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

Data lines have the following formats:

0000906 2R0,Dm=271D,Sm=1.8M,Ta=29.3C,Ua=83.2P,Pa=1003.3H,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
0001716 2R1,Dm=269D,Sm=1.5M
            wdir    wspd
0001742 2R2,Ta=29.3C,Ua=83.2P,Pa=1003.4H
            Tair     RH       Pmb
0007913 2R3,Rc=0.00M,Rd=0s,Ri=0.0M,Hc=0.0M,Hd=0s,Hi=0.0M
            cum_rain  dur   mm/hr  cum_hail dur  intens
0009945 2R5,Th=28.9C,Vh=12.0#,Vs=12.2V,Vr=3.518V
             Theata  status   supplyV  refV

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

source file names follow the convention: met3yydddhh_raw.txt
output file names follow the convention: wxt0yydddhh_raw.txt

Byron Blomquist, NOAA/ESRL/PSD, Aug 2019
"""
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('met3')]

# parse data files and save into txt files
for file in files:
    # open output file, new name begins with wxt0
    outp = 'wxt0'+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