Need to change of file

Shell Programming and Scripting
1

I need to change the filed named from

3479238.conf.bad
3274293.conf.bad
2389749.conf.bad
3242397.conf.bad

Need to move as

3479238.conf
3274293.conf
2389749.conf
3242397.conf

Via shell script or any find command.

2

use man cut

3

Assuming the only files present with the name *.conf.bad are the ones to be renamed then:

for SFILE in *.conf.bad; do
  TFILE=`echo ${SFILE} | sed 's/.bad//'`
  mv ${SFILE} ${TFILE}
done

Would do the job.

4

Thank you, it worked out...

5

That's useless use of echo & cut

for FILE in *.conf.bad
do
   mv ${FILE} ${FILE%.*}
done
6

Not "useless", just not as efficient (and it was sed not cut :wink: ).

7

"TonyFullerMalv" solution is cool because it follows logical lines. "danmero" solution is naive.

Slight improvement if we quote every filename and every ".", and allow for no files:

ls -1 *\.conf\.bad 2>/dev/null | while read SFILE
do
  TFILE=`echo "${SFILE}" | sed 's/.bad//'`
  mv "${SFILE}" "${TFILE}"
done
8

Can you explaing why? Let's see the difference on 10000 test files.

time for SFILE in *.conf.bad; do   TFILE=`echo ${SFILE} | sed 's/.bad//'`;   mv ${SFILE} ${TFILE}; done

real    1m54.197s
user    0m28.088s
sys     1m20.431s

And now using shell built-in parameter expansion.

# time for FILE in *.conf.bad; do mv ${FILE} ${FILE%.*}; done

real    0m32.753s
user    0m7.041s
sys     0m24.094s

Look like the shell only is almost 4 time faster that your solution. You should think about that :rolleyes: