From: Tara L Andrews <tla@mit.edu>
Date: Fri, 15 Jun 2012 13:39:43 +0000 (+0200)
Subject: now using Perseus-generated Latin treetagger params
X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?p=scpubgit%2Fstemmatology.git;a=commitdiff_plain;h=ad2a2c4722a8c6bc324a8a19bee0990c7992c070

now using Perseus-generated Latin treetagger params
---

diff --git a/lib/Text/Tradition/Language/Base.pm b/lib/Text/Tradition/Language/Base.pm
index dab12f2..d0d1caf 100644
--- a/lib/Text/Tradition/Language/Base.pm
+++ b/lib/Text/Tradition/Language/Base.pm
@@ -229,7 +229,7 @@ sub _treetag_string {
 	}
 	# OK, we can run it then.
 	# First upgrade to UTF8 for necessary languages.
-	my @utf8_supported = qw/ French /;
+	my @utf8_supported = qw/ French Latin Greek /;
 	my %ttopts = ( 'language' => $lang, 'options' => [ qw/ -token -lemma / ] );
 	if( grep { $_ eq $lang } @utf8_supported ) {
 		$ttopts{'use_utf8'} = 1;
diff --git a/lib/Text/Tradition/Language/Latin.pm b/lib/Text/Tradition/Language/Latin.pm
index 1d278b5..5643351 100644
--- a/lib/Text/Tradition/Language/Latin.pm
+++ b/lib/Text/Tradition/Language/Latin.pm
@@ -108,46 +108,7 @@ sub morphology_tags {
 			}
 		}
 	}
-		
-	# TODO special case:
-	#  passive verbs (-or)
-	#  T sapientia -> sapientia
-	#  T primus -> unus
-	#  T occulta -> occultus (with occulo in next field, hmm...)
-	#  T carne -> carnis
-	#  T melius -> bonus
-	
-	my %excep = (
-		'absens' => 'absum',
-		'aperte' => 'apertus',
-		'evolvo' => 'exvolvo',
-		'inquiam' => 'inquam',
-		'intelligo' => 'intellego',
-		'itaque' => 'ita',
-		'iuste' => 'iustus',
-		'longe' => 'longus',
-		'male' => 'malus|malum',
-		'multum' => 'multus',
-		'nec' => 'neque',
-		'nos' => 'ego',
-		'occultum' => 'occultus',
-		'peregrinans' => 'peregrinor',
-		'perfectus' => 'perficio',
-		'potius' => 'potis',
-		'praesente' => 'praesens',
-		'prius' => 'prior',
-		'quotidianus' => 'cottidianus',
-		'se' => 'sui',
-		'septem' => 'septimus',
-		'Spiritum' => 'spiritus',
-		'viriliter' => 'virilis', # TODO special case -iter?
-		'vos' => 'tu',
-		
-		'datum' => 'do|data|datus',
-		'forte' => 'fors|fortis',
-		'vere' => 'verum|verus',
-		);
-		
+				
 	sub _perseus_lookup_tt {
 		my( $orig, $pos, $lemma ) = split( /\t/, $_[0] );
 		_morph_connect();
@@ -155,16 +116,9 @@ sub morphology_tags {
 		# Discard results that don't match the lemma, unless lemma is unknown
 		my $lookupopts = {};
 		unless( $lemma eq '<unknown>' || $lemma =~ /^\W+$/ ) {
-			# TODO Perseus lemma might have a number on the end, yuck.
-			#  multiple lemmata separated with |
-			$lemma =~ s/[^\w|]//g;
-			$lemma = $excep{$lemma} if exists $excep{$lemma};
-			$lemma =~ s/j/i/g;
-			if( $lemma ) { # if we have anything left...
-				my %lems;
-				map { $lems{$_} = 1; $lems{lc($_)} = 1 } split( /\|/, $lemma );
-				$lookupopts->{'lemma'} = [ keys %lems ];
-			}
+			my %lems;
+			map { $lems{$_} = 1; $lems{lc($_)} = 1 } split( /\|/, $lemma );
+			$lookupopts->{'lemma'} = [ keys %lems ];
 		}
 		$lookupopts->{'ttpos'} = $pos if $pos;