Source code for pycif.plugins.datastreams.fluxes.chimere.read

import datetime
import os

import numpy as np
import xarray as xr
import resource

from .....utils.netcdf import readnc
from .....utils.check.errclass import CifError
from .....utils.hdf5 import _hdf5_lock




def read(
        self,
        name,
        varnames,
        dates,
        files,
        interpol_flx=False,
        tracer=None,
        model=None,
        ddi=None,
        **kwargs
):
    """Get fluxes from pre-computed CHIMERE AEMISSIONS/BEMISSIONS files.

    For each requested date/file pair, determines the hour index to extract
    either from ``ddi`` (if given) or by parsing the file's ``Times``
    variable, then reads that hourly slice of the requested variable. If
    ``tracer.diurnal_factor`` is set, the diurnal cycle of the flux is
    rescaled toward its daily mean while conserving the daily total
    (EXPERIMENTAL).

    Args:
        self: the fluxes Plugin.
        name (str): name of the component.
        varnames (list[str] or str): variable name(s) to read; ``name`` is
            used if ``varnames`` is empty.
        dates (list): list of the date intervals to extract.
        files (list): list of files matching ``dates``.
        interpol_flx (bool): unused, kept for interface compatibility.
        tracer: the tracer Plugin, giving access to ``diurnal_factor``.
        model: unused, kept for interface compatibility.
        ddi (datetime.datetime, optional): reference start date of the file,
            used to compute the hour index directly; if ``None``, the hour
            is instead derived from the file's ``Times`` variable.
        **kwargs: unused, kept for interface compatibility.

    Returns:
        xr.DataArray: the flux data with dimensions
        ``(time, lev, lat, lon)``.

    Raises:
        CifError: if ``ddi`` is not given and the file has no ``Times``
            variable to derive the hour index from.
    """

    var2extract = varnames if varnames != "" else name

    # Reading required fluxes files
    trcr_flx = []
    for period, ff in zip(dates, files):
        with _hdf5_lock:
            ds = xr.open_dataset(ff)

            # Force conversion to float64
            data = ds[var2extract].values.astype(float)

            # Get the correct hour in the file
            if ddi is not None:
                hour = int((period[0] - ddi).total_seconds() // 3600)

            else:
                if "Times" not in ds:
                    raise CifError("Times variables is not available and I have "
                                    "no information on the file date")

                times = ds["Times"].values
                times = [
                    datetime.datetime.strptime(
                        str(b"".join(s), "utf-8"), "%Y-%m-%d_%H:%M:%S"
                    )
                    for s in times
                ]
                hour = int((period[0] - times[0]).total_seconds() // 3600)

        # Scale the diurnal cycle
        if hasattr(tracer, "diurnal_factor"):
            data = tracer.diurnal_factor * data \
                   + (1 - tracer.diurnal_factor) * data.mean(axis=0)

        trcr_flx.append(data[hour, ...])
    
    # Building a xarray
    xmod = xr.DataArray(
        trcr_flx, coords={"time": np.array(dates)[:, 0]},
        dims=("time", "lev", "lat", "lon")
    )
    return xmod