def _basic_init(self):
self._page = rget(self._url)
if self._page.text[:800].find("Data_fundSharesPositions") >= 0:
raise FundTypeError(
"This code seems to be a fund, use fundinfo instead")
l = eval(
re.match(r".*Data_millionCopiesIncome = ([^;]*);.*",
self._page.text).groups()[0])
self.name = re.match(r".*fS_name = \"([^;]*)\";.*",
self._page.text).groups()[0]
tz_bj = dt.timezone(dt.timedelta(hours=8))
datel = [
dt.datetime.fromtimestamp(int(d[0]) / 1e3,
tz=tz_bj).replace(tzinfo=None) for d in l
]
ratel = [float(d[1]) for d in l]
netvalue = [1]
for dailyrate in ratel:
netvalue.append(netvalue[-1] * (1 + dailyrate * 1e-4))
netvalue.remove(1)
df = pd.DataFrame(
data={
"date": datel,
"netvalue": netvalue,
"totvalue": netvalue,
"comment": [0 for _ in datel],
})
df = df[df["date"].isin(opendate)]
df = df.reset_index(drop=True)
self.price = df[df["date"] <= yesterdaydash()]
def _basic_init(self):
self._page = rget(self._url)
if self._page.status_code == 404:
raise ParserFailure(
"Unrecognized fund, please check fund code you input.")
if self._page.text[:800].find("Data_millionCopiesIncome") >= 0:
raise FundTypeError(
"This code seems to be a mfund, use mfundinfo instead")
l = re.match(r"[\s\S]*Data_netWorthTrend = ([^;]*);[\s\S]*",
self._page.text).groups()[0]
l = l.replace("null", "None") # 暂未发现基金净值有 null 的基金，若有，其他地方也很可能出问题！
l = eval(l)
ltot = re.match(r"[\s\S]*Data_ACWorthTrend = ([^;]*);[\s\S]*",
self._page.text).groups()[0] # .* doesn't match \n
ltot = ltot.replace("null", "None") ## 096001 总值数据中有 null！
ltot = eval(ltot)
## timestamp transform tzinfo must be taken into consideration
tz_bj = dt.timezone(dt.timedelta(hours=8))
infodict = {
"date": [
dt.datetime.fromtimestamp(int(d["x"]) / 1e3,
tz=tz_bj).replace(tzinfo=None)
for d in l
],
"netvalue": [float(d["y"]) for d in l],
"comment": [_nfloat(d["unitMoney"]) for d in l],
}
if len(l) == len(ltot): # 防止总值和净值数据量不匹配，已知有该问题的基金：502010
infodict["totvalue"] = [d[1] for d in ltot]
try:
rate = float(
eval(
re.match(r"[\s\S]*fund_Rate=([^;]*);[\s\S]*",
self._page.text).groups()[0]))
except ValueError:
rate = 0
logger.info(
"warning: this fund has no data for rate") # know cases: ETF
name = eval(
re.match(r"[\s\S]*fS_name = ([^;]*);[\s\S]*",
self._page.text).groups()[0])
self.rate = rate
# shengou rate in tiantianjijin, daeshengou rate discount is not considered
self.name = name # the name of the fund
df = pd.DataFrame(data=infodict)
df = df[df["date"].isin(opendate)]
df = df.reset_index(drop=True)
if len(df) == 0:
raise ParserFailure("no price table found for this fund %s" %
self.code)
self.price = df[df["date"] <= yesterdaydash()]
# deal with the redemption fee attrs finally
if not self.priceonly:
self._feepreprocess()
def _basic_init(self):
self._page = _download(self._url)
if self._page.text[:800].find("Data_millionCopiesIncome") >= 0:
raise FundTypeError(
"This code seems to be a mfund, use mfundinfo instead")
l = eval(
re.match(r".*Data_netWorthTrend = ([^;]*);.*",
self._page.text).groups()[0])
ltot = eval(
re.match(r".*Data_ACWorthTrend = ([^;]*);.*",
self._page.text).groups()[0])
## timestamp transform tzinfo must be taken into consideration
tz_bj = dt.timezone(dt.timedelta(hours=8))
infodict = {
"date": [
dt.datetime.fromtimestamp(int(d["x"]) / 1e3,
tz=tz_bj).replace(tzinfo=None)
for d in l
],
"netvalue": [float(d["y"]) for d in l],
"comment": [_nfloat(d["unitMoney"]) for d in l],
}
if len(l) == len(ltot): # 防止总值和净值数据量不匹配，已知有该问题的基金：502010
infodict["totvalue"] = [d[1] for d in ltot]
try:
rate = float(
eval(
re.match(r".*fund_Rate=([^;]*);.*",
self._page.text).groups()[0]))
except ValueError:
rate = 0
if _do_print_warning:
print("warning: this fund has no data for rate"
) # know cases: 510030
name = eval(
re.match(r".*fS_name = ([^;]*);.*", self._page.text).groups()[0])
self.rate = rate
# shengou rate in tiantianjijin, daeshengou rate discount is not considered
self.name = name # the name of the fund
df = pd.DataFrame(data=infodict)
df = df[df["date"].isin(opendate)]
df = df.reset_index(drop=True)
self.price = df[df["date"] <= yesterdaydash()]
# deal with the redemption fee attrs finally
self._feepreprocess()
def _basic_init(self):
self._page = _download(self._url)
if self._page.text[:800].find("Data_fundSharesPositions") >= 0:
raise FundTypeError("This code seems to be a fund, use fundinfo instead")
parser = Parser()
tree = parser.parse(self._page.text)
nodenet = [
node.children()[0].children()[1]
for node in nodevisitor.visit(tree)
if isinstance(node, ast.VarStatement)
and node.children()[0].children()[0].value == "Data_millionCopiesIncome"
][0]
name = [
node.children()[0].children()[1]
for node in nodevisitor.visit(tree)
if isinstance(node, ast.VarStatement)
and (node.children()[0].children()[0].value == "fS_name")
][0]
self.name = name.value.strip('"')
tz_bj = dt.timezone(dt.timedelta(hours=8))
datel = [
dt.datetime.fromtimestamp(
int(nodenet.children()[i].children()[0].value) / 1e3, tz=tz_bj
).replace(tzinfo=None)
for i in range(len(nodenet.children()))
]
ratel = [
float(nodenet.children()[i].children()[1].value)
for i in range(len(nodenet.children()))
]
netvalue = [1]
for dailyrate in ratel:
netvalue.append(netvalue[-1] * (1 + dailyrate * 1e-4))
netvalue.remove(1)
df = pd.DataFrame(
data={
"date": datel,
"netvalue": netvalue,
"totvalue": netvalue,
"comment": [0 for _ in datel],
}
)
df = df[df["date"].isin(opendate)]
df = df.reset_index(drop=True)
self.price = df[df["date"] <= yesterdaydash()]
def __init__(
self,
code,
round_label=0,
dividend_label=0,
fetch=False,
save=False,
path="",
form="csv",
priceonly=False,
):
if round_label == 1 or (code in droplist):
label = 1 # the scheme of round down on share purchase
else:
label = 0
if code.startswith("F") and code[1:].isdigit():
code = code[1:]
elif code.startswith("M") and code[1:].isdigit():
raise FundTypeError(
"This code seems to be a mfund, use ``mfundinfo`` instead")
self._url = ("http://fund.eastmoney.com/pingzhongdata/" + code + ".js"
) # js url api for info of certain fund
self._feeurl = ("http://fund.eastmoney.com/f10/jjfl_" + code + ".html"
) # html url for trade fees info of certain fund
self.priceonly = priceonly
super().__init__(
code,
fetch=fetch,
save=save,
path=path,
form=form,
round_label=label,
dividend_label=dividend_label,
)
self.special = self.price[self.price["comment"] != 0]
self.specialdate = list(self.special["date"])
# date with nonvanishing comment, usually fenhong or zhesuan
try:
self.fenhongdate = list(
self.price[self.price["comment"] > 0]["date"])
self.zhesuandate = list(
self.price[self.price["comment"] < 0]["date"])
except TypeError:
print("There are still string comments for the fund!")
def _fetch_sql(self, path):
"""
fetch the information and pricetable from sql, not recommend to use manually,
just set the fetch label to be true when init the object
:param path: engine object from sqlalchemy
"""
try:
content = pd.read_sql("xa" + self.code, path)
pricetable = content.iloc[1:]
commentl = [float(com) for com in pricetable.comment]
self.price = pricetable[["date", "netvalue", "totvalue"]]
self.price["comment"] = commentl
saveinfo = json.loads(content.iloc[0].comment)
if not isinstance(saveinfo, dict):
raise FundTypeError("This csv doesn't looks like from fundinfo")
self.segment = saveinfo["segment"]
self.feeinfo = saveinfo["feeinfo"]
self.name = saveinfo["name"]
self.rate = saveinfo["rate"]
except exc.ProgrammingError as e:
# print('no saved copy of %s' % self.code)
raise e
def _fetch_csv(self, path):
"""
fetch the information and pricetable from path+code.csv, not recommend to use manually,
just set the fetch label to be true when init the object
:param path: string of folder path
"""
try:
content = pd.read_csv(path + self.code + ".csv")
pricetable = content.iloc[1:]
datel = list(pd.to_datetime(pricetable.date))
self.price = pricetable[["netvalue", "totvalue", "comment"]]
self.price["date"] = datel
saveinfo = json.loads(content.iloc[0].date)
if not isinstance(saveinfo, dict):
raise FundTypeError("This csv doesn't looks like from fundinfo")
self.segment = saveinfo["segment"]
self.feeinfo = saveinfo["feeinfo"]
self.name = saveinfo["name"]
self.rate = saveinfo["rate"]
except FileNotFoundError as e:
# print('no saved copy of fund %s' % self.code)
raise e
def _basic_init(self):
self._page = _download(self._url)
if self._page.text[:800].find("Data_millionCopiesIncome") >= 0:
raise FundTypeError("This code seems to be a mfund, use mfundinfo instead")
parser = Parser() # parse the js text of API page using slimit module
tree = parser.parse(self._page.text)
nodenet = [
node.children()[0].children()[1]
for node in nodevisitor.visit(tree)
if isinstance(node, ast.VarStatement)
and node.children()[0].children()[0].value == "Data_netWorthTrend"
][0]
nodetot = [
node.children()[0].children()[1]
for node in nodevisitor.visit(tree)
if isinstance(node, ast.VarStatement)
and node.children()[0].children()[0].value == "Data_ACWorthTrend"
][0]
## timestamp transform tzinfo must be taken into consideration
tz_bj = dt.timezone(dt.timedelta(hours=8))
infodict = {
"date": [
dt.datetime.fromtimestamp(
int(nodenet.children()[i].children()[0].right.value) / 1e3, tz=tz_bj
).replace(tzinfo=None)
for i in range(len(nodenet.children()))
],
"netvalue": [
float(nodenet.children()[i].children()[1].right.value)
for i in range(len(nodenet.children()))
],
"comment": [
_nfloat(nodenet.children()[i].children()[3].right.value)
for i in range(len(nodenet.children()))
],
}
if len(nodenet.children()) == len(
nodetot.children()
): # 防止总值和净值数据量不匹配，已知有该问题的基金：502010
infodict["totvalue"] = [
float(nodetot.children()[i].children()[1].value)
for i in range(len(nodenet.children()))
]
rate = [
node.children()[0].children()[1]
for node in nodevisitor.visit(tree)
if isinstance(node, ast.VarStatement)
and (node.children()[0].children()[0].value == "fund_Rate")
][0]
name = [
node.children()[0].children()[1]
for node in nodevisitor.visit(tree)
if isinstance(node, ast.VarStatement)
and (node.children()[0].children()[0].value == "fS_name")
][0]
self.rate = float(
rate.value.strip('"')
) # shengou rate in tiantianjijin, daeshengou rate discount is not considered
self.name = name.value.strip('"') # the name of the fund
df = pd.DataFrame(data=infodict)
df = df[df["date"].isin(opendate)]
df = df.reset_index(drop=True)
self.price = df[df["date"] <= yesterdaydash()]
# deal with the redemption fee attrs finally
self._feepreprocess()
