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

import datetime
import glob
import os
import calendar 
import numpy as np




def find_valid_file(ref_dir, file_format, dd):
    
    # Get all files and dates matching the file and format
    #print('Here, basic listing of all available files')
    list_files_orig = os.listdir(ref_dir)
    list_files_avail = []
    for f in list_files_orig:
      try:
         check = datetime.datetime.strptime(f, file_format)
         list_files_avail.append(f)
      except:
            continue
    #print('All availables files',list_files_avail)
    
  
    list_files = []
    list_dates = []    
    # XXXXXXX sans closest year pour le moment, A AJOUTER PLUS TARD XXXXXX
    # actually in each file, information every month
    for f in list_files_avail:
      year_beg_file = (f.split('-')[0]).split('.')[-1]
      #date_beg = datetime.datetime(year = int(year_beg_file), month = 1, day = 15)
      year_end_file = (f.split('-')[1]).split('.')[0]
      #date_end = datetime.datetime(year = int(year_end_file), month = 12, day = 15)
      nb_deltas = (int(year_end_file) - int(year_beg_file) + 1 ) * 12
      list_dates_covered = []
      yy = int(year_beg_file)
      mm =1 
      for k in range(nb_deltas):
        list_dates_covered.append(datetime.datetime(year = yy , month = mm , day =15))
        mm = mm + 1 
        if mm == 13 :
          yy = yy + 1 
          mm =1 
      for d in list_dates_covered:
         list_files.append(f)
         list_dates.append(d)

    list_files = np.array(list_files)
    list_dates = np.array(list_dates)
    # Sorting along dates
    isort = np.argsort(list_dates)
    list_dates = list_dates[isort]
    list_files = list_files[isort]
    if list_files == []:
        raise Exception("Did not find any valid GCP flux files in {} "
                        "with format {}. Please check your yml file"
                        .format(ref_dir, file_format))

   
    # Compute deltas
    mask = (list_dates - dd) <= datetime.timedelta(0)
    #print(mask)
    file_ref1 = ref_dir + list_files[mask][np.argmax(list_dates[mask])]
    date_ref1 = list_dates[mask].max()
    #print('111111111',file_ref1, date_ref1)
    mask = (list_dates - dd) >= datetime.timedelta(0)
    #print('mmmm',mask)
    file_ref2 = ref_dir + list_files[mask][np.argmin(list_dates[mask])]
    date_ref2 = list_dates[mask].min()
    #print('2222222222',file_ref2, date_ref2)

    return  [file_ref1, file_ref2], [date_ref1, date_ref2]