"""parseDatalogUserFiles.py
Parses data from Picarro Datalog_User raw files into hourly files.

Datalog_User files have 22 tab-delimited columns:

1  DATE
2  TIME
3  FRAC_DAYS_SINCE_JAN1
4  FRAC_HRS_SINCE_JAN1
5  JULIAN_DAYS
6  EPOCH_TIME
7  ALARM_STATUS
8  INST_STATUS
9  CavityPressure
10 CavityTemp
11 DasTemp
12 EtalonTemp
13 WarmBoxTemp
14 species
15 MPVPosition
16 OutletValve
17 solenoid_valves
18 CH4
19 CH4_dry
20 CO2
21 CO2_dry
22 H2O

raw file names are: CFHADS2043-YYYYMMDD-hhmmssZ-DataLog_User.dat
where the time indicated is approximately the start time, usually a
few seconds after the first data point in the file.

the script assumes current directory contains the raw files, and
output files are written to the same directory

output file names follow PSD style: ghg0YYDDDHH.txt

Byron Blomquist, NOAA/ESRL/PSD, Nov 2019
"""
import os
import string

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

# parse data files and save into txt files
HH_prior = '';
for file in files:
    print(file)
    fin = open(file)
    line1 = fin.readline()  # read header line
    for line in fin:  # read data lines sequentially
        cleanLine = line.strip()
        fields = cleanLine.split()
        ddd = fields[4]
        idx = ddd.find('.')
        DDD = ddd[0:idx]      # day of year string
        YY = fields[0][2:4]     # 2 digit year string
        HH = fields[1][0:2]     # 2 digit hour string
        if HH != HH_prior:  # new outp if hour changes or first file
            HH_prior = fields[1][0:2]  # remember current hour of day
            # open output file, new name begins with ghg0
            outp = 'ghg0'+YY+DDD+HH+'_raw.txt'
            fghg = open(outp, 'w')
            # first, write header line to outp
            cleanLine = line1.strip()
            vars = cleanLine.split()
            header = ''
            for item in vars:
                header += item + ','
            headerLine = header[0:-1] + '\n'
            fghg.write(headerLine)
            # then write data line to outp
            lineOut = ''
            for item in fields:   # concatenate numeric fields to output
                lineOut += item + ','
            printLine = lineOut[0:-1] + '\n'    # add return character
            fghg.write(printLine) # write comma delimited line to output file
        else:  # just write data line to outp
            lineOut = ''
            for item in fields:
                lineOut += item + ','
            printLine = lineOut[0:-1] + '\n'
            fghg.write(printLine)
fghg.close()
# end