Shell Programming and Scripting

Hi,
I am writing a while_do_done loop in ksh script to read records from a file.
===============
while read record
do
if newrec = oldrec
dosomething
else
save the old data to a file
done < file
================

My intension is to read every record, if the new record is the same as the previous one, then dosomething, else save the data of the previous record and continue read the next record from file.
My problem is I only save previous data to a new file when I read next different record. So if I only have 3 same records, their data will not be save because this loop will not read next different record( because I don't have any). But how do I save my data in this kind of situation?


Thanks !

Here's a kludge:

Code:

{ while read record
  do echo rec "$record"
  done < file
  echo eof
} | while read meta data
  do
    ...
  done

This is a good thought. But variable eof is not a system variable. How can this get the last record?

eof is not a variable.

Code:

$ cat > test.txt
hello
there
hehe
^D
$ { while read record
>   do echo rec "$record"
>   done < test.txt
>   echo eof
> } | while read meta data
>   do
>     echo "[$meta] $data"
>     case "$meta" in
>     eof) echo "that was the last record" ;;
>     esac
>   done
[rec] hello
[rec] there
[rec] hehe
[eof]
that was the last record

Maybe try using the "uniq" utility to remove duplicate lines from a sorted file.

i have test.txt like this ...
this is x
this is x
this is x

1) sort -u test.txt gives
this is x
2).



#!/usr/bin/ksh
>toFile

oldrecord=""

exec < test.txt
while read reco ;
do
if [ "$reco" = "$oldrecord" ]
then
oldrecord=$reco
else
echo $reco >> toFile
fi
oldrecord=$reco
done


gives the same result.

I think your script may drop the last record.
Add to your script

if [[ $record != "" ]] #check if records have been read
then
print $record
fi