#! /usr/bin/env python
from __future__ import annotations
from typing import TYPE_CHECKING
from roman_datamodels import datamodels as rdm
from romancal.stpipe import RomanStep
if TYPE_CHECKING:
from typing import ClassVar
__all__ = ["DarkCurrentStep"]
[docs]
class DarkCurrentStep(RomanStep):
"""
DarkCurrentStep: Performs dark current correction by subtracting
dark current reference data from the input science data model.
"""
class_alias = "dark"
spec = """
dark_output = output_file(default = None) # Dark corrected model
"""
reference_file_types: ClassVar = ["dark"]
[docs]
def process(self, input):
if isinstance(input, rdm.DataModel):
input_model = input
else:
# Open the input data model
input_model = rdm.open(input)
# Get the name of the dark reference file to use
self.dark_name = self.get_reference_file(input_model, "dark")
# Check for a valid reference file
if self.dark_name == "N/A":
self.log.warning("No DARK reference file found")
self.log.warning("Dark current step will be skipped")
result = input_model
result.meta.cal_step.dark = "SKIPPED"
return result
self.log.info("Using DARK reference file: %s", self.dark_name)
# Open dark model
with rdm.open(self.dark_name) as dark_model:
# Temporary patch to utilize stcal dark step until MA table support
# is fully implemented
if "nresultants" not in dark_model.meta.exposure:
dark_model.meta.exposure["nresultants"] = dark_model.data.shape[0]
# Do the dark correction
out_model = input_model
nresultants = len(input_model.meta.exposure["read_pattern"])
out_model.data -= dark_model.data[:nresultants]
out_model.pixeldq |= dark_model.dq
out_model.meta.cal_step.dark = "COMPLETE"
# Save dark data to file
if self.dark_output is not None:
dark_model.save(self.dark_output)
# not clear to me that this makes any sense for Roman
if self.save_results:
try:
self.suffix = "darkcurrent"
except AttributeError:
self["suffix"] = "darkcurrent"
return out_model