What the heck does this mean?

[vonman]

            if(  m#^coclli=(.*)# && $coclli eq "" ) {
               $coclli = $1;
               }

I have inherited a piece of code that "supposedly" works.

There are no variables or subroutines named m. What hould this code do? I see it as m then a comment for the rest of the line. Any ideas?

Thanks

Edit 2001-03-12 by tye

[japhy]

Running it through the deparser, I get:

if (/^coclli=(.*)/ and $coclli eq '') {
    $coclli = $1;
}

It's using a regex to assign to $coclli if the variable isn't already assigned.

The deparser is available in Perl 5.005 and up, as:

japhy% perl -MO=Deparse program_name



japhy -- Perl and Regex Hacker

[Falkkin]

Basically, in English, this code reduces to:

If the string $_ starts with "coclli=", and then "some characters", and the variable $coclli is equal to the empty string, set the variable $coclli to the (any-amount-of-characters).

[vonman]

There should be a newline after the eq"")and the $1;. Txs

[myocom]

perldoc -f m tells you to look at the perlop manpage. In short, it's the match operator.

[vonman]

I think what is happening is that the original code must not be displaying characters correctly on my machine. So what should I replace m and # with?

[japhy]

What are you talking about? m// is the pattern match operator, and you can use characters other than / for the delimiter. Thus, the code can be:

/^coclli=(.*)/
# or
m!^coclli=(.*)!
# or
m<^coclli=(.*)>

Perl can be as eccentric as the programmers who use it.

japhy -- Perl and Regex Hacker

[astanley]

perl allows you to specify any character as the regexp seperator. / is just a suggestion and I guess is most widely used? Either way the perldocs say you can use m## just as easy as m// if, for instance you wanted to match a string that had a lot of /'s in it and didn't want to \/ them all - m#///# would work and is a little easier to read than m/\/\/\//.
Adam