def writeRSIToDB(period, stockCode, stockName, rsi_df):
dataList = []
for index, rsi in rsi_df.iterrows():
rsiDate = rsi['date']
if period == "day":
rsiObj = DayRSI(stockCode, stockName)
elif period == "week":
rsiObj = WeekRSI(stockCode, stockName)
elif period == "month":
rsiObj = MonthRSI(stockCode, stockName)
elif period == "5m":
rsiObj = FiveMinRSI(stockCode, stockName)
elif period == "15m":
rsiObj = FiftyMinRSI(stockCode, stockName)
elif period == "30m":
rsiObj = ThirtyMinRSI(stockCode, stockName)
elif period == "60m":
rsiObj = SixtyMinRSI(stockCode, stockName)
rsiObj.date = rsiDate
rsiObj.rsi_6 = rsi['rsi_6']
rsiObj.rsi_12 = rsi['rsi_12']
rsiObj.rsi_24 = rsi['rsi_24']
rsiObj.overBuy = rsi['overBuyFlag']
rsiObj.overSell = rsi['overSellFlag']
dataList.append(rsiObj.__dict__)
mydb = dbutil.connectDB()
collection = mydb[chooseRSICollection(period)]
if len(dataList) > 0:
collection.insert_many(dataList)
else:
raise RuntimeError("RSI数据为空")
def clearRSI(period):
mydb = dbutil.connectDB()
collection = mydb[chooseRSICollection(period)]
collection.delete_many({})
indexes = collection.index_information()
if "code_1_date_1" in indexes.keys():
collection.drop_index("code_1_date_1")
def findLatestRSIDate(period):
mydb = dbutil.connectDB()
collection = mydb[chooseRSICollection(period)]
cursor = collection.find().sort("date", -1).limit(1)
df = pd.DataFrame(list(cursor))
if df.empty:
return "1970-01-01"
return df["date"][0]
def clearKlineData(period, startDate):
mydb = dbutil.connectDB()
collection = mydb[chooseKlineCollection(period)]
startDate = datetime.strptime(startDate + "T00:00:00.000Z", "%Y-%m-%dT%H:%M:%S.000Z")
collection.delete_many({"date":{"$gte":startDate}})
indexes = collection.index_information()
if "code_1_date_1" in indexes.keys():
collection.drop_index( "code_1_date_1" )
def allStocks():
global stocks
if stocks != None:
return stocks
mydb = dbutil.connectDB()
cursor = mydb["Stock"].find({})
df = pd.DataFrame(list(cursor))
df = df.set_index("code")
# set_trace()
stocks = df.to_dict('index')
return stocks
def readStockKline(code, period, startDate, endDate):
mydb = dbutil.connectDB()
cursor = None
periodCollection = chooseKlineCollection(period)
startDate = datetime.strptime(startDate + "T00:00:00.000Z", "%Y-%m-%dT%H:%M:%S.000Z")
endDate = datetime.strptime(endDate + "T23:59:59.000Z", "%Y-%m-%dT%H:%M:%S.000Z")
query = {"date":{"$gte":startDate, "$lte":endDate}}
if code != None and len(code) > 0:
query["code"] = code
cursor = mydb[chooseKlineCollection(period)].find(query)
df = pd.DataFrame(list(cursor))
return df
def initAccount(acctCount, equity):
if acctCount <= 0:
raise RuntimeError("初始化账户数得大于0")
dateStr = datetime.datetime.strftime(datetime.datetime.now(), "%m%d")
mydb = dbutil.connectDB()
collection = mydb["Account"]
dataList = []
for i in range(0, acctCount):
newAcct = Account(
str(uuid.uuid1()).replace("-", ""),
dateStr + "account" + str(i + 1))
newAcct.initiatedEquity = equity
newAcct.equity = equity
dataList.append(newAcct.__dict__)
collection.insert_many(dataList)
return True
def readRSI(period, stockCode, startDate, endDate):
mydb = dbutil.connectDB()
collection = mydb[chooseRSICollection(period)]
if type(startDate) == str:
startDate = datetime.datetime.strptime(startDate + "T00:00:00.000Z",
"%Y-%m-%dT%H:%M:%S.000Z")
endDate = datetime.datetime.strptime(endDate + "T23:59:59.000Z",
"%Y-%m-%dT%H:%M:%S.000Z")
cursor = collection.find({
"code": stockCode,
"date": {
"$gte": startDate,
"$lte": endDate
}
})
df = pd.DataFrame(list(cursor))
return df
def downloadAllStocks(tradeDate):
customLogin()
# set_trace()
# stock_rs = bs.query_all_stock(tradeDate)
stock_rs = bs.query_stock_basic()
stock_df = stock_rs.get_data()
dataList = []
for index,stock in stock_df.iterrows():
stockObj = Stock(stock["code"], stock["code_name"])
stockObj.stockType = stock["type"]
dataList.append(stockObj.__dict__)
mydb = dbutil.connectDB()
mydb["Stock"].delete_many({})
mydb["Stock"].insert_many(dataList)
customLogout()
return True
def loadAccounts():
mydb = dbutil.connectDB()
collection = mydb["Account"]
cursor = collection.find({})
df = pd.DataFrame(list(cursor))
return df
def createIndex(period):
mydb = dbutil.connectDB()
collection = mydb[chooseRSICollection(period)]
collection.create_index([("code", 1), ("date", 1)])
import dbutil
print('start')
con = None
cursor = None
try:
# 1. SQLite에 접속한다.
dbutil.connectDB('addr.db')
print('SQLite Connected')
# 2. Table을 생성한다.
dbutil.makeTable()
print('Table created')
# 3. 사용자 정보를 입력한다.
user = ['id06', 'pwd06', '홍말숙', '01077776666', '경기', 20]
dbutil.insertUser(user)
# 4. 사용자 정보를 조회한다.
allusers = dbutil.selectUser()
for user in allusers:
print('%s %s %s %s %s %d' % (user[0], user[1], user[2], user[3], user[4], user[5]))
except:
print('Error')
finally:
# 5. SQLite를 close 한다.
dbutil.closeDB()
import sqlite3
import dbutil
print('Start ....')
while True:
dbutil.connectDB('hb.db')
try:
menu = input('***DB만들기***\n'
'테이블만들기(m)\n'
'사용자정보추가(i)\n'
'사용자전체정보조회(a)\n'
'사용자정보조회(s)\n'
'사용자정보삭제(d)\n'
'사용자정보수정(u)\n'
'나가기(q)')
menu = menu.lower()
if menu == 'q':
print('Bye')
break
#테이블 만들기
if menu == 'm':
name = input('만드실 테이블 명을 적어주십시오')
dbutil.makeTable(name)
#사용자정보추가
if menu == 'i':
user = input('테이블이름과 추가할 사용자의 정보를 입력하세요\n'
'tableName,id,pwd,name,phone,addr,age\n'
'(띄어쓰기기준)')
user = user.split(' ')
userdata = []
for u in user:
def writeKlineToDb(period, stockCode, stockName, resultSet):
dataList = []
#分钟线多了个time字段，偏移量+1
offset = 1 if period.endswith("m") else 0
while (resultSet.error_code == '0') & resultSet.next():
# 获取一条记录，将记录合并在一起
# data_list.append(rs.get_row_data())
row = resultSet.get_row_data()
kline = None
recordDate = None
if period == "day":
kline = DayKline(stockCode, stockName)
elif period == "week":
kline = WeekKline(stockCode, stockName)
elif period == "month":
kline = MonthKline(stockCode, stockName)
elif period == "5m":
kline = FiveMinKline(stockCode, stockName)
elif period == "15m":
kline = FiftyMinKline(stockCode, stockName)
elif period == "30m":
kline = ThirtyMinKline(stockCode, stockName)
elif period == "60m":
kline = SixtyMinKline(stockCode, stockName)
else:
raise RuntimeError("还不支持这个周期：" + period)
# recordDate = datetime.strptime(row[0], "%Y-%m-%d")
kline.openPrice = row[1 + offset]
kline.highPrice = row[2 + offset]
kline.lowPrice = row[3 + offset]
kline.closePrice = row[4 + offset]
kline.volume = row[5 + offset]
kline.amount = row[6 + offset]
kline.adjustflag = row[7 + offset]
# set_trace()
# 日K、月K、周K有专有属性
if period == "day" or period == "week" or period == "month":
recordDate = datetime.strptime(row[0 + offset], "%Y-%m-%d")
kline.turn = row[8 + offset]
kline.changePercent = row[9 + offset]
else:
recordDate = datetime.strptime(row[0 + offset], "%Y%m%d%H%M%S000")
# 日K专有属性
if period == "day":
kline.preClosePrice = row[10]
kline.tradeStatus = row[11]
kline.isST = row[12]
kline.date = recordDate
dataList.append(kline.__dict__)
mydb = dbutil.connectDB()
collection = mydb[chooseKlineCollection(period)]
if len(dataList) > 0:
collection.insert_many(dataList)
if period == "day":
latestKlineDay = dataList[len(dataList) - 1]
query = {"code":latestKlineDay["code"]}
newvalues = { "$set": { "isST": latestKlineDay["isST"] == "1"} }
mydb["Stock"].update_one(query, newvalues)
else:
raise RuntimeError("数据为空")