Python renaming file with regex

Question

Hi I want to rename files with one source pattern (for example IMG_20190401_235959.jpg) to a destination pattern (for example 2019-04-01_23_59_59.jpg)

I'm trying to do that in python but i can't find out how to use a regex to build the new filename :

#!/usr/bin/python

import os, glob, sys, re    
os.chdir(sys.argv[1])
for filename in glob.glob("IMG_*.jpg"):
    newfilename = re.sub(?????
    try:
       os.rename(filename,newfilename)
    except OSError,e:
       print e

Do you actually need a regex? I'm sure most users could provide you with a good non-regex solution very quickly. — KuboMD
– KuboMD, Commented Apr 25, 2019 at 18:57
If all of the files have the exact same length, why use RegEx? A solution off the top of my head would be to .split('') the filename, add the characters you need, and rejoin. — 101arrowz
– 101arrowz, Commented Apr 25, 2019 at 18:57
Why do you want to shift it 3 days forward? (or is it a mistake? 20190401 -> 2019-04-04) — Nir Alfasi
– Nir Alfasi, Commented Apr 25, 2019 at 18:57
You can start by splitting the file-name (without ext) by underscore, then you'll have three parts, the first you can discard and parse the other two parts. I don't see any advantage for regex here, you can parse these fields either as date/time or as strings. — Nir Alfasi
– Nir Alfasi, Commented Apr 25, 2019 at 19:00

Khalid Ali · Accepted Answer · 2019-04-25 19:19:49Z

6

import re

regex = re.compile(r'^IMG_(\d{4})(\d{2})(\d{2})_(\d{2})(\d{2})(\d{2})\.jpeg$')

oldStr = 'IMG_20190401_235959.jpeg';

match = regex.match(oldStr)

newStr = '{}-{}-{}_{}_{}.jpg'.format(*match.groups())

print(newStr) # 2019-04-01_23_59.jpg

answered Apr 25, 2019 at 19:19

Khalid Ali

1,2241 gold badge10 silver badges12 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Laurent LAPORTE · Accepted Answer · 2019-04-25 19:25:12Z

What you can do is to precompile the RegEx before using it. You can proceed that way:

import re

sub_name = re.compile(r"IMG_(\d{4})(\d{2})(\d{2})_(\d{2})(\d{2})(\d{2})", flags=re.I).sub

Here, sub_name is a function that you can use later in you for loop to replace the name of each image.

Note: Ignoring the case (upper/lower-case) can be useful under Windows, but you also need to adapt the call to glob.glob.

Below is a solution which use glob.glob but you can also use os.walk to browse a directory, searching all images…

# coding: utf-8
import glob
import os
import re
import sys

sub_name = re.compile(r"IMG_(\d{4})(\d{2})(\d{2})_(\d{2})(\d{2})(\d{2})", flags=re.I).sub

work_dir = sys.argv[1]

for old_path in glob.glob(os.path.join(work_dir, "IMG_*.jpg")):
    dirname, old_name = os.path.split(old_path)
    new_name = sub_name("\\1-\\2-\\3_\\4_\\5_\\6", old_name)
    new_path = os.path.join(dirname, new_name)
    try:
        os.rename(old_path, new_path)
    except OSError as exc:
        print(exc)

I noticed that you use the print statement, and the Python 2.6 syntax for exception. It is preferable to use the new syntax. If you use Python 2.7, you can add the directive:

from __future__ import print_function

Put it at the top of your imports…

Ajax1234 · Accepted Answer · 2019-04-25 19:02:47Z

1

You can use re.findall to grab the necessary groups from the file path, and the rejoin:

import re
def new_path(s):
  _, a, b, f_type = re.findall('[a-zA-Z0-9]+', s)
  new_b = '_'.join(b[i:i+2] for i in range(0, len(b), 2))
  return f'{a[:4]}-{a[4:6]}-{a[6:]}_{new_b}.{f_type}'

print(new_path('IMG_20190401_235959.jpg'))

Output:

'2019-04-01_23_59_59.jpg'

Then:

import os, glob, sys, re    
os.chdir(sys.argv[1])
for filename in glob.glob("IMG_*.jpg"):
  try:
    os.rename(filename, new_path(filename))
  except OSError,e:
    print(e)

answered Apr 25, 2019 at 19:02

Ajax1234

71.7k9 gold badges67 silver badges110 bronze badges

Comments

Alec · Accepted Answer · 2019-04-25 19:06:10Z

1

Not sure if regex is the best option here. You can split it up and use basic string operations pretty easily:

original = 'IMG_20190401_235959.jpg'
ol = original.split('_')
date = f'{ol[1][:4]}-{ol[1][4:6]}-{ol[1][6:8]}'
time = f'{ol[2][:2]}_{ol[2][2:4]}_{ol[2][4:6]}'
new = f'{date}_{time}.jpg'
print(new)

answered Apr 25, 2019 at 19:06

Alec

9,7338 gold badges44 silver badges71 bronze badges

1 Comment

SanV Over a year ago

I like this approach. simple and efficient.

SanV · Accepted Answer · 2019-04-25 19:08:12Z

0

If your input is consistent, this should work:

import re
pattern = r"IMG_(\d{4})(\d{2})(\d{2})_(\d{2})(\d{2})(\d{2})"
test_str = "IMG_20190401_235959.jpg"
subst = "\\1-\\2-\\3_\\4_\\5_\\6"
result = re.sub(pattern, subst, test_str, 0, re.MULTILINE)
if result:
    print (result)

# 2019-04-01_23_59_59.jpg

answered Apr 25, 2019 at 19:08

SanV

9459 silver badges17 bronze badges

Comments

schilli · Accepted Answer · 2019-04-25 19:35:49Z

0

The following code worked for me, however it only uses regex to remove the IMG_ in the filename, so you can also just drop the regex completely.

newfilename = re.sub('IMG_', '', filename)
newfilename = newfilename[0:4] + '-' + newfilename[4:6] + '-' + newfilename[6:11] + '_' + newfilename[11:13] + '_' + newfilename[13:]

edited Apr 25, 2019 at 19:35

answered Apr 25, 2019 at 19:06

schilli

1,8581 gold badge12 silver badges18 bronze badges

2 Comments

101arrowz Over a year ago

Instead of using RegEx at all, why not just do newfilename = newfilename[4:] instead of the re.sub?

schilli Over a year ago

Ya, that's what I mean with the comment "...so you can also just drop the regex completely."

Collectives™ on Stack Overflow

Python renaming file with regex

6 Answers 6

Comments

Comments

Comments

1 Comment

Comments

2 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

6 Answers 6

Comments

Comments

Comments

1 Comment

Comments

2 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related