pg_updatedicts

#!/usr/bin/perl -w

# Create dictionaries and affix rules palatable for PostgreSQL, using installed
# myspell and hunspell dictionaries.
#
# (C) 2008-2009 Martin Pitt <mpitt@debian.org>
#
#  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.
#
#  This program is distributed in the hope that it will be useful,
#  but WITHOUT ANY WARRANTY; without even the implied warranty of
#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#  GNU General Public License for more details.

my @srcdirs = ('/usr/share/hunspell', '/usr/share/myspell/dicts');
my $cachedir = '/var/cache/postgresql/dicts';
my $pgsharedir = '/usr/share/postgresql/';

use lib '/usr/share/postgresql-common';
use PgCommon;

# determine encoding of an .aff file
sub get_encoding {
    open F, $_[0] or die "cannot open $_[0]: $!";
    while (<F>) {
        if (/^SET ([\w-]+)\s*$/) { return $1; }
    }
    return undef;
}

if ((system 'mkdir', '-p', $cachedir) != 0) {
    exit 1;
}

print "Building PostgreSQL dictionaries from installed myspell/hunspell packages...\n";
my ($dic, $enc, $locale);
for my $d (@srcdirs) {
    for my $aff (glob "$d/*.aff") {
	next if -l $aff; # ignore symlinks
	$dic = substr($aff, 0, -3) . 'dic';
	if (! -f $dic) {
	    print STDERR "ERROR: $aff does not have corresponding $dic, ignoring\n";
	    next;
	}

	$enc = get_encoding $aff;
	if (!$enc) {
	    print STDERR "ERROR: no ecoding defined in $aff, ignoring\n";
	    next;
	}

	$locale = substr ((split '/', $aff)[-1], 0, -4);
	$locale =~ tr /A-Z/a-z/;

	# convert to UTF-8 and write to cache dir
	print "  $locale\n";
	if ((system 'iconv', '-f', $enc, '-t', 'UTF-8', '-o',
	    "$cachedir/$locale.affix", $aff) != 0) {
	    unlink "$cachedir/$locale.affix";
	    print STDERR "ERROR: Conversion of $aff failed\n";
	    next;
	}
	if ((system 'iconv', '-f', $enc, '-t', 'UTF-8', '-o',
	    "$cachedir/$locale.dict", $dic) != 0) {
	    unlink "$cachedir/$locale.affix";
	    unlink "$cachedir/$locale.dict";
	    print STDERR "ERROR: Conversion of $dic failed\n";
	    next;
	}
	chmod 0644, "$cachedir/$locale.affix", "$cachedir/$locale.dict";

	# install symlinks to all versions >= 8.3
	foreach my $v (get_versions) {
	    next if $v lt '8.3';
	    my $dest = "$pgsharedir/$v/tsearch_data/";
	    next if ! -d $dest;
	    next if -e "$dest/$locale.affix" && ! -l "$dest/$locale.affix";
	    next if -e "$dest/$locale.dict" && ! -l "$dest/$locale.dict";
	    unlink "$dest/$locale.affix";
	    unlink "$dest/$locale.dict";
	    symlink "$cachedir/$locale.affix", "$dest/$locale.affix";
	    symlink "$cachedir/$locale.dict", "$dest/$locale.dict";
	}
    }
}

__END__

=head1 NAME

pg_updatedicts - build PostgreSQL dictionaries from myspell/hunspell ones

=head1 SYNOPSIS

B<pg_updatedicts>

=head1 DESCRIPTION

B<pg_updatedicts> makes dictionaries and affix files from installed myspell
and hunspell dictionary packages available to PostgreSQL for usage with tsearch
and word stem support. In particular, it takes all I<*.dic> and I<*.aff> files
from /usr/share/myspell/dicts/, converts them to UTF-8, puts them into
/var/cache/postgresql/dicts/ with I<*.dict> and I<*.affix> suffixes, and
symlinks them into /usr/share/postgresql/I<version>/tsearch_data/, where
PostgreSQL looks for them.

Through postgresql-common's dpkg trigger, this program is automatically run
whenever a myspell or hunspell dictionary package is installed or upgraded.

=head1 AUTHOR

Martin Pitt L<E<lt>mpitt@debian.orgE<gt>>