def data(self) -> List[CalVT]:
self.check()
# If cache is enabled, then return cache directly
if self.enable_read_cache:
key = "orig_file" + str(self.uri)
if key not in H["c"]:
H["c"][key] = self._read_calendar()
_calendar = H["c"][key]
else:
_calendar = self._read_calendar()
if Freq(self._freq_file) != Freq(self.freq):
_calendar = resam_calendar(
np.array(list(map(pd.Timestamp, _calendar))), self._freq_file,
self.freq, self.region)
return _calendar
def _freq_file(self) -> str:
"""the freq to read from file"""
if not hasattr(self, "_freq_file_cache"):
freq = Freq(self.freq)
if freq not in self.support_freq:
# NOTE: uri
# 1. If `uri` does not exist
# - Get the `min_uri` of the closest `freq` under the same "directory" as the `uri`
# - Read data from `min_uri` and resample to `freq`
freq = Freq.get_recent_freq(freq, self.support_freq)
if freq is None:
raise ValueError(f"can't find a freq from {self.support_freq} that can resample to {self.freq}!")
self._freq_file_cache = freq
return self._freq_file_cache
def data(self) -> List[CalVT]:
# NOTE: uri
# 1. If `uri` does not exist
# - Get the `min_uri` of the closest `freq` under the same "directory" as the `uri`
# - Read data from `min_uri` and resample to `freq`
try:
self.check()
_calendar = self._read_calendar()
except ValueError:
freq_list = self._get_storage_freq()
_freq = Freq.get_recent_freq(self.freq, freq_list)
if _freq is None:
raise ValueError(
f"can't find a freq from {freq_list} that can resample to {self.freq}!"
)
self.file_name = f"{_freq}_future.txt" if self.future else f"{_freq}.txt".lower(
)
# The cache is useful for the following cases
# - multiple frequencies are sampled from the same calendar
cache_key = self.uri
if cache_key not in H["c"]:
H["c"][cache_key] = self._read_calendar()
_calendar = H["c"][cache_key]
_calendar = resam_calendar(
np.array(list(map(pd.Timestamp, _calendar))), _freq, self.freq)
return _calendar
def support_freq(self) -> List[str]:
_v = "_support_freq"
if hasattr(self, _v):
return getattr(self, _v)
if len(self.provider_uri) == 1 and C.DEFAULT_FREQ in self.provider_uri:
freq_l = filter(
lambda _freq: not _freq.endswith("_future"),
map(lambda x: x.stem, self.dpm.get_data_uri(C.DEFAULT_FREQ).joinpath("calendars").glob("*.txt")),
)
else:
freq_l = self.provider_uri.keys()
freq_l = [Freq(freq) for freq in freq_l]
setattr(self, _v, freq_l)
return freq_l
def set_params(self, tdx_files, data_dir, freq):
if self.limit_nums is None:
self.csv_files = tdx_files
else:
self.csv_files = tdx_files[:int(self.limit_nums)]
self.freq = str(Freq(freq))
self.calendar_format = self.DAILY_FORMAT if freq == "day" else self.HIGH_FREQ_FORMAT
self._calendars_list = []
self._calendars_dir = data_dir.joinpath(self.CALENDARS_DIR_NAME)
self._features_dir = data_dir.joinpath(self.FEATURES_DIR_NAME)
self._instruments_dir = data_dir.joinpath(self.INSTRUMENTS_DIR_NAME)
def uri(self) -> Path:
freq = self.freq
if freq not in self.support_freq:
# NOTE: uri
# 1. If `uri` does not exist
# - Get the `min_uri` of the closest `freq` under the same "directory" as the `uri`
# - Read data from `min_uri` and resample to `freq`
freq = Freq.get_recent_freq(freq, self.support_freq)
if freq is None:
raise ValueError(
f"can't find a freq from {self.support_freq} that can resample to {self.freq}!"
)
self.resample_freq = freq
return self.dpm.get_data_uri(self.use_freq).joinpath(
f"{self.storage_name}s", self.file_name)