| 1 | ## -*- Mode: CPerl -*- | ||
| 2 | ## File: DTA::CAB::Chain::DTA.pm | ||
| 3 | ## Author: Bryan Jurish <moocow@cpan.org> | ||
| 4 | ## Description: robust analysis: default chain | ||
| 5 | |||
| 6 | package DTA::CAB::Chain::DTA; | ||
| 7 | use DTA::CAB::Datum ':all'; | ||
| 8 | use DTA::CAB::Chain::Multi; | ||
| 9 | |||
| 10 | ##-- sub-analyzers | ||
| 11 | use DTA::CAB::Analyzer::ExLex; | ||
| 12 | use DTA::CAB::Analyzer::Cache::Static; | ||
| 13 | use DTA::CAB::Analyzer::TokPP; | ||
| 14 | use DTA::CAB::Analyzer::LTS; | ||
| 15 | use DTA::CAB::Analyzer::Morph; | ||
| 16 | use DTA::CAB::Analyzer::Morph::Latin; | ||
| 17 | use DTA::CAB::Analyzer::Morph::Extra::OrtLexHessen; ##-- auxilliary DB: Siedlungsplaetze_Hessen, from HLGL Marburg (optional) | ||
| 18 | use DTA::CAB::Analyzer::MorphSafe; | ||
| 19 | use DTA::CAB::Analyzer::Rewrite; | ||
| 20 | use DTA::CAB::Analyzer::Moot; | ||
| 21 | use DTA::CAB::Analyzer::Moot::Boltzmann; | ||
| 22 | use DTA::CAB::Analyzer::EqPho; | ||
| 23 | use DTA::CAB::Analyzer::EqPhoX; | ||
| 24 | use DTA::CAB::Analyzer::EqRW; | ||
| 25 | use DTA::CAB::Analyzer::RewriteSub; | ||
| 26 | use DTA::CAB::Analyzer::DmootSub; | ||
| 27 | use DTA::CAB::Analyzer::LangId::Simple; | ||
| 28 | use DTA::CAB::Analyzer::MootSub; | ||
| 29 | use DTA::CAB::Analyzer::EqLemma; | ||
| 30 | use DTA::CAB::Analyzer::DTAMapClass; | ||
| 31 | use DTA::CAB::Analyzer::SynCoPe::NER; | ||
| 32 | use DTA::CAB::Analyzer::GermaNet::Hypernyms; | ||
| 33 | use DTA::CAB::Analyzer::GermaNet::Hyponyms; | ||
| 34 | use DTA::CAB::Analyzer::GermaNet::Synonyms; | ||
| 35 | use DTA::CAB::Analyzer::DTAClean; | ||
| 36 | use DTA::CAB::Analyzer::Null; | ||
| 37 | |||
| 38 | use IO::File; | ||
| 39 | use Carp; | ||
| 40 | |||
| 41 | use strict; | ||
| 42 | |||
| 43 | ##============================================================================== | ||
| 44 | ## Constants | ||
| 45 | ##============================================================================== | ||
| 46 | our @ISA = qw(DTA::CAB::Chain::Multi); | ||
| 47 | |||
| 48 | ## supported date-optimized rewrite-ranges | ||
| 49 | our @RW_RANGES = qw(1600-1700 1700-1800 1800-1900); | ||
| 50 | |||
| 51 | ##============================================================================== | ||
| 52 | ## Constructors etc. | ||
| 53 | ##============================================================================== | ||
| 54 | |||
| 55 | ## $obj = CLASS_OR_OBJ->new(%args) | ||
| 56 | ## + object structure: HASH | ||
| 57 | sub new { | ||
| 58 | my $that = shift; | ||
| 59 | return $that->SUPER::new | ||
| 60 | ( | ||
| 61 | ##-- analyzers | ||
| 62 | static => DTA::CAB::Analyzer::Cache::Static->new(typeKeys=>[qw(eqphox errid exlex f lts mlatin morph msafe pnd rw xlit lang)]), | ||
| 63 | exlex => DTA::CAB::Analyzer::ExLex->new(), | ||
| 64 | tokpp => DTA::CAB::Analyzer::TokPP->new(), | ||
| 65 | xlit => DTA::CAB::Analyzer::Unicruft->new(), | ||
| 66 | ## | ||
| 67 | lts => DTA::CAB::Analyzer::LTS->new(), | ||
| 68 | ## | ||
| 69 | morph => DTA::CAB::Analyzer::Morph->new(), | ||
| 70 | mlatin=> DTA::CAB::Analyzer::Morph::Latin->new(), | ||
| 71 | mhessen => DTA::CAB::Analyzer::Morph::Extra::OrtLexHessen->new(), | ||
| 72 | mhessengeo => DTA::CAB::Analyzer::Morph::Extra::GeoLexHessen->new(), | ||
| 73 | msafe => DTA::CAB::Analyzer::MorphSafe->new(), | ||
| 74 | rw => DTA::CAB::Analyzer::Rewrite->new(), | ||
| 75 | (map {("rw.$_" => DTA::CAB::Analyzer::Rewrite->new())} @RW_RANGES), | ||
| 76 | rwsub => DTA::CAB::Analyzer::RewriteSub->new(), | ||
| 77 | ## | ||
| 78 | eqphox => DTA::CAB::Analyzer::EqPhoX->new(), | ||
| 79 | eqpho => DTA::CAB::Analyzer::EqPho->new(), | ||
| 80 | eqrw => DTA::CAB::Analyzer::EqRW->new(), | ||
| 81 | ## | ||
| 82 | ## | ||
| 83 | dmoot => DTA::CAB::Analyzer::Moot::Boltzmann->new(), ##-- moot n-gram disambiguator ((n>=1)-grams) | ||
| 84 | dmoot1 => DTA::CAB::Analyzer::Moot::Boltzmann->new(), ##-- moot n-gram disambiguator (1-grams only) | ||
| 85 | dmootsub => DTA::CAB::Analyzer::DmootSub->new(), ##-- moot n-gram disambiguator: sub-morph | ||
| 86 | moot => DTA::CAB::Analyzer::Moot->new(), ##-- moot tagger (on dmoot output; (n>1)-grams) | ||
| 87 | moot1 => DTA::CAB::Analyzer::Moot->new(), ##-- moot tagger (on dmoot output; 1-grams only) | ||
| 88 | mootsub => DTA::CAB::Analyzer::MootSub->new(), ##-- moot tagger, post-processing hacks | ||
| 89 | mapclass => DTA::CAB::Analyzer::DTAMapClass->new(), ##-- mapping class (post-moot) | ||
| 90 | ## | ||
| 91 | langid => DTA::CAB::Analyzer::LangId::Simple->new(), ##-- language-guesser (stopword-based; between msafe and rw) | ||
| 92 | ## | ||
| 93 | ner => DTA::CAB::Analyzer::SynCoPe::NER->new(), ##-- ne-recognizer (post-moot) | ||
| 94 | ## | ||
| 95 | eqlemma => DTA::CAB::Analyzer::EqLemma->new(), ##-- eqlemma (best only) | ||
| 96 | ## | ||
| 97 | 'gn-syn' => DTA::CAB::Analyzer::GermaNet::Synonyms->new(), ##-- GermaNet synonyms | ||
| 98 | 'gn-isa' => DTA::CAB::Analyzer::GermaNet::Hypernyms->new(), ##-- GermaNet hyperyms (superclasses) | ||
| 99 | 'gn-asi' => DTA::CAB::Analyzer::GermaNet::Hyponyms->new(), ##-- GermaNet hyponyms (subclasses) | ||
| 100 | ## | ||
| 101 | 'ot-syn' => DTA::CAB::Analyzer::GermaNet::Synonyms->new(label=>'ot-syn'), ##-- OpenThesaurus synonyms | ||
| 102 | 'ot-isa' => DTA::CAB::Analyzer::GermaNet::Hypernyms->new(label=>'ot-isa'), ##-- OpenThesaurus hypernyms (superclasses) | ||
| 103 | 'ot-asi' => DTA::CAB::Analyzer::GermaNet::Hyponyms->new(label=>'ot-asi'), ##-- OpenThesaurus hyponyms (subclasses) | ||
| 104 | ## | ||
| 105 | clean => DTA::CAB::Analyzer::DTAClean->new(), | ||
| 106 | ## | ||
| 107 | null => DTA::CAB::Analyzer::Null->new(), ##-- null analyzer (for 'null' chain) | ||
| 108 | |||
| 109 | ##-- security | ||
| 110 | autoClean => 0, ##-- always run 'clean' analyzer regardless of options; checked in both doAnalyze(), analyzeClean() | ||
| 111 | defaultChain => 'default', | ||
| 112 | |||
| 113 | ##-- user args | ||
| 114 | @_, | ||
| 115 | |||
| 116 | ##-- overrides | ||
| 117 | chains => undef, ##-- see setupChains() method | ||
| 118 | chain => undef, ##-- see setupChains() method | ||
| 119 | ); | ||
| 120 | } | ||
| 121 | |||
| 122 | ##============================================================================== | ||
| 123 | ## Methods: Chain selection | ||
| 124 | ##============================================================================== | ||
| 125 | |||
| 126 | ## $ach = $ach->setupChains() | ||
| 127 | ## + setup default named sub-chains in $ach->{chains} | ||
| 128 | ## + override | ||
| 129 | ## + dta ddc and django cab demo both call 'all' chain: | ||
| 130 | ## - ddc web search: (client = 194.95.188.36 = services.dwds.de) | ||
| 131 | ## 2010-11-19 13:15:42 [24194] (DEBUG) DTA.CAB.Server.XmlRpc.Procedure: dta.cab.all.analyzeToken(): client=194.95.188.36 | ||
| 132 | ## - django cab demo (client = 194.95.188.22 = www.deutschestextchiv.de) | ||
| 133 | ## 2010-11-19 13:15:09 [24194] (DEBUG) DTA.CAB.Server.XmlRpc.Procedure: dta.cab.all.analyzeToken(): client=194.95.188.22 | ||
| 134 | sub setupChains { | ||
| 135 | my $ach = shift; | ||
| 136 | $ach->{rwsub}{chain} = [@$ach{qw(lts morph)}]; | ||
| 137 | $ach->{dmootsub}{chain} = [@$ach{qw(morph mlatin)}]; #mhessen mhessengeo | ||
| 138 | my @akeys = grep {UNIVERSAL::isa($ach->{$_},'DTA::CAB::Analyzer')} keys(%$ach); | ||
| 139 | my $chains = $ach->{chains} = | ||
| 140 | { | ||
| 141 | (map {("sub.$_"=>[$ach->{$_}])} @akeys), ##-- sub.xlit, sub.lts, ... | ||
| 142 | #(map {("$_"=>[$ach->{$_}])} @akeys), ##-- xlit, lts, ... | ||
| 143 | ## | ||
| 144 | 'sub.expand' =>[@$ach{qw(eqpho eqrw eqlemma)}], | ||
| 145 | 'sub.sent' =>[@$ach{qw(dmoot dmootsub moot mootsub)}], | ||
| 146 | 'sub.sent1' =>[@$ach{qw(dmoot1 dmootsub moot1 mootsub)}], | ||
| 147 | 'sub.gn' =>[@$ach{qw(gn-syn gn-isa gn-asi)}], | ||
| 148 | 'sub.ot' =>[@$ach{qw(ot-syn ot-isa ot-asi)}], | ||
| 149 | ## | ||
| 150 | 'default.static' =>[@$ach{qw(static)}], | ||
| 151 | 'default.exlex' =>[@$ach{qw(exlex)}], | ||
| 152 | 'default.tokpp' =>[@$ach{qw(tokpp)}], | ||
| 153 | 'default.xlit' =>[@$ach{qw(xlit)}], | ||
| 154 | 'default.lts' =>[@$ach{qw(xlit lts)}], | ||
| 155 | 'default.eqphox' =>[@$ach{qw(tokpp xlit lts eqphox)}], | ||
| 156 | 'default.morph' =>[@$ach{qw(tokpp xlit morph)}], | ||
| 157 | 'default.mlatin' =>[@$ach{qw(tokpp xlit mlatin)}], | ||
| 158 | 'default.mhessen'=>[@$ach{qw(tokpp xlit mhessen)}], | ||
| 159 | 'default.mhessengeo'=>[@$ach{qw(tokpp xlit mhessengeo)}], | ||
| 160 | 'default.msafe' =>[@$ach{qw(tokpp xlit morph mlatin msafe)}], | ||
| 161 | 'default.langid' =>[@$ach{qw(tokpp xlit morph mlatin msafe langid)}], | ||
| 162 | 'default.rw' =>[@$ach{qw(tokpp xlit rw)}], | ||
| 163 | 'default.rw.safe'=>[@$ach{qw(tokpp xlit morph mlatin msafe langid rw)}], | ||
| 164 | 'default.dmoot' =>[@$ach{qw(tokpp xlit lts eqphox morph mlatin msafe langid rw dmoot)}], | ||
| 165 | 'default.dmoot1' =>[@$ach{qw(tokpp xlit lts eqphox morph mlatin msafe langid rw dmoot1)}], | ||
| 166 | 'default.moot' =>[@$ach{qw(tokpp xlit lts eqphox morph mlatin msafe langid rw dmoot dmootsub moot)}], | ||
| 167 | 'default.moot1' =>[@$ach{qw(tokpp xlit lts eqphox morph mlatin msafe langid rw dmoot1 dmootsub moot1)}], | ||
| 168 | 'default.lemma' =>[@$ach{qw(tokpp xlit lts eqphox morph mlatin msafe langid rw dmoot1 dmootsub moot mootsub)}], | ||
| 169 | 'default.lemma1' =>[@$ach{qw(tokpp xlit lts eqphox morph mlatin msafe langid rw dmoot1 dmootsub moot1 mootsub)}], | ||
| 170 | 'default.ner' =>[@$ach{qw(tokpp xlit lts eqphox morph mlatin msafe langid rw dmoot dmootsub moot mootsub ner)}], | ||
| 171 | 'default.base' =>[@$ach{qw(static exlex tokpp xlit lts morph mlatin msafe langid)}], | ||
| 172 | 'default.type' =>[@$ach{qw(static exlex tokpp xlit lts morph mlatin msafe langid rw rwsub)}], | ||
| 173 | ## | ||
| 174 | 'expand.old' =>[@$ach{qw(static exlex xlit lts morph mlatin msafe rw eqpho eqrw)}], | ||
| 175 | 'expand.ext' =>[@$ach{qw(static exlex xlit lts morph mlatin msafe rw eqpho eqrw eqphox)}], | ||
| 176 | 'expand.all' =>[@$ach{qw(static exlex xlit lts morph mlatin msafe rw eqpho eqrw eqphox dmoot1 dmootsub moot1 mootsub eqlemma)}], | ||
| 177 | 'expand.eqpho' =>[@$ach{qw(static exlex xlit lts eqpho)}], | ||
| 178 | 'expand.eqrw' =>[@$ach{qw(static exlex xlit lts morph mlatin msafe rw eqrw)}], | ||
| 179 | 'expand.eqlemma' =>[@$ach{qw(static exlex xlit lts morph mlatin msafe rw eqphox dmoot1 dmootsub moot1 mootsub eqlemma)}], | ||
| 180 | 'expand.gn-syn' =>[@$ach{qw(static exlex xlit lts morph mlatin msafe rw eqphox dmoot1 dmootsub moot1 mootsub gn-syn)}], | ||
| 181 | 'expand.gn-isa' =>[@$ach{qw(static exlex xlit lts morph mlatin msafe rw eqphox dmoot1 dmootsub moot1 mootsub gn-isa)}], | ||
| 182 | 'expand.gn-asi' =>[@$ach{qw(static exlex xlit lts morph mlatin msafe rw eqphox dmoot1 dmootsub moot1 mootsub gn-asi)}], | ||
| 183 | 'expand.gn' =>[@$ach{qw(static exlex xlit lts morph mlatin msafe rw eqphox dmoot1 dmootsub moot1 mootsub gn-syn gn-isa gn-asi)}], | ||
| 184 | 'expand.ot-syn' =>[@$ach{qw(static exlex xlit lts morph mlatin msafe rw eqphox dmoot1 dmootsub moot1 mootsub ot-syn)}], | ||
| 185 | 'expand.ot-isa' =>[@$ach{qw(static exlex xlit lts morph mlatin msafe rw eqphox dmoot1 dmootsub moot1 mootsub ot-isa)}], | ||
| 186 | 'expand.ot-asi' =>[@$ach{qw(static exlex xlit lts morph mlatin msafe rw eqphox dmoot1 dmootsub moot1 mootsub ot-asi)}], | ||
| 187 | 'expand.ot' =>[@$ach{qw(static exlex xlit lts morph mlatin msafe rw eqphox dmoot1 dmootsub moot1 mootsub ot-syn ot-isa ot-asi)}], | ||
| 188 | ## | ||
| 189 | 'norm' =>[@$ach{qw(static exlex tokpp xlit lts morph mlatin msafe langid rw eqphox dmoot dmootsub moot mootsub)}], | ||
| 190 | 'norm1' =>[@$ach{qw(static exlex tokpp xlit lts morph mlatin msafe langid rw eqphox dmoot1 dmootsub moot1 mootsub)}], | ||
| 191 | 'ner' =>[@$ach{qw(static exlex tokpp xlit lts morph mlatin msafe langid rw eqphox dmoot dmootsub moot mootsub ner)}], | ||
| 192 | 'caberr' =>[@$ach{qw(static exlex tokpp xlit lts morph mlatin msafe langid rw eqphox dmoot dmootsub moot mootsub mapclass)}], | ||
| 193 | 'caberr1' =>[@$ach{qw(static exlex tokpp xlit lts morph mlatin msafe langid rw eqphox dmoot1 dmootsub moot1 mootsub mapclass)}], | ||
| 194 | 'all' =>[@$ach{qw(static exlex tokpp xlit lts morph mlatin msafe langid rw rwsub eqpho eqrw eqphox dmoot dmootsub moot mootsub eqlemma)}], ##-- old dta clients use 'all'! | ||
| 195 | 'clean' =>[@$ach{qw(clean)}], | ||
| 196 | 'null' =>[$ach->{null}], | ||
| 197 | }; | ||
| 198 | #$chains->{'default'} = [map {@{$chains->{$_}}} qw(default.type sub.sent)]; | ||
| 199 | |||
| 200 | ##-- chain aliases | ||
| 201 | $chains->{'default'} = $chains->{lemma} = $chains->{'norm'}; | ||
| 202 | $chains->{'default1'} = $chains->{lemma1} = $chains->{'norm1'}; | ||
| 203 | $chains->{'expand'} = $chains->{'expand.all'}; | ||
| 204 | |||
| 205 | ##-- BEGIN TEMPORARY custom chain(s): "hlgl": place-name recognition Hessen | ||
| 206 | $chains->{"norm.hlgl"} = [map {($_,$_ eq $ach->{mlatin} ? $ach->{mhessen} : qw())} @{$chains->{norm}}]; | ||
| 207 | $chains->{"norm1.hlgl"} = [map {($_,$_ eq $ach->{mlatin} ? $ach->{mhessen} : qw())} @{$chains->{norm1}}]; | ||
| 208 | ## | ||
| 209 | $chains->{"norm.hlgl.geo"} = [map {($_,$_ eq $ach->{mlatin} ? $ach->{mhessengeo} : qw())} @{$chains->{norm}}]; | ||
| 210 | $chains->{"norm1.hlgl.geo"} = [map {($_,$_ eq $ach->{mlatin} ? $ach->{mhessengeo} : qw())} @{$chains->{norm1}}]; | ||
| 211 | ##-- END TEMPORARY custom chain(s) | ||
| 212 | |||
| 213 | ##-- date-dependent chains | ||
| 214 | foreach my $rng (@RW_RANGES) { | ||
| 215 | if ($ach->{"rw.$rng"} && ($ach->{"rw.$rng"}{enabled}//1)) { | ||
| 216 | foreach my $key (qw(norm norm1 lemma lemma1 default default1 expand)) { | ||
| 217 | $chains->{"$key.$rng"} = [map {$_ eq $ach->{rw} ? $ach->{"rw.$rng"} : $_} @{$chains->{$key}}]; | ||
| 218 | } | ||
| 219 | } else { | ||
| 220 | $ach->warn("optimized rewrite cascade rw.$rng not available: disabling derived chains for range $rng"); | ||
| 221 | delete $ach->{"rw.$rng"}; | ||
| 222 | delete $chains->{"sub.rw.$rng"}; | ||
| 223 | } | ||
| 224 | } | ||
| 225 | |||
| 226 | ##-- sanitize chains | ||
| 227 | foreach (values %{$ach->{chains}}) { | ||
| 228 | @$_ = grep {ref($_)} @$_; | ||
| 229 | } | ||
| 230 | |||
| 231 | ##-- set default chain | ||
| 232 | $ach->{chain} = $ach->{chains}{$ach->{defaultChain}}; | ||
| 233 | |||
| 234 | ##-- force default labels | ||
| 235 | foreach (grep {UNIVERSAL::isa($ach->{$_},'DTA::CAB::Analyzer')} keys(%$ach)) { | ||
| 236 | next if ($_ =~ /^(?:langid|rw\.[0-9\-]+|mhessen(?:geo)?)$/); ##-- keep labels for these analyzers | ||
| 237 | ($ach->{$_}{label} = $_) =~ s/1$//; ##-- truncate '1' suffix for label (e.g. dmoot1, moot1) | ||
| 238 | } | ||
| 239 | return $ach; | ||
| 240 | } | ||
| 241 | |||
| 242 | ## \@analyzers = $ach->chain() | ||
| 243 | ## \@analyzers = $ach->chain(\%opts) | ||
| 244 | ## + get selected analyzer chain | ||
| 245 | ## + inherited from DTA::CAB::Chain::Multi | ||
| 246 | ## - calls setupChains() if $ach->{chain} is empty | ||
| 247 | ## - checks for $opts{chain} and returns $ach->{chains}{ $opts{chain} } if available | ||
| 248 | |||
| 249 | ## $ach = $ach->ensureChain() | ||
| 250 | ## + checks for $ach->{chain}, calls $ach->setupChains() if needed | ||
| 251 | ## + inherited from DTA::CAB::Chain::Multi | ||
| 252 | |||
| 253 | ##============================================================================== | ||
| 254 | ## Methods: I/O | ||
| 255 | ##============================================================================== | ||
| 256 | |||
| 257 | ## $bool = $ach->ensureLoaded() | ||
| 258 | ## + ensures analysis data is loaded from default files | ||
| 259 | ## + inherited DTA::CAB::Chain::Multi override calls ensureChain() before inherited method | ||
| 260 | sub ensureLoaded { | ||
| 261 | my $ach = shift; | ||
| 262 | $ach->SUPER::ensureLoaded(@_) || return 0; | ||
| 263 | |||
| 264 | ##-- hack: copy chain members AFTER loading for sub-analyzers, setting 'enabled' if appropriate | ||
| 265 | my ($subkey); | ||
| 266 | foreach $subkey (qw(rwsub dmootsub)) { | ||
| 267 | if (ref($ach->{$subkey})) { | ||
| 268 | foreach (grep {!$_->{"_${subkey}"}} @{$ach->{$subkey}{chain}}) { | ||
| 269 | $_ = bless( {%$_}, ref($_) ); | ||
| 270 | $_->{label} = $subkey.'_'.$_->{label}; | ||
| 271 | $_->{enabled} = $ach->{$subkey}{enabled}; | ||
| 272 | $_->{"_$subkey"} = 1; | ||
| 273 | } | ||
| 274 | } | ||
| 275 | } | ||
| 276 | |||
| 277 | return 1; | ||
| 278 | } | ||
| 279 | |||
| 280 | ##============================================================================== | ||
| 281 | ## Methods: Persistence | ||
| 282 | ##============================================================================== | ||
| 283 | |||
| 284 | ##====================================================================== | ||
| 285 | ## Methods: Persistence: Perl | ||
| 286 | |||
| 287 | ## @keys = $class_or_obj->noSaveKeys() | ||
| 288 | ## + returns list of keys not to be saved | ||
| 289 | ## + default just greps for CODE-refs | ||
| 290 | ## + inherited from DTA::CAB::Chain::Multi: override appends {chain},{chains} | ||
| 291 | |||
| 292 | ## $saveRef = $obj->savePerlRef() | ||
| 293 | ## + return reference to be saved (top-level objects only) | ||
| 294 | ## + inherited from DTA::CAB::Persistent | ||
| 295 | |||
| 296 | ## $loadedObj = $CLASS_OR_OBJ->loadPerlRef($ref) | ||
| 297 | ## + default implementation just clobbers $CLASS_OR_OBJ with $ref and blesses | ||
| 298 | ## + inherited from DTA::CAB::Persistent | ||
| 299 | |||
| 300 | ##============================================================================== | ||
| 301 | ## Methods: Analysis | ||
| 302 | ##============================================================================== | ||
| 303 | |||
| 304 | ##------------------------------------------------------------------------ | ||
| 305 | ## Methods: Analysis: v1.x: Utils | ||
| 306 | |||
| 307 | ## $bool = $anl->doAnalyze(\%opts, $name) | ||
| 308 | ## + alias for $anl->can("analyze${name}") && (!exists($opts{"doAnalyze${name}"}) || $opts{"doAnalyze${name}"}) | ||
| 309 | ## + override checks $anl->{autoClean} flag | ||
| 310 | sub doAnalyze { | ||
| 311 | my ($anl,$opts,$name) = @_; | ||
| 312 | return 1 if ($anl->{autoClean} && $name eq 'Clean'); | ||
| 313 | return $anl->SUPER::doAnalyze($opts,$name); | ||
| 314 | } | ||
| 315 | |||
| 316 | |||
| 317 | ##------------------------------------------------------------------------ | ||
| 318 | ## Methods: Analysis: v1.x: API | ||
| 319 | |||
| 320 | ## $doc = $ach->analyzeDocument($doc,\%opts) | ||
| 321 | ## + analyze a DTA::CAB::Document $doc | ||
| 322 | ## + top-level API routine | ||
| 323 | ## + INHERITED from DTA::CAB::Analyzer | ||
| 324 | |||
| 325 | ## $doc = $ach->analyzeTypes($doc,$types,\%opts) | ||
| 326 | ## + perform type-wise analysis of all (text) types in $doc->{types} | ||
| 327 | ## + Chain default calls $a->analyzeTypes for each analyzer $a in the chain | ||
| 328 | ## + INHERITED from DTA::CAB::Chain | ||
| 329 | |||
| 330 | ## $doc = $ach->analyzeTokens($doc,\%opts) | ||
| 331 | ## + perform token-wise analysis of all tokens $doc->{body}[$si]{tokens}[$wi] | ||
| 332 | ## + default implementation just shallow copies tokens in $doc->{types} | ||
| 333 | ## + INHERITED from DTA::CAB::Analyzer | ||
| 334 | |||
| 335 | ## $doc = $ach->analyzeSentences($doc,\%opts) | ||
| 336 | ## + perform sentence-wise analysis of all sentences $doc->{body}[$si] | ||
| 337 | ## + Chain default calls $a->analyzeSentences for each analyzer $a in the chain | ||
| 338 | ## + INHERITED from DTA::CAB::Chain | ||
| 339 | |||
| 340 | ## $doc = $ach->analyzeLocal($doc,\%opts) | ||
| 341 | ## + perform local document-level analysis of $doc | ||
| 342 | ## + Chain default calls $a->analyzeLocal for each analyzer $a in the chain | ||
| 343 | ## + INHERITED from DTA::CAB::Chain | ||
| 344 | |||
| 345 | ## $doc = $ach->analyzeClean($doc,\%opts) | ||
| 346 | ## + cleanup any temporary data associated with $doc | ||
| 347 | ## + Chain default calls $a->analyzeClean for each analyzer $a in the chain, | ||
| 348 | ## then superclass Analyzer->analyzeClean | ||
| 349 | sub analyzeClean { | ||
| 350 | my ($ach,$doc,$opts) = @_; | ||
| 351 | $ach->SUPER::analyzeClean($doc,$opts); ##-- inherited from DTA::CAB::Chain (chain-local cleanup) | ||
| 352 | $ach->analyzeClean_rm_undef($doc,$opts); ##-- remove keys with undef values from tokens | ||
| 353 | return $doc if (!$ach->{autoClean} && !exists($opts->{doAnalyzeClean})); ##-- don't "clean" by default | ||
| 354 | return $ach->{clean}->analyzeClean($doc,$opts); | ||
| 355 | } | ||
| 356 | |||
| 357 | |||
| 358 | ##------------------------------------------------------------------------ | ||
| 359 | ## Methods: Analysis: v1.x: Wrappers | ||
| 360 | |||
| 361 | ## $tok = $ach->analyzeToken($tok_or_string,\%opts) | ||
| 362 | ## + perform type- and token-analyses on $tok_or_string | ||
| 363 | ## + wrapper for $ach->analyzeDocument() | ||
| 364 | ## + INHERITED from DTA::CAB::Analyzer | ||
| 365 | |||
| 366 | ## $tok = $ach->analyzeSentence($sent_or_array,\%opts) | ||
| 367 | ## + perform type-, token-, and sentence-analyses on $sent_or_array | ||
| 368 | ## + wrapper for $ach->analyzeDocument() | ||
| 369 | ## + INHERITED from DTA::CAB::Analyzer | ||
| 370 | |||
| 371 | ## $rpc_xml_base64 = $anl->analyzeData($data_str,\%opts) | ||
| 372 | ## + analyze a raw (formatted) data string $data_str with internal parsing & formatting | ||
| 373 | ## + wrapper for $anl->analyzeDocument() | ||
| 374 | ## + INHERITED from DTA::CAB::Analyzer | ||
| 375 | |||
| 376 | ##============================================================================== | ||
| 377 | ## Methods: XML-RPC | ||
| 378 | ## + INHERITED from DTA::CAB::Chain::Multi | ||
| 379 | |||
| 380 | 1; ##-- be happy | ||
| 381 | |||
| 382 | __END__ |