how to NOT read_csv if csv is empty

Question

Using Python 2.7 and Pandas

I have to parse through my directory and plot a bunch of CSVs. If the CSV is empty, the script breaks and produces the error message:

pandas.io.common.EmptyDataError: No columns to parse from file

If I have my file paths stored in

file_paths=[]

how do I read through each one and only plot the non empty CSVs? If I have an empty dataframe defined as df=[] I attempt the following code

for i in range(0,len(file_paths)):
   if pd.read_csv(file_paths[i] == ""):
      print "empty"
   else df.append(pd.read_csv(file_paths[i],header=None))

Answer 1

I would just catch the appropriate exception, as a catch all is not recommended in python:

import pandas.io.common

for i in range(0,len(file_paths)):
   try:
      pd.read_csv(file_paths[i])
   except pandas.io.common.EmptyDataError:
      print file_paths[i], " is empty"

Answer 2

Note, as of pandas 0.22.0 (that I can be sure of) , the exception raised for empty csv is pandas.errors.EmptyDataError. And if you're importing pandas like import pandas as pd, then use pd instead of pandas.

If your csv filenames are in an array manyfiles, then

import pandas as pd
for filename in manyfiles:
    try:
        df = pd.read_csv(filename)

    except pd.errors.EmptyDataError:
        print('Note: filename.csv was empty. Skipping.')
        continue # will skip the rest of the block and move to next file

    # operations on df

I'm not sure if pandas.io.common.EmptyDataError is still valid or not. Can't find it in reference docs. And I also would advise against the catch-all except: as you won't be able to know if it's something else causing the issue.