i have to replace all the ? with X in unix except some valid?

user_15894604 0
Viewed 79

It kind of some complicated question. I want to replace all the ? in the file with X. But the problem is there are some valid ? also there in input file.

eg:

input:

HELLO ?, WELCOME TO THE NEW WORLD??23, and you are most ? valid person.

output:

HELLO X, WELCOME TO THE NEW WORLDX?23, and you are most X valid person.

here, ? comes before 23 is valid one.. like ?23, many values are there. ?24,?33,?45, etc.,

I tried with sed script, but able to find the exact command.

Script which i used:

LINE_NUM=0 

while IFS= read -r LINE    
do  
LINE_NUM=$?   
EXTRACTED=`echo "${LINE}" |grep '?23' |  sed 's|^.*\?23||; s|\?[0-9].*$||'`

if [ -n "$EXTRACTED" ]    
then    
UPDATED=`echo "$EXTRACTED" | sed 's/?/X/g'`    
UPDATED_1=`echo "$UPDATED" | awk '{gsub("/","%",$0); print}'`    
if [ $EXTRACTED != $UPDATED ]   
then    
LATEST_VALUE=`echo  "${LINE}" | sed "s|${EXTRACTED}|${UPDATED}|g"`    
fi

LATEST_VALUE=`echo  "${LINE}"`    
echo "$LATEST_VALUE"  >> outputfile.txt   
else    
echo "$LINE" >> outputfile.txt    
fi   
done<inputfile.txt
sed awk unix
2 Answers
$ echo "HELLO ?, WELCOME TO THE NEW WORLD??23, and you are most ? valid person." | 
  sed -E 's/\?([^0-9]|$)/X\1/g'


HELLO X, WELCOME TO THE NEW WORLDX?23, and you are most X valid person.

here it escapes ? followed by a digit (or end of line). If your list is more restricted change the regex there.

user_14358690

With your shown samples, please try following. In GNU awk please try following. Simple explanation would be, setting record separator as 1 ? followed by 1 digit, using global substitution to substitute ? with X in current records; setting correct output record separator as RT, then print current line.

awk -v RS='[?][0-9]' '{gsub(/\?/,"X");ORS=RT} 1' Input_file
user_58665800
Related