#! /usr/bin/perl -w
use lib '/usr/lib/perl'; use INN::Config;

##############################################################################
# send-uucp.pl	create and send UUCP news batches from the outgoing files
#
# Author:	Edvard Tuinder <ed@elm.net>
#
# Copyright (C) 1994 Edvard Tuinder - ELM Consultancy B.V.
# Copyright (C) 1995-1997 Miquel van Smoorenburg - Cistron Internet Services
#
# Copyright (C) 2003 Marco d'Itri <md@linux.it>
#   Nearly rewritten. Added syslog support, real errors checking and more.
#
# This program is free software; you can redistribute it and/or modify it
# under the terms of the GNU General Public License as published by the Free
# Software Foundation; either version 2 of the License, or (at your option)
# any later version.
##############################################################################

use strict;
use INN::Utils::Shlock;

# for compatibility with earlier versions of INN
$INN::Config::pathetc ||= '/etc/news';
$INN::Config::syslog_facility ||= 'news';
$INN::Config::uux ||= 'uux';

# some default values
my $MAXSIZE = 500000;
my $MAXJOBS = 200;

my %UNBATCHER = (
    compress    => 'cunbatch',
    bzip2       => 'bunbatch',
    gzip        => 'gunbatch',
);

my $UUX_FLAGS = '- -z -r -gd';
my $BATCHER_FLAGS = '';

##############################################################################
my $config_file = $INN::Config::pathetc . '/send-uucp.cf';
my $lockfile = $INN::Config::locks . '/LOCK.send-uucp';

END {
    # In case we bail out, while holding a lock.
    INN::Utils::Shlock::releaselocks();
}

my $use_syslog = 0;

eval { require Sys::Syslog; import Sys::Syslog; $use_syslog = 1; };

if ($use_syslog) {
    if ($Sys::Syslog::VERSION < 0.15) {
        eval "sub Sys::Syslog::_PATH_LOG { '/dev/log' }" if $^O eq 'dec_osf';
        Sys::Syslog::setlogsock('unix') if $^O =~ /linux|dec_osf|freebsd|darwin/;
    }
    openlog('send-uucp', 'pid', $INN::Config::syslog_facility);
}

my @sitelist;
if (@ARGV) {
    foreach my $site (@ARGV) {
        my @cfg = read_cf($config_file, $site);
        if (not @cfg) {
            logmsg("site $site not found in the configuration", 'err');
            next;
        }
        push @sitelist, @cfg;
    }
} else {
    @sitelist = read_cf($config_file, undef);
}

if (not @sitelist) {
    logmsg('nothing to do', 'debug');
    exit 0;
}

chdir $INN::Config::batch or logdie("Can't access $INN::Config::batch: $!", 'crit');

# Acquire a lock.
INN::Utils::Shlock::lock($lockfile, 60) or logdie("cannot create lockfile $lockfile");

run_site($_) foreach @sitelist;

# Unlock.
INN::Utils::Shlock::unlock($lockfile) or logdie("cannot unlock $lockfile");

exit 0;

##############################################################################
sub read_cf {
    my ($conf_file, $site_wanted) = @_;

    my $hour = (localtime time)[2];

    my @sites;
    open(CF, $conf_file) or logdie("cannot open $conf_file: $!", 'crit');
    while (<CF>) {
        chop;
        s/\s*\#.*$//;
        next if /^$/;

        my ($sitespec, $compress, $size, $time) = split(/\s+/);
        next if not $sitespec;

        my ($site, $host, $funnel) = split(/:/, $sitespec);
        $host = $site if not $host;
        $funnel = $site if not $funnel;

        $compress =~ s/_/ /g if $compress;

        if ($site_wanted) {
            if ($site eq $site_wanted) {
                push @sites, [$site, $host, $funnel, $compress, $size];
                last;
            }
            next;
        }

        if ($time) {
            foreach my $time (split(/,/, $time)) {
                next if $time != $hour;
                push @sites, [$site, $host, $funnel, $compress, $size];
            }
        } else {
            push @sites, [$site, $host, $funnel, $compress, $size];
        }
    }
    close CF;
    return @sites;
}

##############################################################################
# count number of jobs in the UUCP queue for a given site
sub count_jobs {
    my ($site) = @_;

    return 0 if not $INN::Config::uustat;
    open(JOBS, "$INN::Config::uustat -s $site 2> /dev/null |") or logdie("cannot fork: $!");
    my $count = grep(/ Executing rnews /, <JOBS>);
    close JOBS;                    # ignore errors, uustat may fail
    return $count;
}

# select the rnews label appropriate for the compressor program used
sub unbatcher {
    my ($compressor) = @_;

    $compressor =~ s%.*/%%;   # Do not keep the complete path.
    $compressor =~ s% .*%%;   # Do not keep the optional parameters.
    return $UNBATCHER{$compressor} || $UNBATCHER{'gzip'};
}

##############################################################################
# batch articles for one site
sub run_site {
    my ($cfg) = @_;
    my ($site, $host, $funnel, $compress, $size) = @$cfg;

    logmsg("checking site $site", 'debug');
    my $maxjobs = '';
    if ($MAXJOBS) {
        my $jobs = count_jobs($site);
        if ($jobs >= $MAXJOBS) {
            logmsg("too many jobs queued for $site");
            return;
        }
        $maxjobs = '-N ' . ($MAXJOBS - $jobs);
    }

    $compress ||= $INN::Config::gzip;
    $size ||= $MAXSIZE;

    # if a .work temp file left by a previous invocation exists, rename
    # it to .work.tmp, we'll append it to the current batch file once it
    # has been renamed and flushed.
    if (-f "$site.work") {
        rename("$site.work", "$site.work.tmp")
            or logdie("cannot rename $site.work: $!", 'crit');
    }

    if (not -f $site and not -f "$site.work.tmp") {
        logmsg("no batch file for site $site", 'err');
        return;
    }

    rename($site, "$site.work") or logdie("cannot rename $site: $!", 'crit');
    logmsg("Flushing $funnel for site $site", 'debug');
    ctlinnd('-t120', 'flush', $funnel);

    # append the old .work temp file to the current batch file if needed
    if (-f "$site.work.tmp") {
        my $err = '';
        open(OUT, ">>$site.work")
            or logdie("cannot open $site.work: $!", 'crit');
        open(IN, "$site.work.tmp")
            or logdie("cannot open $site.work.tmp: $!", 'crit');
        print OUT while <IN>;
        close IN;
        close OUT or logdie("cannot close $site.work: $!");;
        unlink "$site.work.tmp"
            or logmsg("cannot delete $site.work.tmp: $!", 'err');
    }

    if (not -s "$site.work") {
        logmsg("no articles for $site", 'debug');
        unlink "$site.work" or logmsg("cannot delete $site.work: $!", 'err');
    } else {
        if ($compress eq 'none') {
            system "$INN::Config::newsbin/batcher -b $size $maxjobs $BATCHER_FLAGS "
                . "-p\"$INN::Config::uux $UUX_FLAGS %s!rnews\" $host $site.work";
        } else {
            system "$INN::Config::newsbin/batcher -b $size $maxjobs $BATCHER_FLAGS "
                . "-p\"{ echo '#! " . unbatcher($compress)
                . "' ; exec $compress; } | "
                . "$INN::Config::uux $UUX_FLAGS %s!rnews\" $host $site.work";
        }
        logmsg("batched articles for $site", 'debug');
    }
}

##############################################################################
sub logmsg {
    my ($msg, $lvl) = @_;

    syslog($lvl || 'notice', '%s', $msg) if ($use_syslog);
}

sub logdie {
    my ($msg, $lvl) = @_;

    logmsg($msg, $lvl || 'err');

    # Unlock.
    INN::Utils::Shlock::unlock($lockfile);
    exit 1;
}

sub ctlinnd {
    my ($cmd, @args) = @_;

    my $st = system("$INN::Config::newsbin/ctlinnd", '-s', $cmd, @args);
    logdie('Cannot run ctlinnd: ' . $!) if $st == -1;
    logdie('ctlinnd returned status ' . ($st & 255)) if $st > 0;
}

__END__

=head1 NAME

send-uucp - Send Usenet articles via UUCP

=head1 SYNOPSIS

B<send-uucp> [I<site> ...]

=head1 DESCRIPTION

The B<send-uucp> program processes batch files written by innd(8) to send
Usenet articles to UUCP sites.  It reads a configuration file to control how
it behaves with various sites.  Normally, it is run periodically out of cron
to put together batches and send them to remote UUCP sites.

It makes it possible to reduce bandwidth usage and to send news to remote
UUCP sites which cannot receive a real-time feed (for instance if they
are over dial-up connections).

=head1 OPTIONS

Any arguments provided to the program are interpreted as a list of sites
specified in F<send-uucp.cf> for which batches should be generated.  If no
arguments are supplied then batches will be generated for all sites listed
in that configuration file.

=head1 CONFIGURATION

The sites to which articles are to be sent must be configured in the
configuration file F<send-uucp.cf> in I<pathetc> as set in F<inn.conf>.  Each
site is specified with a line of the form:

    site[:host[:funnel]] [compressor [maxsize [batchtime]]]

=over 4

=item I<site>

The news site name being configured.  This must match a site name 
from newsfeeds(5).

=item I<host>

The UUCP host name to which batches should be sent for this site.
If omitted, the news site name will be used as the UUCP host name.

=item I<funnel>

In the case of a site configured as a funnel, B<send-uucp> needs to flush
the channel (or exploder) being used as the target of the funnel instead of
flushing the site.  This is the way to tell B<send-uucp> the name of the
channel or exploder to flush for this site.  If not specified, default to
flushing the site.

=item I<compressor>

The compression method to use for batches.  This should be one of C<bzip2>,
C<compress>, C<gzip> or C<none>.  Arguments for the compression command may be
specified by using C<_> instead of spaces.  For example, C<gzip_-9>.
The default value is C<gzip>.

=item I<maxsize>

The maximum size in bytes of a single batch I<before> compression.  The default
value is C<500000> bytes.

=item I<batchtime>

A comma separated list of hours during which batches should be generated for
a given site.  When B<send-uucp> runs, a site will only be processed if the
current hour matches one of the hours in I<batchtime>.  The default is no
limitation on when to generate batches.

=back

Fields are separated by spaces and only the site name needs to be specified,
with defaults being used for unspecified values.  If the first character on
a line is a hash sign (C<#>) then the rest of the line is ignored.

=head1 EXAMPLE

Here is an example for the F<send-uucp.cf> configuration file:

    zoetermeer      gzip            1048576         5,18,22
    hoofddorp       gzip            1048576         5,18,22
    pa3ebv          gzip            1048576         5,18,22
    drinkel         bzip2           1048576         5,6,18,20,22,0,2
    manhole         compress        1048576         5,18,22
    owl             compress        1048576
    able
    pern::MYFUNNEL!

This defines eight UUCP sites.  The first three and the last two use C<gzip>
compression, the fourth site (C<drinkel>) uses C<bzip2> and the remaining sites
(C<manhole> and C<owl>) use C<compress>.  The first six use a batch size of
S<1 MB>, and the two last sites (C<able> and C<pern>) use the default of
500,000 bytes.  The C<zoetermeer>, C<hoofddorp>, C<pa3ebv>, and C<manhole> sites
will only have batches generated for them during the hours of 05:00, 18:00,
and 22:00, and the C<drinkel> site will only have batches generated during those
hours and 06:00, 20:00, 00:00, and 02:00.  There are no restrictions on when
batches will be generated for C<owl>, C<able> and C<pern>.

The C<pern> site is configured as a funnel into C<MYFUNNEL!>.  B<send-uucp> will
issue C<ctlinnd flush MYFUNNEL!> instead of C<ctlinnd flush pern>.

As for the F<newsfeeds> file, the usual flags used for a UUCP feed are
C<Tf,Wnb>.  Here is a typical entry for C<zoetermeer>, where the batching
is kept between S<4 KB> and S<1 KB>:

    zoetermeer\
        :*,!junk,!control,!control.*/!foo\
        :Tf,Wnb,B4096/1024:

=head1 FILES

=over 4

=item I<pathbin>/send-uucp

The Perl script itself used to create news batches from the outgoing files.

=item I<pathetc>/send-uucp.cf

The configuration file which specifies a list of sites to be processed.

=back

=head1 HISTORY

This program was originally written by Edvard Tuinder <ed@elm.net> and then
maintained and extended by Miquel van Smoorenburg <miquels@cistron.nl>.
Marco d'Itri <md@linux.it> cleaned up the code for inclusion in INN.  This
manual page was written by Mark Brown <broonie@sirena.org.uk>.

=head1 SEE ALSO

innd(8), newsfeeds(5), uucp(8).

=cut
