Am Mittwoch, 13. April 2005 10.46 schrieb Ramprasad A Padmanabhan:
> On Wed, 2005-04-13 at 12:32, Thomas BÃtzler wrote:
> > Ramprasad A Padmanabhan <[EMAIL PROTECTED]> asked:
> > > I want to write a perl script like "gnu less".
> > >
> > > My perl script accepts input on STDIN or filename(s) as arguments.
> > > If both are missing it should print the error message. How do
> > > I do this ?
> >
> > There is always input on STDIN - even a straight eof could be input,
> > like in "cat empty_file | ramprasads_less". You should probably try
> > and check for filename arguments and revert to STDIN processing if
> > there are none, i.e.
> >
> > my @files;
> >
> > if( @ARGV ){
> > @files = @ARGV;
> > } else {
> > @files = ("-");
> > }
> >
> > foreach my $file (@files){
> > # transparent gzip decompression
> > $file = "/usr/bin/gzip -d < $file|" if $file =~ m/\.gz$/;
> >
> > open( IN, $file ) or die "Cam't open '$file': $!";
> > # do something
> > close( IN );
> > }
> >
> > HTH,
> > Thomas
>
> That is perfectly ok .. but I require to print usage if there is no
> input How do I do this ?
>
> Thanks
> Ram
The following code abbreviates the STDIN/filename issue, but of course the
assignement to $show_usage within the loop (for every input line) is not
optimal...
The script can be called by
$ script filename1 filename2
or
$ script < filename1
or simply (but rather senseless)
$ script
===
#!usr/bin/perl
use warnings; use strict;
my $show_usage=1;
while (<>) {
$show_usage=0;
# do something with the lines of the content
# provided by filenames or direct input from stdin
# Lines are provided in $_ .
}
usage if $show_usage;
sub usage {
print <<EOF
The usage here
EOF
}
===
>
>
> ----------------------------------------------------------
> Netcore Solutions Pvt. Ltd.
> Website: http://www.netcore.co.in
> Spamtraps: http://cleanmail.netcore.co.in/directory.html
> ----------------------------------------------------------
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
<http://learn.perl.org/> <http://learn.perl.org/first-response>
