scale.py

# coding=utf-8
from earthdiagnostics.diagnostic import *
from earthdiagnostics.utils import Utils
from earthdiagnostics.modelingrealm import ModelingRealm
import math


class Scale(Diagnostic):
    """
    Scales a variable by the given value also adding at offset
    Can be useful to correct units or other known errors
    (think of a tas file declaring K as units but with the data stored as Celsius)

    :original author: Javier Vegas-Regidor<javier.vegas@bsc.es>

    :created: July 2016

    :param data_manager: data management object
    :type data_manager: DataManager
    :param startdate: startdate
    :type startdate: str
    :param member: member number
    :type member: int
    :param chunk: chunk's number
    :type chunk: int    :
    :param variable: variable's name
    :type variable: str
    :param domain: variable's domain
    :type domain: ModelingRealm
    """

    alias = 'scale'
    "Diagnostic alias for the configuration file"

    def __init__(self, data_manager, startdate, member, chunk, value, offset, domain, variable, grid,
                 min_limit, max_limit, frequency):
        Diagnostic.__init__(self, data_manager)
        self.startdate = startdate
        self.member = member
        self.chunk = chunk
        self.variable = variable
        self.domain = domain
        self.grid = grid
        self.value = value
        self.offset = offset
        self.min_limit = min_limit
        self.max_limit = max_limit
        self.frequency = frequency

        self.original_values = None

    def __str__(self):
        return 'Scale output Startdate: {0} Member: {1} Chunk: {2} ' \
               'Scale value: {5} Offset: {6} Variable: {3}:{4} ' \
               'Frequency: {7}'.format(self.startdate, self.member, self.chunk, self.domain, self.variable,
                                       self.value, self.offset, self.frequency)

    def __eq__(self, other):
        return self.startdate == other.startdate and self.member == other.member and self.chunk == other.chunk and \
            self.domain == other.domain and self.variable == other.variable and self.frequency == other.frequency

    @classmethod
    def generate_jobs(cls, diags, options):
        """
            Creates a job for each chunk to compute the diagnostic

            :param diags: Diagnostics manager class
            :type diags: Diags
            :param options: variable, domain, grid
            :type options: list[str]
            :return:
            """
        options_available = (DiagnosticDomainOption(),
                             DiagnosticVariableOption(diags.data_manager.config.var_manager),
                             DiagnosticFloatOption('value'),
                             DiagnosticFloatOption('offset'),
                             DiagnosticOption('grid', ''),
                             DiagnosticFloatOption('min_limit', float('nan')),
                             DiagnosticFloatOption('max_limit', float('nan')),
                             DiagnosticListFrequenciesOption('frequencies', [diags.config.frequency]))
        options = cls.process_options(options, options_available)
        job_list = list()
        for frequency in options['frequencies']:
            for startdate, member, chunk in diags.config.experiment.get_chunk_list():
                job_list.append(Scale(diags.data_manager, startdate, member, chunk,
                                      options['value'], options['offset'], options['domain'], options['variable'],
                                      options['grid'], options['min_limit'], options['max_limit'], frequency))
        return job_list

    def request_data(self):
        self.variable_file = self.request_chunk(self.domain, self.variable, self.startdate, self.member, self.chunk,
                                                grid=self.grid, frequency=self.frequency, to_modify=True)

    def declare_data_generated(self):
        self.corrected = self.declare_chunk(self.domain, self.variable, self.startdate, self.member, self.chunk,
                                            grid=self.grid, frequency=self.frequency)

    def compute(self):
        """
        Runs the diagnostic
        """
        variable_file = self.variable_file.local_file

        handler = Utils.openCdf(variable_file)
        var_handler = handler.variables[self.variable]
        self.original_values = var_handler[:]
        if self._check_limits():
            var_handler[:] = self.original_values * self.value + self.offset
        handler.close()
        self.corrected.set_local_file(self.variable_file.local_file, self)

    def _check_limits(self):
        if not math.isnan(self.min_limit) and (self.original_values.min() < self.min_limit):
            return False
        if not math.isnan(self.max_limit) and (self.original_values.max() > self.max_limit):
            return False
        return True