Why can't one read and write both csv's? about pytzwhere HOT 4 CLOSED

The import/export options of the script are mainly there to convert between different types of input files if you want to do that, so it doesn't make a lot of sense to go from CSV to CSV file. The script options are not particularly useful from an end-user perspective, but they make it relatively painless to ship with the most up-to-date version of the timezone map (which reminds me that I have to update that), and see whether you broke anything before you commit.

I am not a 100% sure I understand what you exactly want to do, but if you could elaborate a bit that would be great. I feel like I am now the one that is missing something obvious :)

You don't want to export polygons with timezone information by any chance?

from pytzwhere.

I simply need an offline way to append local time to a csv file with
lat/lon/timestamp (UTC). (Actually in my case it's a bunch of csv files at
once so I use glob.) I am doing analysis in R, and so in/out csv means I
can call it from R with 'system(python tzwhere.py)' and then call
'read_csv()" again easily to get the results back in. I've done a quick and
dirty implementation like so:

from tzwhere import tzwhere
import glob
import pandas as pd

def reader(f):
  d = pd.read_csv(f, names=['id', 'latitude', 'longitude', 'timestamp'])
  return d


def main():
  files = glob.glob('*.csv')
  data = pd.concat([reader(f) for f in files], ignore_index=True)
  tz = tzwhere.tzwhere(shapely=True)
  data['local_tz'] = data.apply(lambda row: tz.tzNameAt(row['latitude'],
                                                row['longitude']), axis=1)

  data.to_csv('to_r.csv', index=False)

if __name__ == "__main__":
  main()

---------- Forwarded message ----------
From: cstich [email protected]
Date: Tue, Nov 17, 2015 at 6:34 PM
Subject: Re: [pytzwhere] Why can't one read and write both csv's? (#30)
To: pegler/pytzwhere [email protected]
Cc: Josie Kressner [email protected]

The import/export options of the script are mainly there to convert between
different types of input files if you want to do that, so it doesn't make a
lot of sense to go from CSV to CSV file. The script options are not
particularly useful from an end-user perspective, but they make it
relatively painless to ship with the most up-to-date version of the
timezone map (which reminds me that I have to update that), and see whether
you broke anything before you commit.

I am not a 100% sure I understand what you exactly want to do, but if you
could elaborate a bit that would be great. I feel like I am now the one
that is missing something obvious :)

You don't want to export polygons with timezone information by any chance?

—
Reply to this email directly or view it on GitHub
#30 (comment).

from pytzwhere.

Your code makes sense to me as it is. Pandas is probably a much better choice for doing all the reading and writing of CSV files than using anything in pytzwhere for that. Does that answer your original question?

from pytzwhere.

Sure thing. I'll continue to use the script with pandas.

from pytzwhere.