Wsrtjtyk

I need to write a script that will look through a folder and strip out the text from a string of an image.

image-w-inch-bob-bob-bob-bob-8820-AV1.jpg

image-w-inch-bob-bob-bob-bob-8820-AV2.jpg

image-w-inch-bob-bob-bob-bob-8820-AV3.jpg

image-w-inch-bob-bob-bob-bob-8820-AV4.jpg

image-w-inch-bob-bob-bob-bob-8820-AV5.jpg

image-w-inch-bob-bob-bob-bob-8820-AV6.jpg

I need this to be renamed to

8820-AV1.jpg

8820-AV2.jpg

8820-AV3.jpg

8820-AV4.jpg

8820-AV5.jpg

8820-AV6.jpg

Does anyone know of a script like this?

The most natural way would be to use regular expressions.

Here's an example of a python implementation:

import re

s = 'image-w-inch-bob-bob-bob-bob-8820-AV1.jpg'

capture_from_first_digit_re = re.compile('D*(d.*)')

print(capture_from_first_digit_re .findall(s)[0])

You can use this as you see fit in your context.

Short explanation:

• D - anything except a digit


• D* - keep going as long as you don't hit a digit


• d - any digit


• .* - anything at all


• (###) - capture the matching ###

Putting it together: capture everything after the first digit you find.

If you're just trying to strip out all text before the first digit is found in the filename, something along these lines with a relatively simple regex should work in python:

import os

import re

# replace with the path to your file:

path = 'test/'

for filename in os.listdir(path):

    os.rename(os.path.join(path, filename),

              os.path.join(path, re.search('d.*',filename).group()))

We can create a test file just for fun to see that it works:

import os

import re



# list out all the filenames to put into our test directory

l=['image-w-inch-bob-bob-bob-bob-8820-AV1.jpg',

   'image-w-inch-bob-bob-bob-bob-8820-AV2.jpg',

   'image-w-inch-bob-bob-bob-bob-8820-AV3.jpg',

   'image-w-inch-bob-bob-bob-bob-8820-AV4.jpg',

   'image-w-inch-bob-bob-bob-bob-8820-AV5.jpg',

   'image-w-inch-bob-bob-bob-bob-8820-AV6.jpg']



# Create Directory

os.mkdir('test')



# add in all the files

for f in l:

    open(f'test/{f}','a').close()



# All the files are there

>>> os.listdir('test')

['image-w-inch-bob-bob-bob-bob-8820-AV5.jpg', 'image-w-inch-bob-bob-bob-bob-8820-AV4.jpg', 'image-w-inch-bob-bob-bob-bob-8820-AV6.jpg', 'image-w-inch-bob-bob-bob-bob-8820-AV3.jpg', 'image-w-inch-bob-bob-bob-bob-8820-AV2.jpg', 'image-w-inch-bob-bob-bob-bob-8820-AV1.jpg']



# rename with the loop provided above:



path = 'test/'

for filename in os.listdir(path):

    os.rename(os.path.join(path, filename),

              os.path.join(path, re.search('d.*',filename).group()))



# all the filenames have changed

>>> os.listdir('test')

['8820-AV1.jpg', '8820-AV3.jpg', '8820-AV2.jpg', '8820-AV6.jpg', '8820-AV5.jpg', '8820-AV4.jpg']

The easiest way to do this would be to locate the python script inside the folder the files are located. Assuming that all the filenames have the same amount of characters before the first number, then you'd need something like:

import os

for file in os.listdir('.'):

    if '.py' not in file: #not the python script

        os.rename(file, file[indexOfFirstNum:])

I haven't tested the script so try it first and modify accordingly before running it in the real folder.

EDIT: Refer to shlomif's answer if you want the script to be more general. Regular expressions are useful for finding and matching patterns in strings.

Here is the Perl one liner solution:

$ ls -l

total 0

-rwxrw-r--+ 1 xxxx xxxx 0 Nov 22 08:23 image-w-inch-bob-bob-bob-bob-8820-AV1.jpg

-rwxrw-r--+ 1 xxxx xxxx 0 Nov 22 08:23 image-w-inch-bob-bob-bob-bob-8820-AV2.jpg

-rwxrw-r--+ 1 xxxx xxxx 0 Nov 22 08:23 image-w-inch-bob-bob-bob-bob-8820-AV3.jpg

-rwxrw-r--+ 1 xxxx xxxx 0 Nov 22 08:23 image-w-inch-bob-bob-bob-bob-8820-AV4.jpg

-rwxrw-r--+ 1 xxxx xxxx 0 Nov 22 08:23 image-w-inch-bob-bob-bob-bob-8820-AV5.jpg

-rwxrw-r--+ 1 xxxx xxxx 0 Nov 22 08:23 image-w-inch-bob-bob-bob-bob-8820-AV6.jpg



$ perl -ne ' BEGIN { foreach $f (glob("image-w*.jpg")) { $newf=$f, $newf=~s/(.[^d]*)-(d+)(.*)/23/g; rename $f, "$newf" } exit } '



$ ls -l

total 0

-rwxrw-r--+ 1 xxxx xxxx 0 Nov 22 08:23 8820-AV1.jpg

-rwxrw-r--+ 1 xxxx xxxx 0 Nov 22 08:23 8820-AV2.jpg

-rwxrw-r--+ 1 xxxx xxxx 0 Nov 22 08:23 8820-AV3.jpg

-rwxrw-r--+ 1 xxxx xxxx 0 Nov 22 08:23 8820-AV4.jpg

-rwxrw-r--+ 1 xxxx xxxx 0 Nov 22 08:23 8820-AV5.jpg

-rwxrw-r--+ 1 xxxx xxxx 0 Nov 22 08:23 8820-AV6.jpg



$

You most likely can do this with rename if your distro provides this command. For example:

$ rename 'image-w-inch-bob-bob-bob-bob-' '' image-w-"

or with PCRE Regex as the expression:

$ rename 's/image-w-inch-bob-bob-bob-bob-//' image-w-*

rename can batch rename a set of files according to a pattern match, including by regular expression. Note that the are several versions of rename which accepts different syntax, you should consult your local man and/or the command's help page to determine which version you are using and the precise syntax to use.