acspo.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# Copyright (c) 2017 Satpy developers
#
# This file is part of satpy.
#
# satpy is free software: you can redistribute it and/or modify it under the
# terms of the GNU General Public License as published by the Free Software
# Foundation, either version 3 of the License, or (at your option) any later
# version.
#
# satpy is distributed in the hope that it will be useful, but WITHOUT ANY
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
# A PARTICULAR PURPOSE.  See the GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along with
# satpy.  If not, see <http://www.gnu.org/licenses/>.
"""ACSPO SST Reader.

See the following page for more information:

https://podaac.jpl.nasa.gov/dataset/VIIRS_NPP-OSPO-L2P-v2.3

"""
import logging
from datetime import datetime

import numpy as np

from satpy.readers.netcdf_utils import NetCDF4FileHandler

LOG = logging.getLogger(__name__)


ROWS_PER_SCAN = {
    'MODIS': 10,
    'VIIRS': 16,
    'AVHRR': None,
}


class ACSPOFileHandler(NetCDF4FileHandler):
    """ACSPO L2P SST File Reader."""

    @property
    def platform_name(self):
        """Get satellite name for this file's data."""
        res = self['/attr/platform']
        if isinstance(res, np.ndarray):
            return str(res.astype(str))
        return res

    @property
    def sensor_name(self):
        """Get instrument name for this file's data."""
        res = self['/attr/sensor']
        if isinstance(res, np.ndarray):
            return str(res.astype(str))
        return res

    def get_shape(self, ds_id, ds_info):
        """Get numpy array shape for the specified dataset.

        Args:
            ds_id (DataID): ID of dataset that will be loaded
            ds_info (dict): Dictionary of dataset information from config file

        Returns:
            tuple: (rows, cols)

        """
        var_path = ds_info.get('file_key', '{}'.format(ds_id['name']))
        if var_path + '/shape' not in self:
            # loading a scalar value
            shape = 1
        else:
            shape = self[var_path + '/shape']
            if len(shape) == 3:
                if shape[0] != 1:
                    raise ValueError("Not sure how to load 3D Dataset with more than 1 time")
                shape = shape[1:]
        return shape

    @staticmethod
    def _parse_datetime(datestr):
        return datetime.strptime(datestr, "%Y%m%dT%H%M%SZ")

    @property
    def start_time(self):
        """Get first observation time of data."""
        return self._parse_datetime(self['/attr/time_coverage_start'])

    @property
    def end_time(self):
        """Get final observation time of data."""
        return self._parse_datetime(self['/attr/time_coverage_end'])

    def get_metadata(self, dataset_id, ds_info):
        """Collect various metadata about the specified dataset."""
        var_path = ds_info.get('file_key', '{}'.format(dataset_id['name']))
        shape = self.get_shape(dataset_id, ds_info)
        units = self[var_path + '/attr/units']
        info = getattr(self[var_path], 'attrs', {})
        standard_name = self[var_path + '/attr/standard_name']
        resolution = float(self['/attr/spatial_resolution'].split(' ')[0])
        rows_per_scan = ROWS_PER_SCAN.get(self.sensor_name) or 0
        info.update(dataset_id.to_dict())
        info.update({
            'shape': shape,
            'units': units,
            'platform_name': self.platform_name,
            'sensor': self.sensor_name,
            'standard_name': standard_name,
            'resolution': resolution,
            'rows_per_scan': rows_per_scan,
            'long_name': self.get(var_path + '/attr/long_name'),
            'comment': self.get(var_path + '/attr/comment'),
        })
        return info

    def get_dataset(self, dataset_id, ds_info):
        """Load data array and metadata from file on disk."""
        var_path = ds_info.get('file_key', '{}'.format(dataset_id['name']))
        metadata = self.get_metadata(dataset_id, ds_info)
        shape = metadata['shape']
        file_shape = self[var_path + '/shape']
        metadata['shape'] = shape

        valid_min = self[var_path + '/attr/valid_min']
        valid_max = self[var_path + '/attr/valid_max']
        # no need to check fill value since we are using valid min/max
        scale_factor = self.get(var_path + '/attr/scale_factor')
        add_offset = self.get(var_path + '/attr/add_offset')

        data = self[var_path]
        data = data.rename({"ni": "x", "nj": "y"})
        if isinstance(file_shape, tuple) and len(file_shape) == 3:
            # can only read 3D arrays with size 1 in the first dimension
            data = data[0]
        data = data.where((data >= valid_min) & (data <= valid_max))
        if scale_factor is not None:
            data = data * scale_factor + add_offset

        if ds_info.get('cloud_clear', False):
            # clear-sky if bit 15-16 are 00
            clear_sky_mask = (self['l2p_flags'][0] & 0b1100000000000000) != 0
            clear_sky_mask = clear_sky_mask.rename({"ni": "x", "nj": "y"})
            data = data.where(~clear_sky_mask)

        data.attrs.update(metadata)
        # Remove these attributes since they are no longer valid and can cause invalid value filling.
        data.attrs.pop('_FillValue', None)
        data.attrs.pop('valid_max', None)
        data.attrs.pop('valid_min', None)
        return data