| line | stmt | bran | cond | sub | pod | time | code | 
| 1 |  |  |  |  |  |  | package Alvis::QueryFilter; | 
| 2 |  |  |  |  |  |  |  | 
| 3 |  |  |  |  |  |  | ######################################################################## | 
| 4 |  |  |  |  |  |  | # | 
| 5 |  |  |  |  |  |  | # A quick'n'dirty query filter | 
| 6 |  |  |  |  |  |  | # | 
| 7 |  |  |  |  |  |  | #   -- Kimmo Valtonen | 
| 8 |  |  |  |  |  |  | # | 
| 9 |  |  |  |  |  |  | ######################################################################## | 
| 10 |  |  |  |  |  |  |  | 
| 11 | 1 |  |  | 1 |  | 32235 | use strict; | 
|  | 1 |  |  |  |  | 2 |  | 
|  | 1 |  |  |  |  | 34 |  | 
| 12 | 1 |  |  | 1 |  | 5 | use warnings; | 
|  | 1 |  |  |  |  | 2 |  | 
|  | 1 |  |  |  |  | 26 |  | 
| 13 |  |  |  |  |  |  |  | 
| 14 | 1 |  |  | 1 |  | 5 | use Carp; | 
|  | 1 |  |  |  |  | 6 |  | 
|  | 1 |  |  |  |  | 95 |  | 
| 15 | 1 |  |  | 1 |  | 3463 | use Data::Dumper; | 
|  | 1 |  |  |  |  | 10678 |  | 
|  | 1 |  |  |  |  | 67 |  | 
| 16 |  |  |  |  |  |  |  | 
| 17 | 1 |  |  | 1 |  | 470 | use Alvis::Treetagger; | 
|  | 1 |  |  |  |  | 3 |  | 
|  | 1 |  |  |  |  | 33 |  | 
| 18 | 1 |  |  | 1 |  | 874 | use CQL::Parser; | 
|  | 1 |  |  |  |  | 37598 |  | 
|  | 1 |  |  |  |  | 51 |  | 
| 19 | 1 |  |  | 1 |  | 884 | use URI::Escape; | 
|  | 1 |  |  |  |  | 2761 |  | 
|  | 1 |  |  |  |  | 77 |  | 
| 20 |  |  |  |  |  |  |  | 
| 21 | 1 |  |  | 1 |  | 7 | use strict; | 
|  | 1 |  |  |  |  | 2 |  | 
|  | 1 |  |  |  |  | 7972 |  | 
| 22 |  |  |  |  |  |  |  | 
| 23 |  |  |  |  |  |  | our $VERSION = '0.3'; | 
| 24 |  |  |  |  |  |  | our $verbose = 0; | 
| 25 |  |  |  |  |  |  |  | 
| 26 |  |  |  |  |  |  | my ($ERR_OK, | 
| 27 |  |  |  |  |  |  | $ERR_CQL_PARSER_INST, | 
| 28 |  |  |  |  |  |  | $ERR_XML_PARSER, | 
| 29 |  |  |  |  |  |  | $ERR_NO_QUERY, | 
| 30 |  |  |  |  |  |  | $ERR_CQL_PARSE, | 
| 31 |  |  |  |  |  |  | $ERR_NONIMP_NODE_TYPE, | 
| 32 |  |  |  |  |  |  | $ERR_NO_SEQ_LIST, | 
| 33 |  |  |  |  |  |  | $ERR_TREETAGGER, | 
| 34 |  |  |  |  |  |  | $ERR_LEMMA_DICT, | 
| 35 |  |  |  |  |  |  | $ERR_APPLYING_TERM_NE, | 
| 36 |  |  |  |  |  |  | $ERR_APPLYING_TYPING, | 
| 37 |  |  |  |  |  |  | $ERR_APPLYING_ONTO, | 
| 38 |  |  |  |  |  |  | $ERR_STRUCT2XML, | 
| 39 |  |  |  |  |  |  | $ERR_CREATING_CAT_LIST, | 
| 40 |  |  |  |  |  |  | $ERR_CREATING_SEQ_LIST, | 
| 41 |  |  |  |  |  |  | $ERR_CREATING_CQL_TAIL, | 
| 42 |  |  |  |  |  |  | $ERR_NEED_BOTH_TERM_AND_NE | 
| 43 |  |  |  |  |  |  | )=(0..16); | 
| 44 |  |  |  |  |  |  |  | 
| 45 |  |  |  |  |  |  | my %ErrMsgs=($ERR_OK=>"", | 
| 46 |  |  |  |  |  |  | $ERR_CQL_PARSER_INST=>"Instantiating CQL::Parser failed.", | 
| 47 |  |  |  |  |  |  | $ERR_XML_PARSER=>"Instantiating the XML parser failed.", | 
| 48 |  |  |  |  |  |  | $ERR_NO_QUERY=>"No query.", | 
| 49 |  |  |  |  |  |  | $ERR_CQL_PARSE=>"CQL parsing failed.", | 
| 50 |  |  |  |  |  |  | $ERR_NONIMP_NODE_TYPE=>'Non-implemented CQL node type.', | 
| 51 |  |  |  |  |  |  | $ERR_NO_SEQ_LIST=> | 
| 52 |  |  |  |  |  |  | "No current data structure representing the " . | 
| 53 |  |  |  |  |  |  | "query expansion. No preceding UI2Zebra() " . | 
| 54 |  |  |  |  |  |  | "call or we're out of sync. You'll get your " . | 
| 55 |  |  |  |  |  |  | "money back.", | 
| 56 |  |  |  |  |  |  | $ERR_TREETAGGER=>"Applying the treetagger failed.", | 
| 57 |  |  |  |  |  |  | $ERR_LEMMA_DICT=>"Applying the lemma dictionary failed.", | 
| 58 |  |  |  |  |  |  | $ERR_APPLYING_TERM_NE=>"Applying the terms and NEs failed.", | 
| 59 |  |  |  |  |  |  | $ERR_APPLYING_TYPING=>"Applying the typing rules failed.", | 
| 60 |  |  |  |  |  |  | $ERR_APPLYING_ONTO=>"Applying the ontology mappings failed.", | 
| 61 |  |  |  |  |  |  | $ERR_STRUCT2XML=>"Converting the data structure into XML failed.", | 
| 62 |  |  |  |  |  |  | $ERR_CREATING_CAT_LIST=> | 
| 63 |  |  |  |  |  |  | "Extracting category part of query failed.", | 
| 64 |  |  |  |  |  |  | $ERR_CREATING_SEQ_LIST=> | 
| 65 |  |  |  |  |  |  | "Creating the list of possible expansions failed.", | 
| 66 |  |  |  |  |  |  | $ERR_CREATING_CQL_TAIL=>"Converting the data structure to a CQL " . | 
| 67 |  |  |  |  |  |  | "tail failed.", | 
| 68 |  |  |  |  |  |  | $ERR_NEED_BOTH_TERM_AND_NE=>"We need both a term and a NE dictionary." | 
| 69 |  |  |  |  |  |  | ); | 
| 70 |  |  |  |  |  |  |  | 
| 71 |  |  |  |  |  |  | sub new | 
| 72 |  |  |  |  |  |  | { | 
| 73 | 0 |  |  | 0 | 1 |  | my $proto=shift; | 
| 74 |  |  |  |  |  |  |  | 
| 75 | 0 |  | 0 |  |  |  | my $class=ref($proto)||$proto; | 
| 76 | 0 |  | 0 |  |  |  | my $parent=ref($proto)&&$proto; | 
| 77 | 0 |  |  |  |  |  | my $self={}; | 
| 78 | 0 |  |  |  |  |  | bless($self,$class); | 
| 79 |  |  |  |  |  |  |  | 
| 80 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_OK); | 
| 81 |  |  |  |  |  |  |  | 
| 82 | 0 |  |  |  |  |  | $self->_init(@_); | 
| 83 |  |  |  |  |  |  |  | 
| 84 | 0 |  |  |  |  |  | $self->{CQLParser}=new CQL::Parser(); | 
| 85 | 0 | 0 |  |  |  |  | if (!defined($self->{CQLParser})) | 
| 86 |  |  |  |  |  |  | { | 
| 87 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_CQL_PARSER_INST); | 
| 88 | 0 |  |  |  |  |  | return undef; | 
| 89 |  |  |  |  |  |  | } | 
| 90 |  |  |  |  |  |  |  | 
| 91 | 0 |  |  |  |  |  | $Alvis::Treetagger::verbose = $verbose; | 
| 92 | 0 |  |  |  |  |  | &Alvis::Treetagger::reopen(); | 
| 93 | 0 |  |  |  |  |  | $self->{termMaxLen} = 0; | 
| 94 | 0 |  |  |  |  |  | $self->{textFields} = "text"; | 
| 95 | 0 |  |  |  |  |  | $self->{tcanon} = \&canonise_def; | 
| 96 | 0 |  |  |  |  |  | $self->{ncanon} = \&canonise_def; | 
| 97 |  |  |  |  |  |  | #  this must match the lemma indexing rules in alvis2index.xsl | 
| 98 | 0 |  |  |  |  |  | $self->{lemmaSearch} = "^[VNJ]"; | 
| 99 |  |  |  |  |  |  |  | 
| 100 |  |  |  |  |  |  |  | 
| 101 | 0 |  |  |  |  |  | return $self; | 
| 102 |  |  |  |  |  |  | } | 
| 103 |  |  |  |  |  |  |  | 
| 104 |  |  |  |  |  |  | sub _init | 
| 105 |  |  |  |  |  |  | { | 
| 106 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 107 |  |  |  |  |  |  |  | 
| 108 | 0 |  |  |  |  |  | $self->{keepLemmas}=1; | 
| 109 |  |  |  |  |  |  |  | 
| 110 | 0 | 0 |  |  |  |  | if (defined(@_)) | 
| 111 |  |  |  |  |  |  | { | 
| 112 | 0 |  |  |  |  |  | my %args=@_; | 
| 113 | 0 |  |  |  |  |  | @$self{ keys %args }=values(%args); | 
| 114 |  |  |  |  |  |  | } | 
| 115 |  |  |  |  |  |  |  | 
| 116 |  |  |  |  |  |  | } | 
| 117 |  |  |  |  |  |  |  | 
| 118 |  |  |  |  |  |  | sub _set_err_state | 
| 119 |  |  |  |  |  |  | { | 
| 120 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 121 | 0 |  |  |  |  |  | my $errcode=shift; | 
| 122 | 0 |  |  |  |  |  | my $errmsg=shift; | 
| 123 |  |  |  |  |  |  |  | 
| 124 | 0 | 0 |  |  |  |  | if (!defined($errcode)) | 
| 125 |  |  |  |  |  |  | { | 
| 126 | 0 |  |  |  |  |  | confess("set_err_state() called with an undefined argument."); | 
| 127 |  |  |  |  |  |  | } | 
| 128 |  |  |  |  |  |  |  | 
| 129 | 0 | 0 |  |  |  |  | if (exists($ErrMsgs{$errcode})) | 
| 130 |  |  |  |  |  |  | { | 
| 131 | 0 | 0 |  |  |  |  | if ($errcode==$ERR_OK) | 
| 132 |  |  |  |  |  |  | { | 
| 133 | 0 |  |  |  |  |  | $self->{errstr}=""; | 
| 134 |  |  |  |  |  |  | } | 
| 135 |  |  |  |  |  |  | else | 
| 136 |  |  |  |  |  |  | { | 
| 137 | 0 |  |  |  |  |  | $self->{errstr}.=" " . $ErrMsgs{$errcode}; | 
| 138 | 0 | 0 |  |  |  |  | if (defined($errmsg)) | 
| 139 |  |  |  |  |  |  | { | 
| 140 | 0 |  |  |  |  |  | $self->{errstr}.=" " . $errmsg; | 
| 141 |  |  |  |  |  |  | } | 
| 142 |  |  |  |  |  |  | } | 
| 143 |  |  |  |  |  |  | } | 
| 144 |  |  |  |  |  |  | else | 
| 145 |  |  |  |  |  |  | { | 
| 146 | 0 |  |  |  |  |  | confess("Internal error: set_err_state() called with an " . | 
| 147 |  |  |  |  |  |  | "unrecognized argument ($errcode).") | 
| 148 |  |  |  |  |  |  | } | 
| 149 |  |  |  |  |  |  | } | 
| 150 |  |  |  |  |  |  |  | 
| 151 |  |  |  |  |  |  | sub errmsg | 
| 152 |  |  |  |  |  |  | { | 
| 153 | 0 |  |  | 0 | 0 |  | my $self=shift; | 
| 154 |  |  |  |  |  |  |  | 
| 155 | 0 |  |  |  |  |  | return $self->{errstr}; | 
| 156 |  |  |  |  |  |  | } | 
| 157 |  |  |  |  |  |  |  | 
| 158 |  |  |  |  |  |  | ############################################################################ | 
| 159 |  |  |  |  |  |  | # | 
| 160 |  |  |  |  |  |  | #          Public methods | 
| 161 |  |  |  |  |  |  | # | 
| 162 |  |  |  |  |  |  | ############################################################################ | 
| 163 |  |  |  |  |  |  |  | 
| 164 |  |  |  |  |  |  | sub set_lemma | 
| 165 |  |  |  |  |  |  | { | 
| 166 | 0 |  |  | 0 | 1 |  | my $self=shift; | 
| 167 | 0 |  |  |  |  |  | $self->{lemmaSearch} = shift; | 
| 168 |  |  |  |  |  |  | } | 
| 169 |  |  |  |  |  |  |  | 
| 170 |  |  |  |  |  |  | sub set_text_fields | 
| 171 |  |  |  |  |  |  | { | 
| 172 | 0 |  |  | 0 | 1 |  | my $self=shift; | 
| 173 | 0 |  |  |  |  |  | my $fields=shift; | 
| 174 | 0 |  |  |  |  |  | $fields =~ s/\s+/ /g; | 
| 175 | 0 |  |  |  |  |  | $fields =~ s/"//g; | 
| 176 | 0 |  |  |  |  |  | $self->{textFields} = $fields; | 
| 177 |  |  |  |  |  |  | } | 
| 178 |  |  |  |  |  |  | sub set_canon | 
| 179 |  |  |  |  |  |  | { | 
| 180 | 0 |  |  | 0 | 1 |  | my $self=shift; | 
| 181 | 0 |  |  |  |  |  | $self->{tcanon} = shift(); | 
| 182 | 0 |  |  |  |  |  | $self->{ncanon} = shift(); | 
| 183 |  |  |  |  |  |  | } | 
| 184 |  |  |  |  |  |  | sub read_dicts | 
| 185 |  |  |  |  |  |  | { | 
| 186 | 0 |  |  | 0 | 1 |  | my $self=shift; | 
| 187 | 0 |  |  |  |  |  | my $lemma_dict_f=shift; | 
| 188 | 0 |  |  |  |  |  | my $term_dict_f=shift; | 
| 189 | 0 |  |  |  |  |  | my $NE_dict_f=shift; | 
| 190 | 0 |  |  |  |  |  | my $typing_rules_f=shift; | 
| 191 | 0 |  |  |  |  |  | my $onto_nodes_f=shift; | 
| 192 | 0 |  |  |  |  |  | my $onto_mapping_f=shift; | 
| 193 |  |  |  |  |  |  |  | 
| 194 | 0 | 0 |  |  |  |  | if (defined($lemma_dict_f)) | 
| 195 |  |  |  |  |  |  | { | 
| 196 | 0 |  |  |  |  |  | $self->{lemma_dict}=$self->_read_lemma_dict($lemma_dict_f); | 
| 197 |  |  |  |  |  |  | } | 
| 198 |  |  |  |  |  |  | else | 
| 199 |  |  |  |  |  |  | { | 
| 200 | 0 |  |  |  |  |  | undef $self->{lemma_dict}; | 
| 201 |  |  |  |  |  |  | } | 
| 202 | 0 | 0 |  |  |  |  | if (defined($term_dict_f)) | 
| 203 |  |  |  |  |  |  | { | 
| 204 | 0 |  |  |  |  |  | $self->{term_dict}=$self->_read_term_dict($term_dict_f); | 
| 205 |  |  |  |  |  |  | } | 
| 206 |  |  |  |  |  |  | else | 
| 207 |  |  |  |  |  |  | { | 
| 208 | 0 |  |  |  |  |  | undef $self->{term_dict}; | 
| 209 |  |  |  |  |  |  | } | 
| 210 | 0 | 0 |  |  |  |  | if (defined($NE_dict_f)) | 
| 211 |  |  |  |  |  |  | { | 
| 212 | 0 |  |  |  |  |  | $self->{NE_dict}=$self->_read_NE_dict($NE_dict_f); | 
| 213 |  |  |  |  |  |  | } | 
| 214 |  |  |  |  |  |  | else | 
| 215 |  |  |  |  |  |  | { | 
| 216 | 0 |  |  |  |  |  | undef $self->{NE_dict}; | 
| 217 |  |  |  |  |  |  | } | 
| 218 | 0 | 0 |  |  |  |  | if (defined($typing_rules_f)) | 
| 219 |  |  |  |  |  |  | { | 
| 220 | 0 |  |  |  |  |  | $self->{typing_rules}=$self->_read_typing_rules($typing_rules_f); | 
| 221 |  |  |  |  |  |  | } | 
| 222 |  |  |  |  |  |  | else | 
| 223 |  |  |  |  |  |  | { | 
| 224 | 0 |  |  |  |  |  | undef $self->{typing_rules}; | 
| 225 |  |  |  |  |  |  | } | 
| 226 | 0 | 0 |  |  |  |  | if (defined($onto_nodes_f)) | 
| 227 |  |  |  |  |  |  | { | 
| 228 | 0 |  |  |  |  |  | $self->{onto_nodes}=$self->_read_onto_nodes($onto_nodes_f); | 
| 229 |  |  |  |  |  |  | } | 
| 230 |  |  |  |  |  |  | else | 
| 231 |  |  |  |  |  |  | { | 
| 232 | 0 |  |  |  |  |  | undef $self->{onto_nodes}; | 
| 233 |  |  |  |  |  |  | } | 
| 234 | 0 | 0 |  |  |  |  | if (defined($onto_mapping_f)) | 
| 235 |  |  |  |  |  |  | { | 
| 236 | 0 |  |  |  |  |  | $self->{onto_paths}=$self->_read_onto_mapping($onto_mapping_f); | 
| 237 |  |  |  |  |  |  | } | 
| 238 |  |  |  |  |  |  | else | 
| 239 |  |  |  |  |  |  | { | 
| 240 | 0 |  |  |  |  |  | undef $self->{onto_paths}; | 
| 241 |  |  |  |  |  |  | } | 
| 242 |  |  |  |  |  |  |  | 
| 243 |  |  |  |  |  |  | # print STDERR " Term dict. check:  'Northern blot' -> " . | 
| 244 |  |  |  |  |  |  | #  $self->{term_dict}->{&canonise_def('Northern blot')} . "\n"; | 
| 245 |  |  |  |  |  |  |  | 
| 246 | 0 |  |  |  |  |  | return 1; | 
| 247 |  |  |  |  |  |  | } | 
| 248 |  |  |  |  |  |  |  | 
| 249 |  |  |  |  |  |  | sub cleanspaces() { | 
| 250 | 0 |  |  | 0 | 0 |  | $_ = shift(); | 
| 251 | 0 |  |  |  |  |  | s/\s+/ /g; | 
| 252 | 0 |  |  |  |  |  | s/^ //g; | 
| 253 | 0 |  |  |  |  |  | s/ $//g; | 
| 254 | 0 |  |  |  |  |  | return $_; | 
| 255 |  |  |  |  |  |  | } | 
| 256 |  |  |  |  |  |  |  | 
| 257 |  |  |  |  |  |  | sub _read_lemma_dict | 
| 258 |  |  |  |  |  |  | { | 
| 259 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 260 | 0 |  |  |  |  |  | my $f=shift; | 
| 261 |  |  |  |  |  |  |  | 
| 262 | 0 |  |  |  |  |  | my %dict=(); | 
| 263 |  |  |  |  |  |  |  | 
| 264 | 0 | 0 |  |  |  |  | if (!defined(open(F,"<:utf8",$f))) | 
| 265 |  |  |  |  |  |  | { | 
| 266 | 0 |  |  |  |  |  | return undef; | 
| 267 |  |  |  |  |  |  | } | 
| 268 |  |  |  |  |  |  |  | 
| 269 | 0 |  |  |  |  |  | while (my $l=) | 
| 270 |  |  |  |  |  |  | { | 
| 271 | 0 |  |  |  |  |  | chomp $l; | 
| 272 | 0 |  |  |  |  |  | my ($form,$lemma,$pos)=split(/\t/,$l,-1); | 
| 273 | 0 |  |  |  |  |  | $form = &cleanspaces($form); | 
| 274 | 0 |  |  |  |  |  | $dict{lc($form)}{lemma}=&cleanspaces($lemma); | 
| 275 | 0 |  |  |  |  |  | $dict{lc($form)}{POS}=$pos; | 
| 276 |  |  |  |  |  |  | } | 
| 277 |  |  |  |  |  |  |  | 
| 278 | 0 |  |  |  |  |  | close(F); | 
| 279 |  |  |  |  |  |  |  | 
| 280 | 0 |  |  |  |  |  | return \%dict; | 
| 281 |  |  |  |  |  |  | } | 
| 282 |  |  |  |  |  |  |  | 
| 283 |  |  |  |  |  |  | #  default method to standardise terms and named entities | 
| 284 |  |  |  |  |  |  | #            lower case, ignore space and '-' | 
| 285 |  |  |  |  |  |  | sub canonise_def { | 
| 286 | 0 |  |  | 0 | 0 |  | $_ = shift(); | 
| 287 | 0 |  |  |  |  |  | s/\s+//g; | 
| 288 | 0 |  |  |  |  |  | s/\-//g; | 
| 289 | 0 |  |  |  |  |  | $_ = lc($_); | 
| 290 |  |  |  |  |  |  | } | 
| 291 |  |  |  |  |  |  |  | 
| 292 |  |  |  |  |  |  | sub _read_term_dict | 
| 293 |  |  |  |  |  |  | { | 
| 294 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 295 | 0 |  |  |  |  |  | my $f=shift; | 
| 296 |  |  |  |  |  |  |  | 
| 297 | 0 |  |  |  |  |  | my %dict=(); | 
| 298 | 0 |  |  |  |  |  | my $term_max_len = 0; | 
| 299 |  |  |  |  |  |  |  | 
| 300 | 0 | 0 |  |  |  |  | if (!defined(open(F,"<:utf8",$f))) | 
| 301 |  |  |  |  |  |  | { | 
| 302 | 0 |  |  |  |  |  | return undef; | 
| 303 |  |  |  |  |  |  | } | 
| 304 |  |  |  |  |  |  |  | 
| 305 | 0 |  |  |  |  |  | while (my $l=) | 
| 306 |  |  |  |  |  |  | { | 
| 307 | 0 |  |  |  |  |  | chomp $l; | 
| 308 | 0 |  |  |  |  |  | my ($form,$can)=split(/\t/,$l,-1); | 
| 309 | 0 |  |  |  |  |  | $form = &cleanspaces($form); | 
| 310 | 0 |  |  |  |  |  | $can = &cleanspaces($can); | 
| 311 | 0 |  |  |  |  |  | my $cf = &{$self->{tcanon}}($form); | 
|  | 0 |  |  |  |  |  |  | 
| 312 | 0 | 0 | 0 |  |  |  | if ( $verbose && defined($dict{$cf}) && $dict{$cf} ne $can ) { | 
|  |  |  | 0 |  |  |  |  | 
| 313 | 0 |  |  |  |  |  | print STDERR "Term of form '$form' has canonical form '$can'\n" | 
| 314 |  |  |  |  |  |  | . "   but maps to the another canonical form '$dict{$cf}'\n"; | 
| 315 |  |  |  |  |  |  | } | 
| 316 | 0 |  |  |  |  |  | $dict{$cf}=$can; | 
| 317 | 0 |  |  |  |  |  | my @tt = split(/ /,$form); | 
| 318 | 0 | 0 |  |  |  |  | if ( scalar(@tt)> $term_max_len) { | 
| 319 | 0 |  |  |  |  |  | $term_max_len = scalar(@tt); | 
| 320 |  |  |  |  |  |  | } | 
| 321 |  |  |  |  |  |  | } | 
| 322 | 0 | 0 |  |  |  |  | if ( $self->{termMaxLen}<$term_max_len ) { | 
| 323 | 0 |  |  |  |  |  | $self->{termMaxLen} = $term_max_len; | 
| 324 |  |  |  |  |  |  | } | 
| 325 |  |  |  |  |  |  |  | 
| 326 | 0 |  |  |  |  |  | close(F); | 
| 327 |  |  |  |  |  |  |  | 
| 328 | 0 |  |  |  |  |  | return \%dict; | 
| 329 |  |  |  |  |  |  | } | 
| 330 |  |  |  |  |  |  |  | 
| 331 |  |  |  |  |  |  | sub _read_NE_dict | 
| 332 |  |  |  |  |  |  | { | 
| 333 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 334 | 0 |  |  |  |  |  | my $f=shift; | 
| 335 |  |  |  |  |  |  |  | 
| 336 | 0 |  |  |  |  |  | my %dict=(); | 
| 337 | 0 |  |  |  |  |  | my $term_max_len = 0; | 
| 338 |  |  |  |  |  |  |  | 
| 339 | 0 | 0 |  |  |  |  | if (!defined(open(F,"<:utf8",$f))) | 
| 340 |  |  |  |  |  |  | { | 
| 341 | 0 |  |  |  |  |  | return undef; | 
| 342 |  |  |  |  |  |  | } | 
| 343 |  |  |  |  |  |  |  | 
| 344 | 0 |  |  |  |  |  | while (my $l=) | 
| 345 |  |  |  |  |  |  | { | 
| 346 | 0 |  |  |  |  |  | chomp $l; | 
| 347 | 0 |  |  |  |  |  | my ($form,$can)=split(/\t/,$l,-1); | 
| 348 | 0 |  |  |  |  |  | $form = &cleanspaces($form); | 
| 349 | 0 |  |  |  |  |  | $can = &cleanspaces($can); | 
| 350 | 0 |  |  |  |  |  | my $cf = &{$self->{ncanon}}($form); | 
|  | 0 |  |  |  |  |  |  | 
| 351 | 0 | 0 | 0 |  |  |  | if (  $verbose && defined($dict{$cf}) && $dict{$cf} ne $can ) { | 
|  |  |  | 0 |  |  |  |  | 
| 352 | 0 |  |  |  |  |  | print STDERR "NE of form '$form' has canonical form '$can'\n" | 
| 353 |  |  |  |  |  |  | . "   but maps to the another canonical form '$dict{$cf}'\n"; | 
| 354 |  |  |  |  |  |  | } | 
| 355 | 0 |  |  |  |  |  | $dict{$cf}=$can; | 
| 356 | 0 |  |  |  |  |  | my @tt = split(/\s+/,$form); | 
| 357 | 0 | 0 |  |  |  |  | if ( scalar(@tt)> $term_max_len) { | 
| 358 | 0 |  |  |  |  |  | $term_max_len = scalar(@tt); | 
| 359 |  |  |  |  |  |  | } | 
| 360 |  |  |  |  |  |  | } | 
| 361 | 0 | 0 |  |  |  |  | if ( $self->{termMaxLen}<$term_max_len ) { | 
| 362 | 0 |  |  |  |  |  | $self->{termMaxLen} = $term_max_len; | 
| 363 |  |  |  |  |  |  | } | 
| 364 |  |  |  |  |  |  |  | 
| 365 | 0 |  |  |  |  |  | close(F); | 
| 366 |  |  |  |  |  |  |  | 
| 367 | 0 |  |  |  |  |  | return \%dict; | 
| 368 |  |  |  |  |  |  | } | 
| 369 |  |  |  |  |  |  |  | 
| 370 |  |  |  |  |  |  | sub _read_typing_rules | 
| 371 |  |  |  |  |  |  | { | 
| 372 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 373 | 0 |  |  |  |  |  | my $f=shift; | 
| 374 |  |  |  |  |  |  |  | 
| 375 | 0 |  |  |  |  |  | my %dict=(); | 
| 376 |  |  |  |  |  |  |  | 
| 377 | 0 | 0 |  |  |  |  | if (!defined(open(F,"<:utf8",$f))) | 
| 378 |  |  |  |  |  |  | { | 
| 379 | 0 |  |  |  |  |  | return undef; | 
| 380 |  |  |  |  |  |  | } | 
| 381 |  |  |  |  |  |  |  | 
| 382 | 0 |  |  |  |  |  | while (my $l=) | 
| 383 |  |  |  |  |  |  | { | 
| 384 | 0 |  |  |  |  |  | chomp $l; | 
| 385 | 0 |  |  |  |  |  | my ($form,$type)=split(/\t/,$l,-1); | 
| 386 | 0 |  |  |  |  |  | $form = &cleanspaces($form); | 
| 387 | 0 |  |  |  |  |  | $type = &cleanspaces($type); | 
| 388 | 0 |  |  |  |  |  | $dict{$form}=$type; | 
| 389 |  |  |  |  |  |  | } | 
| 390 |  |  |  |  |  |  |  | 
| 391 | 0 |  |  |  |  |  | close(F); | 
| 392 |  |  |  |  |  |  |  | 
| 393 | 0 |  |  |  |  |  | return \%dict; | 
| 394 |  |  |  |  |  |  | } | 
| 395 |  |  |  |  |  |  |  | 
| 396 |  |  |  |  |  |  | sub _read_onto_nodes | 
| 397 |  |  |  |  |  |  | { | 
| 398 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 399 | 0 |  |  |  |  |  | my $f=shift; | 
| 400 |  |  |  |  |  |  |  | 
| 401 | 0 |  |  |  |  |  | my %dict=(); | 
| 402 |  |  |  |  |  |  |  | 
| 403 | 0 | 0 |  |  |  |  | if (!defined(open(F,"<:utf8",$f))) | 
| 404 |  |  |  |  |  |  | { | 
| 405 | 0 |  |  |  |  |  | return undef; | 
| 406 |  |  |  |  |  |  | } | 
| 407 |  |  |  |  |  |  |  | 
| 408 | 0 |  |  |  |  |  | while (my $l=) | 
| 409 |  |  |  |  |  |  | { | 
| 410 | 0 |  |  |  |  |  | chomp $l; | 
| 411 | 0 |  |  |  |  |  | my ($form,$onto_node)=split(/\t/,$l,-1); | 
| 412 | 0 |  |  |  |  |  | $form = &cleanspaces($form); | 
| 413 | 0 |  |  |  |  |  | $onto_node = &cleanspaces($onto_node); | 
| 414 | 0 |  |  |  |  |  | $dict{$form}=$onto_node; | 
| 415 |  |  |  |  |  |  | } | 
| 416 |  |  |  |  |  |  |  | 
| 417 | 0 |  |  |  |  |  | close(F); | 
| 418 |  |  |  |  |  |  |  | 
| 419 | 0 |  |  |  |  |  | return \%dict; | 
| 420 |  |  |  |  |  |  | } | 
| 421 |  |  |  |  |  |  |  | 
| 422 |  |  |  |  |  |  | sub _read_onto_mapping | 
| 423 |  |  |  |  |  |  | { | 
| 424 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 425 | 0 |  |  |  |  |  | my $f=shift; | 
| 426 |  |  |  |  |  |  |  | 
| 427 | 0 |  |  |  |  |  | my %dict=(); | 
| 428 |  |  |  |  |  |  |  | 
| 429 | 0 | 0 |  |  |  |  | if (!defined(open(F,"<:utf8",$f))) | 
| 430 |  |  |  |  |  |  | { | 
| 431 | 0 |  |  |  |  |  | return undef; | 
| 432 |  |  |  |  |  |  | } | 
| 433 |  |  |  |  |  |  |  | 
| 434 | 0 |  |  |  |  |  | while (my $l=) | 
| 435 |  |  |  |  |  |  | { | 
| 436 | 0 |  |  |  |  |  | chomp $l; | 
| 437 | 0 |  |  |  |  |  | my ($node,$path)=split(/\t/,$l,-1); | 
| 438 | 0 |  |  |  |  |  | $node = &cleanspaces($node); | 
| 439 | 0 |  |  |  |  |  | $path = &cleanspaces($path); | 
| 440 | 0 |  |  |  |  |  | $dict{$node}=$path; | 
| 441 |  |  |  |  |  |  | } | 
| 442 |  |  |  |  |  |  |  | 
| 443 | 0 |  |  |  |  |  | close(F); | 
| 444 |  |  |  |  |  |  |  | 
| 445 | 0 |  |  |  |  |  | return \%dict; | 
| 446 |  |  |  |  |  |  | } | 
| 447 |  |  |  |  |  |  |  | 
| 448 |  |  |  |  |  |  | sub transform   # just for testing and debugging | 
| 449 |  |  |  |  |  |  | { | 
| 450 | 0 |  |  | 0 | 0 |  | my $self=shift; | 
| 451 | 0 |  |  |  |  |  | my $query=shift; # list of word forms | 
| 452 |  |  |  |  |  |  |  | 
| 453 | 0 |  |  |  |  |  | my $expanded_query_struct=$self->_expand_qword_list($query); | 
| 454 |  |  |  |  |  |  |  | 
| 455 | 0 |  |  |  |  |  | $self->{queryForm} = $query; | 
| 456 | 0 |  |  |  |  |  | $self->{finalForm} = ""; | 
| 457 |  |  |  |  |  |  |  | 
| 458 | 0 |  |  |  |  |  | my $query_XML=$self->_data_struct2XML($expanded_query_struct); | 
| 459 |  |  |  |  |  |  |  | 
| 460 | 0 |  |  |  |  |  | return $query_XML; | 
| 461 |  |  |  |  |  |  | } | 
| 462 |  |  |  |  |  |  |  | 
| 463 |  |  |  |  |  |  | # | 
| 464 |  |  |  |  |  |  | # Given a list of word forms, expand | 
| 465 |  |  |  |  |  |  | # | 
| 466 |  |  |  |  |  |  | sub _expand_qword_list | 
| 467 |  |  |  |  |  |  | { | 
| 468 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 469 | 0 |  |  |  |  |  | my $query=shift; # list of word forms | 
| 470 |  |  |  |  |  |  |  | 
| 471 |  |  |  |  |  |  | # print STDERR "Q: " . Dumper($query) . "\n"; | 
| 472 |  |  |  |  |  |  |  | 
| 473 | 0 |  |  |  |  |  | my $lemmatized_by_tagger=$self->_apply_treetagger($query); | 
| 474 | 0 | 0 |  |  |  |  | if (!defined($lemmatized_by_tagger)) | 
| 475 |  |  |  |  |  |  | { | 
| 476 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_TREETAGGER); | 
| 477 | 0 |  |  |  |  |  | return undef; | 
| 478 |  |  |  |  |  |  | } | 
| 479 |  |  |  |  |  |  |  | 
| 480 |  |  |  |  |  |  | # print STDERR "LEM: " . Dumper($lemmatized_by_tagger) . "\n"; | 
| 481 |  |  |  |  |  |  |  | 
| 482 | 0 |  |  |  |  |  | my $lemmatized= | 
| 483 |  |  |  |  |  |  | $self->_apply_lemma_dict($lemmatized_by_tagger); # if one exists | 
| 484 | 0 | 0 |  |  |  |  | if (!defined($lemmatized)) | 
| 485 |  |  |  |  |  |  | { | 
| 486 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_LEMMA_DICT); | 
| 487 | 0 |  |  |  |  |  | return undef; | 
| 488 |  |  |  |  |  |  | } | 
| 489 |  |  |  |  |  |  |  | 
| 490 |  |  |  |  |  |  | # print STDERR "LEMTAG: " . Dumper($lemmatized) . "\n"; | 
| 491 |  |  |  |  |  |  |  | 
| 492 | 0 |  |  |  |  |  | my $term_NE_expanded=$self->_apply_terms_and_NEs($lemmatized); | 
| 493 | 0 | 0 |  |  |  |  | if (!defined($term_NE_expanded)) | 
| 494 |  |  |  |  |  |  | { | 
| 495 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_APPLYING_TERM_NE); | 
| 496 | 0 |  |  |  |  |  | return undef; | 
| 497 |  |  |  |  |  |  | } | 
| 498 |  |  |  |  |  |  | # print STDERR "TERM: " . Dumper($term_NE_expanded) . "\n"; | 
| 499 |  |  |  |  |  |  |  | 
| 500 |  |  |  |  |  |  |  | 
| 501 | 0 |  |  |  |  |  | my $typing_expanded=$self->_apply_typing_rules($term_NE_expanded); | 
| 502 | 0 | 0 |  |  |  |  | if (!defined($typing_expanded)) | 
| 503 |  |  |  |  |  |  | { | 
| 504 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_APPLYING_TYPING); | 
| 505 | 0 |  |  |  |  |  | return undef; | 
| 506 |  |  |  |  |  |  | } | 
| 507 |  |  |  |  |  |  |  | 
| 508 | 0 |  |  |  |  |  | my $onto_expanded=$self->_apply_onto($typing_expanded); | 
| 509 | 0 | 0 |  |  |  |  | if (!defined($onto_expanded)) | 
| 510 |  |  |  |  |  |  | { | 
| 511 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_APPLYING_ONTO); | 
| 512 | 0 |  |  |  |  |  | return undef; | 
| 513 |  |  |  |  |  |  | } | 
| 514 |  |  |  |  |  |  | # print STDERR "FINAL: " . Dumper($onto_expanded) . "\n"; | 
| 515 |  |  |  |  |  |  |  | 
| 516 | 0 |  |  |  |  |  | return $onto_expanded; | 
| 517 |  |  |  |  |  |  | } | 
| 518 |  |  |  |  |  |  |  | 
| 519 |  |  |  |  |  |  | #  extract query from SRU | 
| 520 |  |  |  |  |  |  | sub UI2Query | 
| 521 |  |  |  |  |  |  | { | 
| 522 | 0 |  |  | 0 | 0 |  | my $self=shift; | 
| 523 | 0 |  |  |  |  |  | my $SRU=shift; | 
| 524 | 0 | 0 |  |  |  |  | if ( /&query=([^\&]*)/ ) { | 
| 525 | 0 |  |  |  |  |  | return $1; | 
| 526 |  |  |  |  |  |  | } | 
| 527 | 0 |  |  |  |  |  | return ""; | 
| 528 |  |  |  |  |  |  | } | 
| 529 |  |  |  |  |  |  |  | 
| 530 |  |  |  |  |  |  | # | 
| 531 |  |  |  |  |  |  | #  UI ---> Zebra  middle man | 
| 532 |  |  |  |  |  |  | # | 
| 533 |  |  |  |  |  |  | sub UI2Zebra | 
| 534 |  |  |  |  |  |  | { | 
| 535 | 0 |  |  | 0 | 1 |  | my $self=shift; | 
| 536 | 0 |  |  |  |  |  | my $SRU=shift; | 
| 537 |  |  |  |  |  |  |  | 
| 538 | 0 |  |  |  |  |  | my @expanded_SRU=(); | 
| 539 |  |  |  |  |  |  |  | 
| 540 |  |  |  |  |  |  | # extract the query | 
| 541 | 0 |  |  |  |  |  | my $query; | 
| 542 | 0 |  |  |  |  |  | my @p=split(/\&/,$SRU,-1); | 
| 543 | 0 |  |  |  |  |  | for my $p (@p) | 
| 544 |  |  |  |  |  |  | { | 
| 545 | 0 | 0 |  |  |  |  | if ($p=~/^query=(.*)$/) | 
| 546 |  |  |  |  |  |  | { | 
| 547 | 0 |  |  |  |  |  | $query=$1; | 
| 548 |  |  |  |  |  |  | } | 
| 549 |  |  |  |  |  |  | else | 
| 550 |  |  |  |  |  |  | { | 
| 551 | 0 |  |  |  |  |  | push(@expanded_SRU,$p); # so we can reconstruct | 
| 552 |  |  |  |  |  |  | } | 
| 553 |  |  |  |  |  |  | } | 
| 554 | 0 | 0 |  |  |  |  | if (!defined($query)) | 
| 555 |  |  |  |  |  |  | { | 
| 556 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_NO_QUERY,"SRU:\"$SRU\""); | 
| 557 | 0 |  |  |  |  |  | return undef; | 
| 558 |  |  |  |  |  |  | } | 
| 559 | 0 |  |  |  |  |  | $self->{queryForm} = $query; | 
| 560 | 0 |  |  |  |  |  | $self->{queryForm} =~ s/\&/\&/g; | 
| 561 | 0 |  |  |  |  |  | $self->{queryForm} =~ s/\</g; | 
| 562 | 0 |  |  |  |  |  | $self->{queryForm} =~ s/>/\>/g; | 
| 563 | 0 |  |  |  |  |  | $self->{finalForm} = ""; | 
| 564 |  |  |  |  |  |  |  | 
| 565 |  |  |  |  |  |  | # decode percentage notation | 
| 566 | 0 |  |  |  |  |  | my $query_copy=$query; | 
| 567 | 0 |  |  |  |  |  | $query_copy=uri_unescape($query_copy); | 
| 568 |  |  |  |  |  |  |  | 
| 569 |  |  |  |  |  |  | # parse the CQL | 
| 570 | 0 |  |  |  |  |  | my $parse_tree; | 
| 571 |  |  |  |  |  |  | eval | 
| 572 | 0 |  |  |  |  |  | { | 
| 573 | 0 |  |  |  |  |  | $parse_tree=$self->{CQLParser}->parse($query_copy); | 
| 574 |  |  |  |  |  |  | }; | 
| 575 | 0 | 0 |  |  |  |  | if ($@) | 
| 576 |  |  |  |  |  |  | { | 
| 577 | 0 |  |  |  |  |  | chomp($query); | 
| 578 | 0 |  |  |  |  |  | $@=~s/(.*) at .* line [0-9]+\n/$1/o; | 
| 579 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_CQL_PARSE,"Query:\"$query\"."); | 
| 580 | 0 |  |  |  |  |  | return undef; | 
| 581 |  |  |  |  |  |  | } | 
| 582 |  |  |  |  |  |  |  | 
| 583 |  |  |  |  |  |  | # Get a list of all possible text query word sequences (so this is | 
| 584 |  |  |  |  |  |  | # implicitly an OR of them) | 
| 585 |  |  |  |  |  |  | # | 
| 586 | 0 |  |  |  |  |  | my $t_qwords=[[]]; # help variable used in the recursion | 
| 587 | 0 |  |  |  |  |  | my $seq_list= | 
| 588 |  |  |  |  |  |  | $self->_get_text_qwords($parse_tree,$t_qwords); | 
| 589 | 0 | 0 |  |  |  |  | if (!defined($seq_list)) | 
| 590 |  |  |  |  |  |  | { | 
| 591 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_CREATING_SEQ_LIST); | 
| 592 | 0 |  |  |  |  |  | return undef; | 
| 593 |  |  |  |  |  |  | } | 
| 594 |  |  |  |  |  |  |  | 
| 595 |  |  |  |  |  |  | # Get the categorising tail anded to the end | 
| 596 |  |  |  |  |  |  | # | 
| 597 | 0 |  |  |  |  |  | my $cats=&get_categories($parse_tree); | 
| 598 | 0 | 0 |  |  |  |  | if (!defined($cats)) | 
| 599 |  |  |  |  |  |  | { | 
| 600 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_CREATING_CAT_LIST); | 
| 601 | 0 |  |  |  |  |  | return undef; | 
| 602 |  |  |  |  |  |  | } | 
| 603 |  |  |  |  |  |  |  | 
| 604 |  |  |  |  |  |  | # Important! Used in the following Zebra2UI, 'cause the | 
| 605 |  |  |  |  |  |  | # SRU response has nothing about the query. | 
| 606 |  |  |  |  |  |  | # So...if used out of sync/with more than one client...kaboom! | 
| 607 |  |  |  |  |  |  | # Not my problem. | 
| 608 | 0 |  |  |  |  |  | $self->{currSeqList}=$seq_list; | 
| 609 |  |  |  |  |  |  |  | 
| 610 |  |  |  |  |  |  | # print STDERR "Term elements: " . Dumper($seq_list) . "\n"; | 
| 611 |  |  |  |  |  |  |  | 
| 612 |  |  |  |  |  |  | # | 
| 613 |  |  |  |  |  |  | # Ok, create the 'tail' i.e. what we AND to the original query | 
| 614 |  |  |  |  |  |  | # as an OR of possible expansions | 
| 615 |  |  |  |  |  |  | # | 
| 616 | 0 |  |  |  |  |  | my $CQL_tail=$self->_data_struct2CQLtail($seq_list); | 
| 617 | 0 | 0 |  |  |  |  | if (!defined($CQL_tail)) | 
| 618 |  |  |  |  |  |  | { | 
| 619 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_CREATING_CQL_TAIL); | 
| 620 | 0 |  |  |  |  |  | return undef; | 
| 621 |  |  |  |  |  |  | } | 
| 622 |  |  |  |  |  |  |  | 
| 623 |  |  |  |  |  |  | # print STDERR "QQ##$query##$cats##$CQL_tail\n"; | 
| 624 |  |  |  |  |  |  |  | 
| 625 |  |  |  |  |  |  | #$query='%28' . $query . '%29%20and%20' $CQL_tail; | 
| 626 | 0 |  |  |  |  |  | $query=$CQL_tail; | 
| 627 | 0 | 0 |  |  |  |  | if ( $cats ) { | 
| 628 | 0 |  |  |  |  |  | $query .= '%20and%20' . $cats; | 
| 629 |  |  |  |  |  |  | } | 
| 630 |  |  |  |  |  |  |  | 
| 631 | 0 |  |  |  |  |  | push(@expanded_SRU,"query=$query"); | 
| 632 |  |  |  |  |  |  |  | 
| 633 | 0 |  |  |  |  |  | $self->{finalForm} = $query; | 
| 634 | 0 |  |  |  |  |  | $self->{finalForm} =~ s/\&/\&/g; | 
| 635 | 0 |  |  |  |  |  | $self->{finalForm} =~ s/\</g; | 
| 636 | 0 |  |  |  |  |  | $self->{finalForm} =~ s/>/\>/g; | 
| 637 |  |  |  |  |  |  |  | 
| 638 | 0 |  |  |  |  |  | return join('&',@expanded_SRU); | 
| 639 |  |  |  |  |  |  | } | 
| 640 |  |  |  |  |  |  |  | 
| 641 |  |  |  |  |  |  | # | 
| 642 |  |  |  |  |  |  | #  Zebra ---> UI  middle man | 
| 643 |  |  |  |  |  |  | # | 
| 644 |  |  |  |  |  |  | sub Zebra2UI | 
| 645 |  |  |  |  |  |  | { | 
| 646 | 0 |  |  | 0 | 1 |  | my $self=shift; | 
| 647 | 0 |  |  |  |  |  | my $SRU_response=shift; | 
| 648 |  |  |  |  |  |  |  | 
| 649 |  |  |  |  |  |  | # We need to know what the query was! It's not in the response. | 
| 650 |  |  |  |  |  |  | # Of course this is bloody dangerous if we get out of sync or | 
| 651 |  |  |  |  |  |  | # have more than 1 client. | 
| 652 |  |  |  |  |  |  | # | 
| 653 | 0 | 0 |  |  |  |  | if (!defined($self->{currSeqList})) | 
| 654 |  |  |  |  |  |  | { | 
| 655 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_NO_SEQ_LIST); | 
| 656 | 0 |  |  |  |  |  | return undef; | 
| 657 |  |  |  |  |  |  | } | 
| 658 |  |  |  |  |  |  |  | 
| 659 |  |  |  |  |  |  | # | 
| 660 |  |  |  |  |  |  | # Just convert it to our XML format to put into extraResponseData | 
| 661 |  |  |  |  |  |  | # I chose to just catenate  elements as an implicit OR.. | 
| 662 |  |  |  |  |  |  | # | 
| 663 | 0 |  |  |  |  |  | my $query_XML=$self->_data_struct2XML($self->{currSeqList}); | 
| 664 | 0 | 0 |  |  |  |  | if (!defined($query_XML)) | 
| 665 |  |  |  |  |  |  | { | 
| 666 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_STRUCT2XML); | 
| 667 | 0 |  |  |  |  |  | return undef; | 
| 668 |  |  |  |  |  |  | } | 
| 669 |  |  |  |  |  |  |  | 
| 670 | 0 |  |  |  |  |  | ${$SRU_response} =~ s/<\/zs:searchRetrieveResponse>/$query_XML<\/zs:extraResponseData><\/zs:searchRetrieveResponse>/; | 
|  | 0 |  |  |  |  |  |  | 
| 671 |  |  |  |  |  |  |  | 
| 672 | 0 |  |  |  |  |  | return 1; | 
| 673 |  |  |  |  |  |  | } | 
| 674 |  |  |  |  |  |  |  | 
| 675 |  |  |  |  |  |  |  | 
| 676 |  |  |  |  |  |  | # | 
| 677 |  |  |  |  |  |  | # Recursive CQL parse tree traversal, results in picking out the relevant | 
| 678 |  |  |  |  |  |  | # text query words in order. Too tired to explain. | 
| 679 |  |  |  |  |  |  | # | 
| 680 |  |  |  |  |  |  | sub _get_text_qwords | 
| 681 |  |  |  |  |  |  | { | 
| 682 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 683 | 0 |  |  |  |  |  | my $CQL_parse_node=shift; | 
| 684 | 0 |  |  |  |  |  | my $text_qwords=shift; | 
| 685 |  |  |  |  |  |  |  | 
| 686 | 0 |  |  |  |  |  | my ($text_qwords_l,$text_qwords_r); | 
| 687 |  |  |  |  |  |  |  | 
| 688 |  |  |  |  |  |  | # print STDERR "ENTRY:",Dumper($text_qwords); | 
| 689 |  |  |  |  |  |  |  | 
| 690 | 0 | 0 |  |  |  |  | if ($CQL_parse_node->isa("CQL::AndNode")) | 
|  |  | 0 |  |  |  |  |  | 
|  |  | 0 |  |  |  |  |  | 
|  |  | 0 |  |  |  |  |  | 
| 691 |  |  |  |  |  |  | { | 
| 692 |  |  |  |  |  |  | #	warn "AND"; | 
| 693 | 0 |  |  |  |  |  | $text_qwords_l=$self->_get_text_qwords($CQL_parse_node->left(), | 
| 694 |  |  |  |  |  |  | $text_qwords); | 
| 695 | 0 |  |  |  |  |  | $text_qwords_r=$self->_get_text_qwords($CQL_parse_node->right(), | 
| 696 |  |  |  |  |  |  | $text_qwords_l); | 
| 697 |  |  |  |  |  |  |  | 
| 698 | 0 |  |  |  |  |  | return $text_qwords_r; | 
| 699 |  |  |  |  |  |  | } | 
| 700 |  |  |  |  |  |  | elsif ($CQL_parse_node->isa("CQL::OrNode")) | 
| 701 |  |  |  |  |  |  | { | 
| 702 |  |  |  |  |  |  | #	warn "OR"; | 
| 703 |  |  |  |  |  |  |  | 
| 704 | 0 |  |  |  |  |  | $text_qwords_l=$self->_get_text_qwords($CQL_parse_node->left(), | 
| 705 |  |  |  |  |  |  | $text_qwords); | 
| 706 | 0 |  |  |  |  |  | $text_qwords_r=$self->_get_text_qwords($CQL_parse_node->right(), | 
| 707 |  |  |  |  |  |  | $text_qwords); | 
| 708 |  |  |  |  |  |  |  | 
| 709 | 0 |  |  |  |  |  | return [@$text_qwords_l,@$text_qwords_r]; | 
| 710 |  |  |  |  |  |  | } | 
| 711 |  |  |  |  |  |  | elsif ($CQL_parse_node->isa("CQL::NotNode")) | 
| 712 |  |  |  |  |  |  | { | 
| 713 | 0 |  |  |  |  |  | $text_qwords_l=$self->_get_text_qwords($CQL_parse_node->left(), | 
| 714 |  |  |  |  |  |  | $text_qwords); | 
| 715 |  |  |  |  |  |  |  | 
| 716 | 0 |  |  |  |  |  | return $text_qwords_l; | 
| 717 |  |  |  |  |  |  |  | 
| 718 |  |  |  |  |  |  | } | 
| 719 |  |  |  |  |  |  | elsif ($CQL_parse_node->isa("CQL::TermNode")) | 
| 720 |  |  |  |  |  |  | { | 
| 721 |  |  |  |  |  |  | #	warn "TERM"; | 
| 722 |  |  |  |  |  |  |  | 
| 723 | 0 |  |  |  |  |  | my $qualifier=$CQL_parse_node->getQualifier(); | 
| 724 | 0 |  |  |  |  |  | my ($index_set_name,$index_name)=split(/\./,$qualifier); | 
| 725 |  |  |  |  |  |  |  | 
| 726 | 0 | 0 |  |  |  |  | if (!defined($index_name)) | 
| 727 |  |  |  |  |  |  | { | 
| 728 | 0 |  |  |  |  |  | $index_set_name=$self->{indexSetName}; | 
| 729 | 0 |  |  |  |  |  | $index_name=$qualifier; | 
| 730 |  |  |  |  |  |  | } | 
| 731 |  |  |  |  |  |  |  | 
| 732 | 0 |  |  |  |  |  | my $term=$CQL_parse_node->getTerm(); | 
| 733 |  |  |  |  |  |  |  | 
| 734 |  |  |  |  |  |  | # Our partial hack solution: if it contains a space, leave as is. | 
| 735 |  |  |  |  |  |  | # Wray's hack on hack - keep space stuff, and deal with it differently | 
| 736 | 0 |  |  |  |  |  | if ( 1 || ( $term!~/\s/) ) | 
| 737 |  |  |  |  |  |  | { | 
| 738 | 0 |  |  |  |  |  | $term =~ s/\s+/\#\#/g; | 
| 739 | 0 | 0 | 0 |  |  |  | if ($qualifier eq 'text' || $qualifier eq 'srw.ServerChoice') | 
| 740 |  |  |  |  |  |  | { | 
| 741 | 0 |  |  |  |  |  | my @update=(); | 
| 742 | 0 |  |  |  |  |  | for my $qwords (@$text_qwords) | 
| 743 |  |  |  |  |  |  | { | 
| 744 | 0 |  |  |  |  |  | my @qw=@$qwords; | 
| 745 | 0 |  |  |  |  |  | push(@qw,$term); | 
| 746 | 0 |  |  |  |  |  | push(@update,[@qw]); | 
| 747 |  |  |  |  |  |  | } | 
| 748 | 0 |  |  |  |  |  | return \@update; | 
| 749 |  |  |  |  |  |  | } | 
| 750 |  |  |  |  |  |  | } | 
| 751 |  |  |  |  |  |  | } | 
| 752 |  |  |  |  |  |  |  | 
| 753 | 0 |  |  |  |  |  | return $text_qwords; | 
| 754 |  |  |  |  |  |  | } | 
| 755 |  |  |  |  |  |  |  | 
| 756 |  |  |  |  |  |  | # | 
| 757 |  |  |  |  |  |  | # Recursive CQL parse tree traversal, results in picking out | 
| 758 |  |  |  |  |  |  | # the ANDed category part at the end | 
| 759 |  |  |  |  |  |  | # | 
| 760 |  |  |  |  |  |  | sub get_categories | 
| 761 |  |  |  |  |  |  | { | 
| 762 | 0 |  |  | 0 | 0 |  | my $CQL_parse_node=shift; | 
| 763 |  |  |  |  |  |  |  | 
| 764 | 0 |  |  |  |  |  | my ($text_catq_l,$text_catq_r); | 
| 765 |  |  |  |  |  |  |  | 
| 766 |  |  |  |  |  |  | #    print STDERR "ENTRY:",Dumper($text_catq); | 
| 767 |  |  |  |  |  |  |  | 
| 768 | 0 | 0 |  |  |  |  | if ($CQL_parse_node->isa("CQL::AndNode")) | 
|  |  | 0 |  |  |  |  |  | 
|  |  | 0 |  |  |  |  |  | 
|  |  | 0 |  |  |  |  |  | 
| 769 |  |  |  |  |  |  | { | 
| 770 |  |  |  |  |  |  | #	warn "AND"; | 
| 771 | 0 |  |  |  |  |  | $text_catq_l=&get_categories($CQL_parse_node->left()); | 
| 772 | 0 |  |  |  |  |  | $text_catq_r=&get_categories($CQL_parse_node->right()); | 
| 773 | 0 | 0 | 0 |  |  |  | if ( $text_catq_l && $text_catq_r ) { | 
| 774 | 0 |  |  |  |  |  | return $text_catq_l . ' and ' .$text_catq_r; | 
| 775 |  |  |  |  |  |  | } | 
| 776 | 0 |  |  |  |  |  | return $text_catq_l . $text_catq_r; | 
| 777 |  |  |  |  |  |  | } | 
| 778 |  |  |  |  |  |  | elsif ($CQL_parse_node->isa("CQL::OrNode")) | 
| 779 |  |  |  |  |  |  | { | 
| 780 | 0 |  |  |  |  |  | return ""; | 
| 781 |  |  |  |  |  |  | } | 
| 782 |  |  |  |  |  |  | elsif ($CQL_parse_node->isa("CQL::NotNode")) | 
| 783 |  |  |  |  |  |  | { | 
| 784 | 0 |  |  |  |  |  | return ""; | 
| 785 |  |  |  |  |  |  |  | 
| 786 |  |  |  |  |  |  | } | 
| 787 |  |  |  |  |  |  | elsif ($CQL_parse_node->isa("CQL::TermNode")) | 
| 788 |  |  |  |  |  |  | { | 
| 789 |  |  |  |  |  |  | #	warn "TERM"; | 
| 790 |  |  |  |  |  |  |  | 
| 791 | 0 |  |  |  |  |  | my $qualifier=$CQL_parse_node->getQualifier(); | 
| 792 |  |  |  |  |  |  |  | 
| 793 |  |  |  |  |  |  | # Our partial hack solution: if it contains a space, leave as is. | 
| 794 |  |  |  |  |  |  | # Wray's hack on hack - keep space stuff, and deal with it differently | 
| 795 |  |  |  |  |  |  |  | 
| 796 | 0 | 0 | 0 |  |  |  | if ($qualifier ne 'text' && $qualifier ne 'srw.ServerChoice') { | 
| 797 | 0 |  |  |  |  |  | return $CQL_parse_node->toCQL(); | 
| 798 |  |  |  |  |  |  | } else { | 
| 799 | 0 |  |  |  |  |  | return ""; | 
| 800 |  |  |  |  |  |  | } | 
| 801 |  |  |  |  |  |  | } | 
| 802 |  |  |  |  |  |  |  | 
| 803 | 0 |  |  |  |  |  | return ""; | 
| 804 |  |  |  |  |  |  | } | 
| 805 |  |  |  |  |  |  |  | 
| 806 |  |  |  |  |  |  | # | 
| 807 |  |  |  |  |  |  | # Converts our expansion data structure to a CQL "tail" | 
| 808 |  |  |  |  |  |  | # | 
| 809 |  |  |  |  |  |  | sub _data_struct2CQLtail | 
| 810 |  |  |  |  |  |  | { | 
| 811 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 812 | 0 |  |  |  |  |  | my $seq_list=shift; | 
| 813 |  |  |  |  |  |  |  | 
| 814 | 0 |  |  |  |  |  | my $query; | 
| 815 | 0 |  |  |  |  |  | my @seq_items=(); | 
| 816 | 0 |  |  |  |  |  | for my $seq (@$seq_list) | 
| 817 |  |  |  |  |  |  | { | 
| 818 | 0 |  |  |  |  |  | my $ds=$self->_expand_qword_list($seq); | 
| 819 |  |  |  |  |  |  |  | 
| 820 | 0 |  |  |  |  |  | my @items=(); | 
| 821 |  |  |  |  |  |  |  | 
| 822 | 0 |  |  |  |  |  | for (my $i=0;$i | 
| 823 |  |  |  |  |  |  | { | 
| 824 | 0 |  |  |  |  |  | my ($token,$POS,$lemma,$max_type,$match_can_form,$pathtype) | 
| 825 | 0 |  |  |  |  |  | = @{$ds->[$i]}; | 
| 826 |  |  |  |  |  |  |  | 
| 827 | 0 | 0 | 0 |  |  |  | if ( $POS eq 'INDEX' && $token =~ /^([a-z0-9\-\_\.]+)=(.*)/ ) { | 
|  |  | 0 | 0 |  |  |  |  | 
|  |  | 0 | 0 |  |  |  |  | 
| 828 | 0 |  |  |  |  |  | push(@items,"$1%3D%22$2%22"); | 
| 829 |  |  |  |  |  |  | } elsif (defined($max_type)) | 
| 830 |  |  |  |  |  |  | { | 
| 831 | 0 | 0 |  |  |  |  | if ($max_type eq 'term_dict') | 
|  |  | 0 |  |  |  |  |  | 
| 832 |  |  |  |  |  |  | { | 
| 833 | 0 |  |  |  |  |  | my $surface_form=$token; | 
| 834 | 0 |  |  |  |  |  | my $can_form=$match_can_form; | 
| 835 | 0 |  |  |  |  |  | my $onto=$pathtype; | 
| 836 | 0 |  |  |  |  |  | my $j; | 
| 837 | 0 |  |  |  |  |  | for ($j=$i+1;$j | 
| 838 |  |  |  |  |  |  | { | 
| 839 | 0 |  |  |  |  |  | my ($token,$POS,$lemma,$max_type,$match_can_form, | 
| 840 |  |  |  |  |  |  | $onto_path) | 
| 841 | 0 |  |  |  |  |  | =@{$ds->[$j]}; | 
| 842 | 0 | 0 | 0 |  |  |  | if (!defined($max_type) || $max_type ne 'term_dict' | 
|  |  |  | 0 |  |  |  |  | 
|  |  |  | 0 |  |  |  |  | 
| 843 |  |  |  |  |  |  | || !defined($match_can_form) || $can_form ne $match_can_form ) | 
| 844 |  |  |  |  |  |  | { | 
| 845 | 0 |  |  |  |  |  | last; | 
| 846 |  |  |  |  |  |  | } | 
| 847 | 0 |  |  |  |  |  | $surface_form.=" $token"; | 
| 848 | 0 |  |  |  |  |  | $onto=$onto_path; | 
| 849 |  |  |  |  |  |  | } | 
| 850 |  |  |  |  |  |  |  | 
| 851 | 0 | 0 | 0 |  |  |  | if ( defined($onto) && $onto ne "" ) | 
| 852 |  |  |  |  |  |  | { | 
| 853 | 0 |  |  |  |  |  | push(@items,"term%3D%22$onto$can_form%22"); # unclear | 
| 854 |  |  |  |  |  |  | } | 
| 855 |  |  |  |  |  |  | else | 
| 856 |  |  |  |  |  |  | { | 
| 857 | 0 |  |  |  |  |  | push(@items,"term%3D%22$can_form%22"); | 
| 858 |  |  |  |  |  |  | } | 
| 859 |  |  |  |  |  |  |  | 
| 860 | 0 |  |  |  |  |  | $i=$j-1; | 
| 861 |  |  |  |  |  |  | } | 
| 862 |  |  |  |  |  |  | elsif ($max_type eq 'NE_dict') | 
| 863 |  |  |  |  |  |  | { | 
| 864 | 0 |  |  |  |  |  | my $surface_form=$token; | 
| 865 | 0 |  |  |  |  |  | my $can_form=$match_can_form; | 
| 866 | 0 |  |  |  |  |  | my $type=$pathtype; | 
| 867 | 0 |  |  |  |  |  | my $j; | 
| 868 | 0 |  |  |  |  |  | for ($j=$i+1;$j | 
| 869 |  |  |  |  |  |  | { | 
| 870 | 0 |  |  |  |  |  | my ($token,$POS,$lemma,$max_type,$match_can_form,$NE_type) | 
| 871 | 0 |  |  |  |  |  | =@{$ds->[$j]}; | 
| 872 | 0 | 0 | 0 |  |  |  | if (!defined($max_type) || $max_type ne 'NE_dict' | 
|  |  |  | 0 |  |  |  |  | 
|  |  |  | 0 |  |  |  |  | 
| 873 |  |  |  |  |  |  | || !defined($match_can_form) || $can_form ne $match_can_form ) | 
| 874 |  |  |  |  |  |  | { | 
| 875 | 0 |  |  |  |  |  | last; | 
| 876 |  |  |  |  |  |  | } | 
| 877 | 0 |  |  |  |  |  | $surface_form.=" $token"; | 
| 878 | 0 |  |  |  |  |  | $type = $NE_type; | 
| 879 |  |  |  |  |  |  | } | 
| 880 |  |  |  |  |  |  |  | 
| 881 | 0 | 0 | 0 |  |  |  | if ( ! defined($type) || $type eq "" ) { | 
|  |  | 0 |  |  |  |  |  | 
| 882 | 0 |  |  |  |  |  | push(@items,"entity%3D%22$can_form%22"); | 
| 883 |  |  |  |  |  |  | } elsif ( $type !~ /\// ) { | 
| 884 | 0 |  |  |  |  |  | push(@items,"entity-$type%3D%22$can_form%22"); | 
| 885 |  |  |  |  |  |  | } else { | 
| 886 | 0 |  |  |  |  |  | push(@items,"entity%3D%22$type$can_form%22"); | 
| 887 |  |  |  |  |  |  | } | 
| 888 |  |  |  |  |  |  |  | 
| 889 |  |  |  |  |  |  |  | 
| 890 | 0 |  |  |  |  |  | $i=$j-1; | 
| 891 |  |  |  |  |  |  | } | 
| 892 |  |  |  |  |  |  | } | 
| 893 |  |  |  |  |  |  | elsif (defined($lemma) && $POS =~ /$self->{lemmaSearch}/o | 
| 894 |  |  |  |  |  |  | && $self->{keepLemmas}) | 
| 895 |  |  |  |  |  |  | { | 
| 896 | 0 |  |  |  |  |  | push(@items,"lemma%3D%22$lemma%22"); | 
| 897 |  |  |  |  |  |  | } | 
| 898 |  |  |  |  |  |  | else { | 
| 899 | 0 |  |  |  |  |  | push(@items,$self->_make_term($token)); | 
| 900 |  |  |  |  |  |  | } | 
| 901 |  |  |  |  |  |  | } | 
| 902 |  |  |  |  |  |  |  | 
| 903 | 0 |  |  |  |  |  | push(@seq_items,"%28" . join('%20and%20',@items)  . "%29"); | 
| 904 |  |  |  |  |  |  | } | 
| 905 | 0 | 0 |  |  |  |  | if ( scalar(@seq_items) <= 1 ) { | 
| 906 | 0 |  |  |  |  |  | $query = $seq_items[0]; | 
| 907 |  |  |  |  |  |  | } else { | 
| 908 | 0 |  |  |  |  |  | $query = "%28" . join('%20or%20',@seq_items)  . "%29"; | 
| 909 |  |  |  |  |  |  | } | 
| 910 |  |  |  |  |  |  |  | 
| 911 | 0 |  |  |  |  |  | return $query; | 
| 912 |  |  |  |  |  |  | } | 
| 913 |  |  |  |  |  |  |  | 
| 914 |  |  |  |  |  |  |  | 
| 915 |  |  |  |  |  |  | sub _make_term | 
| 916 |  |  |  |  |  |  | { | 
| 917 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 918 | 0 |  |  |  |  |  | my $term=shift; | 
| 919 | 0 | 0 | 0 |  |  |  | if ( $term !~/^\"/ || $term !~/\"$/ ) { | 
| 920 | 0 |  |  |  |  |  | $term="\"$term\""; | 
| 921 |  |  |  |  |  |  | } | 
| 922 | 0 | 0 |  |  |  |  | if ( $self->{textFields} =~/ / ) { | 
| 923 | 0 |  |  |  |  |  | my $result = ""; | 
| 924 | 0 |  |  |  |  |  | foreach my $f ( split(/ /,$self->{textFields}) )  { | 
| 925 | 0 |  |  |  |  |  | $result .= " or $f%3D$term"; | 
| 926 |  |  |  |  |  |  | } | 
| 927 | 0 |  |  |  |  |  | $result =~ s/^ or //; | 
| 928 | 0 |  |  |  |  |  | return "($result)"; | 
| 929 |  |  |  |  |  |  | } | 
| 930 | 0 |  |  |  |  |  | return $self->{textFields} . "%3D$term"; | 
| 931 |  |  |  |  |  |  | } | 
| 932 |  |  |  |  |  |  |  | 
| 933 |  |  |  |  |  |  | # | 
| 934 |  |  |  |  |  |  | # Converts our expansion data structure to XML that fits extraResponseData | 
| 935 |  |  |  |  |  |  | # | 
| 936 |  |  |  |  |  |  | sub _data_struct2XML | 
| 937 |  |  |  |  |  |  | { | 
| 938 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 939 | 0 |  |  |  |  |  | my $seq_list=shift; | 
| 940 |  |  |  |  |  |  |  | 
| 941 | 0 |  |  |  |  |  | my $XML = "\n " . $self->{queryForm} . "\n"; | 
| 942 |  |  |  |  |  |  |  | 
| 943 |  |  |  |  |  |  | #    Why was this here in the first place? | 
| 944 |  |  |  |  |  |  | #    $XML.="\n"; | 
| 945 |  |  |  |  |  |  |  | 
| 946 | 0 |  |  |  |  |  | my @seq_items=(); | 
| 947 | 0 |  |  |  |  |  | for my $seq (@$seq_list) | 
| 948 |  |  |  |  |  |  | { | 
| 949 | 0 |  |  |  |  |  | my $ds=$self->_expand_qword_list($seq); | 
| 950 |  |  |  |  |  |  |  | 
| 951 | 0 |  |  |  |  |  | $XML.=" | 
| 952 | 0 |  |  |  |  |  | $XML.="     form=\"" . join(' ',@$seq) . "\" >\n"; | 
| 953 |  |  |  |  |  |  |  | 
| 954 | 0 |  |  |  |  |  | for (my $i=0;$i | 
| 955 |  |  |  |  |  |  | { | 
| 956 | 0 |  |  |  |  |  | my ($token,$POS,$lemma,$max_type,$match_can_form,$pathtype) | 
| 957 | 0 |  |  |  |  |  | = @{$ds->[$i]}; | 
| 958 |  |  |  |  |  |  |  | 
| 959 | 0 | 0 | 0 |  |  |  | if (defined($max_type)) | 
|  |  | 0 | 0 |  |  |  |  | 
| 960 |  |  |  |  |  |  | { | 
| 961 | 0 | 0 |  |  |  |  | if ($max_type eq 'term_dict') | 
|  |  | 0 |  |  |  |  |  | 
| 962 |  |  |  |  |  |  | { | 
| 963 | 0 |  |  |  |  |  | my $surface_form=$token; | 
| 964 | 0 |  |  |  |  |  | my $can_form=$match_can_form; | 
| 965 | 0 |  |  |  |  |  | my $onto=$pathtype; | 
| 966 | 0 |  |  |  |  |  | my $j; | 
| 967 | 0 |  |  |  |  |  | for ($j=$i+1;$j | 
| 968 |  |  |  |  |  |  | { | 
| 969 | 0 |  |  |  |  |  | my ($token,$POS,$lemma,$max_type,$match_can_form, | 
| 970 |  |  |  |  |  |  | $onto_path) | 
| 971 | 0 |  |  |  |  |  | =@{$ds->[$j]}; | 
| 972 | 0 | 0 | 0 |  |  |  | if (!defined($max_type) || $max_type ne 'term_dict' | 
|  |  |  | 0 |  |  |  |  | 
| 973 |  |  |  |  |  |  | || $match_can_form ne $can_form) | 
| 974 |  |  |  |  |  |  | { | 
| 975 | 0 |  |  |  |  |  | last; | 
| 976 |  |  |  |  |  |  | } | 
| 977 | 0 |  |  |  |  |  | $surface_form.=" $token"; | 
| 978 | 0 |  |  |  |  |  | $onto=$onto_path; | 
| 979 |  |  |  |  |  |  | } | 
| 980 | 0 |  |  |  |  |  | $XML.="\n"; | 
| 981 | 0 |  |  |  |  |  | $XML.="  \n"; | 
| 982 | 0 |  |  |  |  |  | $XML.="  $can_form\n"; | 
| 983 | 0 | 0 | 0 |  |  |  | if ( defined($onto) && $onto ne "" ) | 
| 984 |  |  |  |  |  |  | { | 
| 985 | 0 |  |  |  |  |  | $onto =~ s/\/$//; | 
| 986 | 0 |  |  |  |  |  | $XML .= "  $onto\n"; | 
| 987 |  |  |  |  |  |  | } | 
| 988 | 0 |  |  |  |  |  | $XML.="\n"; | 
| 989 | 0 |  |  |  |  |  | $i=$j-1; | 
| 990 |  |  |  |  |  |  | } | 
| 991 |  |  |  |  |  |  | elsif ($max_type eq 'NE_dict') | 
| 992 |  |  |  |  |  |  | { | 
| 993 | 0 |  |  |  |  |  | my $surface_form=$token; | 
| 994 | 0 |  |  |  |  |  | my $can_form=$match_can_form; | 
| 995 | 0 |  |  |  |  |  | my $type=$pathtype; | 
| 996 | 0 |  |  |  |  |  | my $j; | 
| 997 | 0 |  |  |  |  |  | for ($j=$i+1;$j | 
| 998 |  |  |  |  |  |  | { | 
| 999 | 0 |  |  |  |  |  | my ($token,$POS,$lemma,$max_type,$match_can_form,$NE_type) | 
| 1000 | 0 |  |  |  |  |  | =@{$ds->[$j]}; | 
| 1001 | 0 | 0 | 0 |  |  |  | if ($max_type ne 'NE_dict' || $match_can_form ne $can_form) | 
| 1002 |  |  |  |  |  |  | { | 
| 1003 | 0 |  |  |  |  |  | last; | 
| 1004 |  |  |  |  |  |  | } | 
| 1005 | 0 |  |  |  |  |  | $surface_form.=" $token"; | 
| 1006 | 0 |  |  |  |  |  | $type=$NE_type; | 
| 1007 |  |  |  |  |  |  | } | 
| 1008 | 0 |  |  |  |  |  | $XML.="\n"; | 
| 1009 | 0 |  |  |  |  |  | $XML.="  \n"; | 
| 1010 | 0 |  |  |  |  |  | $XML.="  $can_form\n"; | 
| 1011 | 0 | 0 | 0 |  |  |  | if ( defined($type) && $type ne "" ) { | 
| 1012 | 0 | 0 |  |  |  |  | if ( $type !~ /\// ) { | 
| 1013 | 0 |  |  |  |  |  | $XML.="  $type\n"; | 
| 1014 |  |  |  |  |  |  | } else { | 
| 1015 | 0 |  |  |  |  |  | $XML.="  $type\n"; | 
| 1016 |  |  |  |  |  |  | } | 
| 1017 |  |  |  |  |  |  | } | 
| 1018 | 0 |  |  |  |  |  | $XML.="\n"; | 
| 1019 | 0 |  |  |  |  |  | $i=$j-1; | 
| 1020 |  |  |  |  |  |  | } | 
| 1021 |  |  |  |  |  |  | } | 
| 1022 |  |  |  |  |  |  | elsif (defined($lemma)  && $POS =~ /$self->{lemmaSearch}/o | 
| 1023 |  |  |  |  |  |  | && $self->{keepLemmas}) | 
| 1024 |  |  |  |  |  |  | { | 
| 1025 | 0 |  |  |  |  |  | $XML.="\n"; | 
| 1026 | 0 |  |  |  |  |  | $XML.="  \n"; | 
| 1027 | 0 |  |  |  |  |  | $XML.="  $lemma\n"; | 
| 1028 | 0 |  |  |  |  |  | $POS=$self->_map_POS($POS); | 
| 1029 | 0 |  |  |  |  |  | $XML.="  $POS\n"; | 
| 1030 | 0 |  |  |  |  |  | $XML.="\n"; | 
| 1031 |  |  |  |  |  |  | } | 
| 1032 |  |  |  |  |  |  | else # no analysis | 
| 1033 |  |  |  |  |  |  | { | 
| 1034 | 0 |  |  |  |  |  | $XML.="$token\n"; | 
| 1035 |  |  |  |  |  |  | } | 
| 1036 |  |  |  |  |  |  | } | 
| 1037 | 0 |  |  |  |  |  | $XML.="\n"; | 
| 1038 |  |  |  |  |  |  | } | 
| 1039 | 0 |  |  |  |  |  | $XML .= " \n"; | 
| 1040 |  |  |  |  |  |  |  | 
| 1041 | 0 |  |  |  |  |  | return $XML; | 
| 1042 |  |  |  |  |  |  | } | 
| 1043 |  |  |  |  |  |  |  | 
| 1044 |  |  |  |  |  |  | sub _map_POS | 
| 1045 |  |  |  |  |  |  | { | 
| 1046 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 1047 | 0 |  |  |  |  |  | my $POS=shift; | 
| 1048 |  |  |  |  |  |  |  | 
| 1049 | 0 | 0 |  |  |  |  | if ($POS=~/^V.*/) | 
|  |  | 0 |  |  |  |  |  | 
| 1050 |  |  |  |  |  |  | { | 
| 1051 | 0 |  |  |  |  |  | $POS='V'; | 
| 1052 |  |  |  |  |  |  | } | 
| 1053 |  |  |  |  |  |  | elsif ($POS=~/^N.*/) | 
| 1054 |  |  |  |  |  |  | { | 
| 1055 | 0 |  |  |  |  |  | $POS='N'; | 
| 1056 |  |  |  |  |  |  | } | 
| 1057 |  |  |  |  |  |  |  | 
| 1058 | 0 |  |  |  |  |  | return $POS; | 
| 1059 |  |  |  |  |  |  | } | 
| 1060 |  |  |  |  |  |  |  | 
| 1061 |  |  |  |  |  |  | sub _apply_nulltagger | 
| 1062 |  |  |  |  |  |  | { | 
| 1063 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 1064 | 0 |  |  |  |  |  | my $query=shift; | 
| 1065 | 0 |  |  |  |  |  | my @lemmatized=($query, 'SENT', undef); | 
| 1066 | 0 |  |  |  |  |  | return \@lemmatized; | 
| 1067 |  |  |  |  |  |  | } | 
| 1068 |  |  |  |  |  |  |  | 
| 1069 |  |  |  |  |  |  | sub _apply_treetagger | 
| 1070 |  |  |  |  |  |  | { | 
| 1071 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 1072 | 0 |  |  |  |  |  | my $query=shift; | 
| 1073 |  |  |  |  |  |  |  | 
| 1074 | 0 |  |  |  |  |  | my $stringified=join(" ",@$query); | 
| 1075 |  |  |  |  |  |  |  | 
| 1076 | 0 |  |  |  |  |  | my $tagged_txt=&Alvis::Treetagger::tag($stringified); | 
| 1077 | 0 |  |  |  |  |  | my @lemmatized=(); | 
| 1078 | 0 |  |  |  |  |  | for my $l (split(/\n/,$tagged_txt)) | 
| 1079 |  |  |  |  |  |  | { | 
| 1080 | 0 |  |  |  |  |  | my ($token,$POS,$lemma)=split(/\t/,$l); | 
| 1081 |  |  |  |  |  |  | #   restore CD lemmas to the form | 
| 1082 | 0 | 0 | 0 |  |  |  | if ( $POS eq "CD" ) { | 
|  |  | 0 |  |  |  |  |  | 
| 1083 | 0 |  |  |  |  |  | $lemma = $token; | 
| 1084 |  |  |  |  |  |  | } | 
| 1085 |  |  |  |  |  |  | #  Treetagger will make some proper nouns lower case, | 
| 1086 |  |  |  |  |  |  | #  so we assume in query text user *only* enters | 
| 1087 |  |  |  |  |  |  | #  upper case on purpose | 
| 1088 |  |  |  |  |  |  | elsif ( $token eq ucfirst($token) | 
| 1089 |  |  |  |  |  |  | && lc($token) eq $lemma ) { | 
| 1090 | 0 |  |  |  |  |  | $lemma = $token; | 
| 1091 |  |  |  |  |  |  | } | 
| 1092 | 0 |  |  |  |  |  | push(@lemmatized,[$token,$POS,$lemma]); | 
| 1093 |  |  |  |  |  |  | } | 
| 1094 |  |  |  |  |  |  |  | 
| 1095 | 0 |  |  |  |  |  | return \@lemmatized; | 
| 1096 |  |  |  |  |  |  | } | 
| 1097 |  |  |  |  |  |  |  | 
| 1098 |  |  |  |  |  |  | sub _apply_lemma_dict | 
| 1099 |  |  |  |  |  |  | { | 
| 1100 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 1101 | 0 |  |  |  |  |  | my $tagger_output=shift; | 
| 1102 |  |  |  |  |  |  |  | 
| 1103 | 0 |  |  |  |  |  | my @lemmatized=(); | 
| 1104 |  |  |  |  |  |  |  | 
| 1105 |  |  |  |  |  |  | # Could be simplified, but don't have the energy any more | 
| 1106 | 0 | 0 |  |  |  |  | if (defined($self->{lemma_dict})) | 
| 1107 |  |  |  |  |  |  | { | 
| 1108 |  |  |  |  |  |  | # not tested btw | 
| 1109 |  |  |  |  |  |  | # | 
| 1110 | 0 |  |  |  |  |  | for my $t (@$tagger_output) | 
| 1111 |  |  |  |  |  |  | { | 
| 1112 | 0 |  |  |  |  |  | my ($token,$POS,$lemma)=@$t; | 
| 1113 |  |  |  |  |  |  |  | 
| 1114 | 0 | 0 |  |  |  |  | if ( $token =~ /^([a-z0-9\-\_\.]+)=(.*)/i ) { | 
|  |  | 0 |  |  |  |  |  | 
|  |  | 0 |  |  |  |  |  | 
|  |  | 0 |  |  |  |  |  | 
| 1115 | 0 |  |  |  |  |  | $token =~ s/\#\#/ /g; | 
| 1116 | 0 |  |  |  |  |  | push(@lemmatized,[$token,'INDEX',undef]); | 
| 1117 |  |  |  |  |  |  | } elsif ( $token =~ /\#\#/ ) { | 
| 1118 |  |  |  |  |  |  | #  Wray's hack on hack | 
| 1119 | 0 |  |  |  |  |  | $token =~ s/\#\#/ /g; | 
| 1120 | 0 |  |  |  |  |  | $token = "\"$token\""; | 
| 1121 | 0 |  |  |  |  |  | push(@lemmatized,[$token,'TEXT',undef]); | 
| 1122 |  |  |  |  |  |  | } elsif ( $token =~ /_/ ) { | 
| 1123 | 0 |  |  |  |  |  | $token =~ s/_+/ /g; | 
| 1124 | 0 |  |  |  |  |  | $token =~ s/ $//; | 
| 1125 | 0 |  |  |  |  |  | $token =~ s/^ //; | 
| 1126 | 0 |  |  |  |  |  | push(@lemmatized,[$token,'TAG',undef]); | 
| 1127 |  |  |  |  |  |  | } elsif ($lemma eq "") # try to fix | 
| 1128 |  |  |  |  |  |  | { | 
| 1129 | 0 | 0 |  |  |  |  | if (defined($self->{lemma_dict}{$token}{lemma})) | 
| 1130 |  |  |  |  |  |  | { | 
| 1131 | 0 |  |  |  |  |  | my $actual_POS; | 
| 1132 | 0 | 0 |  |  |  |  | if (defined($self->{lemma_dict}{$token}{POS})) | 
| 1133 |  |  |  |  |  |  | { | 
| 1134 | 0 |  |  |  |  |  | $actual_POS=$self->{lemma_dict}{$token}{POS}; | 
| 1135 |  |  |  |  |  |  | } | 
| 1136 |  |  |  |  |  |  |  | 
| 1137 | 0 |  |  |  |  |  | push(@lemmatized,[$token,$actual_POS, | 
| 1138 |  |  |  |  |  |  | $self->{lemma_dict}{$token}{lemma}]); | 
| 1139 |  |  |  |  |  |  | } | 
| 1140 |  |  |  |  |  |  | else # mark as lacking | 
| 1141 |  |  |  |  |  |  | { | 
| 1142 | 0 |  |  |  |  |  | push(@lemmatized,[$token,$POS,undef]); | 
| 1143 |  |  |  |  |  |  | } | 
| 1144 |  |  |  |  |  |  | } | 
| 1145 |  |  |  |  |  |  | else | 
| 1146 |  |  |  |  |  |  | { | 
| 1147 | 0 |  |  |  |  |  | push(@lemmatized,[$token,$POS,$lemma]); | 
| 1148 |  |  |  |  |  |  | } | 
| 1149 |  |  |  |  |  |  | } | 
| 1150 |  |  |  |  |  |  | } | 
| 1151 |  |  |  |  |  |  | else # no lemma dictionary so mark as lacking | 
| 1152 |  |  |  |  |  |  | { | 
| 1153 | 0 |  |  |  |  |  | for my $t (@$tagger_output) | 
| 1154 |  |  |  |  |  |  | { | 
| 1155 | 0 |  |  |  |  |  | my ($token,$POS,$lemma)=@$t; | 
| 1156 |  |  |  |  |  |  |  | 
| 1157 | 0 | 0 |  |  |  |  | if ( $token =~ /^([a-z0-9\-\_\.]+)=(.*)/i ) { | 
|  |  | 0 |  |  |  |  |  | 
|  |  | 0 |  |  |  |  |  | 
|  |  | 0 |  |  |  |  |  | 
| 1158 | 0 |  |  |  |  |  | $token =~ s/\#\#/ /g; | 
| 1159 | 0 |  |  |  |  |  | push(@lemmatized,[$token,'INDEX',undef]); | 
| 1160 |  |  |  |  |  |  | } elsif ( $token =~ /\#\#/ ) { | 
| 1161 |  |  |  |  |  |  | #  Wray's hack on hack | 
| 1162 | 0 |  |  |  |  |  | $token =~ s/\#\#/ /g; | 
| 1163 | 0 |  |  |  |  |  | $token = "\"$token\""; | 
| 1164 | 0 |  |  |  |  |  | push(@lemmatized,[$token,'TEXT',undef]); | 
| 1165 |  |  |  |  |  |  | } elsif ( $token =~ /_/ ) { | 
| 1166 | 0 |  |  |  |  |  | $token =~ s/_+/ /g; | 
| 1167 | 0 |  |  |  |  |  | $token =~ s/ $//; | 
| 1168 | 0 |  |  |  |  |  | $token =~ s/^ //; | 
| 1169 | 0 |  |  |  |  |  | push(@lemmatized,[$token,'TAG',undef]); | 
| 1170 |  |  |  |  |  |  | } elsif ($lemma eq "") { | 
| 1171 | 0 |  |  |  |  |  | push(@lemmatized,[$token,$POS,undef]); | 
| 1172 |  |  |  |  |  |  | } | 
| 1173 |  |  |  |  |  |  | else | 
| 1174 |  |  |  |  |  |  | { | 
| 1175 | 0 |  |  |  |  |  | push(@lemmatized,[$token,$POS,$lemma]); | 
| 1176 |  |  |  |  |  |  | } | 
| 1177 |  |  |  |  |  |  | } | 
| 1178 |  |  |  |  |  |  | } | 
| 1179 |  |  |  |  |  |  |  | 
| 1180 | 0 |  |  |  |  |  | return \@lemmatized; | 
| 1181 |  |  |  |  |  |  | } | 
| 1182 |  |  |  |  |  |  |  | 
| 1183 |  |  |  |  |  |  | sub _apply_terms_and_NEs | 
| 1184 |  |  |  |  |  |  | { | 
| 1185 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 1186 | 0 |  |  |  |  |  | my $lemmatized=shift; | 
| 1187 |  |  |  |  |  |  |  | 
| 1188 | 0 |  |  |  |  |  | my @exp=(); | 
| 1189 |  |  |  |  |  |  |  | 
| 1190 | 0 | 0 | 0 |  |  |  | if (!defined($self->{term_dict}) || !defined($self->{NE_dict})) | 
| 1191 |  |  |  |  |  |  | { | 
| 1192 | 0 |  |  |  |  |  | $self->_set_err_state($ERR_NEED_BOTH_TERM_AND_NE); | 
| 1193 | 0 |  |  |  |  |  | return undef; | 
| 1194 |  |  |  |  |  |  | } | 
| 1195 |  |  |  |  |  |  | else | 
| 1196 |  |  |  |  |  |  | { | 
| 1197 | 0 |  |  |  |  |  | for (my $start=0; $start | 
| 1198 |  |  |  |  |  |  | { | 
| 1199 | 0 |  |  |  |  |  | my ($token,$POS,$lemma)=@{$lemmatized->[$start]}; | 
|  | 0 |  |  |  |  |  |  | 
| 1200 |  |  |  |  |  |  |  | 
| 1201 |  |  |  |  |  |  | #  dont check the special cases | 
| 1202 | 0 | 0 | 0 |  |  |  | if ( $POS eq 'INDEX' || $POS eq 'TEXT' ) { | 
| 1203 | 0 |  |  |  |  |  | $start++; | 
| 1204 | 0 |  |  |  |  |  | next; | 
| 1205 |  |  |  |  |  |  | } | 
| 1206 |  |  |  |  |  |  |  | 
| 1207 |  |  |  |  |  |  | # find the longest match | 
| 1208 | 0 |  |  |  |  |  | my $max=0; | 
| 1209 | 0 |  |  |  |  |  | my $max_type; | 
| 1210 |  |  |  |  |  |  | my $max_form; | 
| 1211 |  |  |  |  |  |  |  | 
| 1212 |  |  |  |  |  |  | #  longest possible match | 
| 1213 | 0 |  |  |  |  |  | my $maxmax = scalar(@$lemmatized)-$start; | 
| 1214 | 0 |  |  |  |  |  | my $term = ""; | 
| 1215 |  |  |  |  |  |  |  | 
| 1216 | 0 | 0 |  |  |  |  | if ( $maxmax > $self->{termMaxLen} ) { | 
| 1217 | 0 |  |  |  |  |  | $maxmax = $self->{termMaxLen}; | 
| 1218 |  |  |  |  |  |  | } | 
| 1219 |  |  |  |  |  |  | #  tags get checked without any following sequence | 
| 1220 | 0 | 0 |  |  |  |  | if ( $POS eq 'TAG' ) { | 
| 1221 | 0 |  |  |  |  |  | $maxmax = 1; | 
| 1222 |  |  |  |  |  |  | } | 
| 1223 | 0 |  |  |  |  |  | for ( my $i=0; $i<$maxmax; $i++) { | 
| 1224 | 0 | 0 |  |  |  |  | if ( $i>0 ) { | 
| 1225 | 0 |  |  |  |  |  | my $tpos = $lemmatized->[$start+$i]->[1]; | 
| 1226 |  |  |  |  |  |  | #  these types mark a hard boundary | 
| 1227 | 0 | 0 | 0 |  |  |  | if ( $tpos eq 'INDEX' || $tpos eq 'TAG' || $tpos eq 'TEXT' ) { | 
|  |  |  | 0 |  |  |  |  | 
| 1228 | 0 |  |  |  |  |  | last; | 
| 1229 |  |  |  |  |  |  | } | 
| 1230 |  |  |  |  |  |  | } | 
| 1231 | 0 |  |  |  |  |  | my $lemma; | 
| 1232 | 0 | 0 |  |  |  |  | if (!defined($lemmatized->[$start+$i]->[2])) | 
| 1233 |  |  |  |  |  |  | { | 
| 1234 | 0 |  |  |  |  |  | $lemma=$lemmatized->[$start+$i]->[0]; | 
| 1235 |  |  |  |  |  |  | } | 
| 1236 |  |  |  |  |  |  | else | 
| 1237 |  |  |  |  |  |  | { | 
| 1238 | 0 |  |  |  |  |  | $lemma=$lemmatized->[$start+$i]->[2]; | 
| 1239 |  |  |  |  |  |  | } | 
| 1240 | 0 |  |  |  |  |  | $term .= " " . $lemma; | 
| 1241 | 0 |  |  |  |  |  | my $max_match; | 
| 1242 | 0 | 0 |  |  |  |  | if ( defined($max_match= | 
|  | 0 | 0 |  |  |  |  |  | 
| 1243 | 0 |  |  |  |  |  | $self->{term_dict}->{&{$self->{tcanon}}($term)}) ) { | 
| 1244 | 0 |  |  |  |  |  | $max = $i+1; | 
| 1245 | 0 |  |  |  |  |  | $max_type = 'term_dict'; | 
| 1246 | 0 |  |  |  |  |  | $max_form = $max_match; | 
| 1247 |  |  |  |  |  |  | } elsif ( defined($max_match= | 
| 1248 |  |  |  |  |  |  | $self->{NE_dict}->{&{$self->{ncanon}}($term)}) ) { | 
| 1249 | 0 |  |  |  |  |  | $max = $i+1; | 
| 1250 | 0 |  |  |  |  |  | $max_type = 'NE_dict'; | 
| 1251 | 0 |  |  |  |  |  | $max_form = $max_match; | 
| 1252 |  |  |  |  |  |  | } | 
| 1253 |  |  |  |  |  |  | } | 
| 1254 |  |  |  |  |  |  |  | 
| 1255 | 0 | 0 |  |  |  |  | if ($max) | 
| 1256 |  |  |  |  |  |  | { | 
| 1257 | 0 |  |  |  |  |  | for (my $i=$start; $i<$start+$max;$i++) | 
| 1258 |  |  |  |  |  |  | { | 
| 1259 | 0 |  |  |  |  |  | my ($token,$POS,$lemma)=@{$lemmatized->[$i]}; | 
|  | 0 |  |  |  |  |  |  | 
| 1260 | 0 |  |  |  |  |  | $lemmatized->[$i]=[$token,$POS,$lemma,$max_type,$max_form]; | 
| 1261 |  |  |  |  |  |  | } | 
| 1262 | 0 |  |  |  |  |  | $start+=$max; | 
| 1263 |  |  |  |  |  |  | } | 
| 1264 |  |  |  |  |  |  | else | 
| 1265 |  |  |  |  |  |  | { | 
| 1266 | 0 |  |  |  |  |  | $start++; | 
| 1267 |  |  |  |  |  |  | } | 
| 1268 |  |  |  |  |  |  | } | 
| 1269 |  |  |  |  |  |  |  | 
| 1270 |  |  |  |  |  |  |  | 
| 1271 |  |  |  |  |  |  | } | 
| 1272 |  |  |  |  |  |  |  | 
| 1273 | 0 |  |  |  |  |  | return $lemmatized; | 
| 1274 |  |  |  |  |  |  | } | 
| 1275 |  |  |  |  |  |  |  | 
| 1276 |  |  |  |  |  |  | sub _apply_typing_rules | 
| 1277 |  |  |  |  |  |  | { | 
| 1278 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 1279 | 0 |  |  |  |  |  | my $lemmatized=shift; | 
| 1280 |  |  |  |  |  |  |  | 
| 1281 | 0 |  |  |  |  |  | my @exp=(); | 
| 1282 |  |  |  |  |  |  |  | 
| 1283 | 0 | 0 |  |  |  |  | if (!defined($self->{typing_rules})) | 
| 1284 |  |  |  |  |  |  | { | 
| 1285 |  |  |  |  |  |  | # not an error | 
| 1286 | 0 |  |  |  |  |  | return $lemmatized; | 
| 1287 |  |  |  |  |  |  | } | 
| 1288 |  |  |  |  |  |  | else | 
| 1289 |  |  |  |  |  |  | { | 
| 1290 | 0 |  |  |  |  |  | for (my $i=0; $i | 
| 1291 |  |  |  |  |  |  | { | 
| 1292 | 0 |  |  |  |  |  | my ($token,$POS,$lemma,$max_type,$match_can_form)= | 
| 1293 | 0 |  |  |  |  |  | @{$lemmatized->[$i]}; | 
| 1294 |  |  |  |  |  |  | # print STDERR "_apply_typing_rules: ($token,$POS,$lemma,$max_type,$match_can_form)\n"; | 
| 1295 |  |  |  |  |  |  | # print STDERR "       maps -> " . $self->{typing_rules}{$match_can_form} . "\n"; | 
| 1296 | 0 | 0 | 0 |  |  |  | if (defined($max_type) && $max_type eq 'NE_dict' && | 
|  |  |  | 0 |  |  |  |  | 
|  |  |  | 0 |  |  |  |  | 
| 1297 |  |  |  |  |  |  | defined($match_can_form) && | 
| 1298 |  |  |  |  |  |  | defined($self->{typing_rules}{$match_can_form})) | 
| 1299 |  |  |  |  |  |  | { | 
| 1300 | 0 |  |  |  |  |  | $lemmatized->[$i]= | 
| 1301 |  |  |  |  |  |  | [$token,$POS,$lemma,$max_type,$match_can_form, | 
| 1302 |  |  |  |  |  |  | $self->{typing_rules}{$match_can_form}]; | 
| 1303 |  |  |  |  |  |  | } | 
| 1304 |  |  |  |  |  |  | } | 
| 1305 |  |  |  |  |  |  |  | 
| 1306 |  |  |  |  |  |  | } | 
| 1307 |  |  |  |  |  |  |  | 
| 1308 | 0 |  |  |  |  |  | return $lemmatized; | 
| 1309 |  |  |  |  |  |  | } | 
| 1310 |  |  |  |  |  |  |  | 
| 1311 |  |  |  |  |  |  | sub _apply_onto | 
| 1312 |  |  |  |  |  |  | { | 
| 1313 | 0 |  |  | 0 |  |  | my $self=shift; | 
| 1314 | 0 |  |  |  |  |  | my $lemmatized=shift; | 
| 1315 |  |  |  |  |  |  |  | 
| 1316 | 0 |  |  |  |  |  | my @exp=(); | 
| 1317 |  |  |  |  |  |  |  | 
| 1318 | 0 | 0 | 0 |  |  |  | if (!(defined($self->{onto_nodes}) && defined($self->{onto_paths}))) | 
| 1319 |  |  |  |  |  |  | { | 
| 1320 |  |  |  |  |  |  | # not an error, but need both or nothing | 
| 1321 | 0 |  |  |  |  |  | return $lemmatized; | 
| 1322 |  |  |  |  |  |  | } | 
| 1323 |  |  |  |  |  |  | else | 
| 1324 |  |  |  |  |  |  | { | 
| 1325 | 0 |  |  |  |  |  | for (my $i=0; $i | 
| 1326 |  |  |  |  |  |  | { | 
| 1327 | 0 |  |  |  |  |  | my ($token,$POS,$lemma,$max_type,$match_can_form)= | 
| 1328 | 0 |  |  |  |  |  | @{$lemmatized->[$i]}; | 
| 1329 | 0 | 0 | 0 |  |  |  | if ( defined($match_can_form) && | 
| 1330 |  |  |  |  |  |  | defined($self->{onto_nodes}{$match_can_form})) | 
| 1331 |  |  |  |  |  |  | { | 
| 1332 | 0 |  |  |  |  |  | my $node=$self->{onto_nodes}{$match_can_form}; | 
| 1333 | 0 | 0 |  |  |  |  | if ( defined($self->{onto_paths}{$node}) ) | 
| 1334 |  |  |  |  |  |  | { | 
| 1335 | 0 |  |  |  |  |  | my $path = $self->{onto_paths}{$node}; | 
| 1336 | 0 | 0 |  |  |  |  | if ( $path ne "" ) { | 
| 1337 | 0 |  |  |  |  |  | $lemmatized->[$i] = | 
| 1338 |  |  |  |  |  |  | [$token,$POS,$lemma,$max_type,$match_can_form,"$path/"]; | 
| 1339 |  |  |  |  |  |  | } | 
| 1340 |  |  |  |  |  |  | } | 
| 1341 |  |  |  |  |  |  | } | 
| 1342 |  |  |  |  |  |  | } | 
| 1343 |  |  |  |  |  |  |  | 
| 1344 |  |  |  |  |  |  |  | 
| 1345 |  |  |  |  |  |  | } | 
| 1346 |  |  |  |  |  |  |  | 
| 1347 | 0 |  |  |  |  |  | return $lemmatized; | 
| 1348 |  |  |  |  |  |  | } | 
| 1349 |  |  |  |  |  |  |  | 
| 1350 |  |  |  |  |  |  |  | 
| 1351 |  |  |  |  |  |  | 1; | 
| 1352 |  |  |  |  |  |  | __END__ |