[df_read_csvs_filepattern] Reads data stored in possibly multiple, but identically formatted csv files identified by matching a path wildcard pattern, into a pandas dataframe #pandas #IO #dataframe #csv

df_read_csvs_filepattern.py

#%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%#
#%%                                                                          %%#
#%%  df_read_csvs_filepattern.py                                             %%#
#%%                                                                          %%#
#%%  Author: Casey Conger <casey.conger@priceline.com>                       %%#
#%%                                                                          %%#
#%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%#

import pandas as pd
import tensorflow as tf
from tensorflow.python.lib.io import file_io

def read_csv_file(filename, read_csv_kwargs):
    """Read csv file from specified location into a DataFrame
    
    Args:
        filename (str): specifies location of csv file
        read_csv_kwargs (dict): arbitrary set of keyword arguments to be passed
          to pd.read_csv
    
    Returns:
        (pandas.DataFrame): DataFrame containing data from file
        
    
    """
    with file_io.FileIO(filename, 'r') as f:
        df = pd.read_csv(f, **read_csv_kwargs)
        return df

def read_csv_files(filename_pattern, **read_csv_kwargs):
    """Read csv files matching a pattern into a DataFrame
    
    Args:
        filename_pattern (str): specifies set of csv files to be read into 
          dataframe
        read_csv_kwargs (dict): arbitrary set of keyword arguments to be passed
          to pd.read_csv
    
    Returns:
        (pandas.DataFrame): DataFrame containing concatenated data from files
          matching filename_pattern
    """
    filenames = tf.gfile.Glob(filename_pattern)
    dataframes = [
        read_csv_file(filename, read_csv_kwargs)
        for filename in filenames
    ]
    return pd.concat(dataframes)