Exploring the fileinput module in Python : change file contents in place and more

FileInput : 

I was in need of modifying the contents of a file in place and i was looking at what my options were.

Quick google search finally helped me land on fileinput and i am really glad , how wonderful it is to use this module for the same purpose and it obviously does more than that.

Contents of the Fastfile looked like below and we wanted to change the version number as part of the automation.
Just ignore the regex i used .. I am anyways not really good at it :-)

import fileinput
import re


def modify_fastfile(full_path, version):
    print(full_path)
    mod_ver = r'  version_number = "{}"'.format(version)
    print(mod_ver)
    pattern = r'\s\sversion_number = \"\d.*\"'
    with fileinput.FileInput(full_path, inplace=True, backup='.bak') as file:
        for line in file:
            line = re.sub(pattern, mod_ver, line.rstrip())
            print(line)
    

But the things to focus really on are the parameters.

• inplace = True 
• backup = '.bak'

And thats it voila we were able to change the contents of the file in place.

You should really go over quickly the documentation on https://docs.python.org/2/library/fileinput.html 

This module implements a helper class and functions to quickly write a loop over standard input or a list of files. If you just want to read or write one file see open().

I am showing couple of ways in which it can be used.

• In first case we are just accepting the input from standard input 
• In second case we are providing list of files 

#! /usr/bin/env python3
import fileinput

with fileinput.input() as f:
    for line in f:
        print(line, end='')

# ls | ./file_input.py
# __pycache__
# file_input.py
# file_input_multi_files.py

#! /usr/bin/env python3
import fileinput

with fileinput.input(files=('file_input.py', 'file_input_multi_files.py')) as f:
    for line in f:
        print(f.filename(), f.lineno(), line, end='')


# ./file_input_multi_files.py

# file_input.py 1 #! /usr/bin/env python3
# file_input.py 2 import fileinput
# file_input.py 3
# file_input.py 4 with fileinput.input() as f:
# file_input.py 5     print()
# file_input.py 6     for line in f:
# file_input.py 7         print(line, end='\n')file_input_multi_files.py 8 #! /usr/bin/env python3
# file_input_multi_files.py 9 import fileinput
# file_input_multi_files.py 10
# file_input_multi_files.py 11 with fileinput.input(files=('file_input.py', 'file_input_multi_files.py')) as f:
# file_input_multi_files.py 12     for line in f:
# file_input_multi_files.py 13         print(f.filename(), f.lineno(), line, end='')%

Now when i had a look at the definition in the docs , i saw openhooks parameter.
Basically this allows us to have compressed and uncompressed files in the list. Thats a great flexibility.

I was thinking of problem that David solved using generators and thought  fileinput can also be used to solve the same problem.

If you have seen http://www.dabeaz.com/generators-uk/GeneratorsUK.pdf from @dabeaz ,  which i think you should absolutely go through cause he is just an amazing Python Instructor.

Well thats it folks i hope you like this post and you explore the  fileinput module further and share your findings with the community.

Python Interpreters for better productivity

Hellos Pythonistas , While we all agree that getting good handle over programming language requires practice and we often want playground sort of experience to get a hang of what we are trying , thats when REPL - Read Evaluate Print Loop helps us

While default python interpreter is really nice , often you cant go back to what you typed if you make typos , and doesn't offer completion or nice GUI , there are couple of options to mitigate this pain and enjoy the pure python fun

• bpython : https://bpython-interpreter.org/screenshots.html
• ptpython : https://github.com/jonathanslenders/ptpython

I personally use bpython and love it so far , look at the screenshot

for the curious , b in the bpython stands for "Bob" 

• Ipython : https://ipython.org/
• A powerful interactive shell.
• A kernel for Jupyter.
• Support for interactive data visualization and use of GUI toolkits.
• Flexible, embeddable interpreters to load into your own projects.
• Easy to use, high performance tools for parallel computing.

• Jupyter Notebook : http://jupyter.org/ 

          Project Jupyter is a project born out of Ipython and its focus is interactive data science and 

          scientific computing

  

Well , thats it folks , the point across is all these options are better than default interpreter so make a switch to one of these for better productivity and lesser typing :-). 

Happy Pythoning , until next time ...........