scripts for generating hungarian charset table for Sphinxsearch.com

scripts/sphinx-charset-generator.hungarian.txt

@@ -0,0 +1,2 @@
+charset_table=a..z,0..9,A..Z->a..z, \
+U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C7->c, U+C8->e, U+C9->e, U+CA->e, U+CC->i, U+CD->i, U+CE->i, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D9->u, U+DA->u, U+DB->u, U+E0->a, U+E1->a, U+E2->a, U+E3->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, U+EC->i, U+ED->i, U+EE->i, U+F2->o, U+F3->o, U+F4->o, U+F5->o, U+F9->u, U+FA->u, U+FB->u, U+102->a, U+103->a, U+15E->s, U+15F->s, U+162->t, U+163->t

scripts/sphinx-charset-generator.pl

@@ -0,0 +1,33 @@
+#!/usr/bin/env perl
+#
+# Generates a charset_table entry for Sphinx
+# 
+
+use strict;
+use warnings;
+use utf8;
+use Text::Unaccent;
+use Encode;
+
+my $map = gen_map_accents($ARGV[0] || 'utf-8');
+
+print "charset_table=a..z,0..9,A..Z->a..z, \\\n";
+print join(', ', map { sprintf('U+%X->%s', unpack('W*'), $map->{$_}) } sort keys %$map);
+print "\n";
+
+sub gen_map_accents {
+  my $charset = shift;
+  my %map;
+
+#  my $src = 'áéíóúàèìòùãõâêîôûç';
+  my $src = 'áéíóúàèìòùãõâêîôûçĂÂÎŢŞăâîţş';
+  $src .= uc($src);  
+  $src = encode($charset, $src) unless $charset =~ m/^utf-?8$/i;
+
+  my $clean = lc(unac_string($charset, $src));
+  @map{split(//, $src)} = split(//, $clean);
+
+  return \%map;
+}
+
+