def collect_stock_info():
ebest = EBest("DEMO")
mongodb = MongoDBHandler()
ebest.login()
code_list = mongodb.find_items({}, "stock", "m_code_info")
target_code = set([item["단축코드"] for item in code_list])
today = datetime.today().strftime("%Y%m%d")
print(today)
collect_list = mongodb.find_items({"날짜":today}, "stock", "price_info") \
.distinct("code")
for col in collect_list:
target_code.remove(col)
for code in target_code:
time.sleep(1)
print("code:", code)
result_price = ebest.get_stock_price_by_code(code, "1")
if len(result_price) > 0:
print(result_price)
mongodb.insert_items(result_price, "stock", "price_info")
result_credit = ebest.get_credit_trend_by_code(code, today)
if len(result_credit) > 0:
mongodb.insert_items(result_credit, "stock", "credit_info")
result_short = ebest.get_short_trend_by_code(code,
sdate=today,
edate=today)
if len(result_short) > 0:
mongodb.insert_items(result_short, "stock", "short_info")
result_agent = ebest.get_agent_trend_by_code(code,
fromdt=today,
todt=today)
if len(result_agent) > 0:
mongodb.insert_items(result_agent, "stock", "agent_info")
class DataCollector(object):
def __init__(self):
self.starttime = time.time(
) # 시작 시간 저장 print("time :", time.time() - start) # 현재시각 - 시작시간 = 실행 시간
self.mongodb = MongoDBHandler()
self.db_name = "stock1"
self.log = Logger()
self.ebest = EBest("DEMO")
self.ebest.login()
self.ebest.change_field_lang('E')
self.ins_date = datetime.today().strftime("%Y%m%d")
self.ins_time = datetime.today().strftime("%H%M%S")
self.day_90 = (datetime.today() -
timedelta(days=90)).strftime("%Y%m%d")
self.day_search_signal = (datetime.today() -
timedelta(days=6)).strftime("%Y%m%d")
self.wd = WorkDate()
self.tradingday = self.wd.find_working_day(self.ins_date, 0)
self.befoneday = self.wd.find_working_day(self.ins_date, 30)
self.momt_dates = []
def __del__(self):
# self.ebest.logout()
print("classs__del__")
def clear_data_base(self):
self.mongodb.delete_items({}, self.db_name, "m_stock_code")
self.mongodb.delete_items({}, self.db_name, "m_code_info")
self.mongodb.delete_items({}, self.db_name, "daily_price")
self.mongodb.delete_items({}, self.db_name, "Naver_news")
def calc_avg_volume(self, shcode, volume):
if int(volume) < 10000: return int(volume)
tot_volume = 0
cond = {
'shcode': shcode,
'date': {
'$gte': self.befoneday,
'$lte': self.tradingday
}
}
price_list = list(
self.mongodb.find_items(cond, self.db_name, "daily_price"))
if len(price_list):
for i, price in enumerate(price_list):
if int(price['jdiff_vol']):
tot_volume = tot_volume + int(price['jdiff_vol'])
if i: avg_volume = round(tot_volume / i)
else: avg_volume = 0
return avg_volume
def get_processing_date(self):
# 모멘텀을 계산을 위해 주식거래일 추출(1일, 20일, 60일 등)
offset = [0, 30, 90, 180, 360]
self.momt_dates = self.wd.get_trading_day_list(self.tradingday, offset)
def calc_bollingerBand_by_code(self, code):
cond = {
'shcode': code['shcode'],
'date': {
'$gte': self.day_90,
'$lte': self.ins_date
}
}
results = list(
self.mongodb.find_items(cond, self.db_name, "daily_price"))
if not len(results): return
# JSON => dataframe으로 변환
df = pd.DataFrame.from_dict(results, orient='columns')
df['close'] = df['close'].astype(float)
df['high'] = df['high'].astype(float)
df['low'] = df['low'].astype(float)
df['volume'] = df['jdiff_vol'].astype(float)
df['MA20'] = df['close'].rolling(window=20).mean()
df['stddev'] = df['close'].rolling(window=20).std()
df['upper'] = df['MA20'] + (df['stddev'] * 2)
df['lower'] = df['MA20'] - (df['stddev'] * 2)
df['PB'] = (df['close'] - df['lower']) / (df['upper'] - df['lower'])
# 고가, 저가, 종가의 합을 3으로 나눠서 중심가격 구함.
df['TP'] = (df['high'] + df['low'] + df['close']) / 3
df['PMF'] = 0
df['NMF'] = 0
# range함수는 마지막 값을 포함하지 않으므로 0부터 종가개수 -2까지 반복
for i in range(len(df.close) - 1):
if df.TP.values[i] < df.TP.values[i + 1]:
# 긍정적 현금흐름 : 중심가격이 전날보다 상승한 날들의 현금흐름의 합
df.PMF.values[i +
1] = df.TP.values[i + 1] * df.volume.values[i +
1]
df.NMF.values[i + 1] = 0
else:
# 긍정적 현금흐름 : 중심가격이 전날보다 하락한 날들의 현금흐름의 합
df.NMF.values[i +
1] = df.TP.values[i + 1] * df.volume.values[i +
1]
df.PMF.values[i + 1] = 0
df['MFR'] = (df.PMF.rolling(window=10).sum() /
df.NMF.rolling(window=10).sum())
df['MFI10'] = 100 - 100 / (1 + df['MFR'])
df = df[19:]
# 상승 하락유무 체크
df_up = df.loc[(df.date.values > self.day_search_signal) &
(df.PB.values > 0.8) & (df.MFI10.values > 80),
["date", "PB", "MFI10"]]
df_dw = df.loc[(df.date.values > self.day_search_signal) &
(df.PB.values < 0.2) & (df.MFI10.values < 20),
["date", "PB", "MFI10"]]
ins_bolband = {}
if not df_up.empty or not df_dw.empty:
signal = 'UP'
if not df_dw.empty:
signal = 'DW'
# 주간, 월간, 3개월 상승율 계산
ret_1w = 0
ret_4w = 0
ret_8w = 0
ret_12w = 0
if len(df.close) - 1 - 5 > 0:
ret_1w = round(
(df.close.values[len(df.close) - 1] /
df.close.values[len(df.close) - 1 - 5] - 1) * 100,
2) # 1주간 수익율
if len(df.close) - 1 - 20 > 0:
ret_4w = round(
(df.close.values[len(df.close) - 1] /
df.close.values[len(df.close) - 1 - 20] - 1) * 100,
2) # 20일 한달 수익율
if len(df.close) - 1 - 40 > 0:
ret_8w = round(
(df.close.values[len(df.close) - 1] /
df.close.values[len(df.close) - 1 - 40] - 1) * 100,
2) # 40일 두달 수익율
if len(df.close) > 0:
ret_12w = round(
(df.close.values[len(df.close) - 1] / df.close.values[0] -
1) * 100, 2) # 60일 두달 수익율
ins_bolband = {
"shcode": code["shcode"],
"hname": code["hname"],
"signal": signal,
"search_date": self.day_search_signal,
"insdate": self.ins_date,
"instime": self.ins_time,
"ret_1w": ret_1w,
"ret_4w": ret_4w,
"ret_8w": ret_8w,
"ret_12w": ret_12w
}
return ins_bolband
def calc_momentum(self, code):
if len(self.momt_dates) != 5:
self.log.error("모멘텀 계산날짜가 잘못 되었습니다. !!df_dates[%d]" %
len(self.momt_dates))
return {}
cond = {"shcode": code['shcode'], 'date': {'$in': self.momt_dates}}
price_list = list(
self.mongodb.find_items(cond, self.db_name, "daily_price"))
if len(price_list) != 5:
self.log.error("모멘텀 계산 - 데이터 부족 !!momt_dates[%s][%d]" %
(code['shcode'], len(price_list)))
return {}
df = pd.DataFrame.from_dict(price_list, orient='columns')
df.date.sort_values()
df['close'] = df['close'].astype(float)
ret_1mon = round((df.close.values[4] / df.close.values[3] - 1) * 100,
2)
ret_3mon = round((df.close.values[4] / df.close.values[2] - 1) * 100,
2)
ret_6mon = round((df.close.values[4] / df.close.values[1] - 1) * 100,
2)
ret_12mon = round((df.close.values[4] / df.close.values[0] - 1) * 100,
2)
ins_code = {}
if ret_1mon > 0 and ret_3mon > 0:
ins_code = {
"shcode": code["shcode"],
"hname": code["hname"],
"ret_1mon": ret_1mon,
"ret_3mon": ret_3mon,
"ret_6mon": ret_6mon,
"ret_12mon": ret_12mon,
"insdate": self.ins_date,
"instime": self.ins_time
}
return ins_code
# ins_list.append(ins_code)
# c += 1
# # print('\ncount :{} \nins_list : {}'.format(c, ins_list))
# if (c % 100) == 0: # 100건 처리하고 등록
# if ins_list :
# self.mongodb.insert_items(ins_list, self.db_name, "a2_momentum")
# print(ins_list)
# ins_list.clear()
# print('\n전체 {} 건 / {} 건 처리 중....'.format(len(code_list), c))
def collect_daily_stock_data_from_ebest(self, init=False, tradingday=None):
if tradingday:
self.tradingday = tradingday
# stock_code 업데이트
if init:
self.get_stock_code_from_ebest()
# code_info, daily_price 업데이트
daily_prices = [] # 일별 거래정보
code_info = list(
self.mongodb.find_items({"insdate": self.ins_date}, self.db_name,
"m_code_info"))
if init or not code_info:
code_list = list(
self.mongodb.find_items({"bu12gubun": "01"}, self.db_name,
"m_stock_code"))
i = 0
str_codes = ""
result_ext_all = []
# 종목별 거래정보 읽어오기
for code in code_list:
str_codes = str_codes + code["shcode"]
i = i + 1
if len(str_codes) >= 300 or len(code_list) == i:
result_ext = self.ebest.get_current_price_by_shcodes(
str_codes)
result_ext_all.extend(result_ext)
self.log.info("result_ext_all 건수[%d]" %
len(result_ext_all))
str_codes = ""
# 일일 종목별 가격 업데이트 (daily_price)
for extend in result_ext_all:
daily_price = {
'date': self.tradingday,
'time': "",
'open': extend['open'],
'high': extend['high'],
'low': extend['low'],
"close": extend['bidho'],
"jdiff_vol": extend['volume'],
"value": extend['value'],
"shcode": extend['shcode'],
"hname": extend['hname'],
"insdate": self.ins_date,
"instime": self.ins_time
}
daily_prices.append(daily_price)
if daily_prices:
self.mongodb.delete_items({'date': self.tradingday},
self.db_name, "daily_price")
self.mongodb.insert_items(daily_prices, self.db_name,
"daily_price")
self.log.info("daily_prices 등록.[%d]" % (len(daily_prices)))
daily_prices.clear()
df_code = pd.DataFrame.from_dict(code_list, orient='columns')
df_exp = pd.DataFrame.from_dict(result_ext_all, orient='columns')
df_exp.drop(['hname'], axis=1, inplace=True)
df_exp['avg_volume'] = df_exp.apply(
lambda x: self.calc_avg_volume(x['shcode'], x['volume']),
axis=1)
df_exp['insdate'] = self.ins_date
df_exp['instime'] = self.ins_time
df_all = pd.merge(df_code, df_exp, how='outer', on='shcode')
code_list = df_all.to_dict(orient='records')
if code_list:
self.mongodb.delete_items({}, self.db_name, "m_code_info")
self.mongodb.insert_items(code_list, self.db_name,
"m_code_info")
if not code_info:
code_info = list(
self.mongodb.find_items({"bu12gubun": "01"}, self.db_name,
"m_code_info"))
# 1. update code_info
# - 주식종목 기본 + 추가 정보(전일거래량, 시가, 종가 등)
# - 거래량이 만주 이하인 종목 제외
if not code_info:
self.log.error("데이터 검색을 위한 자료가 없습니다.")
return
self.log.info("************데이터 분석 시작*********************")
# 데이터 검색을 위한 날짜 초기화
self.get_processing_date()
daily_prices = [] # 일별 거래정보
inc_vol_codes = [] # 거래량 증가 정보
inc_bol_bands = [] # 볼린저밴트 종목 검색
inc_momentums = [] # 종목 모멘텀 분석
# 코드정보와 주식현재가 추가, 일별 주가정보(daily_price) 업데이트
for icnt, code in enumerate(code_info):
# 거래량 만주 이하인 종목은 분석하지 않음.
if code['avg_volume'] is NaN: continue
if code['avg_volume'] is np.NaN: continue
avg_volume = int(code['avg_volume'])
if avg_volume < 10000: continue
# 1. 거래량 증가 종목 검색
if int(code['avg_volume']) > 0:
inc_rate = int(code['volume']) / int(code['avg_volume'])
else:
inc_rate = 0
if inc_rate > 5:
inc_code = {
"shcode": code["shcode"],
"hname": code["hname"],
"sdate": self.tradingday,
"avg_volume": int(avg_volume),
"volume": int(code['volume']),
"insdate": self.ins_date,
"instime": self.ins_time
}
inc_vol_codes.append(inc_code)
# 2. 볼린저밴드 종목 검색
inc_bol_band = self.calc_bollingerBand_by_code(code)
if inc_bol_band:
inc_bol_bands.append(inc_bol_band)
# 3. 모멘텀 종목 검색
inc_momentum = self.calc_momentum(code)
if inc_momentum:
inc_momentums.append(inc_momentum)
self.log.debug(
"종목 검색 중..[%d]/[%d][%s][%s]" %
(icnt, len(code_info), code["shcode"], code["hname"]))
if inc_bol_bands:
self.mongodb.delete_items({'insdate': self.ins_date}, self.db_name,
"a1_bollband")
self.mongodb.insert_items(inc_bol_bands, self.db_name,
"a1_bollband")
self.log.info("inc_bol_bands 등록.[%d]" % (len(inc_bol_bands)))
if inc_momentums:
self.mongodb.delete_items({'insdate': self.ins_date}, self.db_name,
"a2_momentum")
self.mongodb.insert_items(inc_momentums, self.db_name,
"a2_momentum")
self.log.info("inc_momentums 등록.[%d]" % (len(inc_momentums)))
if inc_vol_codes:
self.mongodb.delete_items({'date': self.tradingday}, self.db_name,
"a3_inc_volume")
self.mongodb.insert_items(inc_vol_codes, self.db_name,
"a3_inc_volume")
self.log.info("inc_vol_codes 등록.[%d]" % (len(inc_vol_codes)))
self.insert_exec_job_list('EB', '종목코드 + 현재가 반영 작업', 'code_info',
str(len(code_info)) + " 건이 정상처리 되었습니다.")
pass
def get_stock_code_from_ebest(self):
"""
:주식 종목 코드 가져오기
"""
# stocks = self.mongodb.find_items({}, self.db_name, "m_stock_code")
results = list(self.ebest.get_code_list("ALL"))
# !!@@나중에 업그레이드 하자, db에 없는 종목만 업데이트
# for i, result in results:
# for stock in stocks:
# if stock['shcode'] == result['shcode']:
# del
self.mongodb.delete_items({}, self.db_name, "m_stock_code")
self.mongodb.insert_items(results, self.db_name, "m_stock_code")
self.insert_exec_job_list('EB', '종목코드 가져오기', 'm_stock_code',
str(len(results)) + " 건이 정상처리 되었습니다.")
def get_code_info_from_ebest(self):
"""
주식 종목에 현재 가격, 거래량 정보 추가
"""
result_cod = self.ebest.get_code_list("ALL")
self.log.info("get_code_list%d" % len(result_cod))
result_ext_all = []
i = 0
if len(result_cod) > 0:
self.log.info("t8407 주식현재가 시작")
str_codes = ""
for code in result_cod:
str_codes = str_codes + code["shcode"]
i = i + 1
if len(str_codes) >= 300 or len(result_cod) == i:
result_ext = self.ebest.get_current_price_by_shcodes(
str_codes)
result_ext_all.extend(result_ext)
self.log.info("result_ext_all 건수[%d]" %
len(result_ext_all))
str_codes = ""
# 코드정보와 주식현재가 병합
for code in result_cod:
for extend in result_ext_all:
if code["shcode"] == extend["shcode"]:
code.update(extend)
self.log.info("종목코드 + 주식 현재가 반영 : {} 건".format(len(result_cod)))
# self.mongodb.delete_items({}, self.db_name, "m_code_info")
# self.mongodb.insert_items(result_cod, self.db_name, "m_code_info")
self.insert_exec_job_list('EB', '종목코드 + 현재가 반영 작업', 'code_info',
str(len(result_cod)) + " 건이 정상처리 되었습니다.")
def search_increase_vol_by_code(self):
""" 거래량 증가 종목 검색
: 한달 평균 거래량 보다 5배 증가한 종목 수집(check_volume)
"""
# 증권그룹 '01'
code_list = list(
self.mongodb.find_items({"bu12gubun": "01"}, self.db_name,
"m_code_info"))
self.log.info("CODE_LIST[%d]" % len(code_list))
today = datetime.today().strftime("%Y%m%d")
fromday = (datetime.today() - timedelta(days=30)).strftime("%Y%m%d")
inc_codes = []
vol_codes = []
loop_cnt = 0
for code in code_list:
if int(code["volume"]) < 10000:
continue
loop_cnt = loop_cnt + 1
if loop_cnt % 100 == 0 and len(inc_codes) > 0:
print(str(loop_cnt % 100) + "//" + str(len(inc_codes)))
print(code["shcode"] + "진행율 : " +
str((loop_cnt / len(code_list) * 100)))
self.mongodb.insert_items(inc_codes, self.db_name,
"check_volume")
inc_codes.clear()
results = self.ebest.get_stock_chart_by_code(
code["shcode"], "2", fromday, today)
time.sleep(1)
if len(results) > 0:
# 평균 거래량 계산
tot_volume = 0
i_count = 0
for result in results:
if int(result['jdiff_vol']) != 0:
tot_volume = tot_volume + int(result['jdiff_vol'])
i_count = i_count + 1
if i_count == 0 or tot_volume == 0:
continue
avg_volume = tot_volume / i_count
inc_rate = int(result['jdiff_vol']) / avg_volume
inc_code = {
"shcode": code["shcode"],
"hname": code["hname"],
"sdate": today,
"avg_volume": int(avg_volume),
"volume": int(result['jdiff_vol']),
"inc_rate": inc_rate
}
vol_codes.append(inc_code)
print("체크 종목 :" + code["shcode"] + " 거래량 [" +
result['jdiff_vol'] + "] 평균 [" + str(avg_volume) +
"] 비율[" + str(inc_rate) + "]")
# 거래량이 5배 이상이면 종목 추가
if inc_rate > 5:
inc_codes.append(inc_code)
print("추가된 종목 :" + code["shcode"] + " 건수 : " +
str(len(inc_codes)))
if len(inc_codes) > 0:
self.mongodb.insert_items(inc_codes, self.db_name, "check_volume")
if len(vol_codes) > 0:
self.mongodb.insert_items(vol_codes, self.db_name, "volume")
self.insert_exec_job_list('TA', '거래량 급증가 종목 검색', 'check_volume',
str(len(vol_codes)) + " 건이 정상처리 되었습니다.")
def insert_exec_job_list(self, jobtype, jobname, tablelist, logmsg):
""" 작업로그 기록
: jobtype -> 'EB' : EBEST, 'FA': 재무분석작업, 'TA' : 기술적 분석 작업,
: 'ML' : 머신러닝 예측작업, 'MA' : 재료분석 작업
"""
jobs = []
today = datetime.today().strftime("%Y%m%d")
totime = datetime.today().strftime("%H%M%S")
job = {
"jobtype": jobtype,
"jobdate": today,
"jobtime": totime,
"jobname": jobname,
"tablelist": tablelist,
"logmsg": logmsg
}
jobs.append(job)
self.mongodb.insert_items(jobs, self.db_name, "job_list")
print(jobs)
@staticmethod
def ins_daily_price_from_ebest(self, start_date, code_list):
daily_price = []
today = datetime.today().strftime("%Y%m%d")
totime = datetime.today().strftime("%H%M%S")
for code in code_list:
results = self.ebest.get_stock_chart_by_code(
code['shcode'], "2", start_date, today)
for result in results:
result['shcode'] = code['shcode']
result['hname'] = code['hname']
result['insdate'] = today
result['instime'] = totime
daily_price.append(result)
if len(daily_price) > 0:
self.mongodb.insert_items(daily_price, self.db_name,
"daily_price")
daily_price.clear()
self.insert_exec_job_list('EB', '일일 주식거래정보(Candle Chart형)',
'daily_price',
str(len(code_list)) + " 건이 정상처리 되었습니다.")
def insert_daily_price_from_ebest(self,
init=None,
start_date=None,
shcode=None):
####
# 일일 주식 거래정보(candle chart형)
# init : 1 -> 모든 종목 초기화 후 재생성, 2 -> 입력받은 코드만 조건대로 시작일 부터 갱신
####
today = datetime.today().strftime("%Y%m%d")
totime = datetime.today().strftime("%H%M%S")
if init == 1:
#데이터 삭제
self.mongodb.delete_items({}, self.db_name, "daily_price")
code_list = list(
self.mongodb.find_items({"bu12gubun": "01"}, self.db_name,
"m_stock_code"))
if len(code_list):
self.ins_daily_price_from_ebest(self, start_date, code_list)
else:
print("No Data!!")
elif init == 2:
if shcode:
cond = {
'shcode': shcode,
'date': {
'$gte': start_date,
'$lte': today
}
}
else:
cond = {'date': {'$gte': start_date, '$lte': today}}
self.mongodb.delete_items(cond, self.db_name, "daily_price")
if shcode:
cond = {'shcode': shcode}
else:
cond = {"bu12gubun": "01"}
print("조건", cond)
code_list = list(
self.mongodb.find_items(cond, self.db_name, "m_stock_code"))
if len(code_list):
self.ins_daily_price_from_ebest(self, start_date, code_list)
else:
print("No Data !!")
else:
print("init error No Data !!")
def search_bollingerBand_by_code(self, search_date, order_type):
"""볼린저 밴드 매수, 매도 시그널이 있는 종목 검색
: search_date : 검색 시작일자, order_type : 1 -> 매수, 2 -> 매도, 0 -> ALL
"""
today = datetime.today().strftime("%Y%m%d")
totime = datetime.today().strftime("%H%M%S")
fromday = (datetime.today() - timedelta(days=90)).strftime("%Y%m%d")
# 기존 작업 삭제
cond = {'insdate': today}
self.mongodb.delete_items(cond, self.db_name, "a1_bollband")
# 증권그룹 '01' 종목리스트 조회
code_list = list(
self.mongodb.find_items({"bu12gubun": "01"}, self.db_name,
"m_stock_code"))
print("CODE_LIST", len(code_list))
c = 0 # count
ins_list = []
for code in code_list:
cond = {
'shcode': code['shcode'],
'date': {
'$gte': fromday,
'$lte': today
}
}
results = list(
self.mongodb.find_items(cond, self.db_name, "daily_price"))
if not len(results): continue
# JSON => dataframe으로 변환
df = pd.DataFrame.from_dict(results, orient='columns')
df['close'] = df['close'].astype(float)
df['high'] = df['high'].astype(float)
df['low'] = df['low'].astype(float)
df['volume'] = df['jdiff_vol'].astype(float)
df['MA20'] = df['close'].rolling(window=20).mean()
df['stddev'] = df['close'].rolling(window=20).std()
df['upper'] = df['MA20'] + (df['stddev'] * 2)
df['lower'] = df['MA20'] - (df['stddev'] * 2)
df['PB'] = (df['close'] - df['lower']) / (df['upper'] -
df['lower'])
# 고가, 저가, 종가의 합을 3으로 나눠서 중심가격 구함.
df['TP'] = (df['high'] + df['low'] + df['close']) / 3
df['PMF'] = 0
df['NMF'] = 0
# range함수는 마지막 값을 포함하지 않으므로 0부터 종가개수 -2까지 반복
for i in range(len(df.close) - 1):
if df.TP.values[i] < df.TP.values[i + 1]:
# 긍정적 현금흐름 : 중심가격이 전날보다 상승한 날들의 현금흐름의 합
df.PMF.values[
i + 1] = df.TP.values[i + 1] * df.volume.values[i + 1]
df.NMF.values[i + 1] = 0
else:
# 긍정적 현금흐름 : 중심가격이 전날보다 하락한 날들의 현금흐름의 합
df.NMF.values[
i + 1] = df.TP.values[i + 1] * df.volume.values[i + 1]
df.PMF.values[i + 1] = 0
df['MFR'] = (df.PMF.rolling(window=10).sum() /
df.NMF.rolling(window=10).sum())
df['MFI10'] = 100 - 100 / (1 + df['MFR'])
df = df[19:]
# 상승 하락유무 체크
df_up = df.loc[(df.date.values > search_date) &
(df.PB.values > 0.8) & (df.MFI10.values > 80),
["date", "PB", "MFI10"]]
df_dw = df.loc[(df.date.values > search_date) &
(df.PB.values < 0.2) & (df.MFI10.values < 20),
["date", "PB", "MFI10"]]
if not df_up.empty or not df_dw.empty:
signal = 'UP'
if not df_dw.empty:
signal = 'DW'
# 주간, 월간, 3개월 상승율 계산
ret_1w = round(
(df.close.values[len(df.close) - 1] /
df.close.values[len(df.close) - 1 - 5] - 1) * 100,
2) # 1주간 수익율
ret_4w = round(
(df.close.values[len(df.close) - 1] /
df.close.values[len(df.close) - 1 - 20] - 1) * 100,
2) # 20일 한달 수익율
ret_8w = round(
(df.close.values[len(df.close) - 1] /
df.close.values[len(df.close) - 1 - 40] - 1) * 100,
2) # 40일 두달 수익율
ret_12w = round(
(df.close.values[len(df.close) - 1] / df.close.values[0] -
1) * 100, 2) # 60일 두달 수익율
ins_code = {
"shcode": code["shcode"],
"hname": code["hname"],
"signal": signal,
"search_date": search_date,
"insdate": today,
"instime": totime,
"ret_1w": ret_1w,
"ret_4w": ret_1w,
"ret_8w": ret_8w,
"ret_12w": ret_12w
}
ins_list.append(ins_code)
# for i in range(len(df.close)):
# if df.PB.values[i] > 0.8 and df.MFI10.values[i] > 80: # ①
# print("매수신호")
# elif df.PB.values[i] < 0.2 and df.MFI10.values[i] < 20: # ③
# print("매도신호")
c += 1
# print('\ncount :{} \nins_list : {}'.format(c, ins_list))
# if ins_list:
# return
if (c % 100) == 0: # 100건 처리하고 등록
if ins_list:
self.mongodb.insert_items(ins_list, self.db_name,
"a1_bollband")
print(ins_list)
ins_list.clear()
print('\n전체 {} 건 / {} 건 처리 중....'.format(len(code_list), c))
self.insert_exec_job_list('TA', '볼린저 밴드 매수 시그날 종목 검색', 'a1_bollband',
str(len(code_list)) + " 건이 정상처리 되었습니다.")
def search_momentum(self):
""" 모멘텀 종목 검색
:
"""
today = datetime.today().strftime("%Y%m%d")
totime = datetime.today().strftime("%H%M%S")
fromday = (datetime.today() - timedelta(days=90)).strftime("%Y%m%d")
# 기존 작업 삭제
cond = {'insdate': today}
self.mongodb.delete_items(cond, self.db_name, "a2_momentum")
# 네이버 조회를 통해 모멘텀 계산할 일자 조회
price_list = list(
self.mongodb.find_items({"shcode": "035420"}, self.db_name,
"daily_price"))
if not price_list:
print("Not found price Data !!")
return
df = pd.DataFrame.from_dict(price_list, orient='columns')
df_dates = df.iloc[[
len(df.close) - 1,
len(df.close) - 21,
len(df.close) - 61,
len(df.close) - 121,
len(df.close) - 241
], ]
if len(df_dates) != 5:
print("Not correct dates !!")
return
# 증권그룹 '01' 종목리스트 조회
code_list = list(
self.mongodb.find_items({"bu12gubun": "01"}, self.db_name,
"m_stock_code"))
print("CODE_LIST", len(code_list))
c = 0 # count
ins_list = []
# 종목별 상승율 계산
for code in code_list:
cond = {
"shcode": code['shcode'],
'date': {
'$in': df_dates.date.values.tolist()
}
}
price_list = list(
self.mongodb.find_items(cond, self.db_name, "daily_price"))
df = pd.DataFrame.from_dict(price_list, orient='columns')
df.date.sort_values()
df['close'] = df['close'].astype(float)
ret_1mon = round(
(df.close.values[4] / df.close.values[3] - 1) * 100, 2)
ret_3mon = round(
(df.close.values[4] / df.close.values[2] - 1) * 100, 2)
ret_6mon = round(
(df.close.values[4] / df.close.values[1] - 1) * 100, 2)
ret_12mon = round(
(df.close.values[4] / df.close.values[0] - 1) * 100, 2)
if ret_1mon > 0 and ret_3mon > 0:
ins_code = {
"shcode": code["shcode"],
"hname": code["hname"],
"insdate": today,
"instime": totime,
"ret_1mon": ret_1mon,
"ret_3mon": ret_3mon,
"ret_6mon": ret_6mon,
"ret_12mon": ret_12mon
}
ins_list.append(ins_code)
c += 1
# print('\ncount :{} \nins_list : {}'.format(c, ins_list))
if (c % 100) == 0: # 100건 처리하고 등록
if ins_list:
self.mongodb.insert_items(ins_list, self.db_name,
"a2_momentum")
print(ins_list)
ins_list.clear()
print('\n전체 {} 건 / {} 건 처리 중....'.format(len(code_list), c))
self.insert_exec_job_list('TA', '모멘텀 상승 종목 검색', 'a2_momentum',
str(len(code_list)) + " 건이 정상처리 되었습니다.")
def collect_stock_info(self):
code_list = self.mongodb.find_items({}, self.db_name, "m_code_info")
target_code = set([item["단축코드"] for item in code_list])
today = datetime.today().strftime("%Y%m%d")
collect_list = self.mongodb.find_items({
"날짜": today
}, self.db_name, "price_info").distinct("code")
for col in collect_list:
target_code.remove(col)
for code in target_code:
result_price = self.ebest.get_stock_price_by_code(code, "1")
time.sleep(1)
if len(result_price) > 0:
self.mongodb.insert_items(result_price, self.db_name,
"price_info")
def collect_theme_info(self):
# 테마 정보 분석
# tm_momt_date : 테마 모멘텀 분석 비교일자
# 테마코드 조회(마스터)
tm_list = self.ebest.get_theme_by_tmcode("", "")
# 특이 테마 조회
tm_momt_list = self.ebest.find_special_theme("1",
self.day_search_signal)
tm_price_list = []
# 테마 마스터와 특이테마 merge
for i, tm in enumerate(tm_list):
for tm_momt in tm_momt_list:
if tm['tmcode'] == tm_momt['tmcode']:
tm.update(tm_momt)
tm.update({
"insdate": self.ins_date,
"instime": self.ins_time
})
if i < 70:
tm_result = self.ebest.get_price_by_theme(tm['tmcode'])
for tm_re in tm_result:
tm_re.update({
"tmcode": tm['tmcode'],
"tmname": tm['tmname'],
"insdate": self.ins_date,
"instime": self.ins_time
})
tm_price_list.extend(tm_result)
# 거래량,
# print(tm_list)
df = pd.DataFrame.from_dict(tm_list, orient='columns')
df_price_list = pd.DataFrame.from_dict(tm_price_list, orient='columns')
df['rank_avgdiff'] = df['avgdiff'].rank(method='min', ascending=False)
df['rank_uprate'] = df['uprate'].rank(method='min', ascending=False)
df['rank_diff_vol'] = df['diff_vol'].rank(method='min',
ascending=False)
df['rank_sum'] = df['rank_avgdiff'] + df['rank_diff_vol'] + df[
'rank_uprate']
df_rank = df.sort_values(by='rank_sum').head(40)
print("df_rank", df_rank)
print("\ndf_price_list", df_price_list)
# tm_code = df_rank['tmcode']
# print("df_rank", tm_code)
# df_ddd = df_price_list[tm_code]
# print("\ndf_rank", df_ddd)
df_merge = pd.merge(df_rank, df_price_list, how='inner', on='tmcode')
df_merge = df_merge.astype({'marketcap': 'float'})
# 중복 컬럼 삭제
df_merge.drop(['tmname_y'], axis=1, inplace=True)
df_merge.drop(['insdate_y'], axis=1, inplace=True)
df_merge.drop(['instime_y'], axis=1, inplace=True)
df_merge.rename(columns={
'tmname_x': 'tmname',
'insdate_x': 'insdate',
'instime_x': 'instime'
},
inplace=True)
# df_rank.join(df_price_list, how='inner')
if not df_merge.empty:
df_merge['rank_by_code'] = df_merge.groupby(
'tmcode')['marketcap'].rank(method='min', ascending=False)
df_tm_rank = df_merge[df_merge['rank_by_code'] < 10]
tm_rank_list = df_tm_rank.to_dict(orient='records')
self.mongodb.delete_items({'insdate': self.ins_date}, self.db_name,
"a4_theme_momentum")
self.mongodb.insert_items(tm_rank_list, self.db_name,
"a4_theme_momentum")
self.log.info("a4_theme_momentum 등록.[%d]" % (len(tm_rank_list)))
tm_list.clear()
tm_list = df.to_dict(orient='records')
if tm_list:
self.mongodb.delete_items({'insdate': self.ins_date}, self.db_name,
"m_theme_code")
self.mongodb.insert_items(tm_list, self.db_name, "m_theme_code")
self.log.info("m_theme_code 등록.[%d]" % (len(tm_list)))
if tm_price_list:
self.mongodb.delete_items({'insdate': self.ins_date}, self.db_name,
"theme_price")
self.mongodb.insert_items(tm_price_list, self.db_name,
"theme_price")
self.log.info("theme_price 등록.[%d]" % (len(tm_price_list)))
# print(df_head)
pass
def counting_event(self, h1, h2, h3, h4):
count = 0
if not h1 is NaN: count += 1
if not h2 is NaN: count += 1
if not h3 is NaN: count += 1
if not h4 is NaN: count += 1
return count
def run_technical_analysis(self):
# self.tradingday = '20210609'
print(self.tradingday)
a1_list = list(
self.mongodb.find_items({
'insdate': self.ins_date,
'signal': 'UP'
}, self.db_name, "a1_bollband"))
a2_list = list(
self.mongodb.find_items({'insdate': self.ins_date}, self.db_name,
"a2_momentum"))
a3_list = list(
self.mongodb.find_items({'insdate': self.ins_date}, self.db_name,
"a3_inc_volume"))
a4_list = list(
self.mongodb.find_items({'insdate': self.ins_date}, self.db_name,
"a4_theme_momentum"))
df_a1 = pd.DataFrame.from_dict(a1_list, orient='columns')
df_a2 = pd.DataFrame.from_dict(a2_list, orient='columns')
df_a3 = pd.DataFrame.from_dict(a3_list, orient='columns')
df_a4 = pd.DataFrame.from_dict(a4_list, orient='columns')
print("a1_list", a1_list)
print("df_a1", df_a1)
df_a1 = df_a1[['shcode', 'hname']]
df_a2 = df_a2[['shcode', 'hname']]
df_a3 = df_a3[['shcode', 'hname', 'volume']]
df_a4 = df_a4[['shcode', 'hname', 'tmcode', 'tmname']]
df_a2.rename(columns={'hname': 'hname_a2'}, inplace=True)
df_a3.rename(columns={'hname': 'hname_a3'}, inplace=True)
df_a4.rename(columns={'hname': 'hname_a4'}, inplace=True)
df_merge = pd.merge(df_a1,
df_a2,
how='outer',
left_on='shcode',
right_on='shcode')
df_merge = pd.merge(df_merge,
df_a3,
how='outer',
left_on='shcode',
right_on='shcode')
df_merge = pd.merge(df_merge,
df_a4,
how='outer',
left_on='shcode',
right_on='shcode')
# df_a1.drop(['search_date'], axis=1, inplace=True)
# df_a1.drop(['insdate'], axis=1, inplace=True)
# df_a1.drop(['instime'], axis=1, inplace=True)
# df_a2.drop(['hname'], axis=1, inplace=True)
# df_a2.drop(['insdate'], axis=1, inplace=True)
# df_a2.drop(['instime'], axis=1, inplace=True)
# df_a3.drop(['hname'], axis=1, inplace=True)
# df_a3.drop(['insdate'], axis=1, inplace=True)
# df_a3.drop(['instime'], axis=1, inplace=True)
# df_a4.drop(['hname'], axis=1, inplace=True)
# df_a4.drop(['insdate'], axis=1, inplace=True)
# df_a4.drop(['instime'], axis=1, inplace=True)
# df_merge = pd.merge(df_a1, df_a2, df_a3, df_a4, how='outer',on='shcode')
df_merge['event_cnt'] = df_merge.apply(lambda x: self.counting_event(
x['hname'], x['hname_a2'], x['hname_a3'], x['hname_a4']),
axis=1)
df_merge.sort_values(by=['event_cnt'],
axis=0,
ascending=False,
inplace=True)
print('df_merge', df_merge.head(30))
pass
def get(self, id=None):
if id:
print("get method id :" + str(id))
# 종목코드 가져오기
if id == 1:
ebest = EBest("DEMO")
ebest.login()
result_cod = ebest.get_code_list("ALL")
print("get_code_list", len(result_cod))
result_ext_all = []
i = 0
if len(result_cod) > 0:
print("t8407 주식현재가 시작")
str_codes = ""
for code in result_cod:
str_codes = str_codes + code["단축코드"]
i = i + 1
if len(str_codes) >= 300 or len(result_cod) == i:
result_ext = ebest.get_current_price_by_shcodes(
str_codes)
result_ext_all.extend(result_ext)
print("result_ext_all 건수", len(result_ext_all))
str_codes = ""
# 코드정보와 주식현재가 병합
for code in result_cod:
for extend in result_ext_all:
if code["단축코드"] == extend["종목코드"]:
code.update(extend)
print("result_cod 건수", len(result_cod))
mongodb = MongoDBHandler()
mongodb.delete_items({}, "stock", "m_code_info")
mongodb.insert_items(result_cod, "stock", "m_code_info")
# collect_code_list()
# collect_stock_info()
return {
"errcode": 0,
"errmsg": str(len(result_cod)) + " 건이 정상처리 되었습니다."
}
# 종목가격정보 가져오기
elif id == 2:
ebest = EBest("DEMO")
ebest.login()
code_list = mongodb.find_items({}, "stock", "m_code_info")
target_code = set([item["단축코드"] for item in code_list])
today = datetime.today().strftime("%Y%m%d")
collect_list = mongodb.find_items({
"날짜": today
}, "stock", "price_info").distinct("code")
for col in collect_list:
target_code.remove(col)
for code in target_code:
result_price = ebest.get_stock_price_by_code(code, "1")
time.sleep(1)
if len(result_price) > 0:
mongodb.insert_items(result_price, "stock",
"price_info")
return {
"errcode": 0,
"errmsg": str(len(result)) + " 건이 정상처리 되었습니다."
}
# 모든 종목정보 가져오기
elif id == 3:
ebest = EBest("DEMO")
mongodb = MongoDBHandler()
ebest.login()
code_list = mongodb.find_items({}, "stock", "m_code_info")
target_code = set([item["단축코드"] for item in code_list])
today = datetime.today().strftime("%Y%m%d")
print(today)
collect_list = mongodb.find_items({
"날짜": today
}, "stock", "price_info").distinct("code")
for col in collect_list:
target_code.remove(col)
for code in target_code:
time.sleep(1)
print("code:", code)
result_price = ebest.get_stock_price_by_code(code, "1")
if len(result_price) > 0:
print(result_price)
mongodb.insert_items(result_price, "stock",
"price_info")
result_credit = ebest.get_credit_trend_by_code(code, today)
if len(result_credit) > 0:
mongodb.insert_items(result_credit, "stock",
"credit_info")
result_short = ebest.get_short_trend_by_code(code,
sdate=today,
edate=today)
if len(result_short) > 0:
mongodb.insert_items(result_short, "stock",
"short_info")
result_agent = ebest.get_agent_trend_by_code(code,
fromdt=today,
todt=today)
if len(result_agent) > 0:
mongodb.insert_items(result_agent, "stock",
"agent_info")
# 일자별 주가정보에서 평균 거래량 가져오기(거래량 증가 종목 찾기)
elif id == 4:
ebest = EBest("DEMO")
ebest.login()
mongodb = MongoDBHandler()
code_list = list(mongodb.find_items({}, "stock",
"m_code_info"))
print(len(code_list))
target_code = set([item["단축코드"] for item in code_list])
# print(target_code)
today = datetime.today().strftime("%Y%m%d")
fromday = (datetime.today() -
timedelta(days=30)).strftime("%Y%m%d")
inc_codes = []
vol_codes = []
loop_cnt = 0
for code in code_list:
if int(code["누적거래량"]) < 10000:
continue
loop_cnt = loop_cnt + 1
if loop_cnt % 100 == 0 and len(inc_codes) > 0:
print(str(loop_cnt % 100) + "//" + str(len(inc_codes)))
print(code["단축코드"] + "진행율 : " +
str((loop_cnt / len(code_list) * 100)))
mongodb.insert_items(inc_codes, "stock",
"check_volume")
inc_codes.clear()
results = ebest.get_stock_chart_by_code(
code["단축코드"], "2", fromday, today)
time.sleep(1)
if len(results) > 0:
# 평균 거래량 계산
tot_volume = 0
i_count = 0
for result in results:
if int(result['거래량']) != 0:
tot_volume = tot_volume + int(result['거래량'])
i_count = i_count + 1
if i_count == 0 or tot_volume == 0:
continue
avg_volume = tot_volume / i_count
inc_rate = int(result['거래량']) / avg_volume
inc_code = {
"code": code["단축코드"],
"종목명": code["종목명"],
"sdate": today,
"avg_volume": int(avg_volume),
"volume": int(result['거래량'])
}
vol_codes.append(inc_code)
print("체크 종목 :" + code["종목명"] + " 거래량 [" +
result['거래량'] + "] 평균 [" + str(avg_volume) +
"] 비율[" + str(inc_rate) + "]")
# 거래량이 5배 이상이면 종목 추가
if inc_rate > 5:
inc_codes.append(inc_code)
print("추가된 종목 :" + code["종목명"] + " 건수 : " +
str(len(inc_codes)))
if len(inc_codes) > 0:
mongodb.insert_items(inc_codes, "stock", "check_volume")
if len(vol_codes) > 0:
mongodb.insert_items(vol_codes, "stock", "volume")
return {
"errcode": 0,
"errmsg": str(len(vol_codes)) + " 건이 정상처리 되었습니다."
}
elif id == 5:
print("Error Id : " + str(id))
else:
print("Error Id : " + str(id))
else:
print("list of users")
# collect_code_list()
# collect_stock_info()
return "list of users"