The reason that the header info is only included in the
first file is most likely a consequence that your total
lines of all index files are less than the value of
$= (current page length - see perlvar), which
defaults to 60. If not, then find the file with the 58th
line (59th, something like that ;)) and you should find a
header.
Top-of-Page Formats are meant for hardcopy devices, but
you can achieve the results you want with a simple hack:
make $= equal to the number of lines you are about to
print plus 2 for the numbers of lines in INDEXFILE_TOP.
Here is an example:
use strict;
my ($prog,$desc,$pages,$date,$file);
my @line1 = (
qq|prog desc pages data file\n|,
qq|foo blah 5 bar foo.txt\n|,
qq|bar blah 5 baz bar.txt\n|,
qq|baz blah 5 qux baz.txt\n|,
);
my @line2 = (
qq|prog desc pages data file\n|,
qq|foo blah 5 bar foo.txt\n|,
);
$~ = 'INDEXFILE';
write_2_file(@line1);
write_2_file(@line2);
sub write_2_file {
$= = 2 + @_;
for (@_) {
chomp;
($prog,$desc,$pages,$date,$file) = split;
write; # to STDOUT for this example
}
}
format INDEXFILE_TOP =
Program Description Pages Date Filename
------------------------------------------------
.
format INDEXFILE =
@<<<<< @<<<<<<<<<<<<<<<<<< @<<<<< @<<<<< @<<<<<<<<<<<
$prog, $desc, $pages, $date, $file
.
The only things that really bugs me about this code is
having to hard code the number of lines in INDEXFILE_TOP.
There might be a way to derive this number, i'll be on
the lookout for it (anybody knows...feel free to share ;))
UPDATE: Doh! Looks like rinceWind has a much cleaner
answer.
Change '$= = 2 + @_' to '$- = 0' in
the above example for a more robust subroutine.
rinceWind++
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)
|