I'm using a process that creates massive amounts of .jpg files. They are
named in numerical order as they are created:

00000001.jpg
00000002.jpg
00000003.jpg
..
..
etc.

I would like a simple(?) one liner or bash script that will delete (or move)
every even numbered file, and then rename the files remaining in correct
order, ie.:

file0001.jpg
file0002.jpg
file0003.jpg
..
..
etc.

I've tried fooling around with reg expressions to do this but I guess I
just don't get it

Thanks,
Geoff
--
** **
G. Sullivan sunfishATshell.gis.net

Geoff Sullivan wrote:

OK, I've tried

mv *[02468].jpg /wherever

and it works good enough for my puposes. Now to change the remaining
filenames. The formula for this would be something like x=y/2+.5 where y is
the original number of the file. I'd want to append x to a four-letter
prefix, followed by the extension .jpg.

Chris F.A. Johnson wrote:

This works great! Now can you EXPLAIN what goes on here so I can learn from
this instead of being a script leech?

Thanks Chris.

On Wed, 25 Feb 2004 at 14:17 GMT, Geoff Sullivan wrote:
rm *[02468].jpg
n=1
z=0000
for file in *.jpg
do
f=$z$n
newfile=file${f: -4}.jpg
mv "$file" "$newfile"
n=$(( $n + 1 ))
done

There are other ways to build the new file name:

newfile=`printf "file%04d.jpg" $n`

f=$z$n
newfile=file${f#${f%????}}.jpg

--
Chris F.A. Johnson http://cfaj.freeshell.org/shell
================================================== =================
My code (if any) in this post is copyright 2004, Chris F.A. Johnson
and may be copied under the terms of the GNU General Public License

In message <c1is26$1jipu2$1@ID-210011.news.uni-berlin.de>, Chris F.A.
Johnson wrote:

Shouldn't that be 'mv "$newfile" "$file"' ?

--
Garry Knight
garryknight@gmx.net ICQ 126351135
Linux registered user 182025

In message <1077738290.93232.0@doris.uk.clara.net>, Garry Knight wrote:

Sorry, forget that: brain in C mode.

--
Garry Knight
garryknight@gmx.net ICQ 126351135
Linux registered user 182025

On Wed, 25 Feb 2004 at 15:51 GMT, Geoff Sullivan wrote:
n=1 ## initialize first file number
z=0000 ## a string for padding with zeroes
for file in *.jpg ## for each file enfing with .jpg
do
f=$z$n ## put the zeroes ahead of the number
newfile=file${f: -4}.jpg ## use the last 4 characters of $f
mv "$file" "$newfile" ## move 'em out
n=$(( $n + 1 )) ## add one to $n
done

Extract the number from the file name:

file=00000009.jpg
num=${file%.jpg} ## remove .jpg

## remove leading zeroes, or the number will be read as octal,
## and the script will barf on 08 or 09
while :
do
case $num in
0*) num=${num#0} ;;
*) break ;;
esac
done

prefix=file
newnum=$(( $num / 2 + 5 ))
padnum=$num ## use any method you like to pad with zeroes
newfile=$prefix$padnum.jpg

--
Chris F.A. Johnson http://cfaj.freeshell.org/shell
================================================== =================
My code (if any) in this post is copyright 2004, Chris F.A. Johnson
and may be copied under the terms of the GNU General Public License

"GS" == Geoff Sullivan <sunfish@REMOVE.shell.gis.net>:
GS> I'm using a process that creates massive amounts of .jpg files. They are
GS> named in numerical order as they are created:
GS> 00000001.jpg
GS> 00000002.jpg
GS> 00000003.jpg
GS> I would like a simple(?) one liner or bash script that will delete (or move)
GS> every even numbered file, and then rename the files remaining in correct
GS> order, ie.:
GS> file0001.jpg
GS> file0002.jpg
GS> file0003.jpg

The folowing bash command will tell you what needs to be done :

c=0;for i in *.jpg;do x=`basename $i .jpg|sed 's@^0*@@';`;if [ 0 = \
$(($x%2)) ];then echo rm $i;else c=$(($c+1));echo mv $i \
`printf file%04s $c`.jpg;fi;done

Removing the 'echo' strings from the above command will silently do it
(probably dangerous without testing first).

Anonymous Sender wrote:

Anyone who would recommend that particular one-liner to
a newbie, even with the echo command in it, is a sadist!

::chuckle::


--
Ñ
"It is impossible to make anything foolproof because fools are so
ingenious" - A. Bloch

Geoff Sullivan wrote:

perl -e'$_%2?push@o,$_:unlink for<*.jpg>;rename$_,sprintf"file%04d.jpg",++$c for@o'


John
--
use Perl;
program
fulfillment

On Thu, 26 Feb 2004 01:33:02 +0000, John W. Krahn wrote:

Now that's just mean.

After about fifteen minutes of poking around the Python online docs,
and another five for testing and debugging, I came up with this:

#!/usr/bin/python
import glob
import os
import string
for unwantedfilename in glob.glob('*[02468].jpg'):
os.remove(unwantedfilename)
newnumber = 0
filenamelist = glob.glob('*[13579].jpg')
filenamelist.sort()
for oldfilename in filenamelist:
newnumber += 1
os.rename(oldfilename, 'file' + string.zfill(str(newnumber), 4) + '.jpg')