def create_inverted_index():
rdx_name_list = g_conf.get("ridx", "ridx_list").split(",")
for rdx_name in rdx_name_list:
ridx_obj = inverted_index_t()
name_list = rdx_name.split("_")
for name in name_list:
if name == "idea":
continue
key_name = name + "_id"
data_file = "../data/" + rdx_name + ".txt"
fp = open(data_file, "r")
file_obj = file_process_t()
line_cnt = 0
indexed_idea_dict = {}
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
idea_id = file_obj.get_value("idea_id", data_list)
if not (idea_id in g_valid_idea_dict):
logging.debug("no need to indexed idea id[%s]" % (idea_id))
continue
key = file_obj.get_value(key_name, data_list)
ridx_obj.insert_data(key, idea_id)
indexed_idea_dict[idea_id] = 1
for idea_id in g_valid_idea_dict:
if not (idea_id in indexed_idea_dict) and rdx_name != "ban_idea":
ridx_obj.insert_data("0", idea_id)
fp.close()
ridx_obj.dump("../index/" + rdx_name + ".ridx")
def process(input_file, output_sql):
now_time_list = datetime.datetime.now().strftime("%H:%M:%S").split(":")
date_today = datetime.datetime.now().strftime("%Y-%m-%d")
hour = int(now_time_list[0])
minute = int(now_time_list[1])
second = int(now_time_list[2])
total_second = hour * 3600 + minute * 60 + second
ratio = float(total_second) / 86400
input_fp = open(input_file, "r")
sql_fp = open(output_sql, "w")
file_obj = file_process_t()
line_cnt = 0
sql_fp.write("use mis;\n")
for ori_line in input_fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
idea_id = file_obj.get_value("id", data_list)
budget = float(file_obj.get_value("budget", data_list))
available_budget = budget * ratio
sql_fp.write(
"insert into consumptions(date,idea_id,smooth) values('%s','%s','%f') on DUPLICATE KEY update smooth='%f';\n"
% (date_today, idea_id, available_budget, available_budget))
input_fp.close()
sql_fp.close()
def create_inverted_index():
rdx_name_list = g_conf.get("ridx", "ridx_list").split(",")
for rdx_name in rdx_name_list:
ridx_obj = inverted_index_t()
name_list = rdx_name.split("_")
for name in name_list:
if name == "idea":
continue
key_name = name + "_id"
data_file = "../data/" + rdx_name + ".txt"
fp = open(data_file, "r")
file_obj = file_process_t()
line_cnt = 0
indexed_idea_dict = {}
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
idea_id = file_obj.get_value("idea_id", data_list)
if not (idea_id in g_valid_idea_dict):
logging.debug("no need to indexed idea id[%s]" % (idea_id))
continue
key = file_obj.get_value(key_name, data_list)
ridx_obj.insert_data(key, idea_id)
indexed_idea_dict[idea_id] = 1
for idea_id in g_valid_idea_dict:
if not (idea_id in indexed_idea_dict) and rdx_name != "ban_idea":
ridx_obj.insert_data("0", idea_id)
fp.close()
ridx_obj.dump("../index/" + rdx_name + ".ridx")
def process(input_file,output_sql):
now_time_list=datetime.datetime.now().strftime("%H:%M:%S").split(":")
date_today=datetime.datetime.now().strftime("%Y-%m-%d")
hour=int(now_time_list[0])
minute=int(now_time_list[1])
second=int(now_time_list[2])
total_second=hour*3600+minute*60+second
ratio=float(total_second)/86400
input_fp=open(input_file,"r")
sql_fp=open(output_sql,"w")
file_obj=file_process_t()
line_cnt=0
sql_fp.write("use mis;\n")
for ori_line in input_fp:
data_list=ori_line.rstrip("\r\n").split("\t")
line_cnt+=1
if line_cnt==1:
file_obj.init_header(data_list)
continue
idea_id=file_obj.get_value("id",data_list)
budget=float(file_obj.get_value("budget",data_list))
available_budget=budget*ratio
sql_fp.write("insert into consumptions(date,idea_id,smooth) values('%s','%s','%f') on DUPLICATE KEY update smooth='%f';\n" %(date_today,idea_id,available_budget,available_budget))
input_fp.close()
sql_fp.close()
def create_index():
idea_fp = open(g_conf.get("file", "ideas"), "r")
index_name_list = g_conf.get("index", "index_list").split(",")
index_obj = index_t()
line_cnt = 0
file_obj = file_process_t()
type_ridx = inverted_index_t()
size_ridx = inverted_index_t()
for ori_line in idea_fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
idea_id = file_obj.get_value("id", data_list)
if not (idea_id in g_valid_idea_dict):
logging.debug("idea id no budget[%s]" % (idea_id))
continue
type = file_obj.get_value("type", data_list)
size_id = file_obj.get_value("size_id", data_list)
type_ridx.insert_data(type, idea_id)
size_ridx.insert_data(size_id, idea_id)
for name in index_name_list:
value = file_obj.get_value(name, data_list)
index_obj.insert_data(idea_id, name, value)
idea_fp.close()
#dump type invert index
type_ridx.dump("../index/type.ridx")
size_ridx.dump("../index/size.ridx")
file_obj = file_process_t()
line_cnt = 0
operate_fp = open(g_conf.get("file", "idea_operate"), "r")
for ori_line in operate_fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
idea_id = file_obj.get_value("idea_id", data_list)
if not (idea_id in g_valid_idea_dict):
logging.debug("idea id no budget[%s]" % (idea_id))
continue
value = file_obj.get_value("basic_bid", data_list)
index_obj.insert_data(idea_id, "basic_bid", value)
operate_fp.close()
index_obj.dump(g_conf.get("index", "idea_idx"))
def create_index():
idea_fp = open(g_conf.get("file", "ideas"), "r")
index_name_list = g_conf.get("index", "index_list").split(",")
index_obj = index_t()
line_cnt = 0
file_obj = file_process_t()
type_ridx = inverted_index_t()
size_ridx = inverted_index_t()
for ori_line in idea_fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
idea_id = file_obj.get_value("id", data_list)
if not (idea_id in g_valid_idea_dict):
logging.debug("idea id no budget[%s]" % (idea_id))
continue
type = file_obj.get_value("type", data_list)
size_id = file_obj.get_value("size_id", data_list)
type_ridx.insert_data(type, idea_id)
size_ridx.insert_data(size_id, idea_id)
for name in index_name_list:
value = file_obj.get_value(name, data_list)
index_obj.insert_data(idea_id, name, value)
idea_fp.close()
# dump type invert index
type_ridx.dump("../index/type.ridx")
size_ridx.dump("../index/size.ridx")
file_obj = file_process_t()
line_cnt = 0
operate_fp = open(g_conf.get("file", "idea_operate"), "r")
for ori_line in operate_fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
idea_id = file_obj.get_value("idea_id", data_list)
if not (idea_id in g_valid_idea_dict):
logging.debug("idea id no budget[%s]" % (idea_id))
continue
value = file_obj.get_value("basic_bid", data_list)
index_obj.insert_data(idea_id, "basic_bid", value)
operate_fp.close()
index_obj.dump(g_conf.get("index", "idea_idx"))
def load_idea_info(idea_operator_file, idea_file):
global g_idea_dict
g_idea_dict = {}
fp = open(idea_operator_file, "r")
file_obj = file_process_t()
line_cnt = 0
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
idea_id = file_obj.get_value("idea_id", data_list)
charge_type = file_obj.get_value("charge_type", data_list)
basic_price = float(file_obj.get_value("basic_price", data_list))
g_idea_dict[idea_id] = {
"charge_type": charge_type,
"basic_price": basic_price
}
fp.close()
fp = open(idea_file, "r")
file_obj = file_process_t()
line_cnt = 0
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
idea_id = file_obj.get_value("id", data_list)
user_id = file_obj.get_value("user_id", data_list)
plan_id = file_obj.get_value("plan_id", data_list)
if idea_id in g_idea_dict:
g_idea_dict[idea_id]["user_id"] = user_id
g_idea_dict[idea_id]["plan_id"] = plan_id
else:
sub_dict = {}
sub_dict = {
"user_id": user_id,
"plan_id": plan_id,
"charge_type": "cpm"
}
g_idea_dict[idea_id] = sub_dict
fp.close()
def load_idea_info(idea_operator_file,idea_file):
global g_idea_dict
g_idea_dict={}
fp=open(idea_operator_file,"r")
file_obj=file_process_t()
line_cnt=0
for ori_line in fp:
data_list=ori_line.rstrip("\r\n").split("\t")
line_cnt+=1
if line_cnt==1:
file_obj.init_header(data_list)
continue
idea_id=file_obj.get_value("idea_id",data_list)
charge_type=file_obj.get_value("charge_type",data_list)
basic_price=float(file_obj.get_value("basic_price",data_list))
g_idea_dict[idea_id]={"charge_type":charge_type,"basic_price":basic_price}
fp.close()
fp=open(idea_file,"r")
file_obj=file_process_t()
line_cnt=0
for ori_line in fp:
data_list=ori_line.rstrip("\r\n").split("\t")
line_cnt+=1
if line_cnt==1:
file_obj.init_header(data_list)
continue
idea_id=file_obj.get_value("id",data_list)
user_id=file_obj.get_value("user_id",data_list)
plan_id=file_obj.get_value("plan_id",data_list)
if idea_id in g_idea_dict:
g_idea_dict[idea_id]["user_id"]=user_id
g_idea_dict[idea_id]["plan_id"]=plan_id
else:
sub_dict={}
sub_dict={"user_id":user_id,"plan_id":plan_id,"charge_type":"cpm"}
g_idea_dict[idea_id]=sub_dict
fp.close()
def __init__(self,idea_operator_file):
self.idea_operator_dict={}
fp=open(idea_operator_file,"r")
line_cnt=0
file_obj=file_process_t()
for line in fp:
data_list=line.rstrip("\r\n").split("\t")
line_cnt+=1
if line_cnt==1:
file_obj.init_header(data_list)
continue
idea_id=file_obj.get_value("idea_id",data_list)
basic_bid=float(file_obj.get_value("basic_bid",data_list))
thr=float(file_obj.get_value("thr",data_list))
self.idea_operator_dict[idea_id]={"basic_bid":basic_bid,"thr":thr}
fp.close()
def process(input_file, output_file, date_str):
input_fp = open(input_file, "r")
output_fp = open(output_file, "w")
date_time = datetime.datetime.strptime(date_str, "%Y-%m-%d-%H:%M:%S")
date_day_str = datetime.datetime.strftime(date_time, "%Y-%m-%d")
file_obj = file_process_t()
line_cnt = 0
cost_dict = {}
price_dict = {}
for ori_line in input_fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
idea_id = file_obj.get_value("idea_id", data_list)
price = float(file_obj.get_value("price", data_list))
cost = float(file_obj.get_value("cost", data_list))
if idea_id in cost_dict:
cost_dict[idea_id] += cost
else:
cost_dict[idea_id] = cost
if idea_id in price_dict:
price_dict[idea_id] += price
else:
price_dict[idea_id] = price
output_fp.write("use mis;set names utf8;\n")
for idea_id in cost_dict:
cost = cost_dict[idea_id]
if idea_id in price_dict:
price = price_dict[idea_id]
else:
price = 0
if not (idea_id in g_idea_dict):
logging.warning("no such idea id[%s] in idea lib" % (idea_id))
continue
idea_info_dict = g_idea_dict[idea_id]
plan_id = idea_info_dict["plan_id"]
user_id = idea_info_dict["user_id"]
output_fp.write(
"insert into consumptions(idea_id,user_id,plan_id,price,cost,date) values('%s','%s','%s','%f','%f','%s') on DUPLICATE KEY update price=price+%f,cost=cost+%f,user_id='%s',plan_id='%s';\n"
% (idea_id, user_id, plan_id, price, cost, date_day_str, price,
cost, user_id, plan_id))
input_fp.close()
output_fp.close()
def process(input_file,output_file,date_str):
input_fp=open(input_file,"r")
output_fp=open(output_file,"w")
date_time=datetime.datetime.strptime(date_str,"%Y-%m-%d-%H:%M:%S")
date_day_str=datetime.datetime.strftime(date_time,"%Y-%m-%d")
file_obj=file_process_t()
line_cnt=0
cost_dict={}
price_dict={}
for ori_line in input_fp:
data_list=ori_line.rstrip("\r\n").split("\t")
line_cnt+=1
if line_cnt==1:
file_obj.init_header(data_list)
continue
idea_id=file_obj.get_value("idea_id",data_list)
price=float(file_obj.get_value("price",data_list))
cost=float(file_obj.get_value("cost",data_list))
if idea_id in cost_dict:
cost_dict[idea_id]+=cost
else:
cost_dict[idea_id]=cost
if idea_id in price_dict:
price_dict[idea_id]+=price
else:
price_dict[idea_id]=price
output_fp.write("use mis;set names utf8;\n")
for idea_id in cost_dict:
cost=cost_dict[idea_id]
if idea_id in price_dict:
price=price_dict[idea_id]
else:
price=0
if not(idea_id in g_idea_dict):
logging.warning("no such idea id[%s] in idea lib" %(idea_id))
continue
idea_info_dict=g_idea_dict[idea_id]
plan_id=idea_info_dict["plan_id"]
user_id=idea_info_dict["user_id"]
output_fp.write("insert into consumptions(idea_id,user_id,plan_id,price,cost,date) values('%s','%s','%s','%f','%f','%s') on DUPLICATE KEY update price=price+%f,cost=cost+%f,user_id='%s',plan_id='%s';\n" %(idea_id,user_id,plan_id,price,cost,date_day_str,price,cost,user_id,plan_id))
input_fp.close()
output_fp.close()
def __init__(self, idea_operator_file):
self.idea_operator_dict = {}
fp = open(idea_operator_file, "r")
line_cnt = 0
file_obj = file_process_t()
for line in fp:
data_list = line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
idea_id = file_obj.get_value("idea_id", data_list)
basic_bid = float(file_obj.get_value("basic_bid", data_list))
thr = float(file_obj.get_value("thr", data_list))
self.idea_operator_dict[idea_id] = {
"basic_bid": basic_bid,
"thr": thr
}
fp.close()
def __init__(self, conf):
self.adx_id_dict = {}
self.adx_id_file_dict = {}
self.adx_id_dict["device"] = {}
self.adx_id_file_dict["device"] = conf.get("file", "device")
self.adx_id_dict["network"] = {}
self.adx_id_file_dict["network"] = conf.get("file", "network")
self.adx_id_dict["classification"] = {}
self.adx_id_file_dict["classification"] = conf.get(
"file", "classification")
self.adx_id_dict["app"] = {}
self.adx_id_file_dict["app"] = conf.get("file", "app")
self.search_type_dict = {}
self.type_size_dict = {}
self.size_dict = {}
self.size_invert_dict = {}
for name in self.adx_id_file_dict:
file_name = self.adx_id_file_dict[name]
map_dict = self.adx_id_dict[name]
fp = open(file_name, "r")
file_obj = file_process_t()
line_cnt = 0
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
id = file_obj.get_value("id", data_list)
adx = file_obj.get_value("adx", data_list)
map_dict[adx] = id
fp = open(conf.get("file", "sizes"), "r")
file_obj = file_process_t()
line_cnt = 0
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
id = file_obj.get_value("id", data_list)
width = file_obj.get_value("width", data_list)
height = file_obj.get_value("height", data_list)
is_full = file_obj.get_value("is_full", data_list)
size_str = width + "x" + height
self.size_invert_dict[size_str] = id
self.size_dict[id] = {"size_str": size_str, "is_full": is_full}
fp = open(conf.get("file", "type_size_join"), "r")
file_obj = file_process_t()
line_cnt = 0
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
type_name = file_obj.get_value("name_en", data_list)
ad_type = (type_name.split("_"))[0]
device_id = file_obj.get_value("device_id", data_list)
size_id = file_obj.get_value("size_id", data_list)
combine_key = ad_type + "_" + device_id
self.search_type_dict[combine_key] = type_name
if type_name in self.type_size_dict:
sub_dict = self.type_size_dict[type_name]
else:
sub_dict = {}
sub_dict[size_id] = 1
self.type_size_dict[type_name] = sub_dict
fp = open(conf.get("file", "manual_operator"), "r")
self.adx_id_dict["operator"] = {}
file_obj = file_process_t()
line_cnt = 0
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
code = file_obj.get_value("code", data_list)
id = file_obj.get_value("id", data_list)
self.adx_id_dict["operator"][code] = id
fp.close()
def create_valid_idea():
global g_valid_idea_dict
g_valid_idea_dict = {}
valid_user_dict = {}
valid_plan_dict = {}
plan_consume_dict = {}
idea_consume_dict = {}
date_str = datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S")
week_day = datetime.datetime.now().weekday()
hour = int(datetime.datetime.now().strftime("%H"))
hour_idx = str(week_day * 24 + hour)
logging.debug("date:%s hour:%d hour_idx:%s" % (date_str, hour, hour_idx))
#load user file
user_file = g_conf.get("file", "users")
fp = open(user_file, "r")
file_obj = file_process_t()
line_cnt = 0
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
user_id = file_obj.get_value("user_id", data_list)
consume = float(file_obj.get_value("consumption_total", data_list))
total = float(file_obj.get_value("total", data_list))
balance = total - consume
if balance <= 0:
logging.debug("user id[%s] no budget" % (balance))
continue
valid_user_dict[user_id] = 1
fp.close()
#load consumption
consumption_file = g_conf.get("file", "consumption")
fp = open(consumption_file, "r")
file_obj = file_process_t()
line_cnt = 0
idea_smooth_dict = {}
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
plan_id = file_obj.get_value("plan_id", data_list)
idea_id = file_obj.get_value("idea_id", data_list)
consume = float(file_obj.get_value("price", data_list))
smooth = float(file_obj.get_value("smooth", data_list))
idea_smooth_dict[idea_id] = smooth
if plan_id in plan_consume_dict:
plan_consume_dict[plan_id] += consume
else:
plan_consume_dict[plan_id] = consume
if idea_id in idea_consume_dict:
idea_consume_dict[idea_id] += consume
else:
idea_consume_dict[idea_id] = consume
fp.close()
#create valid plan
plan_file = g_conf.get("file", "plans")
fp = open(plan_file, "r")
file_obj = file_process_t()
line_cnt = 0
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
plan_id = file_obj.get_value("plan_id", data_list)
user_id = file_obj.get_value("user_id", data_list)
budget = float(file_obj.get_value("budget", data_list))
start_time = file_obj.get_value("start_time", data_list)
end_time = file_obj.get_value("end_time", data_list)
if date_str < start_time or date_str > end_time:
logging.debug("date not in date range idea_id[%s]" % (plan_id))
continue
if not (user_id in valid_user_dict):
logging.debug("plan id[%s] user id[%s] no budget" %
(plan_id, user_id))
continue
if plan_id in plan_consume_dict:
plan_consume = plan_consume_dict[plan_id]
else:
plan_consume = 0
#print budget,plan_consume,plan_id
if budget - plan_consume <= 0:
logging.debug("plan[%s] no budget" % (plan_id))
continue
valid_plan_dict[plan_id] = 1
fp.close()
#create valid idea
idea_file = g_conf.get("file", "ideas")
fp = open(idea_file, "r")
file_obj = file_process_t()
line_cnt = 0
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
idea_id = file_obj.get_value("id", data_list)
plan_id = file_obj.get_value("plan_id", data_list)
budget = float(file_obj.get_value("budget", data_list))
status = file_obj.get_value("status", data_list)
if status != "3":
continue
time_range_list = file_obj.get_value("timerange", data_list)
if time_range_list != None and time_range_list != "NULL" and len(
time_range_list) > 0:
time_range_list = time_range_list.split(",")
dict_temp = {}
for target_hour in time_range_list:
dict_temp[target_hour] = 1
if not (hour_idx in dict_temp):
logging.debug("hour idx[%s] not in hour range" % (hour_idx))
continue
if not (plan_id in valid_plan_dict):
logging.debug("plan id[%s] no budget" % (plan_id))
continue
if idea_id in idea_consume_dict:
idea_consume = idea_consume_dict[idea_id]
else:
idea_consume = 0
if idea_id in idea_smooth_dict:
smooth = idea_smooth_dict[idea_id]
if idea_consume > smooth:
logging.debug(
"idea id[%s] no smooth budget[%f] idea consume[%f]" %
(idea_id, smooth, idea_consume))
continue
if budget - idea_consume <= 0:
logging.debug("idea id[%s] no budget idea consume[%f]" %
(idea_id, idea_consume))
continue
g_valid_idea_dict[idea_id] = 1
fp.close()
def create_valid_idea():
global g_valid_idea_dict
g_valid_idea_dict = {}
valid_user_dict = {}
valid_plan_dict = {}
plan_consume_dict = {}
idea_consume_dict = {}
date_str = datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S")
week_day = datetime.datetime.now().weekday()
hour = int(datetime.datetime.now().strftime("%H"))
hour_idx = str(week_day * 24 + hour)
logging.debug("date:%s hour:%d hour_idx:%s" % (date_str, hour, hour_idx))
# load user file
user_file = g_conf.get("file", "users")
fp = open(user_file, "r")
file_obj = file_process_t()
line_cnt = 0
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
user_id = file_obj.get_value("user_id", data_list)
consume = float(file_obj.get_value("consumption_total", data_list))
total = float(file_obj.get_value("total", data_list))
balance = total - consume
if balance <= 0:
logging.debug("user id[%s] no budget" % (balance))
continue
valid_user_dict[user_id] = 1
fp.close()
# load consumption
consumption_file = g_conf.get("file", "consumption")
fp = open(consumption_file, "r")
file_obj = file_process_t()
line_cnt = 0
idea_smooth_dict = {}
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
plan_id = file_obj.get_value("plan_id", data_list)
idea_id = file_obj.get_value("idea_id", data_list)
consume = float(file_obj.get_value("price", data_list))
smooth = float(file_obj.get_value("smooth", data_list))
idea_smooth_dict[idea_id] = smooth
if plan_id in plan_consume_dict:
plan_consume_dict[plan_id] += consume
else:
plan_consume_dict[plan_id] = consume
if idea_id in idea_consume_dict:
idea_consume_dict[idea_id] += consume
else:
idea_consume_dict[idea_id] = consume
fp.close()
# create valid plan
plan_file = g_conf.get("file", "plans")
fp = open(plan_file, "r")
file_obj = file_process_t()
line_cnt = 0
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
plan_id = file_obj.get_value("plan_id", data_list)
user_id = file_obj.get_value("user_id", data_list)
budget = float(file_obj.get_value("budget", data_list))
start_time = file_obj.get_value("start_time", data_list)
end_time = file_obj.get_value("end_time", data_list)
if date_str < start_time or date_str > end_time:
logging.debug("date not in date range idea_id[%s]" % (plan_id))
continue
if not (user_id in valid_user_dict):
logging.debug("plan id[%s] user id[%s] no budget" % (plan_id, user_id))
continue
if plan_id in plan_consume_dict:
plan_consume = plan_consume_dict[plan_id]
else:
plan_consume = 0
# print budget,plan_consume,plan_id
if budget - plan_consume <= 0:
logging.debug("plan[%s] no budget" % (plan_id))
continue
valid_plan_dict[plan_id] = 1
fp.close()
# create valid idea
idea_file = g_conf.get("file", "ideas")
fp = open(idea_file, "r")
file_obj = file_process_t()
line_cnt = 0
for ori_line in fp:
data_list = ori_line.rstrip("\r\n").split("\t")
line_cnt += 1
if line_cnt == 1:
file_obj.init_header(data_list)
continue
idea_id = file_obj.get_value("id", data_list)
plan_id = file_obj.get_value("plan_id", data_list)
budget = float(file_obj.get_value("budget", data_list))
status = file_obj.get_value("status", data_list)
if status != "3":
continue
time_range_list = file_obj.get_value("timerange", data_list)
if time_range_list != None and time_range_list != "NULL" and len(time_range_list) > 0:
time_range_list = time_range_list.split(",")
dict_temp = {}
for target_hour in time_range_list:
dict_temp[target_hour] = 1
if not (hour_idx in dict_temp):
logging.debug("hour idx[%s] not in hour range" % (hour_idx))
continue
if not (plan_id in valid_plan_dict):
logging.debug("plan id[%s] no budget" % (plan_id))
continue
if idea_id in idea_consume_dict:
idea_consume = idea_consume_dict[idea_id]
else:
idea_consume = 0
if idea_id in idea_smooth_dict:
smooth = idea_smooth_dict[idea_id]
if idea_consume > smooth:
logging.debug("idea id[%s] no smooth budget[%f] idea consume[%f]" % (idea_id, smooth, idea_consume))
continue
if budget - idea_consume <= 0:
logging.debug("idea id[%s] no budget idea consume[%f]" % (idea_id, idea_consume))
continue
g_valid_idea_dict[idea_id] = 1
fp.close()
def __init__(self,conf):
self.adx_id_dict={}
self.adx_id_file_dict={}
self.adx_id_dict["device"]={}
self.adx_id_file_dict["device"]=conf.get("file","device")
self.adx_id_dict["network"]={}
self.adx_id_file_dict["network"]=conf.get("file","network")
self.adx_id_dict["classification"]={}
self.adx_id_file_dict["classification"]=conf.get("file","classification")
self.adx_id_dict["app"]={}
self.adx_id_file_dict["app"]=conf.get("file","app")
self.search_type_dict={}
self.type_size_dict={}
self.size_dict={}
self.size_invert_dict={}
for name in self.adx_id_file_dict:
file_name=self.adx_id_file_dict[name]
map_dict=self.adx_id_dict[name]
fp=open(file_name,"r")
file_obj=file_process_t()
line_cnt=0
for ori_line in fp:
data_list=ori_line.rstrip("\r\n").split("\t")
line_cnt+=1
if line_cnt==1:
file_obj.init_header(data_list)
continue
id=file_obj.get_value("id",data_list)
adx=file_obj.get_value("adx",data_list)
map_dict[adx]=id
fp=open(conf.get("file","sizes"),"r")
file_obj=file_process_t()
line_cnt=0
for ori_line in fp:
data_list=ori_line.rstrip("\r\n").split("\t")
line_cnt+=1
if line_cnt==1:
file_obj.init_header(data_list)
continue
id=file_obj.get_value("id",data_list)
width=file_obj.get_value("width",data_list)
height=file_obj.get_value("height",data_list)
is_full=file_obj.get_value("is_full",data_list)
size_str=width+"x"+height
self.size_invert_dict[size_str]=id
self.size_dict[id]={"size_str":size_str,"is_full":is_full}
fp=open(conf.get("file","type_size_join"),"r")
file_obj=file_process_t()
line_cnt=0
for ori_line in fp:
data_list=ori_line.rstrip("\r\n").split("\t")
line_cnt+=1
if line_cnt==1:
file_obj.init_header(data_list)
continue
type_name=file_obj.get_value("name_en",data_list)
ad_type=(type_name.split("_"))[0]
device_id=file_obj.get_value("device_id",data_list)
size_id=file_obj.get_value("size_id",data_list)
combine_key=ad_type+"_"+device_id
self.search_type_dict[combine_key]=type_name
if type_name in self.type_size_dict:
sub_dict=self.type_size_dict[type_name]
else:
sub_dict={}
sub_dict[size_id]=1
self.type_size_dict[type_name]=sub_dict
fp=open(conf.get("file","manual_operator"),"r")
self.adx_id_dict["operator"]={}
file_obj=file_process_t()
line_cnt=0
for ori_line in fp:
data_list=ori_line.rstrip("\r\n").split("\t")
line_cnt+=1
if line_cnt==1:
file_obj.init_header(data_list)
continue
code=file_obj.get_value("code",data_list)
id=file_obj.get_value("id",data_list)
self.adx_id_dict["operator"][code]=id
fp.close()