about the bash shell command

naihe2010 · 10-25-2005, 09:19 PM

In a dir, I have lots of file like these :

abcdfasdfdsdfdsfs.mp3,
abdddsfdsfdafsdfdl.mp3,
ds125dsdfsddsd7dsfdsaf.pdd.mp3,
etc.

I want to rename them once.

for examble, I want to make them turned to be

abc_sfs.mp3,
abd_fdl.mp3,
ds1_pdd.mp3,
etc.

If the bash shell can do this ?

nilleso · 10-25-2005, 09:46 PM

yeah, I'm sure there is a shell script for that... I am just not aware of one at the moment.

However, if you are running kde, their is a built-in tool called krename which works great even for many hundreds of files.

ilikejam · 10-26-2005, 06:18 PM

Hi.

I thought that sounded a bit interesting, so I had a bash myself:

Code:

#!/bin/bash

for FILE in *
  do
  PREFIX=`echo "$FILE" | cut -b 1-3`
  EXTENSION=`echo "$FILE" | sed 's/.*\.//'`
  BASE=`basename "$FILE" $EXTENSION`
  BASECHARS=`echo "$BASE" | wc | awk '{ print $3 }'`
  CUTAT=`expr $BASECHARS - 4`
  POSTFIX=`echo $BASE | cut -b $CUTAT-$BASECHARS`
  mv "$FILE" "$PREFIX"_"$POSTFIX$EXTENSION"
done

That will rename every file in the current directory in the way described.

Try it out on a test directory before you use it on anything important.

Dave

naihe2010 · 10-26-2005, 09:14 PM

I think because there are lots of Unicode chractors in my file name,they turned to be like :

?????.wma,
????????.mp3,
etc

Thank you all the same.

ilikejam · 10-26-2005, 10:14 PM

Ah, Unicode.

The only statements in the above code that should be affected by Unicode are the 'cut' statements.
Try this one instead - I've replaced the 'cut's with 'awk'.

Code:

#!/bin/bash

for FILE in *
  do
  PREFIX=`echo "$FILE" | awk -F "" '{ OFS=""; i=1; print $i,$++i,$++i }'`
  EXTENSION=`echo "$FILE" | sed 's/.*\.//'`
  BASE=`basename "$FILE" $EXTENSION`
  BASECHARS=`echo $BASE | wc -m`
  ((BASECHARS--))
  CUTAT=`expr $BASECHARS - 3`
  POSTFIX=`echo $BASE | awk -F "" '{ OFS=""; i='$CUTAT'; print $i,$++i,$++i }'`
  mv "$FILE" "$PREFIX"_"$POSTFIX"."$EXTENSION"
done

Dave

freegianghu · 10-26-2005, 10:34 PM

Yet another way:

Code:

for mp3 in `ls *.mp3 | sed s/.mp3//g`; do
    mv $mp3.mp3 ${mp3:0:3}_${mp3:`expr length $mp3 - 3`:3}.mp3
done

Cheers,

ilikejam · 10-26-2005, 11:34 PM

Quote:

Originally posted by freegianghu
Yet another way:

Code:

for mp3 in `ls *.mp3 | sed s/.mp3//g`; do mv $mp3.mp3 ${mp3:0:3}_${mp3:`expr length $mp3 - 3`:3}.mp3 done

Cheers,

What happens if you have a file called 'hello.mp3.mp3'? ;-)

Actually, my script above does weird things with very short filenames (e.g. 'a.mp3') and files with no extension...

I like the ${mp3:0:3} notation, though. I've never seen that before - should prove usefull at some point.

Dave

freegianghu · 10-27-2005, 12:40 AM

Quote:

Originally posted by ilikejam
What happens if you have a file called 'hello.mp3.mp3'? ;-)

Actually, my script above does weird things with very short filenames (e.g. 'a.mp3') and files with no extension...

I like the ${mp3:0:3} notation, though. I've never seen that before - should prove usefull at some point.

Dave

hello.mp3.mp3 => hel_mp3.mp3
ds125dsdfsddsd7dsfdsaf.pdd.mp3 => ds1_pdd.mp3

Whats about you?

I havent tested with short filename, thanks for your comment!

Cheers up,