice_time = choice_time.replace(hour=23, minute=59, second=59, microsecond=999)
match_stage = MatchStage({'updated_dt': {'$gte': choice_time, '$lt': max_choice_time}})
stage_list = [match_stage]
if city_code_list:
stage_list.append(MatchStage({'city_code': {'$in': city_code_list}}))
group_stage = GroupStage('province_code', t_total={'$sum': '$total'}, t_correct={'$sum': '$correct'})
add_fields_stage = AddFieldsStage(t_accuracy={
'$cond':
{
'if': {'$eq': ['$t_total', 0]},
'then': 0,
'else':
{
'$divide': ['$t_correct', '$t_total']
}
}
})
sort_stage = SortStage([('t_accuracy', DESC)])
lookup_stage = LookupStage(AdministrativeDivision, '_id', 'post_code', 'ad_list')
stage_list.extend([group_stage, add_fields_stage, sort_stage, lookup_stage])
province_stat_list = MemberSubjectStatistics.sync_aggregate(stage_list)
province_dict = {}
while True:
try:
province_stat = province_stat_list.next()
if province_stat:
province_code = province_stat.id if province_stat.id else '000000'
total = province_stat.t_total if province_stat.t_total else 0
correct = province_stat.t_correct if province_stat.t_correct else 0
title = 'undefined'
ad_list = province_stat.ad_list
if ad_list:
ad: FacadeO = ad_list[0]
if ad:
title = ad.title.replace('省', '').replace('市', '')
province_dict[province_code] = {
'code': province_code,
'title': title,
'correct': correct,
'total': total,
'data': round(correct / total * 100 if total > 0 else 0, 2)
}
except StopIteration:
break
# 合并城市统计信息
do_merge_city_stat_accuracy(province_dict, city_code_list)
data = [v for v in province_dict.values()]
RedisCache.set(cache_key, msgpack.packb(data))
if not data:
early_warning_empty("start_statistics_member_accuracy", cache_key, city_code_list, '学习近况中正确率数据为空，请检查！')
return data
def do_statistics_subject_cross(cache_key, main_dimension_code, second_dimension_code, m_city_code_list,
province_code_list, city_code_list, gender_list, age_group_list,
education_list):
"""
:param cache_key:
:param main_dimension_code:
:param second_dimension_code:
:param m_city_code_list:
:param province_code_list:
:param city_code_list:
:param gender_list:
:param age_group_list:
:param education_list:
:return:
"""
RedisCache.set(cache_key, KEY_CACHE_REPORT_DOING_NOW, 5 * 60)
main_dimension = SubjectDimension.sync_find_one(
dict(code=main_dimension_code, status=STATUS_SUBJECT_DIMENSION_ACTIVE))
main_sub_dimension_list = SubjectDimension.sync_find(dict(parent_cid=main_dimension.cid)).sort(
[('ordered', ASC)]).to_list(None)
second_dimension = SubjectDimension.sync_find_one(
dict(code=second_dimension_code, status=STATUS_SUBJECT_DIMENSION_ACTIVE))
second_sub_dimension_list = SubjectDimension.sync_find(dict(parent_cid=second_dimension.cid)).sort(
[('ordered', ASC)]).to_list(None)
data = []
for index, m_dimen in enumerate(main_sub_dimension_list):
sub_data_list = []
for s_dimen in second_sub_dimension_list:
stage_list = []
# 取前一天凌晨12点之前的数据
time_match = get_yesterday()
stage_list.append(MatchStage({'updated_dt': {'$lt': time_match}}))
match_dict = {'dimension.%s' % main_dimension.cid: m_dimen.cid,
'dimension.%s' % second_dimension.cid: s_dimen.cid}
if m_city_code_list:
match_dict['city_code'] = {'$in': m_city_code_list}
stage_list.append(MatchStage(match_dict))
query_dict = {}
if province_code_list:
query_dict['province_code'] = {'$in': province_code_list}
if city_code_list:
query_dict['city_code'] = {'$in': city_code_list}
if gender_list:
query_dict['gender'] = {'$in': [int(s_gender) for s_gender in gender_list]}
if age_group_list:
query_dict['age_group'] = {'$in': [int(s_age_group) for s_age_group in age_group_list]}
if education_list:
query_dict['education'] = {'$in': [int(s_education) for s_education in education_list]}
if query_dict:
stage_list.append(MatchStage(query_dict))
# 分组
group_params = {
'total': {'$sum': '$total'},
'correct': {'$sum': '$correct'}
}
stage_list.append(GroupStage(None, **group_params))
stat_result = MemberSubjectStatistics.sync_aggregate(
stage_list).to_list(None)
tmp_data = {
'code': s_dimen.code,
'title': s_dimen.title,
'ordered': s_dimen.ordered,
'correct': stat_result[0].correct if stat_result else 0,
'total': stat_result[0].total if stat_result else 0
}
sub_data_list.append(tmp_data)
main_data = {
'code': str(index + 1),
'title': m_dimen.title,
'ordered': index + 1,
'sub': sub_data_list
}
data.append(main_data)
if data:
data.sort(key=lambda x: x.get('ordered', 0))
if not data:
early_warning_empty("start_statistics_subject_parameter_cross", cache_key, locals(), '获取维度正确率统计数据为空，请检查！')
RedisCache.set(cache_key, msgpack.packb(data))
def do_statistics_member_top_n(cache_key, m_city_code_list, stat_type, top_n,
time_range):
"""
:param cache_key:
:param m_city_code_list:
:param stat_type:
:param top_n:
:param time_range:
:return:
"""
RedisCache.set(cache_key, KEY_CACHE_REPORT_DOING_NOW, 5 * 60)
stage_list = []
# 取前一天凌晨12点之前的数据
time_match = get_yesterday()
stage_list.append(MatchStage({'updated_dt': {'$lt': time_match}}))
if m_city_code_list:
stage_list.append(MatchStage({'city_code': {'$in': m_city_code_list}}))
s_code = ''
e_code = ''
if time_range:
suffix = time_range[-1:]
range_num = int(time_range.replace(suffix, ''))
delta = None
if suffix.upper() == 'D':
delta = datetime.timedelta(days=range_num)
elif suffix.upper() == 'M':
delta = datetime.timedelta(days=range_num * 30)
elif suffix.upper() == 'Y':
delta = datetime.timedelta(days=range_num * 365)
if delta:
s_code = datetime2str(datetime.datetime.now() - delta,
date_format='%Y%m%d000000')
e_code = datetime2str(datetime.datetime.now(),
date_format='%Y%m%d000000')
if s_code and e_code:
stage_list.append(
MatchStage({'daily_code': {
'$gte': s_code,
'$lte': e_code
}}))
stage_list.extend([
GroupStage({
'daily_code': '$daily_code',
'member_cid': '$member_cid',
'province_code': '$province_code'
} if stat_type == 1 else {
'daily_code': '$daily_code',
'member_cid': '$member_cid',
'province_code': '$province_code',
'city_code': '$city_code'
},
learn_times={'$sum': '$learn_times'}),
GroupStage({
'daily_code': '$_id.daily_code',
'province_code': '$_id.province_code'
} if stat_type == 1 else {
'daily_code': '$_id.daily_code',
'province_code': '$_id.province_code',
'city_code': '$_id.city_code'
},
count={'$sum': 1},
times={'$sum': '$learn_times'}),
LookupStage(AdministrativeDivision, '_id.province_code', 'post_code',
'province_list'),
LookupStage(AdministrativeDivision, '_id.city_code', 'post_code',
'city_list'),
ProjectStage(
**{
'_id':
False,
'daily_code':
'$_id.daily_code',
'count':
'$count',
'times':
'$times',
'province_code':
'$_id.province_code',
'province_title':
'$province_list.title',
'ad_code':
'$_id.province_code' if stat_type == 1 else '$_id.city_code',
'ad_title':
'$province_list.title' if stat_type ==
1 else '$city_list.title'
}),
SortStage([('daily_code', ASC), ('count', DESC)])
])
# 检索数据
data = []
stat_cursor = MemberDailyStatistics.sync_aggregate(stage_list)
t_code, t_list = None, None
top_n_found = False
while True:
try:
daily_stat = stat_cursor.next()
if daily_stat:
daily_code = daily_stat.daily_code
if not daily_code == t_code:
t_code = daily_code
t_list = []
top_n_found = False
print(t_code)
if len(t_list) < top_n:
t_list.append({
'date':
daily_code[:8],
'ad_code':
daily_stat.ad_code,
'province_title':
daily_stat.province_title[0] if
daily_stat.province_title and stat_type == 2 else '',
'title':
daily_stat.ad_title[0]
if daily_stat.ad_title else 'undefined',
'quantity':
daily_stat.count,
'times':
daily_stat.times
})
elif not top_n_found:
if t_code is not None:
data.append(t_list)
top_n_found = True
except StopIteration:
break
if not data:
early_warning_empty("start_statistics_member_top_n", cache_key,
locals(), '每日参与TOP5统计数据为空，请检查！')
RedisCache.set(cache_key, msgpack.packb(data))
def do_statistics_member_active(cache_key, m_city_code_list,
province_code_list, city_code_list,
gender_list, age_group_list, education_list):
"""
:param cache_key:
:param m_city_code_list:
:param province_code_list:
:param city_code_list:
:param gender_list:
:param age_group_list:
:param education_list:
:return:
"""
RedisCache.set(cache_key, KEY_CACHE_REPORT_DOING_NOW, 5 * 60)
# 统计数据
stage_list = []
if m_city_code_list:
stage_list.append(MatchStage({'city_code': {'$in': m_city_code_list}}))
# 取前一天凌晨12点之前的数据
time_match = get_yesterday()
stage_list.append(MatchStage({'updated_dt': {'$lt': time_match}}))
query_dict = {}
if province_code_list:
query_dict['province_code'] = {'$in': province_code_list}
if city_code_list:
query_dict['city_code'] = {'$in': city_code_list}
if gender_list:
query_dict['gender'] = {
'$in': [int(s_gender) for s_gender in gender_list]
}
if age_group_list:
query_dict['age_group'] = {
'$in': [int(s_age_group) for s_age_group in age_group_list]
}
if education_list:
query_dict['education'] = {
'$in': [int(s_education) for s_education in education_list]
}
if query_dict:
stage_list.append(MatchStage(query_dict))
stage_list.extend([
GroupStage('learning_code', quantity={'$sum': 1}),
SortStage([('_id', ASC)]),
LimitStage(8)
])
mld_stat_cursor = MemberLearningDayStatistics.sync_aggregate(stage_list)
data = []
while True:
try:
mld_stat = mld_stat_cursor.next()
if mld_stat:
data.append({
'days': mld_stat.id,
'quantity': mld_stat.quantity
})
except StopIteration:
break
all_members = sum([d.get('quantity') for d in data])
data_dict = {'data_list': data, 'all_members': all_members}
if not data_dict:
early_warning_empty("start_statistics_member_active", cache_key,
locals(), '答题活跃度统计数据为空，请检查！')
RedisCache.set(cache_key, msgpack.packb(data_dict))
def do_statistics_quiz_trends(cache_key, stat_type, m_city_code_list,
province_code_list, city_code_list, gender_list,
age_group_list, education_list, time_range):
"""
:param cache_key:
:param stat_type:
:param m_city_code_list:
:param province_code_list:
:param city_code_list:
:param gender_list:
:param age_group_list:
:param education_list:
:param time_range:
:return:
"""
RedisCache.set(cache_key, KEY_CACHE_REPORT_DOING_NOW, 20 * 60)
basic_stages = []
if m_city_code_list:
basic_stages.append(
MatchStage({'city_code': {
'$in': m_city_code_list
}}))
if province_code_list:
basic_stages.append(
MatchStage({'province_code': {
'$in': province_code_list
}}))
if city_code_list:
basic_stages.append(MatchStage({'city_code': {'$in': city_code_list}}))
if gender_list:
basic_stages.append(MatchStage({'sex': {"$in": gender_list}}))
if age_group_list:
basic_stages.append(MatchStage({'age_group': {'$in': age_group_list}}))
if education_list:
basic_stages.append(MatchStage({'education': {'$in': education_list}}))
yesterday = get_yesterday()
time_match = MatchStage({'created_dt': {'$lte': yesterday}})
if time_range:
suffix = time_range[-1:]
range_num = int(time_range.replace(suffix, ''))
delta = None
if suffix.upper() == 'D':
delta = datetime.timedelta(days=range_num)
elif suffix.upper() == 'M':
delta = datetime.timedelta(days=range_num * 30)
elif suffix.upper() == 'Y':
delta = datetime.timedelta(days=range_num * 365)
start_dt = yesterday - delta
time_match = MatchStage(
{'created_dt': {
'$gt': start_dt,
'$lt': yesterday
}})
data = []
game_data = do_stat_in_history(MemberGameHistory, time_match, basic_stages,
stat_type)
ckpt_data = do_stat_in_history(MemberCheckPointHistory, time_match,
basic_stages, stat_type)
for k, v in game_data.items():
try:
ckpt_data[k] += v
except KeyError:
ckpt_data[k] = v
data.append({k: ckpt_data[k]})
if not data:
early_warning_empty("start_statistics_quiz_trends", cache_key,
locals(), '答题趋势统计数据为空，请检查！')
RedisCache.set(cache_key, msgpack.packb(data))
def do_statistics_learning_situation(cache_key,
chart_type=None,
m_city_code_list=None,
gender_list=None,
province_code_list=None,
city_code_list=None,
age_group_list=None,
education_list=None,
dimension=None,
time_range=None,
dimension_code=None):
"""
:param cache_key:
:param chart_type:
:param m_city_code_list:
:param gender_list:
:param province_code_list:
:param city_code_list:
:param age_group_list:
:param education_list:
:param dimension:
:param time_range:
:param dimension_code:
:return:
"""
RedisCache.set(cache_key, KEY_CACHE_REPORT_DOING_NOW, 5 * 60)
stage_list = []
# 取前一天凌晨12点之前的数据
time_match = get_yesterday()
stage_list.append(MatchStage({'created_dt': {'$lt': time_match}}))
s_code, e_code = '', ''
if chart_type == 1:
s_code, e_code = get_daily_code_range(time_range)
if city_code_list:
stage_list.append(MatchStage({'city_code': {'$in': m_city_code_list}}))
parent_dimension_cid, dimension_cid = get_dimension(dimension_code)
if parent_dimension_cid and dimension_cid:
stage_list.append(
MatchStage({'dimension.%s' % parent_dimension_cid: dimension_cid}))
query_dict = {}
if s_code and e_code:
query_dict['daily_code'] = {'$gte': s_code, '$lte': e_code}
if province_code_list:
query_dict['province_code'] = {'$in': province_code_list}
if city_code_list:
query_dict['city_code'] = {'$in': city_code_list}
if gender_list:
query_dict['gender'] = {
'$in': [int(s_gender) for s_gender in gender_list]
}
if age_group_list:
query_dict['age_group'] = {
'$in': [int(s_age_group) for s_age_group in age_group_list]
}
if education_list:
query_dict['education'] = {
'$in': [int(s_education) for s_education in education_list]
}
if dimension:
try:
s_dimension = json.loads(dimension)
for k, v in s_dimension.items():
query_dict['dimension.%s' % k] = {'$in': v}
except Exception:
pass
if query_dict:
stage_list.append(MatchStage(query_dict))
stage_list.append(
GroupStage('daily_code' if chart_type == 1 else 'learning_code',
total={'$sum': '$subject_total_quantity'},
correct={'$sum': '$subject_correct_quantity'}))
stage_list.append(SortStage([('_id', ASC)]))
if chart_type == 2:
stage_list.append(MatchStage({'_id': {'$lte': 20}}))
data = []
# 检索数据
if chart_type == 1:
stat_cursor = MemberDailyDimensionStatistics.sync_aggregate(stage_list)
else:
stat_cursor = MemberLearningDayDimensionStatistics.sync_aggregate(
stage_list)
while True:
try:
md_stat = stat_cursor.next()
if md_stat:
data.append({
md_stat.id[:8] if chart_type == 1 else md_stat.id: {
'total': md_stat.total,
'correct': md_stat.correct
}
})
except StopIteration:
break
if not data:
early_warning_empty("start_statistics_learning_situation", cache_key,
locals(), '学习效果中数据为空，请检查！')
RedisCache.set(cache_key, msgpack.packb(data))
