Python questions (How covert txt files to .csv files?)

[Mick]

My company asks to convert txt files to .csv files. Could you advise how to write a Python script for ArcGIS?  

[DanPatterson]

The bigger question is the *.txt files in comma separated values format?

Otherwise it would be a simple renaming operation.

And in any event, the filename extension doesn't need to be *.csv for it to be read as one, it is the contents that are important

A previous answer provides the solution, which just needs to be put within a loop

Solved: Using Python to split and rename file - Esri Community

[Mick]

There is no comma.

[DanPatterson]

Then a sample of the input would be needed, otherwise you have to read the file, parse it on spaces or tabs and reassemble

[Mick]

I have only two spaces between coordinates in my files. Can you recommend a code to replace two spaces for a comma in the script?

[DanPatterson]

I use numpy

fname = r"c:\temp\space.txt"
a = np.loadtxt(fname, delimiter=" ")

outname = fname[:-3] + "csv"
np.savetxt(outname, a, fmt='%f', delimiter=", ")

a text file

1.0 2.0 3.0
4.0 5.0 6.0

as csv file

1.000000, 2.000000, 3.000000
4.000000, 5.000000, 6.000000

you can format as you see fit. 

[Mick]

Thank you, can you recommend any extra options with arcpy?

[JoshuaBixby]

NumPy has been a standard package in Esri's Python deployment for years.  If your Python deployment has ArcPy, it has NumPy installed too.

[JoeBorgione · Accepted solution]

You won't use arcpy as it is nothing more than a python module but you can use the re (regular expression) module to do it:

import re
myFile = r'C:\temp\Coords.txt'

with open(myFile,'r+') as f:
    text = f.read()
    text = re.sub('  ',',',text)
    f.seek(0)
    f.write(text)
    f.truncate()
    
f.close()

Assuming you have the file C:\temp\Coords.txt that looks like this:

-111.45  40.57
-112.05  40.06

I stole this right from https://www.kite.com/python/answers/how-to-update-and-replace-text-in-a-file-in-python

[Mick]

Can you recommend how to use not only one file to convert but a folder with one hundred files?