/usr/local/CPAN/FreeHAL/AI/FreeHAL/Module/Tagger.pm
package AI::FreeHAL::Module::Tagger;
BEGIN {
use Exporter ();
our ( $VERSION, @ISA, @EXPORT, @EXPORT_OK, %EXPORT_TAGS );
# set the version for version checking
$VERSION = 0.01;
@ISA = qw(Exporter);
# functions
@EXPORT = qw(
&pos_prop
&pos_of
);
%EXPORT_TAGS = (); # eg: TAG => [ qw!name1 name2! ],
# your exported package globals go here,
# as well as any optionally exported functions
@EXPORT_OK = qw();
}
our @EXPORT_OK;
our $data = {};
our %config;
use AI::Util;
use AI::POS;
use Data::Dumper;
our @functions = qw{
sql_get_part_of_speech_property
pos_prop
short_save_in_memory
pos_of
find_word_type
};
sub sql_get_part_of_speech_property {
my ($word) = @_;
return { type => '', genus => '' } if !$data->{modes}{use_sql};
#say 'word', $word;
eval q{
use DBI;
my $db_string = "DBI:mysql:database=" . $config{'mysql'}{'database'};
if ( $config{'mysql'}{'host'} ) {
$db_string .= ';host=' . $config{'mysql'}{'host'};
}
my $user = $config{'mysql'}{'user'};
my $password = $config{'mysql'}{'password'};
my $dbh = DBI->connect( $db_string, $user, $password )
or (
say(
"not connected to ",
$config{'mysql'}{'database'},
", user ",
$config{'mysql'}{'user'},
", password ",
$config{'mysql'}{'user'},
": "
)
and return { type => '', genus => '' }
);
my $sql = qq{SELECT type, genus FROM part_of_speech WHERE word = '$word' };
my $sth = $dbh->prepare($sql);
$sth->execute();
while ( $sth && ( my $data = $sth->fetchrow_arrayref ) ) {
my $return_hash = { type => $data->[0], genus => $data->[1] };
print Dumper $return_hash;
$sth->finish();
$dbh->disconnect();
return $return_hash;
}
$sth->finish();
$dbh->disconnect();
} if !$::batch;
return { type => '', genus => '' };
}
sub pos_prop {
my ( $CLIENT_ref, $word, $count, $not_guess, $not_ask,
$write_it_into_database )
= @_;
if ( !$data->{abilities}->{'tagger'} ) {
my $sock = ${ connect_to( data => *data, name => 'tagger' ) };
print {$sock} 'GET<;;>genus<;;>', join( '<;;>', @_ ) . "\n";
my $result = <$sock>;
chomp $result;
close $sock;
eval {
local $SIG{'__DIE__'};
$result = thaw( r_unescape($result) );
};
if ($@) {
say 'error in that(): ', $@;
}
return $result;
}
AI::FreeHAL::Engine::try_use_lowlatency_mode();
my $CLIENT = undef;
eval '$CLIENT = ${$CLIENT_ref};';
$count ||= 0;
$write_it_into_database ||= 0;
#$word =~ s/^(.ein.?.?)qq/$1_/igm;
$word =~ s/^ein(.?)[_\s]//igm;
$word =~ s/[-]//gm;
$word =~ s/^(und|or|and|oder)\s+//gm;
$word =~ s/_/ /igm;
$word =~ s/^\s+//igm;
$word =~ s/\s+$//igm;
#if ( $word ne lc $word ) {
$word = ucfirst lc $word;
#}
if ( $has_no_genus{$word} ) {
part_of_speech_get_memory()->{$word}->{'genus'} = q{q};
}
part_of_speech_get_memory()->{$word} = part_of_speech_get_entry($word)
if !$::batch && !part_of_speech_get_memory()->{$word};
my $last_word = ( split /\s|[_]/, $word )[-1];
if ( $last_word
&& $word ne $last_word )
{
my $props_for_last_word =
pos_prop( $CLIENT_ref, $last_word, 0, $not_guess, $not_ask,
$write_it_into_database );
if ( $props_for_last_word->{'genus'} ) {
return $props_for_last_word;
}
}
#say '$not_guess: ', $not_guess || 0;
# $CLIENT_ref, $word, $at_beginning_of_sentence, $noun_automatism,
# $do_not_ask_user, $sentence, $do_not_guess, $do_save
my $type_str = q{}; # empty
if ( !$not_guess
&& pos_of( $CLIENT_ref, $word, 0, 1, 0, undef, 0, 0 ) == $data->{const}{NOUN}
&& !$has_no_genus{ lc $word } )
{
if (
(
!defined part_of_speech_get_memory()->{$word}->{'genus'}
|| (
part_of_speech_get_memory()->{$word}->{'genus'}
? part_of_speech_get_memory()->{$word}->{'genus'}
: q{-}
) eq '-'
|| (
length( part_of_speech_get_memory()->{$word}->{'genus'} ) ==
0
&& defined part_of_speech_get_memory()->{$word}->{'genus'} )
# experimental
# ask if german mode and there is a "q" in db
|| ( LANGUAGE() eq 'de'
&& part_of_speech_get_memory()->{$word}->{'genus'} eq 'q' )
)
&& ( !( ( sql_get_part_of_speech_property $word)->{genus} )
&& !( ( sql_get_part_of_speech_property lc $word )->{genus} ) )
)
{
if ( !$has_no_genus{ lc $word } ) {
my $first_word = ( split /\s|[_]/, $word )[0];
$first_word =~ s/_$//igm;
$last_word =~ s/_$//igm;
my $wt =
pos_of( $CLIENT_ref, $word, 0, 1, 0, undef, 0, 0 );
my $first_wt =
$word eq $last_word
? $wt
: pos_of( $CLIENT_ref, $first_word, 0, 1, 0, undef, 0,
0 );
my $last_wt =
$word eq $last_word
? $wt
: pos_of( $CLIENT_ref, $last_word, 0, 1, 0, undef, 0,
0 );
if ( $count > 5 ) {
$type_str = 'perhaps_s';
}
elsif ( ( $first_wt == $data->{const}{ADJ} || $first_wt == $ART )
&& $first_word =~ /^[a-zA-Z]+nen$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ernet$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+urm$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+tter$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ek$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+urti$/ )
{
$type_str = 'perhaps_f';
}
elsif ( ( $first_wt == $data->{const}{ADJ} || $first_wt == $ART )
&& $first_word =~ /^[a-zA-Z]+nem$/ )
{
$type_str = 'perhaps_m';
}
elsif ( ( $first_wt == $data->{const}{ADJ} || $first_wt == $ART )
&& $first_word =~ /^[a-zA-Z]+ner$/ )
{
$type_str = 'perhaps_f';
}
elsif ( $last_word =~ /^[a-zA-Z]+re$/ ) {
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+adt$/ )
{
$type_str = 'perhaps_f';
}
elsif ( $first_word =~ /^[a-zA-Z]+ige$/ ) {
$type_str = 'perhaps_f';
}
#elsif ( ( $first_wt == $data->{const}{ADJ} || $first_wt == $ART )
#&& $first_word =~ /^[a-zA-Z]+nen$/ )
#{
#(my $modified = $last_word ) =~ s/nen$/n/;
#my $revert_changes = 0;
#if ( !part_of_speech_get_memory()->{$modified} ) {
#$revert_changes = 1;
#}
## $CLIENT_ref, $word, $count, $not_guess, $not_ask, $write_it_into_database
#$type_str = pos_prop($CLIENT_ref, $modified, 0, 0, 1, 5)->{genus};
#say 'reduzed to ', $modified, ' => ', $type_str;
#if ( $revert_changes ) {
#part_of_speech_get_memory()->{$word} = undef;
#}
#$revert_changes = 0;
#$modified .= 'e';
#if ( !part_of_speech_get_memory()->{$modified} ) {
#$revert_changes = 1;
#}
#$type_str = pos_prop($CLIENT_ref, $modified, 0, 0, 1, 5)->{genus};
#say 'reduzed to ', $modified, ' => ', $type_str;
#if ( $revert_changes ) {
#part_of_speech_get_memory()->{$word} = undef;
#}
#$type_str ||= 'perhaps_f';
#}
#elsif ( $last_word =~ /^[a-zA-Z]+[ndgmr]e[nsrm]$/ )
#{
#(my $modified = $last_word ) =~ s/.e.$/n/;
#my $revert_changes = 0;
#if ( !part_of_speech_get_memory()->{$modified} ) {
#$revert_changes = 1;
#}
#$type_str = pos_prop($CLIENT_ref, $modified, 0, 0, 1, 5)->{genus};
#say 'reduzed to ', $modified, ' => ', $type_str;
#if ( $revert_changes ) {
#part_of_speech_get_memory()->{$word} = undef;
#}
#$revert_changes = 0;
#$modified .= 'e';
#if ( !part_of_speech_get_memory()->{$modified} ) {
#$revert_changes = 1;
#}
#$type_str = pos_prop($CLIENT_ref, $modified, 0, 0, 1, 5)->{genus};
#say 'reduzed to ', $modified, ' => ', $type_str;
#if ( $revert_changes ) {
#part_of_speech_get_memory()->{$word} = undef;
#}
#$type_str ||= 'perhaps_f';
#}
#elsif ( $last_word =~ /^[a-zA-Z]+[nsmr]$/ )
#{
#(my $modified = $last_word ) =~ s/.$/n/;
#my $revert_changes = 0;
#if ( !part_of_speech_get_memory()->{$modified} ) {
#$revert_changes = 1;
#}
#$type_str = pos_prop($CLIENT_ref, $modified, 0, 0, 1, 5)->{genus}
#if $last_word ne $modified;
#say 'reduzed to ', $modified, ' => ', $type_str;
#if ( $revert_changes ) {
#part_of_speech_get_memory()->{$word} = undef;
#}
#$revert_changes = 0;
#$modified .= 'e';
#if ( !part_of_speech_get_memory()->{$modified} ) {
#$revert_changes = 1;
#}
#$type_str = pos_prop($CLIENT_ref, $modified, 0, 0, 1, 5)->{genus}
#if $last_word ne $modified;
#say 'reduzed to ', $modified, ' => ', $type_str;
#if ( $revert_changes ) {
#part_of_speech_get_memory()->{$word} = undef;
#}
#$type_str ||= 'perhaps_f';
#}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+in$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+innen$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+cken$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+iz$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ahn$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+aus$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+rom$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+amm$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+uhr$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ann$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+aph$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ext$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+od$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ohn$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+plan$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ahlen$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ua$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+uehr$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ot$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+erb$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ity$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ey$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+weg$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ind$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*?hut$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*?mut$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ut$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+na$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+aetz$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ruck$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ssi$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ail$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+til$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+opf$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*lauch$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*auch$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*och$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ip$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ell$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*fehl$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*ie$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*?[bt]uch$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*?ucht$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+mpf$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ift$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ausch$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*(st|r|erb)and$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*(l)and$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*and$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*ient$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+urst$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+unst$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+eil$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+wort$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+sort$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+to$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*zug$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*ad$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*al$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*koll$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*att$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*iki$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*fest$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*schuh$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+all$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+lust$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ust$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+uehl$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+rief$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+rst$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+epp$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+kt$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+nsa$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*zeit$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*bahn$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+beit$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ilz$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+eis$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ei[gk]$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+(ta|um)$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+off$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ohl$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+[ae]rz$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+a[xgst]$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ef$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+do$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+go$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+iff$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+eld$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ied$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+rist$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ist$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+af$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ock$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+uck$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*?ha[n]t$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+raut$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*?luft$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+huft$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+aut$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+urz$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+esen$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+acken$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+uhl$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+uv$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+orf$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+olk$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+raft$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+id$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ack$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+weck$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ueck$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+atz$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ank$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+and$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+itz$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ett$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+la$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ma$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+und$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ereich$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+rich$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+eich$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ard$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*ast$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+uff$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+iet$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*zeug$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*ing$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*isch$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*[rw]icht$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*ap$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*[d]icht$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*icht$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*acht$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*leid$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ol$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+nch$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ahl$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+aeck$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+net$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+et$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+eo$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ex$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+erl$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+nia$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+olg$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*utsch$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+log$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ang$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+amt$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+yp$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ko$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+alg$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+[nkr]itt$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+tab$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]*[rtmg]at$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+aupt$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+a[l]$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+olt$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+old$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ald$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+el$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ien$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+auf$/ )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+rma$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ra$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ert$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ngen$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+erk$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+gen$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+[rhs]o$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ben$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+ten$/ )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+olf$/ )
{
$type_str = 'perhaps_m';
}
elsif (LANGUAGE() eq 'de'
&& ( $first_wt == $data->{const}{ADJ} || $first_wt == $ART )
&& $first_word =~ /^[a-zA-Z]+ne$/ )
{
$type_str = 'perhaps_f';
}
elsif (LANGUAGE() eq 'de'
&& $first_wt == $ART
&& lc $first_word eq 'der' )
{
$type_str = 'perhaps_m';
}
elsif (LANGUAGE() eq 'de'
&& $first_wt == $ART
&& lc $first_word eq 'die' )
{
$type_str = 'perhaps_f';
}
elsif (LANGUAGE() eq 'de'
&& $first_wt == $ART
&& lc $first_word eq 'das' )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /.+?(ismus|ling|or|ant)$/i )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /.+?e$/i )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /.+?er$/i )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /.+?ers$/i )
{
$type_str = 'perhaps_m';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /.+?ens$/i )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /.+?(on|um)$/i )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /.+?(on|um)s$/i )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /.+?ung$/i )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /.+?au$/i )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /.+?nis$/i )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& length($last_word) == 1 )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~
/.+?(ung|heit|keit|schaft|ei|enz|ie|ik|ion|taet|ur)$/i )
{
$type_str = 'perhaps_f';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+st$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+e$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /^[a-zA-Z]+en$/ )
{
$type_str = 'perhaps_s';
}
elsif ( LANGUAGE() eq 'de'
&& $last_word =~ /.+?(chen|lein|um|ment)$/i )
{
$type_str = 'perhaps_s';
}
elsif ( $::batch || !$AI::SemanticNetwork::initialized ) {
$type_str = 'perhaps_s';
}
elsif ( $last_wt == $data->{const}{NOUN} ) {
my $downloaded_genus = download_genus($last_word);
if ($downloaded_genus) {
$type_str = $downloaded_genus;
}
elsif ( LANGUAGE() eq 'en'
&& !$AI::SemanticNetwork::initialized )
{
$type_str = 'perhaps_s';
}
elsif ( $last_word eq lc $last_word ) {
$type_str = 'perhaps_s';
}
elsif ( !$AI::SemanticNetwork::initialized && $in_cgi_mode )
{
$type_str = 'perhaps_s';
}
elsif ($not_ask) {
$type_str = 'q';
}
else {
$write_it_into_database ||= 1;
my $line = impl_get_genus( $CLIENT_ref, $last_word );
# return 'EXIT' if $line eq 'EXIT';
$type_str =
$line == 1 ? 'm'
: $line == 2 ? 'f'
: $line == 3 ? 's'
: '';
}
}
else {
$type_str = 'perhaps_s';
}
}
if ( !$type_str && !$has_no_genus{$word} ) {
print "Redoing: pos_prop( $CLIENT_ref, $word )\n";
return pos_prop( $CLIENT_ref, $word, $count + 1 );
}
if ( $write_it_into_database && $write_it_into_database != 5 ) {
if ($use_sql) {
eval q{
use DBI;
my $db_string = "DBI:mysql:database=" . $config{'mysql'}{'database'};
if ( $config{'mysql'}{'host'} ) {
$db_string .= ';host=' . $config{'mysql'}{'host'};
}
my $user = $config{'mysql'}{'user'};
my $password = $config{'mysql'}{'password'};
my $dbh = DBI->connect( $db_string, $user, $password )
or say(
"not connected to ",
$config{'mysql'}{'database'},
", user ",
$config{'mysql'}{'user'},
", password ",
$config{'mysql'}{'user'},
": "
);
my $sql =
qq{SELECT word FROM part_of_speech WHERE word = '$word'};
my $sth = $dbh->prepare($sql);
$sth->execute();
if ( !( $sth && ( my $data = $sth->fetchrow_arrayref ) ) ) {
$dbh->do(
qq{INSERT INTO part_of_speech( word, type, genus ) VALUES ( '$word', '', '' )}
);
}
$sql =
qq{UPDATE part_of_speech SET genus = '$type_str' WHERE word = '$word'};
my $sth = $dbh->prepare($sql);
$sth->execute();
$dbh->disconnect();
} if !$::batch;
}
else {
part_of_speech_get_memory()->{$word}->{'genus'} = $type_str;
delete part_of_speech_get_memory()->{''};
delete part_of_speech_get_memory()->{' '};
}
}
}
}
# if ( part_of_speech_get_memory()->{$word}->{'genus'} eq "w" ) {
# part_of_speech_get_memory()->{$word}->{'genus'} = "f";
# }
my $return_value = part_of_speech_get_memory()->{$word};
return {
type => (
( part_of_speech_get_memory()->{$word} || {} )->{type}
|| ( ( sql_get_part_of_speech_property $word) || {} )->{type}
|| ( ( sql_get_part_of_speech_property lc $word ) || {} )->{type}
),
genus => (
$type_str
|| ( part_of_speech_get_memory()->{$word} || {} )->{genus}
|| ( ( sql_get_part_of_speech_property $word) || {} )->{genus}
|| ( ( sql_get_part_of_speech_property lc $word ) || {} )->{genus}
),
};
}
sub short_save_in_memory {
my ( $do_save, $word, $type, $always ) = @_;
say "$config{'features'}{'tagger'}: ", $config{'features'}{'tagger'};
if ( part_of_speech_get_memory()->{$word}->{'type'}
&& part_of_speech_get_memory()->{$word}->{'type'} ne 'q'
&& (!$always||!$config{'features'}{'tagger'}) )
{
return $type;
}
if ( !$do_save ) {
return $type;
}
part_of_speech_get_memory()->{$word}->{'new'} = 1;
print "short_save_in_memory\n";
print join(', ', @_), "\n";
my $val = $data->{lang}{string_to_constant}{ part_of_speech_get_memory()->{$word}->{'type'} =
$data->{lang}{constant_to_string}{$type} };
print "val: ", $val, "\n";
return $val;
}
sub pos_of {
my (
$CLIENT_ref, $word,
$at_beginning_of_sentence, $noun_automatism,
$do_not_ask_user, $sentence,
$do_not_guess, $do_save,
$no_autoguess
) = @_;
my $CLIENT = undef;
eval '$CLIENT = ${$CLIENT_ref};';
$do_save = 1 if !defined $do_save;
$do_not_guess ||= 0;
return $data->{const}{NO_POS} if !defined $word;
return $data->{const}{NO_POS} if !$word;
$word =~ tr/-/ /;
if ( !$sentence ) {
$sentence = $word;
}
$noun_automatism = 1 if !defined $noun_automatism;
if ( $word ne lc $word ) {
$word = ucfirst lc $word;
}
my $user_defined_word_type = undef;
if ( $_[1] =~ /[{][{][{]/ ) {
( $_[1], my $type ) = split /[{][{][{]/, $_[1];
$type = ( split /[}]/, $type )[0];
my $type_value = undef;
my $exec_str = '$type_value = $' . uc $type . ';';
say $exec_str;
eval $exec_str;
if ( $type_value && !$@ ) {
say 'ok. '
. $_[1] . "->"
. $type_value . "->"
. $data->{lang}{constant_to_string}{$type_value};
part_of_speech_get_memory()->{ $_[1] }->{'type'} =
$data->{lang}{constant_to_string}{$type_value};
say 'ok...';
if ($at_beginning_of_sentence) {
say 'added to cache_noun_or_not';
$cache_noun_or_not{ $_[1] } = $type_value;
$cache_noun_or_not{ lc $_[1] } = $type_value;
}
#delete part_of_speech_get_memory()->{''};
#delete part_of_speech_get_memory()->{' '};
# foreach my $key ( keys %{part_of_speech_get_memory()} ) {
# delete part_of_speech_get_memory()->{$key}
# if $key =~ /['"*+\-)(]|(^[\s_])/;
# }
say 'ok... ...';
# return $type_value;
return pos_of( $CLIENT_ref, $_[1], $at_beginning_of_sentence,
$noun_automatism, $do_not_ask_user, $sentence, $do_not_guess );
}
}
if ( !$data->{abilities}->{'tagger'} ) {
my $sock = ${ connect_to( data => *data, name => 'tagger' ) };
print {$sock} 'GET<;;>type<;;>' . join( '<;;>', @_ ) . "\n";
my $result = <$sock>;
while ( $result =~ /GET_WORD_TYPE:/ ) {
print $CLIENT $result;
while ( defined( $result = <$CLIENT> ) ) {
if ( $result =~ /HERE_IS/ ) {
print $sock $result;
last;
}
}
$result = <$sock>;
}
chomp $result;
#print 'result:', Dumper $result;
close $sock;
$result ||= r_escape( nfreeze( \undef ) );
eval {
local $SIG{'__DIE__'};
$result = ${ thaw( r_unescape($result) ) };
};
if ($@) {
say 'error in thaw(): ', $@;
}
return $result;
}
AI::FreeHAL::Engine::try_use_lowlatency_mode();
$word =~ s/^ein(.?)[_\s]//igm;
$word =~ s/^(und|or|and|oder)\s+//gm;
$word =~ s/[{]//gm;
$word =~ s/[}]//gm;
$word =~ s{/}{}gm;
$word =~ s/|//gm;
$word =~ s/[-]//gm;
#$word =~ s/^(.ein.?.?)qq/$1_/igm;
chomp $word;
if ( $word ne lc $word ) {
$word = ucfirst lc $word;
}
return $data->{const}{ADJ} if $word =~ /^[-\s+]?\d+$/;
return $data->{const}{ADJ}
if $word =~ /^[+-]?(\d+\.\d+|\d+\.|\.\d+|\d+)([eE][+-]?\d+)?$/;
return $data->{const}{ADJ} if $word =~ /^\d+$/;
return $data->{const}{ADJ} if $word =~ /^\d+[_]\d+[_]\d+$/;
return $data->{const}{ADJ} if $word =~ /_komma_/;
return $data->{const}{INTER} if $word eq '+' || $word eq '-';
return $data->{const}{NO_POS} if not $word;
return $data->{const}{NO_POS} if $word =~ /^[?,;]/;
return $data->{const}{ADJ} if $word =~ /[(]/;
return $data->{const}{NOUN} if $word =~ /http[:]/i;
return $data->{const}{NOUN} if $word =~ /['"]/;
return $data->{const}{ADJ} if $word =~ /zum_/;
return $data->{const}{ADJ} if $word =~ /zur_/;
return $data->{const}{NOUN} if $word =~ /_/;
return $data->{const}{NOUN} if $word =~ / /;
return $data->{const}{NOUN} if $word =~ /^\$\$/;
return $data->{const}{NOUN} if $word =~ /^\s/;
return $data->{const}{ADJ} if $word =~ /[%]/;
$word =~ s/\.|[?!,;]//gm;
my $word_low = lc $word;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word eq $word_low && $word =~ /los$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB}, 1 )
if $word eq $word_low && $word =~ /machen$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB}, 1 )
if $word eq $word_low && $word =~ /mache$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB}, 1 )
if $word eq $word_low && $word =~ /machst$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB}, 1 )
if $word eq $word_low && $word =~ /macht$/;
return $ART if $word_low eq q{a} && LANGUAGE() eq 'en';
#say 'pos_of(' . $word . ')';
if ( $word_low =~ /\s/ ) {
my $last_word = ( split /\s+/, $word_low )[-1];
return (
pos_of(
$CLIENT_ref, $last_word,
$at_beginning_of_sentence, $noun_automatism,
$do_not_ask_user, $sentence
)
);
}
say '.';
my $builtin_table = AI::FreeHAL::Engine::build_builtin_table( LANGUAGE() );
#print Dumper $builtin_table;
say ',';
if ( defined $builtin_table->{$word_low} ) {
return short_save_in_memory( $do_save, $word,
$builtin_table->{$word_low} );
}
{
my $word_low_stripped = $word_low;
$word_low_stripped =~ s/.$//;
if ( defined $builtin_table->{$word_low_stripped} ) {
return short_save_in_memory( $do_save, $word,
$builtin_table->{$word_low_stripped} );
}
$word_low_stripped =~ s/.$//;
if ( defined $builtin_table->{$word_low_stripped} ) {
return short_save_in_memory( $do_save, $word,
$builtin_table->{$word_low_stripped} );
}
$word_low_stripped =~ s/.$//;
if ( defined $builtin_table->{$word_low_stripped} ) {
return short_save_in_memory( $do_save, $word,
$builtin_table->{$word_low_stripped} );
}
}
if ( !$no_autoguess && $config{'features'}{'tagger'} ) {
if ( LANGUAGE() eq 'de' ) {
return short_save_in_memory( $do_save, $word, $data->{const}{NOUN}, 1 ) #always
if $word =~ /kern$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 ) #always
if $word =~ /.farben$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 ) #always
if $word =~ /.quent$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if ( $word eq lc $word || $at_beginning_of_sentence )
&& $word =~ /(der|al|haft|lich|ig|bar|isch|ich)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 ) #always
if ( $word ne lc $word || $at_beginning_of_sentence )
&& $word =~ /.(haft|lich|ig|isch|ich)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{QUESTIONWORD} )
if ( $word eq lc $word || $at_beginning_of_sentence )
&& $word =~ /^wor/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if ( $word eq lc $word || $at_beginning_of_sentence )
&& $word =~ /(der|al|ich|bar)e$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if ( $word eq lc $word || $at_beginning_of_sentence )
&& $word =~ /(der|al|ich|bar)es$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if ( $word eq lc $word || $at_beginning_of_sentence )
&& $word =~ /(der|al|ich|bar)en$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if ( $word eq lc $word || $at_beginning_of_sentence )
&& $word =~ /(der|al|ich|bar)em$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if ( $word eq lc $word || $at_beginning_of_sentence )
&& $word =~ /(der|al|ich|bar)er$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if ( $word eq lc $word || $at_beginning_of_sentence )
&& $word =~ /(ss|mm|nn|gg)er$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB}, 1 )
if $word =~ /^(ge|$regex_str_verb_prefixes).*?t$/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~ /^(ge|ver).*?ten$/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{NOUN}, 1 )
if $word =~ /(heit|keit)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word =~ /(haft|lich|ig|isch|dnet)e?$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word =~ /(haft|lich|ig|isch|dnet)es$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word =~ /(haft|lich|ig|isch|dnet)en$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word =~ /(haft|lich|ig|isch|dnet)em$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word =~ /(haft|lich|ig|isch|dnet)er$/;
return short_save_in_memory( $do_save, $word, $data->{const}{NOUN} )
if $word =~ /angst$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word =~ /reich$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word eq lc $word && $word =~ /reit$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word eq lc $word && $word =~ /iert$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 ) #always
if $word eq lc $word && $word =~ /ierte$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word eq lc $word && $word =~ /ierter$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word eq lc $word && $word =~ /iertes$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word eq lc $word && $word =~ /ierten$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word eq lc $word && $word =~ /iertem$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word eq lc $word && $word =~ /ene[nmsr]?$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word eq lc $word && $word =~ /on$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /on$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /iv$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /iv(e|en|er|es|em|en)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /ig(e|es|en|em|er|)?$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /^meist/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /h(e|es|en|em|er)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /ht(e|es|en|em|er)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /er(e|en|er|es|em|en)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word =~ /lich$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word =~ /te(t|te|ter|te|ten|tem|tes)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /ens$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /st(e|en|er|es|em|en)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word =~ /lich(e|es|en|em)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /.det(e|es|en|em)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /.haft$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~
/^(ge|$regex_str_verb_prefixes).*?nn(($)|((e|en|er|es|em|en)$))/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~
/^(ge|$regex_str_verb_prefixes).*?[lb]o[gt](($)|((e|en|er|es|em|en)$))/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~
/^(ge|$regex_str_verb_prefixes).*?b(($)|((e|en|er|es|em|en)$))/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~
/^(ge|$regex_str_verb_prefixes).*?s(($)|((e|en|er|es|em|en)$))/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~
/^(ge|$regex_str_verb_prefixes).*?mm(($)|((e|en|er|es|em|en)$))/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~ /^be.*?t$/ && LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~ /^ver.*?t$/ && LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word eq lc $word
&& $word =~ /ier(t|en|st|e)$/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~ /^ge.*?t$/ && LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~ /^be.*?t$/ && LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~ /^ver.*?t$/ && LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word =~
/^(un|in|an|a|im|$regex_str_verb_prefixes)?ge.*?[tn]et(($)|((e|en|er|es|em|en)$))/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word =~
/^(un|in|an|a|im|$regex_str_verb_prefixes)?be.*?[tn]et(($)|((e|en|er|es|em|en)$))/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word =~
/^(un|in|an|a|im|$regex_str_verb_prefixes)?ge.*?(es|er|em|en|e)$/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 )
if $word =~
/^(un|in|an|a|im|$regex_str_verb_prefixes)?ge.*?[fsrmnlp]t$/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word =~
/^(un|in|an|a|im|$regex_str_verb_prefixes)?ge.*?[tn]et$/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word =~
/^(un|in|an|a|im|$regex_str_verb_prefixes)?be.*?[tn]et$/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word eq lc $word
&& $word =~
/.(ier|eid|ied|und|uch|ueh|eh|erb|arb|ng|eb|eis|s|ieg|bau|iess|eig|esch|isch|ruh|err|mm|mpf|ink|ohl)en$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB}, 1 ) #always
if $word eq lc $word
&& $word =~
/.(ier|eid|ied|und|uch|ueh|eh|erb|arb|ng|eb|eis|s|ieg|bau|ies|eig|esch|isch|ruh|err|mm|mpf|ink|ohl)st$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB}, 1 ) #always
if $word eq lc $word
&& $word =~
/.(ier|eid|ied|und|uch|ueh|eh|erb|arb|ng|eb|eis|s|ieg|bau|iess|eig|esch|isch|ruh|err|mm|mpf|ink|ohl)e$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB}, 1 ) #always
if $word eq lc $word
&& $word =~
/.(ier|eid|ied|und|uch|ueh|eh|erb|arb|ng|eb|eis|s|ieg|ruh|isch|err|mm|mpf|ieh|sch|ink|ohl)t$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB}, 1 ) #always
if $word eq lc $word
&& $word =~
/.(ier|eid|ied|und|uch|ueh|eh|erb|arb|ng|eb|eis|s|ieg|ruh)et$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word eq lc $word
&& $word =~
/.(ier|eid|ied|und|uch|ueh|eh|erb|arb|ng|eb|eis|s|ieg|ruh)test$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word eq lc $word
&& $word =~
/.(ier|eid|ied|und|uch|ueh|eh|erb|arb|ng|eb|eis|s|ieg|ruh)est$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word eq lc $word
&& $word =~
/.(ier|eid|ied|und|uch|ueh|eh|erb|arb|ng|eb|eis|s|ieg|ruh)tst$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word eq lc $word
&& $word =~
/.(ier|eid|ied|und|uch|ueh|eh|erb|arb|ng|eb|eis|s|ieg|ruh)te$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~ /...eck(t|en|e|st)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word eq lc $word && $word =~ /.ern$/;
return short_save_in_memory( $do_save, $word, $data->{const}{NOUN} )
if $word =~ /.[hk]eit$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word eq $word_low
&& $word =~ /^.*?[nskz]en$/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word eq $word_low
&& $word =~ /^.*?eln$/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~ /^(ver|be|ge|ent|er).+?et$/;
return short_save_in_memory( $do_save, $word, $data->{const}{VERB} )
if $word =~ /^(zer|ver|be|ge|ent|er).+?et$/;
return short_save_in_memory( $do_save, $word, $data->{const}{NOUN} )
if $word =~ /[ts]ion$/;
return short_save_in_memory( $do_save, $word, $data->{const}{NOUN} )
if $word ne lc $word && $word =~ /...([gftlhsbkdm]|ier)er$/;
return short_save_in_memory( $do_save, $word, $data->{const}{NOUN} )
if $word ne lc $word && $word =~ /...([gftlhsbkdm]|ier)erin$/;
return short_save_in_memory( $do_save, $word, $data->{const}{NOUN} )
if $word ne lc $word && $word =~ /...([gftlhsbkdm]|ier)erinnen$/;
return short_save_in_memory( $do_save, $word, $data->{const}{NOUN} )
if $word =~ /.([sk])um$/;
return short_save_in_memory( $do_save, $word, $data->{const}{NOUN} )
if $word ne lc $word && $word =~ /...([gftlhsbkdm]|ier)ero$/;
return short_save_in_memory( $do_save, $word, $data->{const}{NOUN} )
if $word ne lc $word && $word =~ /...([gftlhsbkdm]|ier)t$/;
return short_save_in_memory( $do_save, $word, $data->{const}{NOUN} )
if $word ne lc $word && $word =~ /...([gftlhsbkdm]|ier)ist/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /st(e|en|er|es|em|en)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 ) #always
if $word =~ /en(e|es|en|em|er)$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word eq lc $word && $word =~ /cht$/;
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ}, 1 ) # always
if $word =~
/^(ge|$regex_str_verb_prefixes).*?(((e|en|er|es|em|en|t)$))/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB}, 1 ) # always
if $word =~ /tzt$/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB}, 1 ) # always
if $word =~ /pft$/
&& LANGUAGE() eq 'de';
return short_save_in_memory( $do_save, $word, $data->{const}{VERB}, 1 ) # always
if $word =~ /ckt$/
&& LANGUAGE() eq 'de';
}
}
if ( $word ne $word_low && $config{'features'}{'tagger'} ) {
return short_save_in_memory( $do_save, $word, $data->{const}{NOUN} );
}
if ( defined $builtin_table->{$word_low} ) {
return $builtin_table->{$word_low};
}
my $word_low_without_prefix = $word_low;
foreach my $prefix ( keys %is_verb_prefix ) {
$word_low_without_prefix =~ s/^$prefix//im;
}
if ( !$no_autoguess && $config{'features'}{'tagger'} ) {
if ( $word_low =~ /end$/ && length($word) >= 6 ) {
my $word_to_check = $word_low;
$word_to_check =~ s/[a-z]$//; # no 'g', only once!
if (
pos_of( $CLIENT_ref, $word_to_check, 0, 0, 0, undef,
undef, 0 ) == $data->{const}{VERB}
)
{
return $data->{const}{ADJ};
}
}
if ( $word eq $word_low && $word_low =~ /e$/ && length($word) >= 4 ) {
my $word_to_check = $word_low . 'n';
if (
pos_of( $CLIENT_ref, $word_to_check, 0, 0, 1, undef,
undef, 0, 1 )
|| 0 == $data->{const}{VERB}
)
{
return $data->{const}{VERB};
}
}
if ( $word eq $word_low && $word_low =~ /s$/ && length($word) >= 4 ) {
my $word_to_check = $word_low;
$word_to_check =~ s/[a-z]$//; # no 'g', only once!
if (
pos_of( $CLIENT_ref, $word_to_check, 0, 1, 1, undef,
undef, 0, 1 )
|| 0 == $data->{const}{VERB}
)
{
return $data->{const}{VERB};
}
}
if ( $word_low =~ /s$/ && length($word) >= 4 ) {
my $word_to_check = $word;
$word_to_check =~ s/[a-z]$//; # no 'g', only once!
if (
pos_of( $CLIENT_ref, $word_to_check, 0, 0, 1, undef,
undef, 0, 1 )
|| 0 == $data->{const}{NOUN}
)
{
return $data->{const}{NOUN};
}
}
if ( $word_low =~ /e[sr]$/ && length($word) >= 4 ) {
my $word_to_check = $word;
$word_to_check =~ s/[a-z][a-z]$//; # no 'g', only once!
if (
pos_of( $CLIENT_ref, $word_to_check, 0, 0, 1, undef,
undef, 0, 1 )
|| 0 == $data->{const}{NOUN}
)
{
return $data->{const}{NOUN};
}
}
if ( $word eq $word_low && $word_low =~ /st$/ && length($word) >= 5 ) {
my $word_to_check = $word_low;
$word_to_check =~ s/st$/en/; # no 'g', only once!
if (
pos_of( $CLIENT_ref, $word_to_check, 0, 0, 1, undef,
undef, 0, 1 ) == $data->{const}{VERB}
)
{
return $data->{const}{VERB};
}
}
if ( $word eq $word_low && $word_low =~ /et$/ && length($word) >= 5 ) {
my $word_to_check = $word_low;
$word_to_check =~ s/et$/en/; # no 'g', only once!
if (
pos_of( $CLIENT_ref, $word_to_check, 0, 0, 1, undef,
undef, 0, 1 ) == $data->{const}{VERB}
)
{
return $data->{const}{VERB};
}
}
if ( $word eq $word_low && $word_low =~ /t$/ && length($word) >= 5 ) {
my $word_to_check = $word_low;
$word_to_check =~ s/t$/en/; # no 'g', only once!
if (
pos_of( $CLIENT_ref, $word_to_check, 0, 0, 1, undef,
undef, 0, 1 ) == $data->{const}{VERB}
)
{
return $data->{const}{VERB};
}
}
if ( $word_low =~ /(ed|ing)$/
&& length($word) >= 6
&& LANGUAGE() eq 'en' )
{
my $word_to_check = $word_low;
$word_to_check =~ s/(ed|ing)$//; # no 'g', only once!
if (
pos_of( $CLIENT_ref, $word_to_check, 0, 0, 0, undef,
undef, 0, 1 ) == $data->{const}{VERB}
)
{
return $data->{const}{ADJ};
}
if (
pos_of( $CLIENT_ref, $word_to_check . 'e',
0, 0, 1, undef, undef, 0, 1 ) == $data->{const}{VERB}
)
{
return $data->{const}{ADJ};
}
}
return short_save_in_memory( $do_save, $word, $data->{const}{ADJ} )
if $word =~ /en(d|de|den|des|dem|der)$/;
}
if ($at_beginning_of_sentence) {
if ( !$AI::SemanticNetwork::initialized ) {
$cache_noun_or_not{$word_low} = $data->{const}{NOUN};
}
my $in_cache = $cache_noun_or_not{$word_low};
if ( !$in_cache ) {
$in_cache = $WORDTYPE_UNKNOWN;
}
return $in_cache if $in_cache == $data->{const}{NOUN};
return $in_cache if $in_cache == $data->{const}{VERB};
return $in_cache if $in_cache == $data->{const}{ADJ};
print "Error in Cache: $in_cache" if $in_cache != $WORDTYPE_UNKNOWN;
#if ( $in_cache != $WORDTYPE_UNKNOWN ) {
my $wt1 = $WORDTYPE_UNKNOWN;
my $wt2 = $WORDTYPE_UNKNOWN;
if ( $word ne lc $word ) {
$wt1 =
pos_of( $CLIENT_ref, lc $word, 0, 0, undef, $sentence );
$wt2 = pos_of( $CLIENT_ref, ucfirst $word, 0, 0, undef,
$sentence );
if ( $wt1 == $data->{const}{QUESTIONWORD} || $wt2 == $data->{const}{QUESTIONWORD} ) {
return $data->{const}{QUESTIONWORD};
}
}
if ( $wt1 != $wt2 && $word ne lc $word ) {
if ( $wt1 == $data->{const}{VERB} || $wt2 == $data->{const}{VERB} ) {
return $data->{const}{VERB};
}
if ( LANGUAGE() eq 'en'
&& ( $wt1 == $data->{const}{NOUN} || $wt2 == $data->{const}{NOUN} )
&& ( $wt1 == $data->{const}{ADJ} || $wt2 == $data->{const}{ADJ} ) )
{
return $data->{const}{NOUN};
}
if ( LANGUAGE() eq 'en'
&& ( $wt1 == $data->{const}{NOUN} || $wt2 == $data->{const}{NOUN} )
&& ( $wt1 == $data->{const}{PREP} || $wt2 == $data->{const}{PREP} ) )
{
return $data->{const}{PREP};
}
if ( LANGUAGE() eq 'en'
&& ( $wt1 == $data->{const}{NOUN} || $wt2 == $data->{const}{NOUN} )
&& ( $wt1 == $ART || $wt2 == $ART ) )
{
return $ART;
}
my $line =
!$AI::SemanticNetwork::initialized
? 1
: impl_get_noun_or_not( $CLIENT_ref, $word );
if ( $line == 1 ) {
$cache_noun_or_not{$word_low} = $data->{const}{NOUN};
}
elsif ( $line == 2 ) {
$cache_noun_or_not{$word_low} = $wt1;
}
else {
print "Illegal Command: HERE_IS_NOUN_OR_NOT:" . $line . "\n";
}
# last;
}
elsif ( $wt1 != $wt2 && $word eq lc $word ) {
return $wt1;
}
#}
}
my $word_without_underscores = $word;
$word_without_underscores =~ s/_/ /igm;
my $type = q{}; # empty
if ($at_beginning_of_sentence) {
my $prop_ref =
part_of_speech_get_memory()->{ lc $word_without_underscores };
if ( $prop_ref->{'type'} ) {
$type = $prop_ref->{'type'};
}
}
my $prop_ref = part_of_speech_get_memory()->{$word_without_underscores};
if ( $prop_ref->{'type'} && !$type ) {
$type = $prop_ref->{'type'};
}
if ( !$type && $at_beginning_of_sentence ) {
my $prop_ref = part_of_speech_get_entry( lc $word_without_underscores );
if ( $prop_ref->{'type'} ) {
$type = $prop_ref->{'type'};
}
}
if ( !$type ) {
$prop_ref = part_of_speech_get_entry($word_without_underscores);
if ( $prop_ref->{'type'} ) {
$type = $prop_ref->{'type'};
}
}
if ($at_beginning_of_sentence) {
my $prop_ref = part_of_speech_get_memory()
->{ '_nosave_' . lc $word_without_underscores };
if ( $prop_ref->{'type'} ) {
$type = $prop_ref->{'type'};
}
}
my $prop_ref =
part_of_speech_get_memory()->{ '_nosave_' . $word_without_underscores };
if ( $prop_ref->{'type'} && !$type ) {
$type = $prop_ref->{'type'};
}
if ( !$type && $at_beginning_of_sentence ) {
my $prop_ref =
part_of_speech_get_entry( '_nosave_' . lc $word_without_underscores );
if ( $prop_ref->{'type'} ) {
$type = $prop_ref->{'type'};
}
}
if ( !$type ) {
$prop_ref =
part_of_speech_get_entry( '_nosave_' . $word_without_underscores );
if ( $prop_ref->{'type'} ) {
$type = $prop_ref->{'type'};
}
}
if ($use_sql) {
if ( ( !$type || $type eq 'q' ) && $at_beginning_of_sentence ) {
$type =
( sql_get_part_of_speech_property lc $word_without_underscores )
->{type};
}
if ( ( !$type || $type eq 'q' ) ) {
$type =
( sql_get_part_of_speech_property $word_without_underscores )
->{type};
}
}
if ( $type eq 'q' ) {
$type = q{}; # empty
}
# if ( $at_beginning_of_sentence && !$type ) {
# $type = word_types_read_only_one_thing( lc $word_without_underscores, 'type' );
# }
# if ( !$type ) {
# $type = word_types_read_only_one_thing( $word_without_underscores, 'type' );
# }
if (
( $word || '' ) ne ( $word_low || '' )
&& !$at_beginning_of_sentence
&& $noun_automatism
&& (
( $data->{lang}{string_to_constant}{ $type || '' } || '' ) ne $data->{const}{NOUN}
&& ( $data->{lang}{string_to_constant}{ $data->{lang}{constant_to_string}{ $type || '' } || '' }
|| '' ) ne $data->{const}{NOUN}
)
)
{
}
else {
if ( $data->{lang}{string_to_constant}{$type} ) {
return $data->{lang}{string_to_constant}{$type};
}
if ( $data->{lang}{string_to_constant}{ $data->{lang}{constant_to_string}{$type} || '' } ) {
return $data->{lang}{string_to_constant}{ $data->{lang}{constant_to_string}{$type} };
}
}
#print "failure: " . $word . '[' . $type . "] not known\n" if is_verbose;
if ($do_not_ask_user) { #&& $downloaded == $data->{const}{NO_POS} ) {
return $data->{const}{NO_POS};
}
my $downloaded =
LANGUAGE() eq 'de' && $config{'features'}{'tagger'}
? download_pos( $CLIENT_ref, $word, $at_beginning_of_sentence )
: $data->{const}{NO_POS};
say "$do_not_guess || $config{'features'}{'tagger'}";
my $word_type_tagged =
$do_not_guess || !$config{'features'}{'tagger'}
? ''
: uc find_word_type( $CLIENT_ref, $word, $at_beginning_of_sentence,
$sentence );
if ( $word_low_without_prefix ne $word_low
&& $word eq $word_low
&& $config{'features'}{'tagger'} )
{
if (
pos_of( $CLIENT_ref, $word_low_without_prefix, 0, 0, 0, undef,
0 ) == $data->{const}{VERB}
)
{
$downloaded = $data->{const}{VERB};
}
# $do_not_guess = 1;
}
if ( !$at_beginning_of_sentence
&& $word_low ne $word
&& $noun_automatism
&& $downloaded == $data->{const}{NO_POS}
&& ( !$word_type_tagged || LANGUAGE() eq 'de' )
&& $config{'features'}{'tagger'} )
{
$downloaded = $data->{const}{NOUN};
}
$downloaded ||= 0;
$downloaded =
$downloaded == $data->{const}{NOUN} ? 2
: $downloaded == $data->{const}{VERB} ? 1
: $downloaded == $data->{const}{PREP} ? 6
: $downloaded == $data->{const}{ADJ} ? 3
: $downloaded == $data->{const}{INTER} ? 7
: $downloaded == $data->{const}{QUESTIONWORD} ? 5
: 0;
my $word_type_tagged_new =
$word_type_tagged eq 'CD' ? 3
: $word_type_tagged eq 'EX' ? 3
: $word_type_tagged eq 'IN' ? 6
: $word_type_tagged eq 'JJ' ? 3
: $word_type_tagged eq 'JJR' ? 3
: $word_type_tagged eq 'JJS' ? 3
: $word_type_tagged eq 'MD' ? 1
: $word_type_tagged eq 'NN' ? 2
: $word_type_tagged eq 'NNS' ? 2
: $word_type_tagged eq 'NNPS' ? 2
: $word_type_tagged eq 'NNP' ? 2
: $word_type_tagged eq 'PDT' ? 3
: $word_type_tagged eq 'PRP' ? 2
: $word_type_tagged eq 'PRPS' ? 3
: $word_type_tagged eq 'RB' ? 3
: $word_type_tagged eq 'RBR' ? 3
: $word_type_tagged eq 'RBS' ? 3
: $word_type_tagged eq 'RP' ? 3
: $word_type_tagged eq 'TO' ? 6
: $word_type_tagged eq 'UH' ? 7
: $word_type_tagged eq 'SYM' ? 2
: $word_type_tagged =~ /^VB/ ? 1
: $word_type_tagged =~ /^W/ ? 5
: 0;
if ( !$word_type_tagged_new ) {
say 'Tagged Word type not known: ', $word_type_tagged;
}
else {
say 'Word type tagged successful: ', $word_type_tagged, ' = ',
$word_type_tagged_new;
}
$word_type_tagged = $word_type_tagged_new;
if ( $do_not_ask_user && !$downloaded && !$word_type_tagged ) {
return $data->{const}{NO_POS};
}
my $line =
($downloaded) ? ($downloaded)
: ($word_type_tagged) ? ($word_type_tagged)
: ( impl_get_word_type( $CLIENT_ref, $word ) );
print '$line: ', $line, "\n";
my $type_str =
$line == 1 ? 'vt'
: $line == 2 ? 'n,'
: $line == 3 ? 'adj'
: $line == 4 ? 'n,'
: $line == 5 ? 'fw'
: $line == 6 ? 'prep'
: $line == 7 ? 'inter'
: 'nothing';
if ( $word ne $word_low ) {
$type_str = 'n,';
}
##if ( !$in_cgi_mode ) {
#my $from_yaml =
#$yaml->read( $dir . 'lang_' . LANGUAGE() . '/word_types.base' );
#foreach my $key ( %{$from_yaml->[0]} ) {
#foreach my $key_2 ( %{$from_yaml->[0]->{$key}} ) {
#part_of_speech_get_memory()->{$key}->{$key_2} = $from_yaml->[0]->{$key}->{$key_2}
#if !part_of_speech_get_memory()->{$key}->{$key_2};
#}
#}
##}
if ( !$do_save ) {
part_of_speech_get_memory()->{ '_nosave_' . $word_without_underscores }
->{'type'} = $type_str;
}
elsif ($use_sql) {
eval q{
use DBI;
my $db_string = "DBI:mysql:database=" . $config{'mysql'}{'database'};
if ( $config{'mysql'}{'host'} ) {
$db_string .= ';host=' . $config{'mysql'}{'host'};
}
my $user = $config{'mysql'}{'user'};
my $password = $config{'mysql'}{'password'};
my $dbh = DBI->connect( $db_string, $user, $password )
or say(
"not connected to ",
$config{'mysql'}{'database'},
", user ",
$config{'mysql'}{'user'},
", password ",
$config{'mysql'}{'user'},
": "
);
my $sql =
qq{SELECT word FROM part_of_speech WHERE word = '$word_without_underscores'};
my $sth = $dbh->prepare($sql);
$sth->execute();
say( $sth->errstr );
say( $dbh->errstr );
if ( !( $sth && ( my $data = $sth->fetchrow_arrayref ) ) ) {
$dbh->do(
qq{INSERT INTO part_of_speech( word, type, genus ) VALUES ( '$word_without_underscores', '', '' )}
);
say( $sth->errstr );
say( $dbh->errstr );
}
$sql =
qq{UPDATE part_of_speech SET type = '$type_str' WHERE word = '$word_without_underscores'};
my $sth = $dbh->prepare($sql);
$sth->execute();
say( $sth->errstr );
say( $dbh->errstr );
} if !$::batch;
# $sth->finish();
# $dbh->disconnect();
#part_of_speech_get_memory()->{$word} ||= {};
}
else {
part_of_speech_get_memory()->{$word_without_underscores}->{'type'} =
$type_str;
part_of_speech_get_memory()->{$word_without_underscores}->{'new'} = 1;
}
#}
#else {
# write_to( $dir . 'lang_' . LANGUAGE() . '/word_types.temp',
# part_of_speech_get_memory() );
#}
# open HANDLE, '>>', $dir . 'lang_' . LANGUAGE() . '/word_types.dic'
# or die 'Cannot write to: ' . $dir . 'lang_'
# . LANGUAGE()
# . '/word_types.dic';
# print HANDLE $word . '|' . $type_str . "\n";
# close HANDLE;
return pos_of( $CLIENT_ref, $word, $at_beginning_of_sentence,
$noun_automatism, $sentence, $do_not_guess, $do_save );
return $data->{const}{NO_POS};
}
sub find_word_type {
my ( $CLIENT_ref, $word, $at_beginning_of_sentence, $sentence ) = @_;
$word =~ s/^[_]//igm;
$word =~ s/[_]$//igm;
$sentence =~ s/^[_]//igm;
$sentence =~ s/[_]$//igm;
$sentence =~ s/\s[_]/ /igm;
$sentence =~ s/[_]\s/ /igm;
return 'JJ'
if $word =~ /^ge.*?tet(($)|((e|en|er|es|em|en)$))/ && LANGUAGE() eq 'de';
return 'JJ'
if $word =~ /^be.*?tet(($)|((e|en|er|es|em|en)$))/ && LANGUAGE() eq 'de';
return 'VB'
if $word =~ /^ge.*?t(($)|((e|en|er|es|em|en)$))/ && LANGUAGE() eq 'de';
return 'VB'
if $word =~ /^be.*?t(($)|((e|en|er|es|em|en)$))/ && LANGUAGE() eq 'de';
return 'VB'
if $word =~ /^ver.*?t(($)|((e|en|er|es|em|en)$))/ && LANGUAGE() eq 'de';
return 'JJ' if $word =~ /^ge.*?(es|er|em|en|e)$/ && LANGUAGE() eq 'de';
return 'VB' if $word eq lc $word && $word =~ /.[gftlrs]en$/;
my $builtin_table = AI::FreeHAL::Engine::build_builtin_table();
foreach my $key ( keys %$builtin_table ) {
part_of_speech_get_memory()->{$key}{'type'} =
$data->{lang}{constant_to_string}{ $builtin_table->{$key} };
part_of_speech_get_memory()->{$key}{rtime} = 'not_new';
part_of_speech_get_memory()->{ ucfirst $key }{'type'} =
$data->{lang}{constant_to_string}{ $builtin_table->{$key} };
part_of_speech_get_memory()->{ ucfirst $key }{rtime} = 'not_new';
}
if ( !$use_sql ) {
open my $old_format_file, '>', $dir . 'lang_de/word_types.taggerinput';
foreach my $key ( sort keys %{ part_of_speech_get_memory() } ) {
print $old_format_file $key . '|'
. part_of_speech_get_memory()->{$key}->{'type'} . "\n"
if part_of_speech_get_memory()->{$key}->{'type'}
&& part_of_speech_get_memory()->{$key}->{'type'} !~ /^n/
&& $key eq lc $key;
}
close $old_format_file;
}
else {
eval q{
use DBI;
my $db_string = "DBI:mysql:database=" . $config{'mysql'}{'database'};
if ( $config{'mysql'}{'host'} ) {
$db_string .= ';host=' . $config{'mysql'}{'host'};
}
my $user = $config{'mysql'}{'user'};
my $password = $config{'mysql'}{'password'};
my $dbh = DBI->connect( $db_string, $user, $password )
or say(
"not connected to ",
$config{'mysql'}{'database'},
", user ",
$config{'mysql'}{'user'},
", password ",
$config{'mysql'}{'user'},
": "
);
my $sql =
qq{create table part_of_speech (word varbinary(120), type char(3), genus char(1))};
$dbh->prepare($sql)->execute();
$sql = qq{ALTER TABLE `part_of_speech` ADD PRIMARY KEY ( `word` )};
$dbh->prepare($sql)->execute();
$sql = qq{SELECT word, type FROM part_of_speech};
my $sth = $dbh->prepare($sql);
$sth->execute();
open my $old_format_file, '>', $dir . 'lang_de/word_types.taggerinput';
while ( $sth && ( my $data = $sth->fetchrow_arrayref ) ) {
print $old_format_file $data->[0] . '|' . $data->[1] . "\n"
if $data->[1]
&& $data->[1] !~ /^n/
&& $data->[0] eq lc $data->[0];
}
close $old_format_file;
$sth->finish();
} if !$::batch;
}
require 'convert-word-types-dic.pl' if LANGUAGE() eq 'de' && !$::batch;
say $sentence;
# exit 0;
my $p =
LANGUAGE() eq 'en'
? new Lingua::EN::Tagger
: new Lingua::DE::Tagger;
my $readable_text = $p->get_readable($sentence);
$readable_text =~ s/[<]\/(.+?)[>]/\/$1 /igm;
$readable_text =~ s/[<].+?[>]//igm;
$readable_text =~ s/[\-]//igm;
my %word_list = map { lc $_ } map { split /\//, $_ } split /\s/,
$readable_text;
foreach my $k ( keys %word_list ) {
$k = lc $k;
}
print Dumper \%word_list;
my $tagged =
$word_list{ lc $word }
|| $word_list{$word}
|| $word_list{ ucfirst $word };
if ( lc $tagged eq 'nn' && $word =~ /(es|er|en|em)$/ ) {
$tagged = 'JJ';
}
return $tagged;
}
1;