Korn shell READ command taking hours to run through simple loop; can this be improved?

Hello

I have developed a script designed to read multiple files and extract spcific text information and write that info to an xml file. However, it takes many many hours (>24 in test). Since all the files are contained locally, I suspect the script itself is at fault.Please have a look at this code and file sample and help me find a more effect method. Kep in mind this is an old system. The version of MKS is unsupported and ancient (still waiting for a exact version from our Admins).

The translation file is built using SQR, so there are lots of odd characters that will not paste correctly into this text box, but the point of the file is to have element markers like such:

90v652H|element_X|0|value_X
90v2044H|element_X_X|1|value_X_X
180v3320H|element_X_X_X|2|value_X_X_X
270v4596H|element_X_X_Y|2|value_X_X_Y
270v4596H|element_X_Y|1|value_X_Y

The logic is that the XML processing job will extract the test between the pipes "|" and use the number as a tree index and the last part of the marker as the element value:

90v652H|element_X|0|value_X
90v2044H|element_X_X|1|value_X_X
180v3320H|element_X_X_X|2|value_X_X_X
270v4596H|element_X_X_Y|2|value_X_X_Y
270v4596H|element_X_Y|1|value_X_Y

Becomes

<element_X>value_X
      <element_X_X>value_X_X
            <element_X_X_X>value_X_X_X</element_X_X_X>
            <element_X_X_Y>value_X_X_Y</element_X_X_Y>
      </element_X_X>
      <element_X_Y></element_X_Y>
</element_X>

Here is the code:
#######Begin Code#######
#!/bin/ksh

USAGE="Usage :: parser.ksh [file name]"

#  Set Jobname
#
export JOBNAME=printing_xml
export FEEDER_DIR=${JOBS_DIR}${JOBNAME}/
REPORT_NAME=$1
export LIS_FILE=${FEEDER_DIR}Report/${REPORT_NAME}.lis
export LOG_FILE=${FEEDER_DIR}Log/${REPORT_NAME}_xml.log

echo `date` >> $LOG_FILE

case $# in
  1) if [ -e ${LIS_FILE} ]
     then
        echo LIS located > $LOG_FILE
     else
        echo Could not find ${LIS_FILE} >> $LOG_FILE
     exit    
     fi
     ;;

  *) # wrong parameters
     echo "Usage error" >> $LOG_FILE
     print $USAGE
     exit
     ;;
esac

echo "Setting up header and xml file" >> $LOG_FILE

XML_FILE=${FEEDER_DIR}Report/filename.xml
exec 3> "$XML_FILE"
echo "<?xml version="\""1.0"\"" encoding="\""ISO-8859-1"\""?>" >&3

#set up array tracker for walking through elements
ELEMENT_INDEX=-1

echo "Processing files" >> $LOG_FILE

LIS_COUNTER=0

#go through each file
while  [ -e $LIS_FILE ]
do
    echo "Reading " $LIS_FILE >> $LOG_FILE
    #Go through each line looking for '|'
   
    while read curr_line
    do
       CODE_VALIDATE=`echo "$curr_line" | grep -c "|"`
       
       if [ $CODE_VALIDATE -eq 1 ]
       then
           #round out the numbers
           EVAL_ELEMENT_VALUE=`echo "$curr_line" | cut -f4 -d"|"`
           if echo $EVAL_ELEMENT_VALUE | egrep '^-{0,1}[0-9]*\.[0-9]+$' >/dev/null 2>&1
           then
               EVAL_ELEMENT_VALUE=$(printf "%3.2f" $EVAL_ELEMENT_VALUE)
           fi
           
           if echo $EVAL_ELEMENT_VALUE | egrep '\&' >/dev/null 2>&1
           then
               EVAL_ELEMENT_VALUE=`echo "$EVAL_ELEMENT_VALUE" | sed -e "s/&/and/g"`
           fi

           #place element in xml file according to their index
           ELEMENT=`echo "$curr_line" | cut -f2 -d"|"`
           ELEMENT_INDEX_TMP=`echo "$curr_line" | cut -f3 -d"|"`
           ELEMENT_VALUE=$EVAL_ELEMENT_VALUE
           
           if [ $ELEMENT_INDEX_TMP -gt $ELEMENT_INDEX ]
           then
               ELEMENT_INDEX=$ELEMENT_INDEX_TMP
               ARR_ELEMENTS[$ELEMENT_INDEX]="$ELEMENT"
               echo "<${ARR_ELEMENTS[$ELEMENT_INDEX]}>${ELEMENT_VALUE}" >&3
           elif [ $ELEMENT_INDEX_TMP -eq $ELEMENT_INDEX ]
           then
               echo "</${ARR_ELEMENTS[$ELEMENT_INDEX]}>" >&3
               ELEMENT_INDEX=$ELEMENT_INDEX_TMP
               ARR_ELEMENTS[$ELEMENT_INDEX]="$ELEMENT"
               echo "<${ARR_ELEMENTS[$ELEMENT_INDEX]}>${ELEMENT_VALUE}" >&3
           else
               while [ $ELEMENT_INDEX_TMP -lt $ELEMENT_INDEX ]
               do
                   echo "</${ARR_ELEMENTS[$ELEMENT_INDEX]}>" >&3
                   ELEMENT_INDEX=$ELEMENT_INDEX-1
               done
               echo "</${ARR_ELEMENTS[$ELEMENT_INDEX]}>" >&3
               ELEMENT_INDEX=$ELEMENT_INDEX_TMP
               ARR_ELEMENTS[$ELEMENT_INDEX]="$ELEMENT"
               echo "<${ARR_ELEMENTS[$ELEMENT_INDEX]}>${ELEMENT_VALUE}" >&3
           fi          
       fi
    done  < "$LIS_FILE"
   
    let "LIS_COUNTER=$LIS_COUNTER+1"
   
    if [ $LIS_COUNTER -gt 9 ]
    then
        LIS_FILE="${FEEDER_DIR}Report/${REPORT_NAME}_${LIS_COUNTER}.lis"
    else
        LIS_FILE="${FEEDER_DIR}Report/${REPORT_NAME}_0${LIS_COUNTER}.lis"
    fi
done

#clean up remaining tags
while [ $ELEMENT_INDEX -gt 0 ]
do
    echo "</${ARR_ELEMENTS[$ELEMENT_INDEX]}>" >&3
    ELEMENT_INDEX=$ELEMENT_INDEX-1
done
echo "</${ARR_ELEMENTS[$ELEMENT_INDEX]}>" >&3

#######END Code#######

Can someone help me optimize this job and reduce it's run time greatly?

Also, for a hero cookie, I noticed the code that puts the closing tags on the end of the document failed to print on a big run (10+ meg), this may be part of the multiple file logic, but some feedback on that would be appreciated too.