1. substr $var, 1;

   perldoc -f substr for more info, on Unix

2. s/\?.*//

   Seearch for: a literal question mark character, followed by any number of any characters; replace all that with: nothing ( i.e., delete it )

Proceessing input files line by line is the best way to do it, unless there are reasons for processing larger chunks. hint: most of the time, when you think you need to process larger chunks, what you really need is to find a different way of thinking about the process.

In the then clause, you chop( $qnum ), removing the last character. Or is this supposed to be the reverse-chop, from part 1? In any case, $qtext continues to end with a newline character.

In the else clause, you chomp( $qspec ). chomp() is like chop(), except it removes only the current $INPUT_RECORD_SEPARATOR, $/, normally "\n". $qspec can't have a newline, since it comes from the middle of a line. But $x ends with a newline.

I would suppose using chomp() right after the while(), to trim newlines from all input.

Update: I forgot the most important part ... You left out "use strict;", and you should localize your variables with my.

The regular expression also removes the question mark, which should remain.

  s/\?.*/?/
[download]

perl -pi -le'$_=substr($_,0,index($_,"?")+1)' file.txt
[download]

L-LL-L--L-LL-L--L-LL-L--
-R--R-RR-R--R-RR-R--R-RR
B--B--B--B--B--B--B--B--
H---H---H---H---H---H---
(the triplet paradiddle with high-hat)

No -i switch if you call perl as a filter from an editor!

Then what about:

#!/usr/bin/perl -pi

s/this/that/g;
[download]

I honestly don't know how to call filters from vi ... so i don't know how to test this. But i think this should take care of the problem.

UPDATE:
Just figured out how to use filters in vi, that wasn't hard at all! Yes, this definetly takes care of the problem. :)

jeffa

L-LL-L--L-LL-L--L-LL-L--
-R--R-RR-R--R-RR-R--R-RR
B--B--B--B--B--B--B--B--
H---H---H---H---H---H---
(the triplet paradiddle with high-hat)

I answer question 1 only. You can use this code:

$string=~ s/(.)//s; $char= $1;
[download]

There's also an other solution:

$char= substr ($string, 0, 1, "");
[download]

Thanks for the hints-- I got something that works And learned a lot in the process. Which is always the goal. In case anyone ever is interested in calling a Perl script from VI, here's my solution (does several things besides those ennumerated in OP):

#!/usr/bin/perl  -w
use strict;
my ($num,$v,$text,$qnum,$qtext,$stext,$spec,$x,$qspec);            
while (<>){
        s/ \t/\t/g;      
        s/9[0-9][0-9][0-9] \t/9999\t/g;
        s/9[0-9][0-9][0-9] /9999\t/g;                              
        if (/^[SCQD]/){
        ($num,$v,$text) = split("\t");
        chop ($num);
        $qnum=substr($_,0,index($_,"."));
        $qtext = substr($text,0,index($text,"?")+1);            
        print "n03;coltxt=___\nn05TOTAL\nL $qnum\nttl$qnum. $qtext\nn1
+0TOTAL BASE\n";
        }
        else {   
        ($stext,$spec,$x) = split("[\t]");       
        chomp ($spec);
        $qspec = substr($spec,1,1);
        unless (/^$/){ print "n01$stext;c=c{$qnum}'$qspec'\n"}
        }                        
}
        print "n03;coltxt=___\nn05TOTAL\n";
[download]

Basically, the key is to clear out some of the less predictable junk Before the if statement, and then use substr to control the more predictable junk in the scalar variables. And then if you want to add a line or two to the end of the section you're editing, print it in after the while loop and it adds the line(s) to the file.

Thanks again for the help.

Pax,

NovMonk

ps TomDLux: You're right about the chomp ($qspec); -- it's there because it Should be the last thing on the line, but wasn't always. I knew I didn't need anything after that field, so I named an extra one to make sure that junk didn't become part of $qspec.