extracting latitude from a string #126

Activity

1. arunaskuciauskas commented on Jan 28, 2022

   

   arunaskuciauskas

   on Jan 28, 2022

   Last edited by dopplershift

   Author

   More actions

   Hi: I am trying to extract the latitude from hundreds of files that contain the following listing:

   %TITLE%
   STC 190313/034000 -0.1,-27.46
   
          LEVEL       HGHT       TEMP       DWPT       WDIR       WSPD
   -------------------------------------------------------------------
   %RAW%
        1011.44,      0.00,     30.37,     24.46,      9.60,      0.01
         986.07,    227.06,     27.65,     22.99,      9.50,      0.01
         958.59,    477.68,     25.43,     21.13,      9.40,      0.01
         931.52,    730.00,     23.50,     19.20,      9.30,      0.01
   

   I perform the following read:
   df = pd.read_csv(fn, sep ='s,', skiprows=0, nrows = 1)

   resulting in output:

   %TITLE%
   0  STC 190313/034000 -0.36,-25.61
   

   where the latitude is -0.36, my intended value. In reading multiple files, the latitude value in the files can vary
   in format from 0.0 (2 char), to -25.67 (6 char)

   I applied a split on the '/' as:

       coord = df.iloc[0,0]
       print('Coord ',coord)

   resulting in output:

   Coord:  STC 190313/034000 -0.36,-25.61
   

   lat = coord.split('/')
   lati = lat[1][7:12]
   print('lati: ', lati)

   resulting in output:
   lati: -0.1,

   The start of the latitude value always starts in the 7th position
   Question: How can I make sure that I grab the entire latitude value?
2. dopplershift commented on Feb 19, 2022

   

   dopplershift

   on Feb 19, 2022

   Member

   More actions

   Apologies for the delayed response. I wouldn't use parse_csv to handle any of that header, only the later data. This is what I wrote to parse what you provided:

   with open(fn, 'rt') as infile:
       # skip "%TITLE%
       infile.readline()
   
       # Get the siteid, date string, and lat/lon group by splitting on whitespace
       site, date, latlon = infile.readline().split()
   
       # Split the lat/lon group on a comma and convert to floats
       lat, lon = map(float, latlon.split(','))
   
       # Skip blank
       infile.readline()
   
       # Get the column names from the names that *don't* have commas
       columns = infile.readline().split()
   
       # Let pandas parse the rest of the file
       df = pd.read_csv(infile, skiprows=2, names=columns)
3. Unidata

   locked and limited conversation to collaborators on Feb 22, 2022
4. dcamron

   converted this issue into a discussion #129 on Feb 22, 2022