Thanks to all who replied.
Clive Eisen suggested the following solution.
It is short and elegant and it works great!
-Mahtaj
====
Untested off the 'top of my head' :-)
while (($operid,$emplid,$course) = $handle->fetchrow_array {
$output{$operid .' ' .$emplid} .= $course . ' ';
}
foreach ( sort keys %output ) {
print "$_ $output{$_}\n";
}
--
Clive
> -----Original Message-----
> From: Steven Lembark [mailto:[EMAIL PROTECTED]
> Sent: Wednesday, December 17, 2003 4:23 PM
> To: Michael Higgins; Khamneian, Mahtaj
> Cc: [EMAIL PROTECTED]
> Subject: Re: Simple Perl
>
>
>
>
> -- Michael Higgins <[EMAIL PROTECTED]>
>
> > Khamneian, Mahtaj wrote:
> >
> >> I have a query that returns the following data from the db:
> >> ...
> >> operid emplid course
> >> -------- ---------- ---------
> > [snip]
> >> ...
> >>
> >> Noting that different courses are associated with the same
> oprid and
> >> emplid, I need to format the output so that a single line
> is printed for
> >> each oprid. e.g.,
> >> SSPT490 12050971 37290 38380 38540 52320
> >>
> >> I have used both fetchall_arrayref() and fetchrow_array()
> methods to
> >> fetch the data. I can think of a quick and dirty way of
> doing this, but
> >> was looking for an efficient/elegant way.
> >>
> >> Any/all help is appreciated.
> >>
> >
> > Maybe push them onto a hashkey?
> >
> > (untested)
> >
> > my %results;
> >
> > foreach (@$fetched){
> >
> > push @{$results{$_->[0].' '.$_->[1]}}, $_->[2];
> >
> > }
> >
> > map {print "$_ @{$results{$_}}\n"} sort keys %results;
>
> If the data can be sorted you can also just walk down the
> list, printing items until the first field changes:
>
> local $\;
> local $,;
>
> my $last = '';
>
> for my $row ( @$rowz )
> {
> if( $last ne $row->[0] )
> {
> print "\n" if $last;
>
> print $last = $row->[0];
> }
>
> print join "\t" '', @{$row}[1,2];
> }
>
> print "\n";
>
> i.e., print a newline if the last record has a different
> $row->[0] than this one, then the new row header. for each
> record print the fields pasted together with tabs and a
> leading tab separator.
>
>
> The hash trick is necessary if your data is not sorted or
> if you have to accumulate it in one place to maniuplate it
> before printing the stuff.
>
> Aside: You could modify the loop above for a speed gain by
> using fetchrow w/ bound values. That would leave you with
> something like "while( fetch... ){ ... }". If the list is
> non-trivial then sorting it and returning the rows singly
> may be less overhead (see the DBI book, DBI pod, and numerous
> postings of Tim's for examples of how to use the bound values).
>
> enjoi
>
> --
> Steven Lembark 2930 W. Palmer
> Workhorse Computing Chicago, IL 60647
> +1 888 359 3508
>
