How to delete a character.

brock_pace · 07-02-2012, 12:38 PM

Im trying to change a filename so that it doesn't have any hyphen/dashes in it. Basically, I have a file like a-1, I just want a1. any suggestions.

earthnet · 07-02-2012, 12:49 PM

Have you tried anything to rename the file?

brock_pace · 07-02-2012, 12:50 PM

Well thats not actually the file name, just an example. But with the real file name ive changed it to make it all lowercase using tr.

earthnet · 07-02-2012, 12:59 PM

So, did you figure it out? Do you still need help?

brock_pace · 07-02-2012, 01:01 PM

yeah, i dont know how to take out a hyphen from a file name

earthnet · 07-02-2012, 01:02 PM

In your example it would be

Code:

mv a-1 a1

brock_pace · 07-02-2012, 01:06 PM

hm, yeah that works great on my file, but for curiosities sake, how would you get rid of the hyphens in a string of randomly named files.

earthnet · 07-02-2012, 01:15 PM

To remove the hyphens out of all the filenames in a directory you'd create a bash script:

Code:

#!/bin/bash
for i in ls
do
  mv $i `echo $i | sed -e 's/-//g'`
done

brock_pace · 07-02-2012, 01:17 PM

im new, and to me that seems complex, could you explain it for me? that would be helpful.

earthnet · 07-02-2012, 01:27 PM

Basically you put those commands into a text file in the directory in question. Change the file to be executable.

Code:

chmod 755 <file>

Then run it.

Code:

./<file>

Here's a breakdown of what is happening:

Code:

#!/bin/bash

This tells the system that it's a bash script.

Code:

for i in `ls`

(it should be `ls` not just ls) Take the output of the "ls" command and store it in the variable $i, repeat what is between "do" and "done" for every file in the list.

Code:

mv $i `echo $i | sed -e 's/-//g'`

mv $i will output "mv <filename>", the "sed" bit outputs the filename without hyphens. So it runs "mv <filename> <filename_without_hyphens>".

brock_pace · 07-02-2012, 01:35 PM

thanks, that worked. ok, since you are awesome, now lets suppose in these filenames there are numbers in order, 1,2,3,4,...etc. how could i then reformat them so that they are each four digits. 0001, 0002, 0003, 0004, etc.

earthnet · 07-02-2012, 01:40 PM

Here's another thread talking about just that.

TobiSGD · 07-02-2012, 02:12 PM

Quote:

Originally Posted by earthnet

Code:

for i in `ls`

Using the output of ls in scripts is considered bad practice, since it can cause problems with filenames that have a space or a newline in them. I would change that to

Code:

for i in *

pixellany · 07-02-2012, 02:35 PM

Quote:

Originally Posted by brock_pace

thanks, that worked. ok, since you are awesome, now lets suppose in these filenames there are numbers in order, 1,2,3,4,...etc. how could i then reformat them so that they are each four digits. 0001, 0002, 0003, 0004, etc.

brock;

If you look back over all the help you've been getting here, this should be pretty easy----e.g. have you read the SED reference we discussed earlier?

In SED, you can replace a pattern explicitly, or you can detect it and then replace it using a back reference. Here's an example of the latter: (eg imagine I'm acting on a string "xyz", and I want it to be "oooxyz")

Code:

sed 's/\(xyz\)/ooo\1/'

now, instead of the literal "xyz", suppose I want to do the same thing for any 3 characters:

Code:

sed 's/\(...\)/ooo\1/'

Now---in your question---do you want to change all files beginning is a single number? What exactly is the rule to be applied?

brock_pace · 07-09-2012, 02:16 PM

I would like for all files to have 4 digit numbers. For instance 1 would be 0001. 257 would be 0257. Im doing this for 240 files.