UNIX for Beginners Questions & Answers

You have an error in your assessment of the problem: what you want is NOT "multiline input" but in fact you want your input one line at a time and work on each line separately. That is, if your input is:

Code:

line1
line2
line3
..

you first want to process "line1", then "line2" and so forth, yes?

You achieve this by reading one line at a time (the "echo" stands for any arbitrary processing you might want to do, replace it with a code block that does what you really want to do):

Code:

while read INPUT ; do
     echo "= $INPUT ="
done

Now save this to a file "script.sh" and call it this way:

Code:

$ ./script.sh < /path/to/input.file

and you will see the content of the file, one line at a time, but surrounded by equal signs (the proof that the script really has read and processed your file).

Note that it is good style to first "correct" the input: trim leading/trailing blanks, tabs, empty lines, maybe filter out comment lines. Suppose your input file looks like this:

Code:

# lines from project A
line-1
    line-2
# line-3   # obsolete

# lines from project B
line-4     
 line-5

What you probably want to process in fact is this:

Code:

line-1
line-2
line-4
line-5

therefore, refine your script like this ("<b>" is a literal blank, "<t>" a tab):

Code:

sed 's/#.*//;s/^[<b><t>]*//;s/[<b><t>]*$//;/^$/d' |\
while read INPUT ; do
     echo "= $INPUT ="
done

I hope this helps.

bakunin