def _queryFromDB(sql):
try:
result = pd.read_sql_query(sql, con=ENGINE)
return result
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("read from db error!")
return None
def readFile(filePath, encoding='utf8'):
try:
fp = open(filePath, mode='r', encoding=encoding)
content = fp.read()
return content
except Exception as ex:
FileLogger.error(ex)
return False
def _refineColumns(datadf, columns=[]):
if datadf.index.name != 'REPORT_DATE':
FileLogger.error("REPORT_DATE must be the index!")
elif len(columns) == 0:
for col in datadf.columns:
if col not in TEXTCOLUMNS:
columns.append(col)
return columns
def downloadFile(url, filePath):
try:
r = requests.get(url)
with open(filePath, "wb") as fp:
fp.write(r.content)
except Exception as ex:
FileLogger.error("downloadfile error on path: %s" % url)
FileLogger.error(ex)
return False
def readFile(filePath):
try:
fp = open(filePath, 'r')
content = fp.read()
return content
except Exception as ex:
FileLogger.error("read file error on path: %s" % filePath)
FileLogger.error(ex)
return False
def write2File(filePath, content, mode="w+") -> bool:
try:
fp = open(filePath, mode)
fp.write(content)
fp.flush()
fp.close()
return True
except Exception as ex:
FileLogger.error("write to file error on path: %s" % filePath)
FileLogger.error(ex)
return False
def queryFromDB(sql):
try:
engine = create_engine(
"mysql+pymysql://root:4401821211@localhost:3306/stock?charset=utf8"
)
result = pd.read_sql_query(sql, con=engine)
return result
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("read from db error!")
return None
def crawlUSStocks():
# 查询语句：select ts_code from usstock.stocklist;
stockList = pd.read_csv("C:/project/Tushare/usstock/code.csv").to_numpy()
for code in stockList:
FileLogger.info("running on code: " + code[0])
try:
crawlHistory(code[0])
time.sleep(1)
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("crawl error on code: %s" % code)
time.sleep(5)
def keepOnlyQuarterData(datadf, quarter):
if datadf.index.name != 'REPORT_DATE':
return None
if quarter == 1:
return datadf[datadf['REPORT_DATE'].str.find('-03-31') != -1]
elif quarter == 2:
return datadf[datadf['REPORT_DATE'].str.find('-06-30') != -1]
elif quarter == 3:
return datadf[datadf['REPORT_DATE'].str.find('-09-30') != -1]
elif quarter == 4:
return datadf[datadf['REPORT_DATE'].str.find('-12-31') != -1]
else:
FileLogger.error("error quarter parameter!")
return None
def gettodayStock():
curDate = time.strftime("%Y%m%d", time.localtime())
tryagain = True
while tryagain:
try:
content = crawlLatestUsStocks()
if content:
path = "C:/project/stockdata/USDay/%s.txt" % curDate
write2File(path, content, mode="w")
FileLogger.info("crawl stock list successfully on date:" +
curDate)
tryagain = False
else:
time.sleep(60)
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("crawl stock list error, retry in 60 seconds")
time.sleep(60)
def crawlHistory(code) -> bool:
link = "https://stock.xueqiu.com/v5/stock/chart/kline.json?symbol=%s&begin=1616585707592&period=day&type=before&count=-100000&indicator=kline,pe,pb,ps,pcf,market_capital,agt,ggt,balance" % code
session = HTMLSession()
r = session.get(link, headers=HEADERS, cookies=COOKIES)
jsonObj = json.loads(r.content)
if jsonObj['error_code'] != 0 or not jsonObj["data"].__contains__("column") or not jsonObj["data"].__contains__("item"):
FileLogger.error("get content error from: %s" % code)
return False
columns = jsonObj["data"]["column"]
items = jsonObj["data"]["item"]
if len(items) > 0:
path = "C:/project/stockdata/UShistory/%s.csv" % code
save2csv(columns, items, path)
FileLogger.info("get %d lines from code: %s" % (len(items), code))
return True
def generateQuarterYRateOnData(code, dataList, columnName):
sqlTemplate = "update `stock`.`incomerate` set `%s_yrate` = %f where `ts_code` = '%s' and `end_date` = '%s' and `report_type` = '1'"
lastAcculatedValue = 1
historyValues = {}
rate = 0
try:
for item in dataList:
date = item[0]
acculatedValue = item[1]
if acculatedValue is None:
continue
# 计算季度比值，要先算出本季度的营收，income报表中都是累加的季度收益
value = acculatedValue - lastAcculatedValue
historyValues[date] = value
endDate = datetime.datetime.strptime(date, "%Y%m%d")
lastYearDate = datetime.datetime(year=endDate.year - 1,
month=endDate.month,
day=endDate.day)
lastYear = lastYearDate.strftime("%Y%m%d")
lastYearValue = historyValues[
lastYear] if lastYear in historyValues else None
if lastYearValue:
rate = (value / lastYearValue - 1) * 100
sql = sqlTemplate % (columnName, rate, code, date)
engine.execute(sql)
# print("%s %d %d %d %d %f" % (date, lastAcculatedValue, acculatedValue, lastYearValue, value, rate))
if endDate.month == 12:
lastAcculatedValue = 0
else:
lastAcculatedValue = acculatedValue
except Exception as ex:
FileLogger.error(ex)
FileLogger.error(
"write to DB for generateQuarterYRateOnData error on sql: %s" %
sql)
def generateQuarterRateOnData(code, dataList, columnName):
sqlTemplate = "update `stock`.`incomerate` set `%s_rate` = %f where `ts_code` = '%s' and `end_date` = '%s' and `report_type` = '1'"
lastDate = datetime.datetime.strptime('19900101', "%Y%m%d")
lastValue = 1
lastAcculatedValue = 0
rate = 0
try:
for item in dataList:
date = item[0]
acculatedValue = item[1]
if acculatedValue is None or math.isnan(acculatedValue):
continue
value = acculatedValue - lastAcculatedValue
endDate = datetime.datetime.strptime(date, "%Y%m%d")
delta = endDate - lastDate
if delta > datetime.timedelta(days=135):
# do nothing, rate should be none
pass
else:
# 计算季度比值，要先算出本季度的营收，income报表中都是累加的季度收益
rate = (value / lastValue - 1) * 100
sql = sqlTemplate % (columnName, rate, code, date)
engine.execute(sql)
print("%d %d %d %d %f" %
(lastAcculatedValue, acculatedValue, lastValue, value, rate))
lastDate = endDate
lastValue = value
if endDate.month == 12:
lastAcculatedValue = 0
else:
lastAcculatedValue = acculatedValue
except Exception as ex:
FileLogger.error(ex)
FileLogger.error(
"write to DB for generateQuarterRateOnData error on sql: %s" % sql)
def generateQuarterData(code, dataList, columnName):
sqlTemplate = "update `stock`.`income` set `%s_qtr` = %f where `ts_code` = '%s' and `end_date` = '%s' and `report_type` = '1'"
try:
for index, row in dataList.iterrows():
endDate = datetime.datetime.strptime(row['end_date'], "%Y%m%d")
lastEndDate = None
lastValue = 0
quarterValue = 0
if endDate.month == 3 and endDate.day == 31:
quarterValue = row[columnName]
elif endDate.month == 6 and endDate.day == 30:
lastEndDate = '%d0331' % endDate.year
elif endDate.month == 9 and endDate.day == 30:
lastEndDate = '%d0630' % endDate.year
elif endDate.month == 12 and endDate.day == 31:
lastEndDate = '%d0930' % endDate.year
if lastEndDate:
lastRow = dataList[dataList['end_date'] ==
lastEndDate].to_numpy()
if len(lastRow) > 0:
lastValue = lastRow[0][1]
quarterValue = row[columnName] - lastValue
print("%s %d %d" %
(row['end_date'], row[columnName], quarterValue))
sql = sqlTemplate % (columnName, quarterValue, code,
row['end_date'])
engine.execute(sql)
except Exception as ex:
FileLogger.error(ex)
FileLogger.error(
"write to DB for generateQuarterRateOnData error on sql: %s" % sql)
FileLogger.info("get %d records on code: %s" % (len(records), code))
if len(records) != 0:
content = json.dumps(records)
path = "C:/project/stockdata/StockNotices/%s.json" % code
write2File(path, content)
if __name__ == "__main__":
stockList = getJsonFromFile("C:/project/stockdata/StockNotices/stock.json")
stockList = stockList["stockList"]
# stockList = [{"orgId":"9900002701","category":"A股","code":"002127","pinyin":"njds","zwjc":"南极电商"}]
for stock in stockList:
FileLogger.info("running on stock: %s(%s)" %
(stock["zwjc"], stock["code"]))
filePath = "C:/project/stockdata/StockNotices/%s.json" % stock['code']
if (os.path.exists(filePath)):
continue
try:
crawlStockNotices(stock["code"], stock["orgId"])
time.sleep(1)
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("crawl balance error on code: %s" % stock["code"])
time.sleep(3)
if __name__ == "__main__":
# http://www.cninfo.com.cn/new/announcement/bulletin_detail?announceId=13519195&flag=true&announceTime=2004-01-17
# retrieveAnualQuarterlyReport()
stockdf = pd.read_csv("C:/project/stockdata/StockNoticesFile/annualreportlist.csv", dtype={'code': np.str, 'year': np.str})
# stockdf = stockdf[stockdf['code'] == '000002']
stockList = stockdf[['code', 'name', 'year', 'announcementId', 'url']].to_numpy()
# stockList = stockList[1:3]
try:
for stock in stockList:
fileName = "[%s]%s年报-%s" % (stock[1], stock[2], stock[3])
savePath = "C:/project/stockdata/StockNoticesFile/pdf/%s.pdf" % fileName
# make sure it's a valid path, no \/:?*"<>|
savePath = savePath.replace("*", "")
unresolvedPath = "C:/project/stockdata/StockNoticesFile/unresolved/%s.pdf" % fileName
url = stock[4]
if os.path.exists(savePath) or os.path.exists(unresolvedPath):
FileLogger.info("file %s exists, skip!" % fileName)
else:
FileLogger.info("downloading file: %s" % fileName)
downloadFile(url, savePath)
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("download error on file: %s" % fileName)
time.sleep(3)
def retrieveAnualQuarterlyReport():
stockList = getJsonFromFile("C:/project/stockdata/StockNotices/stock.json")
stockList = stockList["stockList"]
stockList = [{"orgId":"9900002701","category":"A股","code":"002127","pinyin":"njds","zwjc":"南极电商"}]
# stockList = [{"orgId":"gssz0000002","category":"A股","code":"000002","pinyin":"njds","zwjc":"万科A"}]
for stock in stockList:
FileLogger.info("running on stock: %s(%s)" % (stock["zwjc"], stock["code"]))
try:
filePath = "C:/project/stockdata/StockNotices/%s.json" % stock['code']
jsonList = getJsonFromFile(filePath)
annualDf = None
for jsonObj in jsonList:
announcementType = jsonObj['announcementType']
fileType = jsonObj['adjunctType']
# 得到公告类型，一季报半年报三季报年报
# 公告类型：{'01030501': 第一季度报全文, '01030701':第三季度报, '01030301': 半年报, '01030101':年报全文}
noticeType = None
if announcementType.find("01030101") != -1:
noticeType = "年报"
elif announcementType.find("01030701") != -1:
noticeType = "三季度报"
elif announcementType.find("01030301") != -1:
noticeType = "半年报"
elif announcementType.find("01030501") != -1:
noticeType = "一季度报"
if noticeType is not None and (fileType == 'PDF' or filePath == 'PDF ' or fileType == 'pdf'):
FileLogger.info("downloading file: %s" % jsonObj["announcementTitle"])
noticeDay = jsonObj['adjunctUrl'][10:20]
url = "http://www.cninfo.com.cn/new/announcement/download?bulletinId=%s&announceTime=%s" % (jsonObj['announcementId'], noticeDay)
annualData = {
'code': jsonObj['secCode'],
'name': jsonObj['secName'],
'announcementId': jsonObj['announcementId'],
'title': jsonObj['announcementTitle'],
'noticeDay': noticeDay,
'fileType': jsonObj['adjunctType'],
'url': url,
'Type': noticeType,
'year': int(noticeDay[0:4])-1 if noticeType == "年报" else int(noticeDay[0:4])
}
if annualDf is None:
annualDf = pd.DataFrame(columns=annualData.keys())
annualDf = annualDf.append(annualData, ignore_index=True)
else:
annualDf = annualDf.append(annualData, ignore_index=True)
time.sleep(0)
# save to DB
from sqlalchemy import create_engine
ENGINE = create_engine("mysql+pymysql://root:4401821211@localhost:3306/eastmoney?charset=utf8")
annualDf.to_sql(name="reportbasic", con=ENGINE, if_exists="append")
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("retrieve error on code: %s" % stock["code"])
time.sleep(3)
try:
for item in incomeList:
endDate = datetime.datetime.strptime(item[0], "%Y%m%d")
delta = endDate - lastDate
if delta > datetime.timedelta(days=135):
# do nothing, rate should be none
pass
else:
sql = sqlTemplate % (40, code, item[0])
print(sql)
engine.execute(sql)
lastDate = endDate
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("write to DB error on sql: %s" % sql)
# %%
income = DBLib.getIncomeFromDB('600176.SH')
sortedIncome = income.sort_values(by='end_date', ascending=True)
dataList = sortedIncome[['end_date', 'total_revenue']]
# %%
for index, row in dataList.iterrows():
print(row['end_date'])
# %%
row = dataList[dataList['end_date']=='20130630'].to_numpy()
gatherColumnInfo(jsonSql, jsonObjects, 'TOTAL_CURRENT_LIAB',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'TOTAL_EQUITY', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'TOTAL_LIAB_EQUITY',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'TOTAL_LIABILITIES',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'TOTAL_NONCURRENT_ASSETS',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'TOTAL_NONCURRENT_LIAB',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'TOTAL_OTHER_PAYABLE',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'TOTAL_OTHER_RECE',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'TOTAL_PARENT_EQUITY',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'TRADE_FINASSET_NOTFVTPL',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'UNASSIGN_RPOFIT', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'USERIGHT_ASSET', 'double')
executeSql(jsonSql)
time.sleep(0.1)
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("parse balance error on code: %s" % code)
time.sleep(3)
def write2File(filePath, content, mode="w+") -> bool:
try:
fp = open(filePath, mode)
fp.write(content)
fp.flush()
fp.close()
return True
except Exception as ex:
FileLogger.error("write to file error on path: %s" % filePath)
FileLogger.error(ex)
return False
if __name__ == "__main__":
# 查询语句：select ts_code from usstock.stocklist;
stockdf = pd.read_csv("C:/project/Tushare/usstock/code.csv")
errordf = pd.read_csv("C:/project/Tushare/usstock/get_error_ts_code.csv")
errorList = errordf['ts_code'].to_numpy()
stockList = stockdf[~stockdf['ts_code'].isin(errorList)]
stockList = stockList['ts_code'].to_numpy()
for code in stockList:
FileLogger.info("running on code: %s" % code)
try:
crawlCashflow(code)
time.sleep(1)
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("crawl cashflow error on code: %s" % code)
time.sleep(3)
gatherColumnInfo(jsonSql, jsonObjects, 'OPERATE_PROFIT', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'NONBUSINESS_INCOME', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'NONBUSINESS_EXPENSE', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'TOTAL_PROFIT', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'INCOME_TAX', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'NETPROFIT', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'CONTINUED_NETPROFIT', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'PARENT_NETPROFIT', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'MINORITY_INTEREST', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'DEDUCT_PARENT_NETPROFIT', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'BASIC_EPS', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'DILUTED_EPS', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'OTHER_COMPRE_INCOME', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'PARENT_OCI', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'MINORITY_OCI', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'TOTAL_COMPRE_INCOME', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'PARENT_TCI', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'MINORITY_TCI', 'double')
executeSql(jsonSql)
time.sleep(0.1)
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("parse income error on code: %s" % code)
time.sleep(3)
def write2File(filePath, content, mode="w+") -> bool:
try:
fp = open(filePath, mode)
fp.write(content)
fp.flush()
fp.close()
return True
except Exception as ex:
FileLogger.error("write to file error on path: %s" % filePath)
FileLogger.error(ex)
return False
if __name__ == "__main__":
# 查询语句：select ts_code from usstock.stocklist;
stockdf = pd.read_csv("C:/project/Tushare/usstock/code.csv")
errordf = pd.read_csv("C:/project/Tushare/usstock/get_error_ts_code.csv")
errorList = errordf['ts_code'].to_numpy()
stockList = stockdf[~stockdf['ts_code'].isin(errorList)]
stockList = stockList['ts_code'].to_numpy()
for code in stockList:
FileLogger.info("running on code: %s" % code)
try:
crawlBalance(code)
time.sleep(1)
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("crawl balance error on code: %s" % code)
time.sleep(3)
try:
fp = open(filePath, mode)
fp.write(content)
fp.flush()
fp.close()
return True
except Exception as ex:
FileLogger.error("write to file error on path: %s" % filePath)
FileLogger.error(ex)
return False
if __name__ == "__main__":
# 查询语句：select ts_code from usstock.stocklist;
stockdf = pd.read_csv("C:/project/Tushare/usstock/code.csv")
errordf = pd.read_csv("C:/project/Tushare/usstock/get_error_ts_code.csv")
errorList = errordf['ts_code'].to_numpy()
stockList = stockdf[~stockdf['ts_code'].isin(errorList)]
stockList = stockList['ts_code'].to_numpy()
for code in stockList:
FileLogger.info("running on code: %s" % code)
try:
crawlDivident(code)
time.sleep(1)
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("crawl income error on code: %s" % code)
time.sleep(3)
incomeBaseDF.to_sql(name='incomebase', con=usEngine, if_exists='append')
def readFile(filePath):
try:
fp = open(filePath, 'r')
content = fp.read()
return content
except Exception as ex:
FileLogger.error("read file error on path: %s" % filePath)
FileLogger.error(ex)
return False
if __name__ == "__main__":
stockdf = pd.read_csv("C:/project/stockdata/USIncome/code.csv")
stockList = stockdf['code'].to_numpy()
# parseIncomeBaseList(stockList)
for code in stockList:
FileLogger.info("running on code: %s" % code)
try:
parseUSIncome(code)
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("write data to Database error on code: %s" % code)
time.sleep(1)
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'FINANCE_EXPENSE', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'INVEST_LOSS', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'DEFER_TAX', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'DT_ASSET_REDUCE', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'DT_LIAB_ADD', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'INVENTORY_REDUCE',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'OPERATE_RECE_REDUCE',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'OPERATE_PAYABLE_ADD',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'OPERATE_NETCASH_OTHERNOTE',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'NETCASH_OPERATENOTE',
'double')
gatherColumnInfo(jsonSql, jsonObjects, 'END_CASH', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'BEGIN_CASH', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'CCE_ADDNOTE', 'double')
gatherColumnInfo(jsonSql, jsonObjects, 'OPINION_TYPE',
'varchar(100)')
executeSql(jsonSql)
time.sleep(0.1)
except Exception as ex:
FileLogger.error(ex)
FileLogger.error("parse cashflow error on code: %s" % code)
time.sleep(3)
