"""parseWXTFiles.py
Parses data from Vaisala WXT saved on the PSD data system during FATIMA 2022.

Data lines have the following formats:

0013354 2R0,Dm=223D,Sm=7.2M,Ta=16.9C,Ua=97.0P,Pa=1008.1H,Rc=0.00M,Rd=0s,Ri=0.0M,Hc=0.0M ,Hd=0s,Hi=0.0M
           ,wdir   ,wspd   ,Tair    ,RH      ,Pmb        cum_rain,dur  ,mm/hr  ,cum_hail,dur  ,intens 
0014161 2R1,Dm=223D,Sm=7.1M
            wdirMax,wdir
0014216 2R2,Ta=16.9C,Ua=97.0P,Pa=1008.1H
            Tair    ,RH      ,Pmb

Fields from 2R0 lines are parsed and saved to a single output file.

source file names follow the convention: wxt0yydddhh_raw.txt
output file names follow the convention: wxt_parsed_yydddhh_raw.txt

Byron Blomquist, NOAA/ESRL/PSD, July 2022 2016
"""
import os
import string

# assume current directory contains raw met3 files,
# dump list of data file names into a list variable
files = [f for f in os.listdir('.') if f.startswith('wxt0')]

# parse data files and save into txt files
for file in files:
    # open output file, new name begins with wxt0
    outp = 'wxt_parsed_'+file[4:]
    fwxt = open(outp, 'w')
    print(file)
    fin = open(file)
    for line in fin:
        cleanLine = line.strip()
        fields = cleanLine.split(',')
        if '2R0' in fields[0]:
            # select PSD timecode from field[0]
            lineOut = fields[0][0:7]+','
            raw = fields[1:]    # remaining fields
            # strip all alphabet and '=' characters
            data = [y.strip('=#'+string.ascii_letters) for y in raw]
            for item in data:   # concatenate numeric fields to output
                lineOut += item + ','
            printLine = lineOut[0:-1] + '\n'    # add return character
            fwxt.write(printLine)
    print(outp+' finished')

fwxt.close()
# end