dc2_dia_object.py

"""
DC2 DIA Object Catalog Reader
"""

import os

import numpy as np

from .dc2_dm_catalog import (DC2DMTractCatalog,
                             convert_nanoJansky_to_mag,
                             convert_flux_err_to_mag_err,
                             create_basic_flag_mask)

__all__ = ['DC2DiaObjectCatalog']


class DC2DiaObjectCatalog(DC2DMTractCatalog):
    r"""DC2 DIA Object Catalog reader

    Parameters
    ----------
    base_dir          (str): Directory of data files being served, required
    filename_pattern  (str): The optional regex pattern of served data files
    use_cache        (bool): Cache read data in memory
    is_dpdd          (bool): File are already in DPDD-format.  No translation.

    Attributes
    ----------
    base_dir          (str): The directory of data files being served

    Notes
    -----
    """
    # pylint: disable=too-many-instance-attributes
    FILE_DIR = os.path.dirname(os.path.abspath(__file__))
    FILE_PATTERN = r'dia_object_tract_\d+\.parquet$'
    META_PATH = os.path.join(FILE_DIR, 'catalog_configs/_dc2_dia_object_meta.yaml')

    def _detect_available_bands(self):
        return [col.partition('_')[2] for col in self._columns if col.startswith('psFluxMean_')]

    @staticmethod
    def _generate_modifiers(dm_schema_version=4, bands='ugrizy', **kwargs):  # pylint: disable=arguments-differ
        """Creates a dictionary relating native and homogenized column names

        Args:
            dm_schema_version (int):  Only DM Schema version 3 or greater supported.
                The current implementation is a no-op for both version 3 and 4.
                No DIA products have been prepared with DM schema v1 and v2.
                Function will raise a Runtime error if dm_schema_version <= 3 specified.
            bands (list): Filter names.  These will be used for defined suffixes to flux quanties.
                E.g., bands='gr' would create just `psFlux_g`, `psFlux_g`, `psFlux_r`, and psFluxErr_r'
                in addition to other band-dependent columns.
                This argument should be left at the default when preparing DC2 DPDD tables.

        Returns:
            A dictionary of the form {<homogenized name>: <native name>, ...}
        """
        if dm_schema_version < 3:
            err_msg = 'DM Schema Version {} not supported.'
            raise RuntimeError(err_msg.format(dm_schema_version))

        # Quantities defined in the DPDD but that we don't know how
        # to calculate yet are commented out in the dict below.
        # Based on the LSST DPDD 2018-10-24 +
        # https://jira.lsstcorp.org/browse/RFC-517
        #    which clarifies DIASource vs DIAForceSource
        #    and how to calculate various DIAObject properties based on DIASources
        modifiers = {
            'diaObjectId': 'diaObjectId',
            'ra': (np.rad2deg, 'coord_ra'),
            'dec': (np.rad2deg, 'coord_dec'),
            # 'radecCov': '',  # A covariance matrix for the uncertainty in ra, dec
            # 'radecTai': 'dateobs',  # I don't know what this is called
            # 'pm': '',
            # 'pmParallaxCov': '',
            # 'pmParallaxLnl': '',
            # 'pmParallaxChi2': '',
            # 'pmParallaxNdata': '',
            # 'totFluxMean': 'ip_diffim_forced_PsfFlux_instFlux',
            # 'totFluxMeanErr': 'ip_diffim_forced_PsfFlux_instFluxErr',
            # 'totFluxSigma': 'ip_diffim_forced_PsfFlux_instFluxErr',
            # 'lcPeriodic': '',
            # 'lcNonPeriodic': '',
            # # These are possible to calculate if you require the Object Table
            # 'nearbyObj': '',
            # 'nearbyObjDist': '',
            # 'nearbyObjLnP': '',
        }

        # modifiers['good'] = (create_basic_flag_mask, )
        # modifiers['clean'] = modifiers['good']

        multiband_columns_to_copy = [
            'psFluxMean', 'psFluxMeanErr',
            'psFluxSigma', 'psFluxChi2', 'psFluxNdata']

        for band in bands:
            for base_col in multiband_columns_to_copy:
                col_name = f'{base_col}_{band}'
                modifiers[col_name] = col_name

            # Create new convenience magnitude columns based on flux values
            modifiers[f'magMean_{band}'] = (convert_nanoJansky_to_mag,
                                            f'psFluxMean_{band}')
            modifiers[f'magMeanErr_{band}'] = (convert_flux_err_to_mag_err,
                                               f'psFluxMean_{band}',
                                               f'psFluxMeanErr_{band}')
            modifiers[f'magMeanStd_{band}'] = (convert_flux_err_to_mag_err,
                                               f'psFluxMean_{band}',
                                               f'psFluxSigma_{band}')

        return modifiers