Source code for pycif.plugins.datastreams.fluxes.GCP_1x1_N2O.fetch

import datetime
import glob
import os
import pandas as pd

import numpy as np

from .....utils import path
from logging import info, debug
from .utils import find_valid_file




def fetch(ref_dir, ref_file, input_interval, target_dir,
          tracer=None, component=None, **kwargs):
    """Fetch GCP N2O multi-year files bracketing the requested dates.

    For each date key in ``input_interval``, calls
    :func:`~pycif.plugins.datastreams.fluxes.GCP_1x1_N2O.utils.find_valid_file`
    for every sub-date to bracket the nearest available files (before and
    after), links all found files into ``target_dir``, and returns
    deduplicated, sorted lists of files/dates per key.

    Args:
        ref_dir (str): directory where the original files are found.
        ref_file (str): (template) name of the original files.
        input_interval (dict): mapping of reference dates to the list of
            sub-dates to simulate within each period.
        target_dir (str): directory where links to the original files are
            created.
        tracer: the tracer Plugin; unused directly but kept for interface
            compatibility.
        component: the component Plugin; unused directly but kept for
            interface compatibility.
        **kwargs: unused, kept for interface compatibility.

    Returns:
        (dict, dict): ``list_files`` and ``list_dates``, keyed like
        ``input_interval``, each value a deduplicated sorted list of file
        paths / bracketing dates found across all sub-dates of that key.
    """

    debug('Display here the information about the plugin')
    debug(tracer.plugin.name)
    debug(tracer.plugin.version)
    debug(",".join(dir(tracer)))
    debug(component.plugin.name)
    debug(component.plugin.version)
    debug(dir(component))

    list_files = {}
    list_dates = {}
    for datei in input_interval:
        tmp_files = []
        tmp_dates = []
        for dd in input_interval[datei]:
            # print('Date to simulate:', dd)
            files_orig, dates_orig = find_valid_file(ref_dir, ref_file, dd)
            tmp_files.extend(files_orig)
            tmp_dates.extend(dates_orig)

        # Fetching
        local_files = []
        for f in tmp_files:
            target_file = f"{target_dir}/{os.path.basename(f)}"
            path.link(f, target_file)
            local_files.append(target_file)

        list_files[datei] = sorted(list(set(local_files)))
        list_dates[datei] = sorted(list(set(tmp_dates)))

    return list_files, list_dates