import os
import numpy as np

def read_obsfcstana_extend_datetime(path, file_name, printflag=False):
    
    # Precision of fortran tag
    int_precision = 'int32'
    # Precision of data in input file
    float_precision = 'float32'
    # Precision of data in input file
    logical_precision = 'int32'

    # Initialize outputs in case file does not exist or is empty
    date_time = []
    obs_assim = []
    obs_species = []
    obs_tilenum = []
    obs_lon = []
    obs_lat = []
    obs_obs = []
    obs_obsvar = []
    obs_fcst = []
    obs_fcstvar = []
    obs_ana = []
    obs_anavar = []

    # Determine machine format
    machfmt = 'b'

    # Get a list of files with a similar name in a directory
    file_ext = '.bin'
    # Use os.walk to recursively traverse all directories and subdirectories under the top-level directory
    files = [os.path.join(root, file) for root, dirs, files in os.walk(path) for file in files if file.startswith(file_name) and file.endswith(file_ext)]

    print("files: ", files)

    if printflag:
        print(files)

    # Open each file in turn
    mode = 'rb' if machfmt == 'b' else 'rl'

    for file in files:
        with open(os.path.join(path, file), mode) as ifp:
            if printflag:
                print ('Reading file ', file, '...')
            
            # Read N_obs and time stamp entry
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            N_obs = np.fromfile(ifp, int_precision, 1)
            N_obs = int(N_obs)
            year = np.fromfile(ifp, int_precision, 1)
            month = np.fromfile(ifp, int_precision, 1)
            day = np.fromfile(ifp, int_precision, 1)
            hour = np.fromfile(ifp, int_precision, 1)
            minute = np.fromfile(ifp, int_precision, 1)
            second = np.fromfile(ifp, int_precision, 1)
            dofyr = np.fromfile(ifp, int_precision, 1)
            pentad = np.fromfile(ifp, int_precision, 1)
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            # Populate date_time structure
            date_time_tmp = {
                'year': year,
                'month': month,
                'day': day,
                'hour': hour,
                'min': minute,
                'sec': second,
                'dofyr': dofyr,
                'pentad': pentad
            }
            # Extend date_time so it has the same length as everything else
            date_time.extend([date_time_tmp] * N_obs) 
            
            # Read observation assim flag
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            tmp_data = np.fromfile(ifp, logical_precision, N_obs)
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            tmp_data2 = np.zeros((N_obs, 1))
            indices = np.where(tmp_data != 0)[0]
            tmp_data2[indices] = 1
            obs_assim = np.append(obs_assim, tmp_data2)
            
            # Read species information
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            obs_species = np.append(obs_species, np.fromfile(ifp, int_precision, N_obs))
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            
            # Read tile number information
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            obs_tilenum = np.append(obs_tilenum, np.fromfile(ifp, int_precision, N_obs))
            fortran_tag = np.fromfile(ifp, int_precision, 1)

            # Read longitude
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            obs_lon = np.append(obs_lon, np.fromfile(ifp, float_precision, N_obs))
            fortran_tag = np.fromfile(ifp, int_precision, 1)

            # Read latitude
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            obs_lat = np.append(obs_lat, np.fromfile(ifp, float_precision, N_obs))
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            
            # Read observation value
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            obs_obs = np.append(obs_obs, np.fromfile(ifp, float_precision, N_obs))
            fortran_tag = np.fromfile(ifp, int_precision, 1)

            # Read observation variance
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            obs_obsvar = np.append(obs_obsvar, np.fromfile(ifp, float_precision, N_obs))
            fortran_tag = np.fromfile(ifp, int_precision, 1)

            # Read observation-space model forecast value
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            obs_fcst = np.append(obs_fcst, np.fromfile(ifp, float_precision, N_obs))
            fortran_tag = np.fromfile(ifp, int_precision, 1)

            # Read observation-space model forecast variance
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            obs_fcstvar = np.append(obs_fcstvar, np.fromfile(ifp, float_precision, N_obs))
            fortran_tag = np.fromfile(ifp, int_precision, 1)

            # Read observation-space analysis value
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            obs_ana = np.append(obs_ana, np.fromfile(ifp, float_precision, N_obs))
            fortran_tag = np.fromfile(ifp, int_precision, 1)

            # Read observation-space analysis variance
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            obs_anavar = np.append(obs_anavar, np.fromfile(ifp, float_precision, N_obs))
            fortran_tag = np.fromfile(ifp, int_precision, 1)
            
            
    # Close file
        ifp.close()
     
    # print('Total number of obs = ',len(obs_assim))

    return date_time, obs_species, obs_tilenum, obs_lon, obs_lat, obs_obs, obs_obsvar, obs_fcst, obs_fcstvar, obs_ana, obs_anavar