=head1 NAME

urifind - find URIs in a document and dump them to STDOUT.

=head1 SYNOPSIS

    $ urifind file

=head1 DESCRIPTION

F<urifind> is a simple script that finds URIs in one or more files
(using C<URI::Find>), and outputs them to to STDOUT.  That's it.

To find all the URIs in F<file1>, use:

    $ urifind file1

To find the URIs in multiple files, simply list them as arguments:

    $ urifind file1 file2 file3

F<urifind> will read from C<STDIN> if no files are given or if a
filename of C<-> is specified:

    $ wget http://www.boston.com/ -O - | urifind

When multiple files are listed, F<urifind> prefixes each found URI
with the file from which it came:

    $ urifind file1 file2
    file1: http://www.boston.com/index.html
    file2: http://use.perl.org/

This can be turned on for single files with the C<-p> ("prefix") switch:

    $urifind -p file3
    file1: http://fsck.com/rt/

It can also be turned off for multiple files with the C<-n> ("no
prefix") switch:

    $ urifind file1 file2
    http://www.boston.com/index.html
    http://use.perl.org/

By default, URIs will be displayed in the order found; to sort them
ascii-betically, use the C<-s> ("sort") option.  To reverse sort them,
use the C<-r> ("reverse") flag (C<-r> implies C<-s>).

    $ urifind -s file1 file2
    http://use.perl.org/
    http://www.boston.com/index.html
    mailto:webmaster@boston.com

    $ urifind -r file1 file2
    mailto:webmaster@boston.com
    http://www.boston.com/index.html
    http://use.perl.org/

Finally, F<urifind> supports limiting the returned URIs by scheme or
by arbitrary pattern, using the C<-S> option (for schemes) and the
C<-P> option.  Both C<-S> and C<-P> can be specified multiple times:

    $ urifind -S mailto file1
    mailto:webmaster@boston.com

    $ urifind -S mailto -S http
    mailto:webmaster@boston.com
    http://www.boston.com/index.html

C<-P> takes an arbitrary Perl regex.  It might need to be protected
from the shell:

    $ urifind -P 's?html?' file1
    http://www.boston.com/index.html

    $ urifind -P '\.org\b' -S http file4
    http://www.gnu.org/software/wget/wget.html

Add a C<-d> to have F<urifind> dump the refexen generated from C<-S>
and C<-P> to C<STDERR>.  C<-D> does the same but exits immediately:

    $ urifind -P '\.org\b' -S http -D 
    $scheme = '^(\bhttp\b):'
    @pats = ('^(\bhttp\b):', '\.org\b')

To remove duplicates from the results, use the C<-u> ("unique")
switch.

=head1 OPTION SUMMARY

=over 4

=item -s

Sort results.

=item -r

Reverse sort results (implies -s).

=item -u

Return unique results only.

=item -n

Don't include filename in output.

=item -p

Include filename in output (0 by default, but 1 if multiple files are
included on the command line).

=item -P $re

Print only lines matching regex '$re' (may be specified multiple times).

=item -S $scheme

Only this scheme (may be specified multiple times).

=item -h

Help summary.

=item -v

Display version and exit.

=item -d

Dump compiled regexes for C<-S> and C<-P> to C<STDERR>.

=item -D

Same as C<-d>, but exit after dumping.

=back

=head1 VERSION

This is F<urifind>, revision $Revision: 1.1.1.1 $.

=head1 AUTHOR

darren chamberlain E<lt>darren@cpan.orgE<gt>

=head1 COPYRIGHT

(C) 2003 darren chamberlain

This library is free software; you may distribute it and/or modify it
under the same terms as Perl itself.

=head1 SEE ALSO

L<Perl>