grammar_police 0.1.0
Sign up to get free protection for your applications and to get access to all the features.
- data/.DS_Store +0 -0
- data/.gitignore +4 -0
- data/Gemfile +4 -0
- data/Rakefile +2 -0
- data/c/.DS_Store +0 -0
- data/c/link-grammar.c +65 -0
- data/c/link-grammar.h +60 -0
- data/c/link-grammar.o +0 -0
- data/c/link-grammar.so +0 -0
- data/c/link-grammar/.DS_Store +0 -0
- data/c/link-grammar/.deps/analyze-linkage.Plo +198 -0
- data/c/link-grammar/.deps/and.Plo +202 -0
- data/c/link-grammar/.deps/api.Plo +244 -0
- data/c/link-grammar/.deps/build-disjuncts.Plo +212 -0
- data/c/link-grammar/.deps/command-line.Plo +201 -0
- data/c/link-grammar/.deps/constituents.Plo +201 -0
- data/c/link-grammar/.deps/count.Plo +202 -0
- data/c/link-grammar/.deps/disjunct-utils.Plo +126 -0
- data/c/link-grammar/.deps/disjuncts.Plo +123 -0
- data/c/link-grammar/.deps/error.Plo +121 -0
- data/c/link-grammar/.deps/expand.Plo +133 -0
- data/c/link-grammar/.deps/extract-links.Plo +198 -0
- data/c/link-grammar/.deps/fast-match.Plo +200 -0
- data/c/link-grammar/.deps/idiom.Plo +200 -0
- data/c/link-grammar/.deps/jni-client.Plo +217 -0
- data/c/link-grammar/.deps/link-parser.Po +1 -0
- data/c/link-grammar/.deps/massage.Plo +202 -0
- data/c/link-grammar/.deps/post-process.Plo +202 -0
- data/c/link-grammar/.deps/pp_knowledge.Plo +202 -0
- data/c/link-grammar/.deps/pp_lexer.Plo +201 -0
- data/c/link-grammar/.deps/pp_linkset.Plo +200 -0
- data/c/link-grammar/.deps/prefix.Plo +102 -0
- data/c/link-grammar/.deps/preparation.Plo +202 -0
- data/c/link-grammar/.deps/print-util.Plo +200 -0
- data/c/link-grammar/.deps/print.Plo +201 -0
- data/c/link-grammar/.deps/prune.Plo +202 -0
- data/c/link-grammar/.deps/read-dict.Plo +223 -0
- data/c/link-grammar/.deps/read-regex.Plo +123 -0
- data/c/link-grammar/.deps/regex-morph.Plo +131 -0
- data/c/link-grammar/.deps/resources.Plo +203 -0
- data/c/link-grammar/.deps/spellcheck-aspell.Plo +1 -0
- data/c/link-grammar/.deps/spellcheck-hun.Plo +115 -0
- data/c/link-grammar/.deps/string-set.Plo +198 -0
- data/c/link-grammar/.deps/tokenize.Plo +160 -0
- data/c/link-grammar/.deps/utilities.Plo +222 -0
- data/c/link-grammar/.deps/word-file.Plo +201 -0
- data/c/link-grammar/.deps/word-utils.Plo +212 -0
- data/c/link-grammar/.libs/analyze-linkage.o +0 -0
- data/c/link-grammar/.libs/and.o +0 -0
- data/c/link-grammar/.libs/api.o +0 -0
- data/c/link-grammar/.libs/build-disjuncts.o +0 -0
- data/c/link-grammar/.libs/command-line.o +0 -0
- data/c/link-grammar/.libs/constituents.o +0 -0
- data/c/link-grammar/.libs/count.o +0 -0
- data/c/link-grammar/.libs/disjunct-utils.o +0 -0
- data/c/link-grammar/.libs/disjuncts.o +0 -0
- data/c/link-grammar/.libs/error.o +0 -0
- data/c/link-grammar/.libs/expand.o +0 -0
- data/c/link-grammar/.libs/extract-links.o +0 -0
- data/c/link-grammar/.libs/fast-match.o +0 -0
- data/c/link-grammar/.libs/idiom.o +0 -0
- data/c/link-grammar/.libs/jni-client.o +0 -0
- data/c/link-grammar/.libs/liblink-grammar-java-symbols.expsym +31 -0
- data/c/link-grammar/.libs/liblink-grammar-java.4.dylib +0 -0
- data/c/link-grammar/.libs/liblink-grammar-java.4.dylib.dSYM/Contents/Info.plist +20 -0
- data/c/link-grammar/.libs/liblink-grammar-java.4.dylib.dSYM/Contents/Resources/DWARF/liblink-grammar-java.4.dylib +0 -0
- data/c/link-grammar/.libs/liblink-grammar-java.a +0 -0
- data/c/link-grammar/.libs/liblink-grammar-java.dylib +0 -0
- data/c/link-grammar/.libs/liblink-grammar-symbols.expsym +194 -0
- data/c/link-grammar/.libs/liblink-grammar.4.dylib +0 -0
- data/c/link-grammar/.libs/liblink-grammar.4.dylib.dSYM/Contents/Info.plist +20 -0
- data/c/link-grammar/.libs/liblink-grammar.4.dylib.dSYM/Contents/Resources/DWARF/liblink-grammar.4.dylib +0 -0
- data/c/link-grammar/.libs/liblink-grammar.a +0 -0
- data/c/link-grammar/.libs/liblink-grammar.dylib +0 -0
- data/c/link-grammar/.libs/liblink-grammar.la +41 -0
- data/c/link-grammar/.libs/liblink-grammar.lai +41 -0
- data/c/link-grammar/.libs/massage.o +0 -0
- data/c/link-grammar/.libs/post-process.o +0 -0
- data/c/link-grammar/.libs/pp_knowledge.o +0 -0
- data/c/link-grammar/.libs/pp_lexer.o +0 -0
- data/c/link-grammar/.libs/pp_linkset.o +0 -0
- data/c/link-grammar/.libs/prefix.o +0 -0
- data/c/link-grammar/.libs/preparation.o +0 -0
- data/c/link-grammar/.libs/print-util.o +0 -0
- data/c/link-grammar/.libs/print.o +0 -0
- data/c/link-grammar/.libs/prune.o +0 -0
- data/c/link-grammar/.libs/read-dict.o +0 -0
- data/c/link-grammar/.libs/read-regex.o +0 -0
- data/c/link-grammar/.libs/regex-morph.o +0 -0
- data/c/link-grammar/.libs/resources.o +0 -0
- data/c/link-grammar/.libs/spellcheck-aspell.o +0 -0
- data/c/link-grammar/.libs/spellcheck-hun.o +0 -0
- data/c/link-grammar/.libs/string-set.o +0 -0
- data/c/link-grammar/.libs/tokenize.o +0 -0
- data/c/link-grammar/.libs/utilities.o +0 -0
- data/c/link-grammar/.libs/word-file.o +0 -0
- data/c/link-grammar/.libs/word-utils.o +0 -0
- data/c/link-grammar/Makefile +900 -0
- data/c/link-grammar/Makefile.am +202 -0
- data/c/link-grammar/Makefile.in +900 -0
- data/c/link-grammar/analyze-linkage.c +1317 -0
- data/c/link-grammar/analyze-linkage.h +24 -0
- data/c/link-grammar/and.c +1603 -0
- data/c/link-grammar/and.h +27 -0
- data/c/link-grammar/api-structures.h +362 -0
- data/c/link-grammar/api-types.h +72 -0
- data/c/link-grammar/api.c +1887 -0
- data/c/link-grammar/api.h +96 -0
- data/c/link-grammar/autoit/.DS_Store +0 -0
- data/c/link-grammar/autoit/README +10 -0
- data/c/link-grammar/autoit/_LGTest.au3 +22 -0
- data/c/link-grammar/autoit/_LinkGrammar.au3 +545 -0
- data/c/link-grammar/build-disjuncts.c +487 -0
- data/c/link-grammar/build-disjuncts.h +21 -0
- data/c/link-grammar/command-line.c +458 -0
- data/c/link-grammar/command-line.h +15 -0
- data/c/link-grammar/constituents.c +1836 -0
- data/c/link-grammar/constituents.h +26 -0
- data/c/link-grammar/corpus/.DS_Store +0 -0
- data/c/link-grammar/corpus/.deps/cluster.Plo +1 -0
- data/c/link-grammar/corpus/.deps/corpus.Plo +1 -0
- data/c/link-grammar/corpus/Makefile +527 -0
- data/c/link-grammar/corpus/Makefile.am +46 -0
- data/c/link-grammar/corpus/Makefile.in +527 -0
- data/c/link-grammar/corpus/README +17 -0
- data/c/link-grammar/corpus/cluster.c +286 -0
- data/c/link-grammar/corpus/cluster.h +32 -0
- data/c/link-grammar/corpus/corpus.c +483 -0
- data/c/link-grammar/corpus/corpus.h +46 -0
- data/c/link-grammar/count.c +828 -0
- data/c/link-grammar/count.h +25 -0
- data/c/link-grammar/disjunct-utils.c +261 -0
- data/c/link-grammar/disjunct-utils.h +27 -0
- data/c/link-grammar/disjuncts.c +138 -0
- data/c/link-grammar/disjuncts.h +13 -0
- data/c/link-grammar/error.c +92 -0
- data/c/link-grammar/error.h +35 -0
- data/c/link-grammar/expand.c +67 -0
- data/c/link-grammar/expand.h +13 -0
- data/c/link-grammar/externs.h +22 -0
- data/c/link-grammar/extract-links.c +625 -0
- data/c/link-grammar/extract-links.h +16 -0
- data/c/link-grammar/fast-match.c +309 -0
- data/c/link-grammar/fast-match.h +17 -0
- data/c/link-grammar/idiom.c +373 -0
- data/c/link-grammar/idiom.h +15 -0
- data/c/link-grammar/jni-client.c +779 -0
- data/c/link-grammar/jni-client.h +236 -0
- data/c/link-grammar/liblink-grammar-java.la +42 -0
- data/c/link-grammar/liblink-grammar.la +41 -0
- data/c/link-grammar/link-features.h +37 -0
- data/c/link-grammar/link-features.h.in +37 -0
- data/c/link-grammar/link-grammar-java.def +31 -0
- data/c/link-grammar/link-grammar.def +194 -0
- data/c/link-grammar/link-includes.h +465 -0
- data/c/link-grammar/link-parser.c +849 -0
- data/c/link-grammar/massage.c +329 -0
- data/c/link-grammar/massage.h +13 -0
- data/c/link-grammar/post-process.c +1113 -0
- data/c/link-grammar/post-process.h +45 -0
- data/c/link-grammar/pp_knowledge.c +376 -0
- data/c/link-grammar/pp_knowledge.h +14 -0
- data/c/link-grammar/pp_lexer.c +1920 -0
- data/c/link-grammar/pp_lexer.h +19 -0
- data/c/link-grammar/pp_linkset.c +158 -0
- data/c/link-grammar/pp_linkset.h +20 -0
- data/c/link-grammar/prefix.c +482 -0
- data/c/link-grammar/prefix.h +139 -0
- data/c/link-grammar/preparation.c +412 -0
- data/c/link-grammar/preparation.h +20 -0
- data/c/link-grammar/print-util.c +87 -0
- data/c/link-grammar/print-util.h +32 -0
- data/c/link-grammar/print.c +1085 -0
- data/c/link-grammar/print.h +16 -0
- data/c/link-grammar/prune.c +1864 -0
- data/c/link-grammar/prune.h +17 -0
- data/c/link-grammar/read-dict.c +1785 -0
- data/c/link-grammar/read-dict.h +29 -0
- data/c/link-grammar/read-regex.c +161 -0
- data/c/link-grammar/read-regex.h +12 -0
- data/c/link-grammar/regex-morph.c +126 -0
- data/c/link-grammar/regex-morph.h +17 -0
- data/c/link-grammar/resources.c +180 -0
- data/c/link-grammar/resources.h +23 -0
- data/c/link-grammar/sat-solver/.DS_Store +0 -0
- data/c/link-grammar/sat-solver/.deps/fast-sprintf.Plo +1 -0
- data/c/link-grammar/sat-solver/.deps/sat-encoder.Plo +1 -0
- data/c/link-grammar/sat-solver/.deps/util.Plo +1 -0
- data/c/link-grammar/sat-solver/.deps/variables.Plo +1 -0
- data/c/link-grammar/sat-solver/.deps/word-tag.Plo +1 -0
- data/c/link-grammar/sat-solver/Makefile +527 -0
- data/c/link-grammar/sat-solver/Makefile.am +29 -0
- data/c/link-grammar/sat-solver/Makefile.in +527 -0
- data/c/link-grammar/sat-solver/clock.hpp +33 -0
- data/c/link-grammar/sat-solver/fast-sprintf.cpp +26 -0
- data/c/link-grammar/sat-solver/fast-sprintf.hpp +7 -0
- data/c/link-grammar/sat-solver/guiding.hpp +244 -0
- data/c/link-grammar/sat-solver/matrix-ut.hpp +79 -0
- data/c/link-grammar/sat-solver/sat-encoder.cpp +2811 -0
- data/c/link-grammar/sat-solver/sat-encoder.h +11 -0
- data/c/link-grammar/sat-solver/sat-encoder.hpp +381 -0
- data/c/link-grammar/sat-solver/trie.hpp +118 -0
- data/c/link-grammar/sat-solver/util.cpp +23 -0
- data/c/link-grammar/sat-solver/util.hpp +14 -0
- data/c/link-grammar/sat-solver/variables.cpp +5 -0
- data/c/link-grammar/sat-solver/variables.hpp +829 -0
- data/c/link-grammar/sat-solver/word-tag.cpp +159 -0
- data/c/link-grammar/sat-solver/word-tag.hpp +162 -0
- data/c/link-grammar/spellcheck-aspell.c +148 -0
- data/c/link-grammar/spellcheck-hun.c +136 -0
- data/c/link-grammar/spellcheck.h +34 -0
- data/c/link-grammar/string-set.c +169 -0
- data/c/link-grammar/string-set.h +16 -0
- data/c/link-grammar/structures.h +498 -0
- data/c/link-grammar/tokenize.c +1049 -0
- data/c/link-grammar/tokenize.h +15 -0
- data/c/link-grammar/utilities.c +847 -0
- data/c/link-grammar/utilities.h +281 -0
- data/c/link-grammar/word-file.c +124 -0
- data/c/link-grammar/word-file.h +15 -0
- data/c/link-grammar/word-utils.c +526 -0
- data/c/link-grammar/word-utils.h +152 -0
- data/data/.DS_Store +0 -0
- data/data/Makefile +511 -0
- data/data/Makefile.am +4 -0
- data/data/Makefile.in +511 -0
- data/data/de/.DS_Store +0 -0
- data/data/de/4.0.affix +7 -0
- data/data/de/4.0.dict +474 -0
- data/data/de/Makefile +387 -0
- data/data/de/Makefile.am +9 -0
- data/data/de/Makefile.in +387 -0
- data/data/en/.DS_Store +0 -0
- data/data/en/4.0.affix +26 -0
- data/data/en/4.0.batch +1002 -0
- data/data/en/4.0.biolg.batch +411 -0
- data/data/en/4.0.constituent-knowledge +127 -0
- data/data/en/4.0.dict +8759 -0
- data/data/en/4.0.dict.m4 +6928 -0
- data/data/en/4.0.enwiki.batch +14 -0
- data/data/en/4.0.fixes.batch +2776 -0
- data/data/en/4.0.knowledge +306 -0
- data/data/en/4.0.regex +225 -0
- data/data/en/4.0.voa.batch +114 -0
- data/data/en/Makefile +554 -0
- data/data/en/Makefile.am +19 -0
- data/data/en/Makefile.in +554 -0
- data/data/en/README +173 -0
- data/data/en/tiny.dict +157 -0
- data/data/en/words/.DS_Store +0 -0
- data/data/en/words/Makefile +456 -0
- data/data/en/words/Makefile.am +78 -0
- data/data/en/words/Makefile.in +456 -0
- data/data/en/words/currency +205 -0
- data/data/en/words/currency.p +28 -0
- data/data/en/words/entities.given-bisex.sing +39 -0
- data/data/en/words/entities.given-female.sing +4141 -0
- data/data/en/words/entities.given-male.sing +1633 -0
- data/data/en/words/entities.locations.sing +68 -0
- data/data/en/words/entities.national.sing +253 -0
- data/data/en/words/entities.organizations.sing +7 -0
- data/data/en/words/entities.us-states.sing +11 -0
- data/data/en/words/units.1 +45 -0
- data/data/en/words/units.1.dot +4 -0
- data/data/en/words/units.3 +2 -0
- data/data/en/words/units.4 +5 -0
- data/data/en/words/units.4.dot +1 -0
- data/data/en/words/words-medical.adv.1 +1191 -0
- data/data/en/words/words-medical.prep.1 +67 -0
- data/data/en/words/words-medical.v.4.1 +2835 -0
- data/data/en/words/words-medical.v.4.2 +2848 -0
- data/data/en/words/words-medical.v.4.3 +3011 -0
- data/data/en/words/words-medical.v.4.4 +3036 -0
- data/data/en/words/words-medical.v.4.5 +3050 -0
- data/data/en/words/words.adj.1 +6794 -0
- data/data/en/words/words.adj.2 +638 -0
- data/data/en/words/words.adj.3 +667 -0
- data/data/en/words/words.adv.1 +1573 -0
- data/data/en/words/words.adv.2 +67 -0
- data/data/en/words/words.adv.3 +157 -0
- data/data/en/words/words.adv.4 +80 -0
- data/data/en/words/words.n.1 +11464 -0
- data/data/en/words/words.n.1.wiki +264 -0
- data/data/en/words/words.n.2.s +2017 -0
- data/data/en/words/words.n.2.s.biolg +1 -0
- data/data/en/words/words.n.2.s.wiki +298 -0
- data/data/en/words/words.n.2.x +65 -0
- data/data/en/words/words.n.2.x.wiki +10 -0
- data/data/en/words/words.n.3 +5717 -0
- data/data/en/words/words.n.t +23 -0
- data/data/en/words/words.v.1.1 +1038 -0
- data/data/en/words/words.v.1.2 +1043 -0
- data/data/en/words/words.v.1.3 +1052 -0
- data/data/en/words/words.v.1.4 +1023 -0
- data/data/en/words/words.v.1.p +17 -0
- data/data/en/words/words.v.10.1 +14 -0
- data/data/en/words/words.v.10.2 +15 -0
- data/data/en/words/words.v.10.3 +88 -0
- data/data/en/words/words.v.10.4 +17 -0
- data/data/en/words/words.v.2.1 +1253 -0
- data/data/en/words/words.v.2.2 +1304 -0
- data/data/en/words/words.v.2.3 +1280 -0
- data/data/en/words/words.v.2.4 +1285 -0
- data/data/en/words/words.v.2.5 +1287 -0
- data/data/en/words/words.v.4.1 +2472 -0
- data/data/en/words/words.v.4.2 +2487 -0
- data/data/en/words/words.v.4.3 +2441 -0
- data/data/en/words/words.v.4.4 +2478 -0
- data/data/en/words/words.v.4.5 +2483 -0
- data/data/en/words/words.v.5.1 +98 -0
- data/data/en/words/words.v.5.2 +98 -0
- data/data/en/words/words.v.5.3 +103 -0
- data/data/en/words/words.v.5.4 +102 -0
- data/data/en/words/words.v.6.1 +388 -0
- data/data/en/words/words.v.6.2 +401 -0
- data/data/en/words/words.v.6.3 +397 -0
- data/data/en/words/words.v.6.4 +405 -0
- data/data/en/words/words.v.6.5 +401 -0
- data/data/en/words/words.v.8.1 +117 -0
- data/data/en/words/words.v.8.2 +118 -0
- data/data/en/words/words.v.8.3 +118 -0
- data/data/en/words/words.v.8.4 +119 -0
- data/data/en/words/words.v.8.5 +119 -0
- data/data/en/words/words.y +104 -0
- data/data/lt/.DS_Store +0 -0
- data/data/lt/4.0.affix +6 -0
- data/data/lt/4.0.constituent-knowledge +24 -0
- data/data/lt/4.0.dict +135 -0
- data/data/lt/4.0.knowledge +38 -0
- data/data/lt/Makefile +389 -0
- data/data/lt/Makefile.am +11 -0
- data/data/lt/Makefile.in +389 -0
- data/grammar_police.gemspec +23 -0
- data/lib/.DS_Store +0 -0
- data/lib/grammar_police.rb +11 -0
- data/lib/grammar_police/.DS_Store +0 -0
- data/lib/grammar_police/dictionary.rb +30 -0
- data/lib/grammar_police/linkage.rb +26 -0
- data/lib/grammar_police/parse_options.rb +32 -0
- data/lib/grammar_police/sentence.rb +44 -0
- data/lib/grammar_police/version.rb +3 -0
- data/tests/.DS_Store +0 -0
- data/tests/count_linkages.rb +29 -0
- data/tests/sentences.txt +86 -0
- metadata +408 -0
@@ -0,0 +1,1887 @@
|
|
1
|
+
/*************************************************************************/
|
2
|
+
/* Copyright (c) 2004 */
|
3
|
+
/* Daniel Sleator, David Temperley, and John Lafferty */
|
4
|
+
/* Copyright 2008, 2009 Linas Vepstas */
|
5
|
+
/* All rights reserved */
|
6
|
+
/* */
|
7
|
+
/* Use of the link grammar parsing system is subject to the terms of the */
|
8
|
+
/* license set forth in the LICENSE file included with this software, */
|
9
|
+
/* and also available at http://www.link.cs.cmu.edu/link/license.html */
|
10
|
+
/* This license allows free redistribution and use in source and binary */
|
11
|
+
/* forms, with or without modification, subject to certain conditions. */
|
12
|
+
/* */
|
13
|
+
/*************************************************************************/
|
14
|
+
|
15
|
+
#ifndef API_C
|
16
|
+
#define API_C
|
17
|
+
|
18
|
+
#include <limits.h>
|
19
|
+
#include <math.h>
|
20
|
+
#include <string.h>
|
21
|
+
|
22
|
+
#include "api.h"
|
23
|
+
#include "disjuncts.h"
|
24
|
+
#include "error.h"
|
25
|
+
#include "preparation.h"
|
26
|
+
#include "read-regex.h"
|
27
|
+
#include "regex-morph.h"
|
28
|
+
#include "sat-solver/sat-encoder.h"
|
29
|
+
#include "corpus/corpus.h"
|
30
|
+
#include "spellcheck.h"
|
31
|
+
|
32
|
+
/***************************************************************
|
33
|
+
*
|
34
|
+
* Routines for setting Parse_Options
|
35
|
+
*
|
36
|
+
****************************************************************/
|
37
|
+
static int VDAL_compare_parse(Linkage_info * p1, Linkage_info * p2)
|
38
|
+
{
|
39
|
+
/* for sorting the linkages in postprocessing */
|
40
|
+
if (p1->N_violations != p2->N_violations) {
|
41
|
+
return (p1->N_violations - p2->N_violations);
|
42
|
+
}
|
43
|
+
else if (p1->unused_word_cost != p2->unused_word_cost) {
|
44
|
+
return (p1->unused_word_cost - p2->unused_word_cost);
|
45
|
+
}
|
46
|
+
else if (p1->fat != p2->fat) {
|
47
|
+
return (p1->fat - p2->fat);
|
48
|
+
}
|
49
|
+
else if (p1->disjunct_cost != p2->disjunct_cost) {
|
50
|
+
return (p1->disjunct_cost - p2->disjunct_cost);
|
51
|
+
}
|
52
|
+
else if (p1->and_cost != p2->and_cost) {
|
53
|
+
return (p1->and_cost - p2->and_cost);
|
54
|
+
}
|
55
|
+
else {
|
56
|
+
return (p1->link_cost - p2->link_cost);
|
57
|
+
}
|
58
|
+
}
|
59
|
+
|
60
|
+
#ifdef USE_CORPUS
|
61
|
+
static int CORP_compare_parse(Linkage_info * p1, Linkage_info * p2)
|
62
|
+
{
|
63
|
+
double diff = p1->corpus_cost - p2->corpus_cost;
|
64
|
+
if (fabs(diff) < 1.0e-5)
|
65
|
+
return VDAL_compare_parse(p1, p2);
|
66
|
+
if (diff < 0.0f) return -1;
|
67
|
+
return 1;
|
68
|
+
}
|
69
|
+
#endif
|
70
|
+
|
71
|
+
/**
|
72
|
+
* Create and initialize a Parse_Options object
|
73
|
+
*/
|
74
|
+
Parse_Options parse_options_create(void)
|
75
|
+
{
|
76
|
+
Parse_Options po;
|
77
|
+
|
78
|
+
init_memusage();
|
79
|
+
po = (Parse_Options) xalloc(sizeof(struct Parse_Options_s));
|
80
|
+
|
81
|
+
/* Here's where the values are initialized */
|
82
|
+
po->verbosity = 1;
|
83
|
+
po->linkage_limit = 100;
|
84
|
+
po->disjunct_cost = MAX_DISJUNCT_COST;
|
85
|
+
po->use_fat_links = FALSE;
|
86
|
+
po->min_null_count = 0;
|
87
|
+
po->max_null_count = 0;
|
88
|
+
po->null_block = 1;
|
89
|
+
po->islands_ok = FALSE;
|
90
|
+
po->use_spell_guess = TRUE;
|
91
|
+
po->use_sat_solver = FALSE;
|
92
|
+
|
93
|
+
#ifdef XXX_USE_CORPUS
|
94
|
+
/* Use the corpus cost model, if available.
|
95
|
+
* It really does a better job at parse ranking.
|
96
|
+
* Err .. sometimes ...
|
97
|
+
*/
|
98
|
+
po->cost_model.compare_fn = &CORP_compare_parse;
|
99
|
+
po->cost_model.type = CORPUS;
|
100
|
+
#else /* USE_CORPUS */
|
101
|
+
po->cost_model.compare_fn = &VDAL_compare_parse;
|
102
|
+
po->cost_model.type = VDAL;
|
103
|
+
#endif /* USE_CORPUS */
|
104
|
+
po->short_length = 6;
|
105
|
+
po->all_short = FALSE;
|
106
|
+
po->twopass_length = 30;
|
107
|
+
po->max_sentence_length = 170;
|
108
|
+
po->resources = resources_create();
|
109
|
+
po->display_short = TRUE;
|
110
|
+
po->display_word_subscripts = TRUE;
|
111
|
+
po->display_link_subscripts = TRUE;
|
112
|
+
po->display_walls = FALSE;
|
113
|
+
po->display_union = FALSE;
|
114
|
+
po->allow_null = TRUE;
|
115
|
+
po->use_cluster_disjuncts = FALSE;
|
116
|
+
po->echo_on = FALSE;
|
117
|
+
po->batch_mode = FALSE;
|
118
|
+
po->panic_mode = FALSE;
|
119
|
+
po->screen_width = 79;
|
120
|
+
po->display_on = TRUE;
|
121
|
+
po->display_postscript = FALSE;
|
122
|
+
po->display_constituents = 0;
|
123
|
+
po->display_bad = FALSE;
|
124
|
+
po->display_disjuncts = FALSE;
|
125
|
+
po->display_links = FALSE;
|
126
|
+
po->display_senses = FALSE;
|
127
|
+
|
128
|
+
return po;
|
129
|
+
}
|
130
|
+
|
131
|
+
int parse_options_delete(Parse_Options opts)
|
132
|
+
{
|
133
|
+
resources_delete(opts->resources);
|
134
|
+
xfree(opts, sizeof(struct Parse_Options_s));
|
135
|
+
return 0;
|
136
|
+
}
|
137
|
+
|
138
|
+
void parse_options_set_cost_model_type(Parse_Options opts, int cm)
|
139
|
+
{
|
140
|
+
switch(cm) {
|
141
|
+
case VDAL:
|
142
|
+
opts->cost_model.type = VDAL;
|
143
|
+
opts->cost_model.compare_fn = &VDAL_compare_parse;
|
144
|
+
break;
|
145
|
+
case CORPUS:
|
146
|
+
#ifdef USE_CORPUS
|
147
|
+
opts->cost_model.type = CORPUS;
|
148
|
+
opts->cost_model.compare_fn = &CORP_compare_parse;
|
149
|
+
#else
|
150
|
+
prt_error("Error: Source code compiled with cost model 'CORPUS' disabled.\n");
|
151
|
+
#endif
|
152
|
+
break;
|
153
|
+
default:
|
154
|
+
prt_error("Error: Illegal cost model: %d\n", cm);
|
155
|
+
}
|
156
|
+
}
|
157
|
+
|
158
|
+
int parse_options_get_cost_model_type(Parse_Options opts)
|
159
|
+
{
|
160
|
+
return opts->cost_model.type;
|
161
|
+
}
|
162
|
+
|
163
|
+
void parse_options_set_verbosity(Parse_Options opts, int dummy)
|
164
|
+
{
|
165
|
+
opts->verbosity = dummy;
|
166
|
+
verbosity = opts->verbosity;
|
167
|
+
/* this is one of the only global variables. */
|
168
|
+
}
|
169
|
+
|
170
|
+
int parse_options_get_verbosity(Parse_Options opts) {
|
171
|
+
return opts->verbosity;
|
172
|
+
}
|
173
|
+
|
174
|
+
void parse_options_set_use_sat_parser(Parse_Options opts, int dummy) {
|
175
|
+
#ifdef USE_SAT_SOLVER
|
176
|
+
opts->use_sat_solver = dummy;
|
177
|
+
#else
|
178
|
+
prt_error("Error: cannot enable the Boolean SAT parser; this "
|
179
|
+
" library was built without SAT solver support.\n");
|
180
|
+
#endif
|
181
|
+
}
|
182
|
+
int parse_options_get_use_sat_parser(Parse_Options opts) {
|
183
|
+
return opts->use_sat_solver;
|
184
|
+
}
|
185
|
+
|
186
|
+
void parse_options_set_use_fat_links(Parse_Options opts, int dummy) {
|
187
|
+
opts->use_fat_links = dummy;
|
188
|
+
}
|
189
|
+
int parse_options_get_use_fat_links(Parse_Options opts) {
|
190
|
+
return opts->use_fat_links;
|
191
|
+
}
|
192
|
+
|
193
|
+
void parse_options_set_linkage_limit(Parse_Options opts, int dummy) {
|
194
|
+
opts->linkage_limit = dummy;
|
195
|
+
}
|
196
|
+
int parse_options_get_linkage_limit(Parse_Options opts) {
|
197
|
+
return opts->linkage_limit;
|
198
|
+
}
|
199
|
+
|
200
|
+
void parse_options_set_disjunct_cost(Parse_Options opts, int dummy) {
|
201
|
+
opts->disjunct_cost = dummy;
|
202
|
+
}
|
203
|
+
void parse_options_set_disjunct_costf(Parse_Options opts, float dummy) {
|
204
|
+
opts->disjunct_cost = dummy;
|
205
|
+
}
|
206
|
+
int parse_options_get_disjunct_cost(Parse_Options opts) {
|
207
|
+
return opts->disjunct_cost;
|
208
|
+
}
|
209
|
+
float parse_options_get_disjunct_costf(Parse_Options opts) {
|
210
|
+
return opts->disjunct_cost;
|
211
|
+
}
|
212
|
+
|
213
|
+
void parse_options_set_min_null_count(Parse_Options opts, int val) {
|
214
|
+
opts->min_null_count = val;
|
215
|
+
}
|
216
|
+
int parse_options_get_min_null_count(Parse_Options opts) {
|
217
|
+
return opts->min_null_count;
|
218
|
+
}
|
219
|
+
|
220
|
+
void parse_options_set_max_null_count(Parse_Options opts, int val) {
|
221
|
+
opts->max_null_count = val;
|
222
|
+
}
|
223
|
+
int parse_options_get_max_null_count(Parse_Options opts) {
|
224
|
+
return opts->max_null_count;
|
225
|
+
}
|
226
|
+
|
227
|
+
|
228
|
+
void parse_options_set_null_block(Parse_Options opts, int dummy) {
|
229
|
+
opts->null_block = dummy;
|
230
|
+
}
|
231
|
+
int parse_options_get_null_block(Parse_Options opts) {
|
232
|
+
return opts->null_block;
|
233
|
+
}
|
234
|
+
|
235
|
+
void parse_options_set_islands_ok(Parse_Options opts, int dummy) {
|
236
|
+
opts->islands_ok = dummy;
|
237
|
+
}
|
238
|
+
|
239
|
+
int parse_options_get_islands_ok(Parse_Options opts) {
|
240
|
+
return opts->islands_ok;
|
241
|
+
}
|
242
|
+
|
243
|
+
void parse_options_set_spell_guess(Parse_Options opts, int dummy) {
|
244
|
+
opts->use_spell_guess = dummy;
|
245
|
+
}
|
246
|
+
|
247
|
+
int parse_options_get_spell_guess(Parse_Options opts) {
|
248
|
+
return opts->use_spell_guess;
|
249
|
+
}
|
250
|
+
|
251
|
+
void parse_options_set_short_length(Parse_Options opts, int short_length) {
|
252
|
+
opts->short_length = short_length;
|
253
|
+
}
|
254
|
+
|
255
|
+
int parse_options_get_short_length(Parse_Options opts) {
|
256
|
+
return opts->short_length;
|
257
|
+
}
|
258
|
+
|
259
|
+
void parse_options_set_all_short_connectors(Parse_Options opts, int val) {
|
260
|
+
opts->all_short = val;
|
261
|
+
}
|
262
|
+
|
263
|
+
int parse_options_get_all_short_connectors(Parse_Options opts) {
|
264
|
+
return opts->all_short;
|
265
|
+
}
|
266
|
+
|
267
|
+
void parse_options_set_max_parse_time(Parse_Options opts, int dummy) {
|
268
|
+
opts->resources->max_parse_time = dummy;
|
269
|
+
}
|
270
|
+
|
271
|
+
int parse_options_get_max_parse_time(Parse_Options opts) {
|
272
|
+
return opts->resources->max_parse_time;
|
273
|
+
}
|
274
|
+
|
275
|
+
void parse_options_set_max_memory(Parse_Options opts, int dummy) {
|
276
|
+
opts->resources->max_memory = dummy;
|
277
|
+
}
|
278
|
+
|
279
|
+
int parse_options_get_max_memory(Parse_Options opts) {
|
280
|
+
return opts->resources->max_memory;
|
281
|
+
}
|
282
|
+
|
283
|
+
void parse_options_set_max_sentence_length(Parse_Options opts, int dummy) {
|
284
|
+
opts->max_sentence_length = dummy;
|
285
|
+
}
|
286
|
+
|
287
|
+
int parse_options_get_max_sentence_length(Parse_Options opts) {
|
288
|
+
return opts->max_sentence_length;
|
289
|
+
}
|
290
|
+
|
291
|
+
void parse_options_set_echo_on(Parse_Options opts, int dummy) {
|
292
|
+
opts->echo_on = dummy;
|
293
|
+
}
|
294
|
+
|
295
|
+
int parse_options_get_echo_on(Parse_Options opts) {
|
296
|
+
return opts->echo_on;
|
297
|
+
}
|
298
|
+
|
299
|
+
void parse_options_set_batch_mode(Parse_Options opts, int dummy) {
|
300
|
+
opts->batch_mode = dummy;
|
301
|
+
}
|
302
|
+
|
303
|
+
int parse_options_get_batch_mode(Parse_Options opts) {
|
304
|
+
return opts->batch_mode;
|
305
|
+
}
|
306
|
+
|
307
|
+
void parse_options_set_panic_mode(Parse_Options opts, int dummy) {
|
308
|
+
opts->panic_mode = dummy;
|
309
|
+
}
|
310
|
+
|
311
|
+
int parse_options_get_panic_mode(Parse_Options opts) {
|
312
|
+
return opts->panic_mode;
|
313
|
+
}
|
314
|
+
|
315
|
+
void parse_options_set_allow_null(Parse_Options opts, int dummy) {
|
316
|
+
opts->allow_null = dummy;
|
317
|
+
}
|
318
|
+
|
319
|
+
int parse_options_get_allow_null(Parse_Options opts) {
|
320
|
+
return opts->allow_null;
|
321
|
+
}
|
322
|
+
|
323
|
+
void parse_options_set_use_cluster_disjuncts(Parse_Options opts, int dummy) {
|
324
|
+
opts->use_cluster_disjuncts = dummy;
|
325
|
+
}
|
326
|
+
|
327
|
+
int parse_options_get_use_cluster_disjuncts(Parse_Options opts) {
|
328
|
+
return opts->use_cluster_disjuncts;
|
329
|
+
}
|
330
|
+
|
331
|
+
void parse_options_set_screen_width(Parse_Options opts, int dummy) {
|
332
|
+
opts->screen_width = dummy;
|
333
|
+
}
|
334
|
+
|
335
|
+
int parse_options_get_screen_width(Parse_Options opts) {
|
336
|
+
return opts->screen_width;
|
337
|
+
}
|
338
|
+
|
339
|
+
|
340
|
+
void parse_options_set_display_on(Parse_Options opts, int dummy) {
|
341
|
+
opts->display_on = dummy;
|
342
|
+
}
|
343
|
+
|
344
|
+
int parse_options_get_display_on(Parse_Options opts) {
|
345
|
+
return opts->display_on;
|
346
|
+
}
|
347
|
+
|
348
|
+
void parse_options_set_display_postscript(Parse_Options opts, int dummy) {
|
349
|
+
opts->display_postscript = dummy;
|
350
|
+
}
|
351
|
+
|
352
|
+
int parse_options_get_display_postscript(Parse_Options opts)
|
353
|
+
{
|
354
|
+
return opts->display_postscript;
|
355
|
+
}
|
356
|
+
|
357
|
+
void parse_options_set_display_constituents(Parse_Options opts, int dummy)
|
358
|
+
{
|
359
|
+
if ((dummy < 0) || (dummy > 3)) {
|
360
|
+
prt_error("Possible values for constituents: \n"
|
361
|
+
" 0 (no display)\n"
|
362
|
+
" 1 (treebank style, multi-line indented)\n"
|
363
|
+
" 2 (flat tree, square brackets)\n"
|
364
|
+
" 3 (flat treebank style)\n");
|
365
|
+
opts->display_constituents = 0;
|
366
|
+
}
|
367
|
+
else opts->display_constituents = dummy;
|
368
|
+
}
|
369
|
+
|
370
|
+
int parse_options_get_display_constituents(Parse_Options opts)
|
371
|
+
{
|
372
|
+
return opts->display_constituents;
|
373
|
+
}
|
374
|
+
|
375
|
+
void parse_options_set_display_bad(Parse_Options opts, int dummy) {
|
376
|
+
opts->display_bad = dummy;
|
377
|
+
}
|
378
|
+
|
379
|
+
int parse_options_get_display_bad(Parse_Options opts) {
|
380
|
+
return opts->display_bad;
|
381
|
+
}
|
382
|
+
|
383
|
+
void parse_options_set_display_disjuncts(Parse_Options opts, int dummy) {
|
384
|
+
opts->display_disjuncts = dummy;
|
385
|
+
}
|
386
|
+
|
387
|
+
int parse_options_get_display_disjuncts(Parse_Options opts) {
|
388
|
+
return opts->display_disjuncts;
|
389
|
+
}
|
390
|
+
|
391
|
+
void parse_options_set_display_links(Parse_Options opts, int dummy) {
|
392
|
+
opts->display_links = dummy;
|
393
|
+
}
|
394
|
+
|
395
|
+
int parse_options_get_display_links(Parse_Options opts) {
|
396
|
+
return opts->display_links;
|
397
|
+
}
|
398
|
+
|
399
|
+
void parse_options_set_display_senses(Parse_Options opts, int dummy) {
|
400
|
+
opts->display_senses = dummy;
|
401
|
+
}
|
402
|
+
|
403
|
+
int parse_options_get_display_senses(Parse_Options opts) {
|
404
|
+
return opts->display_senses;
|
405
|
+
}
|
406
|
+
|
407
|
+
void parse_options_set_display_walls(Parse_Options opts, int dummy) {
|
408
|
+
opts->display_walls = dummy;
|
409
|
+
}
|
410
|
+
|
411
|
+
int parse_options_get_display_walls(Parse_Options opts) {
|
412
|
+
return opts->display_walls;
|
413
|
+
}
|
414
|
+
|
415
|
+
int parse_options_get_display_union(Parse_Options opts) {
|
416
|
+
return opts->display_union;
|
417
|
+
}
|
418
|
+
|
419
|
+
void parse_options_set_display_union(Parse_Options opts, int dummy) {
|
420
|
+
opts->display_union = dummy;
|
421
|
+
}
|
422
|
+
|
423
|
+
int parse_options_timer_expired(Parse_Options opts) {
|
424
|
+
return resources_timer_expired(opts->resources);
|
425
|
+
}
|
426
|
+
|
427
|
+
int parse_options_memory_exhausted(Parse_Options opts) {
|
428
|
+
return resources_memory_exhausted(opts->resources);
|
429
|
+
}
|
430
|
+
|
431
|
+
int parse_options_resources_exhausted(Parse_Options opts) {
|
432
|
+
return (resources_timer_expired(opts->resources) || resources_memory_exhausted(opts->resources));
|
433
|
+
}
|
434
|
+
|
435
|
+
void parse_options_reset_resources(Parse_Options opts) {
|
436
|
+
resources_reset(opts->resources);
|
437
|
+
}
|
438
|
+
|
439
|
+
|
440
|
+
/***************************************************************
|
441
|
+
*
|
442
|
+
* Routines for manipulating Dictionary
|
443
|
+
*
|
444
|
+
****************************************************************/
|
445
|
+
|
446
|
+
/* Units will typically have a ".u" at the end. Get
|
447
|
+
* rid of it, as otherwise stipping is messed up. */
|
448
|
+
static inline char * deinflect(const char * str)
|
449
|
+
{
|
450
|
+
char * s = strdup(str);
|
451
|
+
char * p = strchr(s, '.');
|
452
|
+
if (p && p != s) *p = 0x0;
|
453
|
+
return s;
|
454
|
+
}
|
455
|
+
|
456
|
+
static void affix_list_create(Dictionary dict)
|
457
|
+
{
|
458
|
+
int i, j, k, l, m;
|
459
|
+
int r_strippable=0, l_strippable=0, u_strippable=0;
|
460
|
+
int s_strippable=0, p_strippable=0;
|
461
|
+
Dict_node * dn, * dn2, * start_dn;
|
462
|
+
|
463
|
+
const char * rpunc_con = "RPUNC";
|
464
|
+
const char * lpunc_con = "LPUNC";
|
465
|
+
const char * units_con = "UNITS";
|
466
|
+
|
467
|
+
/* Hmm SUF and PRE do not seem to be used at this time ... */
|
468
|
+
const char * suf_con = "SUF";
|
469
|
+
const char * pre_con = "PRE";
|
470
|
+
|
471
|
+
dict->strip_left = NULL;
|
472
|
+
dict->strip_right = NULL;
|
473
|
+
dict->strip_units = NULL;
|
474
|
+
dict->prefix = NULL;
|
475
|
+
dict->suffix = NULL;
|
476
|
+
|
477
|
+
/* Load affixes from the affix table.
|
478
|
+
*/
|
479
|
+
start_dn = list_whole_dictionary(dict->root, NULL);
|
480
|
+
for (dn = start_dn; dn != NULL; dn = dn->right)
|
481
|
+
{
|
482
|
+
if (word_has_connector(dn, rpunc_con, 0)) r_strippable++;
|
483
|
+
if (word_has_connector(dn, lpunc_con, 0)) l_strippable++;
|
484
|
+
if (word_has_connector(dn, units_con, 0)) u_strippable++;
|
485
|
+
if (word_has_connector(dn, suf_con, 0)) s_strippable++;
|
486
|
+
if (word_has_connector(dn, pre_con, 0)) p_strippable++;
|
487
|
+
}
|
488
|
+
dict->strip_right = (const char **) xalloc(r_strippable * sizeof(char *));
|
489
|
+
dict->strip_left = (const char **) xalloc(l_strippable * sizeof(char *));
|
490
|
+
dict->strip_units = (const char **) xalloc(u_strippable * sizeof(char *));
|
491
|
+
dict->suffix = (const char **) xalloc(s_strippable * sizeof(char *));
|
492
|
+
dict->prefix = (const char **) xalloc(p_strippable * sizeof(char *));
|
493
|
+
|
494
|
+
dict->r_strippable = r_strippable;
|
495
|
+
dict->l_strippable = l_strippable;
|
496
|
+
dict->u_strippable = u_strippable;
|
497
|
+
dict->p_strippable = p_strippable;
|
498
|
+
dict->s_strippable = s_strippable;
|
499
|
+
|
500
|
+
i = 0;
|
501
|
+
j = 0;
|
502
|
+
k = 0;
|
503
|
+
l = 0;
|
504
|
+
m = 0;
|
505
|
+
dn = start_dn;
|
506
|
+
|
507
|
+
while (dn != NULL)
|
508
|
+
{
|
509
|
+
if (word_has_connector(dn, rpunc_con, 0))
|
510
|
+
{
|
511
|
+
dict->strip_right[i] = deinflect(dn->string);
|
512
|
+
i++;
|
513
|
+
}
|
514
|
+
if (word_has_connector(dn, lpunc_con, 0))
|
515
|
+
{
|
516
|
+
dict->strip_left[j] = deinflect(dn->string);
|
517
|
+
j++;
|
518
|
+
}
|
519
|
+
if (word_has_connector(dn, units_con, 0))
|
520
|
+
{
|
521
|
+
dict->strip_units[m] = deinflect(dn->string);
|
522
|
+
m++;
|
523
|
+
}
|
524
|
+
if (word_has_connector(dn, suf_con, 0))
|
525
|
+
{
|
526
|
+
dict->suffix[k] = dn->string;
|
527
|
+
k++;
|
528
|
+
}
|
529
|
+
if (word_has_connector(dn, pre_con, 0))
|
530
|
+
{
|
531
|
+
dict->prefix[l] = dn->string;
|
532
|
+
l++;
|
533
|
+
}
|
534
|
+
dn2 = dn->right;
|
535
|
+
dn->right = NULL;
|
536
|
+
xfree(dn, sizeof(Dict_node));
|
537
|
+
dn = dn2;
|
538
|
+
}
|
539
|
+
}
|
540
|
+
|
541
|
+
static void affix_list_delete(Dictionary dict)
|
542
|
+
{
|
543
|
+
int i;
|
544
|
+
for (i=0; i<dict->l_strippable; i++)
|
545
|
+
{
|
546
|
+
free((char *)dict->strip_left[i]);
|
547
|
+
}
|
548
|
+
for (i=0; i<dict->r_strippable; i++)
|
549
|
+
{
|
550
|
+
free((char *)dict->strip_right[i]);
|
551
|
+
}
|
552
|
+
for (i=0; i<dict->u_strippable; i++)
|
553
|
+
{
|
554
|
+
free((char *)dict->strip_units[i]);
|
555
|
+
}
|
556
|
+
xfree(dict->strip_right, dict->r_strippable * sizeof(char *));
|
557
|
+
xfree(dict->strip_left, dict->l_strippable * sizeof(char *));
|
558
|
+
xfree(dict->strip_units, dict->u_strippable * sizeof(char *));
|
559
|
+
xfree(dict->suffix, dict->s_strippable * sizeof(char *));
|
560
|
+
xfree(dict->prefix, dict->p_strippable * sizeof(char *));
|
561
|
+
}
|
562
|
+
|
563
|
+
/**
|
564
|
+
* The following function is dictionary_create with an extra
|
565
|
+
* paramater called "path". If this is non-null, then the path
|
566
|
+
* used to find the file is taken from that path. Otherwise,
|
567
|
+
* the path is taken from the dict_name. This is only needed
|
568
|
+
* because an affix_file is opened by a recursive call to this
|
569
|
+
* function.
|
570
|
+
*/
|
571
|
+
static Dictionary
|
572
|
+
dictionary_six(const char * lang, const char * dict_name,
|
573
|
+
const char * pp_name, const char * cons_name,
|
574
|
+
const char * affix_name, const char * regex_name)
|
575
|
+
{
|
576
|
+
const char * t;
|
577
|
+
Dictionary dict;
|
578
|
+
Dict_node *dict_node;
|
579
|
+
|
580
|
+
init_memusage();
|
581
|
+
|
582
|
+
dict = (Dictionary) xalloc(sizeof(struct Dictionary_s));
|
583
|
+
memset(dict, 0, sizeof(struct Dictionary_s));
|
584
|
+
|
585
|
+
dict->string_set = string_set_create();
|
586
|
+
|
587
|
+
dict->lang = lang;
|
588
|
+
t = strrchr (lang, '/');
|
589
|
+
if (t) dict->lang = string_set_add(t+1, dict->string_set);
|
590
|
+
dict->name = string_set_add(dict_name, dict->string_set);
|
591
|
+
|
592
|
+
dict->max_cost = 1000;
|
593
|
+
dict->num_entries = 0;
|
594
|
+
dict->is_special = FALSE;
|
595
|
+
dict->already_got_it = '\0';
|
596
|
+
dict->line_number = 1;
|
597
|
+
dict->root = NULL;
|
598
|
+
dict->word_file_header = NULL;
|
599
|
+
dict->exp_list = NULL;
|
600
|
+
dict->affix_table = NULL;
|
601
|
+
dict->recursive_error = FALSE;
|
602
|
+
|
603
|
+
/* To disable spell-checking, just set the cheker to NULL */
|
604
|
+
dict->spell_checker = spellcheck_create(dict->lang);
|
605
|
+
|
606
|
+
dict->fp = dictopen(dict->name, "r");
|
607
|
+
if (dict->fp == NULL)
|
608
|
+
{
|
609
|
+
prt_error("Error: Could not open dictionary %s\n", dict_name);
|
610
|
+
goto failure;
|
611
|
+
}
|
612
|
+
|
613
|
+
if (!read_dictionary(dict))
|
614
|
+
{
|
615
|
+
fclose(dict->fp);
|
616
|
+
goto failure;
|
617
|
+
}
|
618
|
+
fclose(dict->fp);
|
619
|
+
|
620
|
+
dict->affix_table = NULL;
|
621
|
+
if (affix_name != NULL)
|
622
|
+
{
|
623
|
+
dict->affix_table = dictionary_six(lang, affix_name, NULL, NULL, NULL, NULL);
|
624
|
+
if (dict->affix_table == NULL)
|
625
|
+
{
|
626
|
+
goto failure;
|
627
|
+
}
|
628
|
+
affix_list_create(dict->affix_table);
|
629
|
+
}
|
630
|
+
|
631
|
+
dict->regex_root = NULL;
|
632
|
+
if (regex_name != NULL)
|
633
|
+
{
|
634
|
+
int rc;
|
635
|
+
rc = read_regex_file(dict, regex_name);
|
636
|
+
if (rc) goto failure;
|
637
|
+
rc = compile_regexs(dict);
|
638
|
+
if (rc) goto failure;
|
639
|
+
}
|
640
|
+
|
641
|
+
#if USE_CORPUS
|
642
|
+
dict->corpus = NULL;
|
643
|
+
if (affix_name != NULL) /* Don't do this for the second time */
|
644
|
+
{
|
645
|
+
dict->corpus = lg_corpus_new();
|
646
|
+
}
|
647
|
+
#endif
|
648
|
+
|
649
|
+
dict->left_wall_defined = boolean_dictionary_lookup(dict, LEFT_WALL_WORD);
|
650
|
+
dict->right_wall_defined = boolean_dictionary_lookup(dict, RIGHT_WALL_WORD);
|
651
|
+
dict->postprocessor = post_process_open(pp_name);
|
652
|
+
dict->constituent_pp = post_process_open(cons_name);
|
653
|
+
|
654
|
+
dict->unknown_word_defined = boolean_dictionary_lookup(dict, UNKNOWN_WORD);
|
655
|
+
dict->use_unknown_word = TRUE;
|
656
|
+
|
657
|
+
#if DONT_USE_REGEX_GUESSING
|
658
|
+
dict->capitalized_word_defined = boolean_dictionary_lookup(dict, PROPER_WORD);
|
659
|
+
dict->pl_capitalized_word_defined = boolean_dictionary_lookup(dict, PL_PROPER_WORD);
|
660
|
+
|
661
|
+
dict->hyphenated_word_defined = boolean_dictionary_lookup(dict, HYPHENATED_WORD);
|
662
|
+
dict->number_word_defined = boolean_dictionary_lookup(dict, NUMBER_WORD);
|
663
|
+
|
664
|
+
dict->ing_word_defined = boolean_dictionary_lookup(dict, ING_WORD);
|
665
|
+
dict->s_word_defined = boolean_dictionary_lookup(dict, S_WORD);
|
666
|
+
dict->ed_word_defined = boolean_dictionary_lookup(dict, ED_WORD);
|
667
|
+
dict->ly_word_defined = boolean_dictionary_lookup(dict, LY_WORD);
|
668
|
+
#endif /* DONT_USE_REGEX_GUESSING */
|
669
|
+
|
670
|
+
if ((dict_node = dictionary_lookup_list(dict, ANDABLE_CONNECTORS_WORD)) != NULL) {
|
671
|
+
dict->andable_connector_set = connector_set_create(dict_node->exp);
|
672
|
+
} else {
|
673
|
+
dict->andable_connector_set = NULL;
|
674
|
+
}
|
675
|
+
free_lookup_list(dict_node);
|
676
|
+
|
677
|
+
if ((dict_node = dictionary_lookup_list(dict, UNLIMITED_CONNECTORS_WORD)) != NULL) {
|
678
|
+
dict->unlimited_connector_set = connector_set_create(dict_node->exp);
|
679
|
+
} else {
|
680
|
+
dict->unlimited_connector_set = NULL;
|
681
|
+
}
|
682
|
+
free_lookup_list(dict_node);
|
683
|
+
|
684
|
+
return dict;
|
685
|
+
|
686
|
+
failure:
|
687
|
+
string_set_delete(dict->string_set);
|
688
|
+
xfree(dict, sizeof(struct Dictionary_s));
|
689
|
+
return NULL;
|
690
|
+
}
|
691
|
+
|
692
|
+
Dictionary
|
693
|
+
dictionary_create(const char * dict_name, const char * pp_name,
|
694
|
+
const char * cons_name, const char * affix_name)
|
695
|
+
{
|
696
|
+
return dictionary_six("en", dict_name, pp_name, cons_name, affix_name, NULL);
|
697
|
+
}
|
698
|
+
|
699
|
+
Dictionary dictionary_create_lang(const char * lang)
|
700
|
+
{
|
701
|
+
Dictionary dictionary;
|
702
|
+
|
703
|
+
if(lang && *lang)
|
704
|
+
{
|
705
|
+
char * dict_name;
|
706
|
+
char * pp_name;
|
707
|
+
char * cons_name;
|
708
|
+
char * affix_name;
|
709
|
+
char * regex_name;
|
710
|
+
|
711
|
+
dict_name = join_path(lang, "4.0.dict");
|
712
|
+
pp_name = join_path(lang, "4.0.knowledge");
|
713
|
+
cons_name = join_path(lang, "4.0.constituent-knowledge");
|
714
|
+
affix_name = join_path(lang, "4.0.affix");
|
715
|
+
regex_name = join_path(lang, "4.0.regex");
|
716
|
+
|
717
|
+
dictionary = dictionary_six(lang, dict_name, pp_name, cons_name,
|
718
|
+
affix_name, regex_name);
|
719
|
+
|
720
|
+
free(regex_name);
|
721
|
+
free(affix_name);
|
722
|
+
free(cons_name);
|
723
|
+
free(pp_name);
|
724
|
+
free(dict_name);
|
725
|
+
}
|
726
|
+
else
|
727
|
+
{
|
728
|
+
prt_error("Error: No language specified!\n");
|
729
|
+
dictionary = NULL;
|
730
|
+
}
|
731
|
+
|
732
|
+
return dictionary;
|
733
|
+
}
|
734
|
+
|
735
|
+
Dictionary dictionary_create_default_lang(void)
|
736
|
+
{
|
737
|
+
Dictionary dictionary;
|
738
|
+
char * lang;
|
739
|
+
|
740
|
+
lang = get_default_locale();
|
741
|
+
if(lang && *lang) {
|
742
|
+
dictionary = dictionary_create_lang(lang);
|
743
|
+
free(lang);
|
744
|
+
} else {
|
745
|
+
/* Default to en when locales are broken (e.g. WIN32) */
|
746
|
+
dictionary = dictionary_create_lang("en");
|
747
|
+
}
|
748
|
+
|
749
|
+
return dictionary;
|
750
|
+
}
|
751
|
+
|
752
|
+
int dictionary_delete(Dictionary dict)
|
753
|
+
{
|
754
|
+
if (verbosity > 0) {
|
755
|
+
prt_error("Info: Freeing dictionary %s\n", dict->name);
|
756
|
+
}
|
757
|
+
|
758
|
+
#if USE_CORPUS
|
759
|
+
lg_corpus_delete(dict->corpus);
|
760
|
+
#endif
|
761
|
+
|
762
|
+
if (dict->affix_table != NULL) {
|
763
|
+
affix_list_delete(dict->affix_table);
|
764
|
+
dictionary_delete(dict->affix_table);
|
765
|
+
}
|
766
|
+
spellcheck_destroy(dict->spell_checker);
|
767
|
+
|
768
|
+
connector_set_delete(dict->andable_connector_set);
|
769
|
+
connector_set_delete(dict->unlimited_connector_set);
|
770
|
+
|
771
|
+
post_process_close(dict->postprocessor);
|
772
|
+
post_process_close(dict->constituent_pp);
|
773
|
+
string_set_delete(dict->string_set);
|
774
|
+
free_regexs(dict);
|
775
|
+
free_dictionary(dict);
|
776
|
+
xfree(dict, sizeof(struct Dictionary_s));
|
777
|
+
|
778
|
+
return 0;
|
779
|
+
}
|
780
|
+
|
781
|
+
int dictionary_get_max_cost(Dictionary dict)
|
782
|
+
{
|
783
|
+
return dict->max_cost;
|
784
|
+
}
|
785
|
+
|
786
|
+
/***************************************************************
|
787
|
+
*
|
788
|
+
* Routines for postprocessing
|
789
|
+
*
|
790
|
+
****************************************************************/
|
791
|
+
|
792
|
+
static Linkage_info * linkage_info_new(int num_to_alloc)
|
793
|
+
{
|
794
|
+
Linkage_info *link_info;
|
795
|
+
link_info = (Linkage_info *) xalloc(num_to_alloc * sizeof(Linkage_info));
|
796
|
+
memset(link_info, 0, num_to_alloc * sizeof(Linkage_info));
|
797
|
+
return link_info;
|
798
|
+
}
|
799
|
+
|
800
|
+
static void linkage_info_delete(Linkage_info *link_info, int sz)
|
801
|
+
{
|
802
|
+
int i,j;
|
803
|
+
|
804
|
+
for (i=0; i<sz; i++)
|
805
|
+
{
|
806
|
+
Linkage_info *lifo = &link_info[i];
|
807
|
+
int nwords = lifo->nwords;
|
808
|
+
for (j=0; j<nwords; j++)
|
809
|
+
{
|
810
|
+
if (lifo->disjunct_list_str[j])
|
811
|
+
free(lifo->disjunct_list_str[j]);
|
812
|
+
}
|
813
|
+
free(lifo->disjunct_list_str);
|
814
|
+
#ifdef USE_CORPUS
|
815
|
+
lg_sense_delete(lifo);
|
816
|
+
#endif
|
817
|
+
}
|
818
|
+
xfree(link_info, sz);
|
819
|
+
}
|
820
|
+
|
821
|
+
static void free_andlists(Sentence sent)
|
822
|
+
{
|
823
|
+
int L;
|
824
|
+
Andlist * andlist, * next;
|
825
|
+
for(L=0; L<sent->num_linkages_post_processed; L++) {
|
826
|
+
/* printf("%d ", sent->link_info[L].canonical); */
|
827
|
+
/* if (sent->link_info[L].canonical==0) continue; */
|
828
|
+
andlist = sent->link_info[L].andlist;
|
829
|
+
while(1) {
|
830
|
+
if(andlist == NULL) break;
|
831
|
+
next = andlist->next;
|
832
|
+
xfree((char *) andlist, sizeof(Andlist));
|
833
|
+
andlist = next;
|
834
|
+
}
|
835
|
+
}
|
836
|
+
/* printf("\n"); */
|
837
|
+
}
|
838
|
+
|
839
|
+
static void free_post_processing(Sentence sent)
|
840
|
+
{
|
841
|
+
if (sent->link_info != NULL) {
|
842
|
+
/* postprocessing must have been done */
|
843
|
+
free_andlists(sent);
|
844
|
+
linkage_info_delete(sent->link_info, sent->num_linkages_alloced);
|
845
|
+
sent->link_info = NULL;
|
846
|
+
}
|
847
|
+
}
|
848
|
+
|
849
|
+
static void post_process_linkages(Sentence sent, Parse_Options opts)
|
850
|
+
{
|
851
|
+
int *indices;
|
852
|
+
int in, block_bottom, block_top;
|
853
|
+
int N_linkages_found, N_linkages_alloced;
|
854
|
+
int N_linkages_post_processed, N_valid_linkages;
|
855
|
+
int N_thin_linkages;
|
856
|
+
int overflowed, only_canonical_allowed;
|
857
|
+
Linkage_info *link_info;
|
858
|
+
int canonical;
|
859
|
+
|
860
|
+
free_post_processing(sent);
|
861
|
+
|
862
|
+
overflowed = build_parse_set(sent, sent->null_count, opts);
|
863
|
+
print_time(opts, "Built parse set");
|
864
|
+
|
865
|
+
if (overflowed && (1 < opts->verbosity))
|
866
|
+
{
|
867
|
+
err_ctxt ec;
|
868
|
+
ec.sent = sent;
|
869
|
+
err_msg(&ec, Warn, "Warning: Count overflow.\n"
|
870
|
+
"Considering a random subset of %d of an unknown and large number of linkages\n",
|
871
|
+
opts->linkage_limit);
|
872
|
+
}
|
873
|
+
N_linkages_found = sent->num_linkages_found;
|
874
|
+
|
875
|
+
if (sent->num_linkages_found == 0)
|
876
|
+
{
|
877
|
+
sent->num_linkages_alloced = 0;
|
878
|
+
sent->num_linkages_post_processed = 0;
|
879
|
+
sent->num_valid_linkages = 0;
|
880
|
+
sent->num_thin_linkages = 0;
|
881
|
+
sent->link_info = NULL;
|
882
|
+
return;
|
883
|
+
}
|
884
|
+
|
885
|
+
if (N_linkages_found > opts->linkage_limit)
|
886
|
+
{
|
887
|
+
N_linkages_alloced = opts->linkage_limit;
|
888
|
+
if (opts->verbosity > 1)
|
889
|
+
{
|
890
|
+
err_ctxt ec;
|
891
|
+
ec.sent = sent;
|
892
|
+
err_msg(&ec, Warn, "Warning: Considering a random subset of %d of %d linkages\n",
|
893
|
+
N_linkages_alloced, N_linkages_found);
|
894
|
+
}
|
895
|
+
}
|
896
|
+
else
|
897
|
+
{
|
898
|
+
N_linkages_alloced = N_linkages_found;
|
899
|
+
}
|
900
|
+
|
901
|
+
link_info = linkage_info_new(N_linkages_alloced);
|
902
|
+
N_valid_linkages = 0;
|
903
|
+
|
904
|
+
/* Generate an array of linkage indices to examine */
|
905
|
+
indices = (int *) xalloc(N_linkages_alloced * sizeof(int));
|
906
|
+
if (overflowed)
|
907
|
+
{
|
908
|
+
for (in=0; in < N_linkages_alloced; in++)
|
909
|
+
{
|
910
|
+
indices[in] = -(in+1);
|
911
|
+
}
|
912
|
+
}
|
913
|
+
else
|
914
|
+
{
|
915
|
+
sent->rand_state = N_linkages_found + sent->length;
|
916
|
+
for (in=0; in<N_linkages_alloced; in++)
|
917
|
+
{
|
918
|
+
double frac = (double) N_linkages_found;
|
919
|
+
frac /= (double) N_linkages_alloced;
|
920
|
+
block_bottom = (int) (((double) in) * frac);
|
921
|
+
block_top = (int) (((double) (in+1)) * frac);
|
922
|
+
indices[in] = block_bottom +
|
923
|
+
(rand_r(&sent->rand_state) % (block_top-block_bottom));
|
924
|
+
}
|
925
|
+
}
|
926
|
+
|
927
|
+
only_canonical_allowed = !(overflowed || (N_linkages_found > 2*opts->linkage_limit));
|
928
|
+
/* When we're processing only a small subset of the linkages,
|
929
|
+
* don't worry about restricting the set we consider to be
|
930
|
+
* canonical ones. In the extreme case where we are only
|
931
|
+
* generating 1 in a million linkages, it's very unlikely
|
932
|
+
* that we'll hit two symmetric variants of the same linkage
|
933
|
+
* anyway.
|
934
|
+
*/
|
935
|
+
/* (optional) first pass: just visit the linkages */
|
936
|
+
/* The purpose of these two passes is to make the post-processing
|
937
|
+
* more efficient. Because (hopefully) by the time you do the
|
938
|
+
* real work in the 2nd pass you've pruned the relevant rule set
|
939
|
+
* in the first pass.
|
940
|
+
*/
|
941
|
+
if (sent->length >= opts->twopass_length)
|
942
|
+
{
|
943
|
+
for (in=0; (in < N_linkages_alloced) &&
|
944
|
+
(!resources_exhausted(opts->resources)); in++)
|
945
|
+
{
|
946
|
+
extract_links(indices[in], sent->null_count, sent->parse_info);
|
947
|
+
if (set_has_fat_down(sent))
|
948
|
+
{
|
949
|
+
if (only_canonical_allowed && !is_canonical_linkage(sent)) continue;
|
950
|
+
analyze_fat_linkage(sent, opts, PP_FIRST_PASS);
|
951
|
+
}
|
952
|
+
else
|
953
|
+
{
|
954
|
+
analyze_thin_linkage(sent, opts, PP_FIRST_PASS);
|
955
|
+
}
|
956
|
+
}
|
957
|
+
}
|
958
|
+
|
959
|
+
/* second pass: actually perform post-processing */
|
960
|
+
N_linkages_post_processed = 0;
|
961
|
+
N_thin_linkages = 0;
|
962
|
+
for (in=0; (in < N_linkages_alloced) &&
|
963
|
+
(!resources_exhausted(opts->resources)); in++)
|
964
|
+
{
|
965
|
+
Linkage_info *lifo = &link_info[N_linkages_post_processed];
|
966
|
+
extract_links(indices[in], sent->null_count, sent->parse_info);
|
967
|
+
if (set_has_fat_down(sent))
|
968
|
+
{
|
969
|
+
canonical = is_canonical_linkage(sent);
|
970
|
+
if (only_canonical_allowed && !canonical) continue;
|
971
|
+
*lifo = analyze_fat_linkage(sent, opts, PP_SECOND_PASS);
|
972
|
+
lifo->fat = TRUE;
|
973
|
+
lifo->canonical = canonical;
|
974
|
+
}
|
975
|
+
else
|
976
|
+
{
|
977
|
+
*lifo = analyze_thin_linkage(sent, opts, PP_SECOND_PASS);
|
978
|
+
lifo->fat = FALSE;
|
979
|
+
lifo->canonical = TRUE;
|
980
|
+
}
|
981
|
+
if (0 == lifo->N_violations)
|
982
|
+
{
|
983
|
+
N_valid_linkages++;
|
984
|
+
if (FALSE == lifo->fat) N_thin_linkages++;
|
985
|
+
}
|
986
|
+
lifo->index = indices[in];
|
987
|
+
lg_corpus_score(sent, lifo);
|
988
|
+
N_linkages_post_processed++;
|
989
|
+
}
|
990
|
+
|
991
|
+
print_time(opts, "Postprocessed all linkages");
|
992
|
+
qsort((void *)link_info, N_linkages_post_processed, sizeof(Linkage_info),
|
993
|
+
(int (*)(const void *, const void *)) opts->cost_model.compare_fn);
|
994
|
+
|
995
|
+
if (!resources_exhausted(opts->resources))
|
996
|
+
{
|
997
|
+
if ((N_linkages_post_processed == 0) &&
|
998
|
+
(N_linkages_found > 0) &&
|
999
|
+
(N_linkages_found < opts->linkage_limit))
|
1000
|
+
{
|
1001
|
+
/* With the current parser, the following sentence will elicit
|
1002
|
+
* this error:
|
1003
|
+
*
|
1004
|
+
* Well, say, Joe, you can be Friar Tuck or Much the miller's
|
1005
|
+
* son, and lam me with a quarter-staff; or I'll be the Sheriff
|
1006
|
+
* of Nottingham and you be Robin Hood a little while and kill
|
1007
|
+
* me.
|
1008
|
+
*/
|
1009
|
+
err_ctxt ec;
|
1010
|
+
ec.sent = sent;
|
1011
|
+
err_msg(&ec, Error, "Error: None of the linkages is canonical\n"
|
1012
|
+
"\tN_linkages_post_processed=%d "
|
1013
|
+
"N_linkages_found=%d\n",
|
1014
|
+
N_linkages_post_processed,
|
1015
|
+
N_linkages_found);
|
1016
|
+
}
|
1017
|
+
}
|
1018
|
+
|
1019
|
+
if (opts->verbosity > 1)
|
1020
|
+
{
|
1021
|
+
err_ctxt ec;
|
1022
|
+
ec.sent = sent;
|
1023
|
+
err_msg(&ec, Info, "Info: %d of %d linkages with no P.P. violations\n",
|
1024
|
+
N_valid_linkages, N_linkages_post_processed);
|
1025
|
+
}
|
1026
|
+
|
1027
|
+
print_time(opts, "Sorted all linkages");
|
1028
|
+
|
1029
|
+
sent->num_linkages_alloced = N_linkages_alloced;
|
1030
|
+
sent->num_linkages_post_processed = N_linkages_post_processed;
|
1031
|
+
sent->num_valid_linkages = N_valid_linkages;
|
1032
|
+
sent->num_thin_linkages = N_thin_linkages;
|
1033
|
+
sent->link_info = link_info;
|
1034
|
+
|
1035
|
+
xfree(indices, N_linkages_alloced * sizeof(int));
|
1036
|
+
/*if(N_valid_linkages == 0) free_andlists(sent); */
|
1037
|
+
}
|
1038
|
+
|
1039
|
+
/***************************************************************
|
1040
|
+
*
|
1041
|
+
* Routines for creating and destroying processing Sentences
|
1042
|
+
*
|
1043
|
+
****************************************************************/
|
1044
|
+
|
1045
|
+
Sentence sentence_create(const char *input_string, Dictionary dict)
|
1046
|
+
{
|
1047
|
+
Sentence sent;
|
1048
|
+
|
1049
|
+
sent = (Sentence) xalloc(sizeof(struct Sentence_s));
|
1050
|
+
memset(sent, 0, sizeof(struct Sentence_s));
|
1051
|
+
sent->dict = dict;
|
1052
|
+
sent->length = 0;
|
1053
|
+
sent->num_linkages_found = 0;
|
1054
|
+
sent->num_linkages_alloced = 0;
|
1055
|
+
sent->num_linkages_post_processed = 0;
|
1056
|
+
sent->num_valid_linkages = 0;
|
1057
|
+
sent->link_info = NULL;
|
1058
|
+
sent->deletable = NULL;
|
1059
|
+
sent->effective_dist = NULL;
|
1060
|
+
sent->num_valid_linkages = 0;
|
1061
|
+
sent->null_count = 0;
|
1062
|
+
sent->parse_info = NULL;
|
1063
|
+
sent->string_set = string_set_create();
|
1064
|
+
|
1065
|
+
sent->q_pruned_rules = FALSE;
|
1066
|
+
sent->is_conjunction = NULL;
|
1067
|
+
|
1068
|
+
sent->dptr = NULL;
|
1069
|
+
sent->deletable = NULL;
|
1070
|
+
|
1071
|
+
/* Make a copy of the input */
|
1072
|
+
sent->orig_sentence = string_set_add (input_string, sent->string_set);
|
1073
|
+
|
1074
|
+
return sent;
|
1075
|
+
}
|
1076
|
+
|
1077
|
+
/* XXX Extreme hack alert -- English-language words are used
|
1078
|
+
* completely naked in the C source code!!! FIXME !!!!
|
1079
|
+
*/
|
1080
|
+
static void set_is_conjunction(Sentence sent)
|
1081
|
+
{
|
1082
|
+
int w;
|
1083
|
+
char * s;
|
1084
|
+
for (w=0; w<sent->length; w++) {
|
1085
|
+
s = sent->word[w].string;
|
1086
|
+
sent->is_conjunction[w] =
|
1087
|
+
(strcmp(s, "and")==0) ||
|
1088
|
+
(strcmp(s, "or" )==0) ||
|
1089
|
+
(strcmp(s, "but")==0) ||
|
1090
|
+
(strcmp(s, "nor")==0);
|
1091
|
+
}
|
1092
|
+
}
|
1093
|
+
|
1094
|
+
int sentence_split(Sentence sent, Parse_Options opts)
|
1095
|
+
{
|
1096
|
+
int i;
|
1097
|
+
Dictionary dict = sent->dict;
|
1098
|
+
|
1099
|
+
/* Cleanup stuff previously allocated. This is because some free
|
1100
|
+
* routines depend on sent-length, which might change in different
|
1101
|
+
* parse-opts settings.
|
1102
|
+
*/
|
1103
|
+
free_deletable(sent);
|
1104
|
+
|
1105
|
+
/* Tokenize */
|
1106
|
+
if (!separate_sentence(sent, opts))
|
1107
|
+
{
|
1108
|
+
return -1;
|
1109
|
+
}
|
1110
|
+
|
1111
|
+
sent->q_pruned_rules = FALSE; /* for post processing */
|
1112
|
+
sent->is_conjunction = (char *) xalloc(sizeof(char)*sent->length);
|
1113
|
+
set_is_conjunction(sent);
|
1114
|
+
initialize_conjunction_tables(sent);
|
1115
|
+
|
1116
|
+
for (i=0; i<sent->length; i++)
|
1117
|
+
{
|
1118
|
+
/* in case we free these before they set to anything else */
|
1119
|
+
sent->word[i].x = NULL;
|
1120
|
+
sent->word[i].d = NULL;
|
1121
|
+
}
|
1122
|
+
|
1123
|
+
if (!(dict->unknown_word_defined && dict->use_unknown_word))
|
1124
|
+
{
|
1125
|
+
if (!sentence_in_dictionary(sent)) {
|
1126
|
+
return -2;
|
1127
|
+
}
|
1128
|
+
}
|
1129
|
+
|
1130
|
+
/* Look up each word in the dictionary, collect up all
|
1131
|
+
* plausible disjunct expressions for each word.
|
1132
|
+
*/
|
1133
|
+
if (!build_sentence_expressions(sent, opts))
|
1134
|
+
{
|
1135
|
+
sent->num_valid_linkages = 0;
|
1136
|
+
return -3;
|
1137
|
+
}
|
1138
|
+
|
1139
|
+
return 0;
|
1140
|
+
}
|
1141
|
+
|
1142
|
+
void sentence_delete(Sentence sent)
|
1143
|
+
{
|
1144
|
+
if (!sent) return;
|
1145
|
+
sat_sentence_delete(sent);
|
1146
|
+
/* free_andlists(sent); */
|
1147
|
+
free_sentence_disjuncts(sent);
|
1148
|
+
free_sentence_expressions(sent);
|
1149
|
+
string_set_delete(sent->string_set);
|
1150
|
+
if (sent->parse_info) free_parse_info(sent->parse_info);
|
1151
|
+
free_post_processing(sent);
|
1152
|
+
post_process_close_sentence(sent->dict->postprocessor);
|
1153
|
+
free_deletable(sent);
|
1154
|
+
free_effective_dist(sent);
|
1155
|
+
free_count(sent);
|
1156
|
+
free_analyze(sent);
|
1157
|
+
if (sent->is_conjunction) xfree(sent->is_conjunction, sizeof(char)*sent->length);
|
1158
|
+
xfree((char *) sent, sizeof(struct Sentence_s));
|
1159
|
+
}
|
1160
|
+
|
1161
|
+
int sentence_length(Sentence sent)
|
1162
|
+
{
|
1163
|
+
if (!sent) return 0;
|
1164
|
+
return sent->length;
|
1165
|
+
}
|
1166
|
+
|
1167
|
+
const char * sentence_get_word(Sentence sent, int index)
|
1168
|
+
{
|
1169
|
+
if (!sent) return NULL;
|
1170
|
+
return sent->word[index].string;
|
1171
|
+
}
|
1172
|
+
|
1173
|
+
const char * sentence_get_nth_word(Sentence sent, int index)
|
1174
|
+
{
|
1175
|
+
if (!sent) return NULL;
|
1176
|
+
return sent->word[index].string;
|
1177
|
+
}
|
1178
|
+
|
1179
|
+
int sentence_null_count(Sentence sent) {
|
1180
|
+
if (!sent) return 0;
|
1181
|
+
return sent->null_count;
|
1182
|
+
}
|
1183
|
+
|
1184
|
+
int sentence_num_thin_linkages(Sentence sent) {
|
1185
|
+
if (!sent) return 0;
|
1186
|
+
return sent->num_thin_linkages;
|
1187
|
+
}
|
1188
|
+
|
1189
|
+
int sentence_num_linkages_found(Sentence sent) {
|
1190
|
+
if (!sent) return 0;
|
1191
|
+
return sent->num_linkages_found;
|
1192
|
+
}
|
1193
|
+
|
1194
|
+
int sentence_num_valid_linkages(Sentence sent) {
|
1195
|
+
if (!sent) return 0;
|
1196
|
+
return sent->num_valid_linkages;
|
1197
|
+
}
|
1198
|
+
|
1199
|
+
int sentence_num_linkages_post_processed(Sentence sent) {
|
1200
|
+
if (!sent) return 0;
|
1201
|
+
return sent->num_linkages_post_processed;
|
1202
|
+
}
|
1203
|
+
|
1204
|
+
int sentence_num_violations(Sentence sent, int i) {
|
1205
|
+
if (!sent) return 0;
|
1206
|
+
|
1207
|
+
/* The sat solver (currently) fails to fill in link_info */
|
1208
|
+
if (!sent->link_info) return 0;
|
1209
|
+
return sent->link_info[i].N_violations;
|
1210
|
+
}
|
1211
|
+
|
1212
|
+
int sentence_and_cost(Sentence sent, int i) {
|
1213
|
+
if (!sent) return 0;
|
1214
|
+
|
1215
|
+
/* The sat solver (currently) fails to fill in link_info */
|
1216
|
+
if (!sent->link_info) return 0;
|
1217
|
+
return sent->link_info[i].and_cost;
|
1218
|
+
}
|
1219
|
+
|
1220
|
+
int sentence_disjunct_cost(Sentence sent, int i) {
|
1221
|
+
if (!sent) return 0;
|
1222
|
+
|
1223
|
+
/* The sat solver (currently) fails to fill in link_info */
|
1224
|
+
if (!sent->link_info) return 0;
|
1225
|
+
return sent->link_info[i].disjunct_cost;
|
1226
|
+
}
|
1227
|
+
|
1228
|
+
int sentence_link_cost(Sentence sent, int i) {
|
1229
|
+
if (!sent) return 0;
|
1230
|
+
|
1231
|
+
/* The sat solver (currently) fails to fill in link_info */
|
1232
|
+
if (!sent->link_info) return 0;
|
1233
|
+
return sent->link_info[i].link_cost;
|
1234
|
+
}
|
1235
|
+
|
1236
|
+
int sentence_nth_word_has_disjunction(Sentence sent, int i)
|
1237
|
+
{
|
1238
|
+
if (!sent) return 0;
|
1239
|
+
prt_error("Warning: sentence_nth_word_has_disjunction() is deprecated!\n");
|
1240
|
+
return (sent->parse_info->chosen_disjuncts[i] != NULL);
|
1241
|
+
}
|
1242
|
+
|
1243
|
+
static void chart_parse(Sentence sent, Parse_Options opts)
|
1244
|
+
{
|
1245
|
+
int nl;
|
1246
|
+
|
1247
|
+
/* Build lists of disjuncts */
|
1248
|
+
prepare_to_parse(sent, opts);
|
1249
|
+
|
1250
|
+
init_fast_matcher(sent);
|
1251
|
+
init_count(sent);
|
1252
|
+
|
1253
|
+
/* A parse set may have been already been built for this sentence,
|
1254
|
+
* if it was previously parsed. If so we free it up before
|
1255
|
+
* building another. */
|
1256
|
+
if (sent->parse_info) free_parse_info(sent->parse_info);
|
1257
|
+
sent->parse_info = parse_info_new(sent->length);
|
1258
|
+
|
1259
|
+
for (nl = opts->min_null_count; nl<=opts->max_null_count ; ++nl)
|
1260
|
+
{
|
1261
|
+
s64 total;
|
1262
|
+
if (resources_exhausted(opts->resources)) break;
|
1263
|
+
sent->null_count = nl;
|
1264
|
+
total = do_parse(sent, sent->null_count, opts);
|
1265
|
+
|
1266
|
+
if (verbosity > 1)
|
1267
|
+
{
|
1268
|
+
prt_error("Info: Total count with %d null links: %lld\n",
|
1269
|
+
sent->null_count, total);
|
1270
|
+
}
|
1271
|
+
|
1272
|
+
/* Give up if the parse count is overflowing */
|
1273
|
+
if (PARSE_NUM_OVERFLOW < total)
|
1274
|
+
{
|
1275
|
+
if (verbosity > 0)
|
1276
|
+
{
|
1277
|
+
prt_error("WARNING: Combinatorial explosion! nulls=%d cnt=%lld\n"
|
1278
|
+
"Consider retrying the parse with the max allowed disjunct cost set lower.\n",
|
1279
|
+
sent->null_count, total);
|
1280
|
+
}
|
1281
|
+
total = (total>INT_MAX) ? INT_MAX : total;
|
1282
|
+
}
|
1283
|
+
|
1284
|
+
sent->num_linkages_found = (int) total;
|
1285
|
+
print_time(opts, "Counted parses");
|
1286
|
+
|
1287
|
+
post_process_linkages(sent, opts);
|
1288
|
+
if (sent->num_valid_linkages > 0) break;
|
1289
|
+
|
1290
|
+
/* If we are here, then no valid linakges were found.
|
1291
|
+
* If there was a parse overflow, give up now. */
|
1292
|
+
if (PARSE_NUM_OVERFLOW < total) break;
|
1293
|
+
}
|
1294
|
+
|
1295
|
+
free_count(sent);
|
1296
|
+
free_fast_matcher(sent);
|
1297
|
+
}
|
1298
|
+
|
1299
|
+
int sentence_parse(Sentence sent, Parse_Options opts)
|
1300
|
+
{
|
1301
|
+
int rc;
|
1302
|
+
|
1303
|
+
verbosity = opts->verbosity;
|
1304
|
+
|
1305
|
+
/* If the sentence has not yet been split, do so now.
|
1306
|
+
* This is for backwards compatibility, for existing programs
|
1307
|
+
* that do not explicitly call the splitter.
|
1308
|
+
*/
|
1309
|
+
if (0 == sent->length)
|
1310
|
+
{
|
1311
|
+
rc = sentence_split(sent, opts);
|
1312
|
+
if (rc) return -1;
|
1313
|
+
}
|
1314
|
+
|
1315
|
+
/* Check for bad sentence length */
|
1316
|
+
if (MAX_SENTENCE <= sent->length)
|
1317
|
+
{
|
1318
|
+
prt_error("Error: sentence too long, contains more than %d words\n",
|
1319
|
+
MAX_SENTENCE);
|
1320
|
+
return -2;
|
1321
|
+
}
|
1322
|
+
|
1323
|
+
/* Initialize/free any leftover garbage */
|
1324
|
+
free_sentence_disjuncts(sent);
|
1325
|
+
resources_reset_space(opts->resources);
|
1326
|
+
|
1327
|
+
if (resources_exhausted(opts->resources)) {
|
1328
|
+
sent->num_valid_linkages = 0;
|
1329
|
+
return 0;
|
1330
|
+
}
|
1331
|
+
|
1332
|
+
init_analyze(sent);
|
1333
|
+
|
1334
|
+
/* Expressions were previously set up during the tokenize stage. */
|
1335
|
+
expression_prune(sent);
|
1336
|
+
print_time(opts, "Finished expression pruning");
|
1337
|
+
if (opts->use_sat_solver)
|
1338
|
+
{
|
1339
|
+
sat_parse(sent, opts);
|
1340
|
+
}
|
1341
|
+
else
|
1342
|
+
{
|
1343
|
+
chart_parse(sent, opts);
|
1344
|
+
}
|
1345
|
+
print_time(opts, "Finished parse");
|
1346
|
+
|
1347
|
+
return sent->num_valid_linkages;
|
1348
|
+
}
|
1349
|
+
|
1350
|
+
/***************************************************************
|
1351
|
+
*
|
1352
|
+
* Routines which allow user access to Linkages.
|
1353
|
+
*
|
1354
|
+
****************************************************************/
|
1355
|
+
|
1356
|
+
Linkage linkage_create(int k, Sentence sent, Parse_Options opts)
|
1357
|
+
{
|
1358
|
+
Linkage linkage;
|
1359
|
+
|
1360
|
+
if (opts->use_sat_solver)
|
1361
|
+
{
|
1362
|
+
//return sat_create_linkage(k, sent, opts);
|
1363
|
+
}
|
1364
|
+
|
1365
|
+
if ((k >= sent->num_linkages_post_processed) || (k < 0)) return NULL;
|
1366
|
+
|
1367
|
+
/* Using exalloc since this is external to the parser itself. */
|
1368
|
+
linkage = (Linkage) exalloc(sizeof(struct Linkage_s));
|
1369
|
+
|
1370
|
+
linkage->num_words = sent->length;
|
1371
|
+
linkage->word = (const char **) exalloc(linkage->num_words*sizeof(char *));
|
1372
|
+
linkage->current = 0;
|
1373
|
+
linkage->num_sublinkages=0;
|
1374
|
+
linkage->sublinkage = NULL;
|
1375
|
+
linkage->unionized = FALSE;
|
1376
|
+
linkage->sent = sent;
|
1377
|
+
linkage->opts = opts;
|
1378
|
+
linkage->info = &sent->link_info[k];
|
1379
|
+
linkage->dis_con_tree = NULL;
|
1380
|
+
|
1381
|
+
extract_links(sent->link_info[k].index, sent->null_count, sent->parse_info);
|
1382
|
+
compute_chosen_words(sent, linkage);
|
1383
|
+
|
1384
|
+
if (set_has_fat_down(sent))
|
1385
|
+
{
|
1386
|
+
extract_fat_linkage(sent, opts, linkage);
|
1387
|
+
}
|
1388
|
+
else
|
1389
|
+
{
|
1390
|
+
extract_thin_linkage(sent, opts, linkage);
|
1391
|
+
}
|
1392
|
+
|
1393
|
+
if (sent->dict->postprocessor != NULL)
|
1394
|
+
{
|
1395
|
+
linkage_post_process(linkage, sent->dict->postprocessor);
|
1396
|
+
}
|
1397
|
+
|
1398
|
+
return linkage;
|
1399
|
+
}
|
1400
|
+
|
1401
|
+
int linkage_get_current_sublinkage(Linkage linkage) {
|
1402
|
+
return linkage->current;
|
1403
|
+
}
|
1404
|
+
|
1405
|
+
int linkage_set_current_sublinkage(Linkage linkage, int index)
|
1406
|
+
{
|
1407
|
+
if ((index < 0) ||
|
1408
|
+
(index >= linkage->num_sublinkages))
|
1409
|
+
{
|
1410
|
+
return 0;
|
1411
|
+
}
|
1412
|
+
linkage->current = index;
|
1413
|
+
return 1;
|
1414
|
+
}
|
1415
|
+
|
1416
|
+
static void exfree_pp_info(PP_info *ppi)
|
1417
|
+
{
|
1418
|
+
if (ppi->num_domains > 0)
|
1419
|
+
exfree(ppi->domain_name, sizeof(const char *)*ppi->num_domains);
|
1420
|
+
ppi->domain_name = NULL;
|
1421
|
+
ppi->num_domains = 0;
|
1422
|
+
}
|
1423
|
+
|
1424
|
+
void linkage_delete(Linkage linkage)
|
1425
|
+
{
|
1426
|
+
int i, j;
|
1427
|
+
Sublinkage *s;
|
1428
|
+
|
1429
|
+
/* Can happen on panic timeout or user error */
|
1430
|
+
if (NULL == linkage) return;
|
1431
|
+
|
1432
|
+
for (i=0; i<linkage->num_words; ++i)
|
1433
|
+
{
|
1434
|
+
exfree((void *) linkage->word[i], strlen(linkage->word[i])+1);
|
1435
|
+
}
|
1436
|
+
exfree(linkage->word, sizeof(char *)*linkage->num_words);
|
1437
|
+
|
1438
|
+
for (i=0; i<linkage->num_sublinkages; ++i)
|
1439
|
+
{
|
1440
|
+
s = &(linkage->sublinkage[i]);
|
1441
|
+
for (j=0; j<s->num_links; ++j) {
|
1442
|
+
exfree_link(s->link[j]);
|
1443
|
+
}
|
1444
|
+
exfree(s->link, sizeof(Link)*s->num_links);
|
1445
|
+
if (s->pp_info != NULL) {
|
1446
|
+
for (j=0; j<s->num_links; ++j) {
|
1447
|
+
exfree_pp_info(&s->pp_info[j]);
|
1448
|
+
}
|
1449
|
+
exfree(s->pp_info, sizeof(PP_info)*s->num_links);
|
1450
|
+
s->pp_info = NULL;
|
1451
|
+
post_process_free_data(&s->pp_data);
|
1452
|
+
}
|
1453
|
+
if (s->violation != NULL) {
|
1454
|
+
exfree((void *) s->violation, sizeof(char)*(strlen(s->violation)+1));
|
1455
|
+
}
|
1456
|
+
}
|
1457
|
+
exfree(linkage->sublinkage, sizeof(Sublinkage)*linkage->num_sublinkages);
|
1458
|
+
if (linkage->dis_con_tree)
|
1459
|
+
free_DIS_tree(linkage->dis_con_tree);
|
1460
|
+
exfree(linkage, sizeof(struct Linkage_s));
|
1461
|
+
}
|
1462
|
+
|
1463
|
+
static int links_are_equal(Link *l, Link *m)
|
1464
|
+
{
|
1465
|
+
return ((l->l == m->l) && (l->r == m->r) && (strcmp(l->name, m->name)==0));
|
1466
|
+
}
|
1467
|
+
|
1468
|
+
static int link_already_appears(Linkage linkage, Link *link, int a)
|
1469
|
+
{
|
1470
|
+
int i, j;
|
1471
|
+
|
1472
|
+
for (i=0; i<a; ++i) {
|
1473
|
+
for (j=0; j<linkage->sublinkage[i].num_links; ++j) {
|
1474
|
+
if (links_are_equal(linkage->sublinkage[i].link[j], link)) return TRUE;
|
1475
|
+
}
|
1476
|
+
}
|
1477
|
+
return FALSE;
|
1478
|
+
}
|
1479
|
+
|
1480
|
+
static PP_info excopy_pp_info(PP_info ppi)
|
1481
|
+
{
|
1482
|
+
PP_info newppi;
|
1483
|
+
int i;
|
1484
|
+
|
1485
|
+
newppi.num_domains = ppi.num_domains;
|
1486
|
+
newppi.domain_name = (const char **) exalloc(sizeof(const char *)*ppi.num_domains);
|
1487
|
+
for (i=0; i<newppi.num_domains; ++i)
|
1488
|
+
{
|
1489
|
+
newppi.domain_name[i] = ppi.domain_name[i];
|
1490
|
+
}
|
1491
|
+
return newppi;
|
1492
|
+
}
|
1493
|
+
|
1494
|
+
|
1495
|
+
static Sublinkage unionize_linkage(Linkage linkage)
|
1496
|
+
{
|
1497
|
+
int i, j, num_in_union=0;
|
1498
|
+
Sublinkage u;
|
1499
|
+
Link *link;
|
1500
|
+
const char *p;
|
1501
|
+
|
1502
|
+
for (i=0; i<linkage->num_sublinkages; ++i) {
|
1503
|
+
for (j=0; j<linkage->sublinkage[i].num_links; ++j) {
|
1504
|
+
link = linkage->sublinkage[i].link[j];
|
1505
|
+
if (!link_already_appears(linkage, link, i)) num_in_union++;
|
1506
|
+
}
|
1507
|
+
}
|
1508
|
+
|
1509
|
+
u.link = (Link **) exalloc(sizeof(Link *)*num_in_union);
|
1510
|
+
u.num_links = num_in_union;
|
1511
|
+
zero_sublinkage(&u);
|
1512
|
+
|
1513
|
+
u.pp_info = (PP_info *) exalloc(sizeof(PP_info)*num_in_union);
|
1514
|
+
u.violation = NULL;
|
1515
|
+
u.num_links = num_in_union;
|
1516
|
+
|
1517
|
+
num_in_union = 0;
|
1518
|
+
|
1519
|
+
for (i=0; i<linkage->num_sublinkages; ++i) {
|
1520
|
+
for (j=0; j<linkage->sublinkage[i].num_links; ++j) {
|
1521
|
+
link = linkage->sublinkage[i].link[j];
|
1522
|
+
if (!link_already_appears(linkage, link, i)) {
|
1523
|
+
u.link[num_in_union] = excopy_link(link);
|
1524
|
+
u.pp_info[num_in_union] = excopy_pp_info(linkage->sublinkage[i].pp_info[j]);
|
1525
|
+
if (((p=linkage->sublinkage[i].violation) != NULL) &&
|
1526
|
+
(u.violation == NULL)) {
|
1527
|
+
char *s = (char *) exalloc((strlen(p)+1)*sizeof(char));
|
1528
|
+
strcpy(s, p);
|
1529
|
+
u.violation = s;
|
1530
|
+
}
|
1531
|
+
num_in_union++;
|
1532
|
+
}
|
1533
|
+
}
|
1534
|
+
}
|
1535
|
+
|
1536
|
+
return u;
|
1537
|
+
}
|
1538
|
+
|
1539
|
+
int linkage_compute_union(Linkage linkage)
|
1540
|
+
{
|
1541
|
+
int i, num_subs=linkage->num_sublinkages;
|
1542
|
+
Sublinkage * new_sublinkage, *s;
|
1543
|
+
|
1544
|
+
if (linkage->unionized) {
|
1545
|
+
linkage->current = linkage->num_sublinkages-1;
|
1546
|
+
return 0;
|
1547
|
+
}
|
1548
|
+
if (num_subs == 1) {
|
1549
|
+
linkage->unionized = TRUE;
|
1550
|
+
return 1;
|
1551
|
+
}
|
1552
|
+
|
1553
|
+
new_sublinkage =
|
1554
|
+
(Sublinkage *) exalloc(sizeof(Sublinkage)*(num_subs+1));
|
1555
|
+
|
1556
|
+
for (i=0; i<num_subs; ++i) {
|
1557
|
+
new_sublinkage[i] = linkage->sublinkage[i];
|
1558
|
+
}
|
1559
|
+
exfree(linkage->sublinkage, sizeof(Sublinkage)*num_subs);
|
1560
|
+
linkage->sublinkage = new_sublinkage;
|
1561
|
+
|
1562
|
+
/* Zero out the new sublinkage, then unionize it. */
|
1563
|
+
s = &new_sublinkage[num_subs];
|
1564
|
+
s->link = NULL;
|
1565
|
+
s->num_links = 0;
|
1566
|
+
zero_sublinkage(s);
|
1567
|
+
linkage->sublinkage[num_subs] = unionize_linkage(linkage);
|
1568
|
+
|
1569
|
+
linkage->num_sublinkages++;
|
1570
|
+
|
1571
|
+
linkage->unionized = TRUE;
|
1572
|
+
linkage->current = linkage->num_sublinkages-1;
|
1573
|
+
return 1;
|
1574
|
+
}
|
1575
|
+
|
1576
|
+
int linkage_get_num_sublinkages(Linkage linkage) {
|
1577
|
+
return linkage->num_sublinkages;
|
1578
|
+
}
|
1579
|
+
|
1580
|
+
int linkage_get_num_words(Linkage linkage)
|
1581
|
+
{
|
1582
|
+
return linkage->num_words;
|
1583
|
+
}
|
1584
|
+
|
1585
|
+
int linkage_get_num_links(Linkage linkage)
|
1586
|
+
{
|
1587
|
+
int current = linkage->current;
|
1588
|
+
return linkage->sublinkage[current].num_links;
|
1589
|
+
}
|
1590
|
+
|
1591
|
+
static inline int verify_link_index(Linkage linkage, int index)
|
1592
|
+
{
|
1593
|
+
if ((index < 0) ||
|
1594
|
+
(index >= linkage->sublinkage[linkage->current].num_links))
|
1595
|
+
{
|
1596
|
+
return 0;
|
1597
|
+
}
|
1598
|
+
return 1;
|
1599
|
+
}
|
1600
|
+
|
1601
|
+
int linkage_get_link_length(Linkage linkage, int index)
|
1602
|
+
{
|
1603
|
+
Link *link;
|
1604
|
+
int word_has_link[MAX_SENTENCE];
|
1605
|
+
int i, length;
|
1606
|
+
int current = linkage->current;
|
1607
|
+
|
1608
|
+
if (!verify_link_index(linkage, index)) return -1;
|
1609
|
+
|
1610
|
+
for (i=0; i<linkage->num_words+1; ++i) {
|
1611
|
+
word_has_link[i] = FALSE;
|
1612
|
+
}
|
1613
|
+
|
1614
|
+
for (i=0; i<linkage->sublinkage[current].num_links; ++i) {
|
1615
|
+
link = linkage->sublinkage[current].link[i];
|
1616
|
+
word_has_link[link->l] = TRUE;
|
1617
|
+
word_has_link[link->r] = TRUE;
|
1618
|
+
}
|
1619
|
+
|
1620
|
+
link = linkage->sublinkage[current].link[index];
|
1621
|
+
length = link->r - link->l;
|
1622
|
+
for (i= link->l+1; i < link->r; ++i) {
|
1623
|
+
if (!word_has_link[i]) length--;
|
1624
|
+
}
|
1625
|
+
return length;
|
1626
|
+
}
|
1627
|
+
|
1628
|
+
int linkage_get_link_lword(Linkage linkage, int index)
|
1629
|
+
{
|
1630
|
+
Link *link;
|
1631
|
+
if (!verify_link_index(linkage, index)) return -1;
|
1632
|
+
link = linkage->sublinkage[linkage->current].link[index];
|
1633
|
+
return link->l;
|
1634
|
+
}
|
1635
|
+
|
1636
|
+
int linkage_get_link_rword(Linkage linkage, int index)
|
1637
|
+
{
|
1638
|
+
Link *link;
|
1639
|
+
if (!verify_link_index(linkage, index)) return -1;
|
1640
|
+
link = linkage->sublinkage[linkage->current].link[index];
|
1641
|
+
return link->r;
|
1642
|
+
}
|
1643
|
+
|
1644
|
+
const char * linkage_get_link_label(Linkage linkage, int index)
|
1645
|
+
{
|
1646
|
+
Link *link;
|
1647
|
+
if (!verify_link_index(linkage, index)) return NULL;
|
1648
|
+
link = linkage->sublinkage[linkage->current].link[index];
|
1649
|
+
return link->name;
|
1650
|
+
}
|
1651
|
+
|
1652
|
+
const char * linkage_get_link_llabel(Linkage linkage, int index)
|
1653
|
+
{
|
1654
|
+
Link *link;
|
1655
|
+
if (!verify_link_index(linkage, index)) return NULL;
|
1656
|
+
link = linkage->sublinkage[linkage->current].link[index];
|
1657
|
+
return link->lc->string;
|
1658
|
+
}
|
1659
|
+
|
1660
|
+
const char * linkage_get_link_rlabel(Linkage linkage, int index)
|
1661
|
+
{
|
1662
|
+
Link *link;
|
1663
|
+
if (!verify_link_index(linkage, index)) return NULL;
|
1664
|
+
link = linkage->sublinkage[linkage->current].link[index];
|
1665
|
+
return link->rc->string;
|
1666
|
+
}
|
1667
|
+
|
1668
|
+
const char ** linkage_get_words(Linkage linkage)
|
1669
|
+
{
|
1670
|
+
return linkage->word;
|
1671
|
+
}
|
1672
|
+
|
1673
|
+
Sentence linkage_get_sentence(Linkage linkage)
|
1674
|
+
{
|
1675
|
+
return linkage->sent;
|
1676
|
+
}
|
1677
|
+
|
1678
|
+
const char * linkage_get_disjunct_str(Linkage linkage, int w)
|
1679
|
+
{
|
1680
|
+
Disjunct *dj;
|
1681
|
+
|
1682
|
+
if (NULL == linkage->info->disjunct_list_str)
|
1683
|
+
{
|
1684
|
+
lg_compute_disjunct_strings(linkage->sent, linkage->info);
|
1685
|
+
}
|
1686
|
+
|
1687
|
+
/* dj will be null if the word wasn't used in the parse. */
|
1688
|
+
dj = linkage->sent->parse_info->chosen_disjuncts[w];
|
1689
|
+
if (NULL == dj) return "";
|
1690
|
+
|
1691
|
+
return linkage->info->disjunct_list_str[w];
|
1692
|
+
}
|
1693
|
+
|
1694
|
+
double linkage_get_disjunct_cost(Linkage linkage, int w)
|
1695
|
+
{
|
1696
|
+
Disjunct *dj = linkage->sent->parse_info->chosen_disjuncts[w];
|
1697
|
+
|
1698
|
+
/* dj may be null, if the word didn't participate in the parse. */
|
1699
|
+
if (dj) return dj->cost;
|
1700
|
+
return 0.0;
|
1701
|
+
}
|
1702
|
+
|
1703
|
+
double linkage_get_disjunct_corpus_score(Linkage linkage, int w)
|
1704
|
+
{
|
1705
|
+
Disjunct *dj = linkage->sent->parse_info->chosen_disjuncts[w];
|
1706
|
+
|
1707
|
+
/* dj may be null, if the word didn't participate in the parse. */
|
1708
|
+
if (NULL == dj) return 99.999;
|
1709
|
+
|
1710
|
+
return lg_corpus_disjunct_score(linkage, w);
|
1711
|
+
}
|
1712
|
+
|
1713
|
+
const char * linkage_get_word(Linkage linkage, int w)
|
1714
|
+
{
|
1715
|
+
return linkage->word[w];
|
1716
|
+
}
|
1717
|
+
|
1718
|
+
int linkage_unused_word_cost(Linkage linkage)
|
1719
|
+
{
|
1720
|
+
/* The sat solver (currently) fails to fill in info */
|
1721
|
+
if (!linkage->info) return 0;
|
1722
|
+
return linkage->info->unused_word_cost;
|
1723
|
+
}
|
1724
|
+
|
1725
|
+
int linkage_disjunct_cost(Linkage linkage)
|
1726
|
+
{
|
1727
|
+
/* The sat solver (currently) fails to fill in info */
|
1728
|
+
if (!linkage->info) return 0;
|
1729
|
+
return (int) floorf(linkage->info->disjunct_cost);
|
1730
|
+
}
|
1731
|
+
|
1732
|
+
int linkage_is_fat(Linkage linkage)
|
1733
|
+
{
|
1734
|
+
/* The sat solver (currently) fails to fill in info */
|
1735
|
+
if (!linkage->info) return 0;
|
1736
|
+
return linkage->info->fat;
|
1737
|
+
}
|
1738
|
+
|
1739
|
+
int linkage_and_cost(Linkage linkage)
|
1740
|
+
{
|
1741
|
+
/* The sat solver (currently) fails to fill in info */
|
1742
|
+
if (!linkage->info) return 0;
|
1743
|
+
return linkage->info->and_cost;
|
1744
|
+
}
|
1745
|
+
|
1746
|
+
int linkage_link_cost(Linkage linkage)
|
1747
|
+
{
|
1748
|
+
/* The sat solver (currently) fails to fill in info */
|
1749
|
+
if (!linkage->info) return 0;
|
1750
|
+
return linkage->info->link_cost;
|
1751
|
+
}
|
1752
|
+
|
1753
|
+
double linkage_corpus_cost(Linkage linkage)
|
1754
|
+
{
|
1755
|
+
/* The sat solver (currently) fails to fill in info */
|
1756
|
+
if (!linkage->info) return 0.0;
|
1757
|
+
return linkage->info->corpus_cost;
|
1758
|
+
}
|
1759
|
+
|
1760
|
+
int linkage_get_link_num_domains(Linkage linkage, int index)
|
1761
|
+
{
|
1762
|
+
PP_info *pp_info;
|
1763
|
+
if (!verify_link_index(linkage, index)) return -1;
|
1764
|
+
pp_info = &linkage->sublinkage[linkage->current].pp_info[index];
|
1765
|
+
return pp_info->num_domains;
|
1766
|
+
}
|
1767
|
+
|
1768
|
+
const char ** linkage_get_link_domain_names(Linkage linkage, int index)
|
1769
|
+
{
|
1770
|
+
PP_info *pp_info;
|
1771
|
+
if (!verify_link_index(linkage, index)) return NULL;
|
1772
|
+
pp_info = &linkage->sublinkage[linkage->current].pp_info[index];
|
1773
|
+
return pp_info->domain_name;
|
1774
|
+
}
|
1775
|
+
|
1776
|
+
const char * linkage_get_violation_name(Linkage linkage)
|
1777
|
+
{
|
1778
|
+
return linkage->sublinkage[linkage->current].violation;
|
1779
|
+
}
|
1780
|
+
|
1781
|
+
int linkage_is_canonical(Linkage linkage)
|
1782
|
+
{
|
1783
|
+
/* The sat solver (currently) fails to fill in info */
|
1784
|
+
if (!linkage->info) return TRUE;
|
1785
|
+
return linkage->info->canonical;
|
1786
|
+
}
|
1787
|
+
|
1788
|
+
int linkage_is_improper(Linkage linkage)
|
1789
|
+
{
|
1790
|
+
/* The sat solver (currently) fails to fill in info */
|
1791
|
+
if (!linkage->info) return FALSE;
|
1792
|
+
return linkage->info->improper_fat_linkage;
|
1793
|
+
}
|
1794
|
+
|
1795
|
+
int linkage_has_inconsistent_domains(Linkage linkage)
|
1796
|
+
{
|
1797
|
+
/* The sat solver (currently) fails to fill in info */
|
1798
|
+
if (!linkage->info) return FALSE;
|
1799
|
+
return linkage->info->inconsistent_domains;
|
1800
|
+
}
|
1801
|
+
|
1802
|
+
void linkage_post_process(Linkage linkage, Postprocessor * postprocessor)
|
1803
|
+
{
|
1804
|
+
int N_sublinkages = linkage_get_num_sublinkages(linkage);
|
1805
|
+
Parse_Options opts = linkage->opts;
|
1806
|
+
Sentence sent = linkage->sent;
|
1807
|
+
Sublinkage * subl;
|
1808
|
+
PP_node * pp;
|
1809
|
+
int i, j, k;
|
1810
|
+
D_type_list * d;
|
1811
|
+
|
1812
|
+
for (i = 0; i < N_sublinkages; ++i)
|
1813
|
+
{
|
1814
|
+
subl = &linkage->sublinkage[i];
|
1815
|
+
if (subl->pp_info != NULL)
|
1816
|
+
{
|
1817
|
+
for (j = 0; j < subl->num_links; ++j)
|
1818
|
+
{
|
1819
|
+
exfree_pp_info(&subl->pp_info[j]);
|
1820
|
+
}
|
1821
|
+
post_process_free_data(&subl->pp_data);
|
1822
|
+
exfree(subl->pp_info, sizeof(PP_info)*subl->num_links);
|
1823
|
+
}
|
1824
|
+
subl->pp_info = (PP_info *) exalloc(sizeof(PP_info)*subl->num_links);
|
1825
|
+
for (j = 0; j < subl->num_links; ++j)
|
1826
|
+
{
|
1827
|
+
subl->pp_info[j].num_domains = 0;
|
1828
|
+
subl->pp_info[j].domain_name = NULL;
|
1829
|
+
}
|
1830
|
+
if (subl->violation != NULL)
|
1831
|
+
{
|
1832
|
+
exfree((void *)subl->violation, sizeof(char)*(strlen(subl->violation)+1));
|
1833
|
+
subl->violation = NULL;
|
1834
|
+
}
|
1835
|
+
|
1836
|
+
if (linkage->info->improper_fat_linkage)
|
1837
|
+
{
|
1838
|
+
pp = NULL;
|
1839
|
+
}
|
1840
|
+
else
|
1841
|
+
{
|
1842
|
+
pp = post_process(postprocessor, opts, sent, subl, FALSE);
|
1843
|
+
/* This can return NULL, for example if there is no
|
1844
|
+
post-processor */
|
1845
|
+
}
|
1846
|
+
|
1847
|
+
if (pp == NULL)
|
1848
|
+
{
|
1849
|
+
for (j = 0; j < subl->num_links; ++j)
|
1850
|
+
{
|
1851
|
+
subl->pp_info[j].num_domains = 0;
|
1852
|
+
subl->pp_info[j].domain_name = NULL;
|
1853
|
+
}
|
1854
|
+
}
|
1855
|
+
else
|
1856
|
+
{
|
1857
|
+
for (j = 0; j < subl->num_links; ++j)
|
1858
|
+
{
|
1859
|
+
k = 0;
|
1860
|
+
for (d = pp->d_type_array[j]; d != NULL; d = d->next) k++;
|
1861
|
+
subl->pp_info[j].num_domains = k;
|
1862
|
+
if (k > 0)
|
1863
|
+
{
|
1864
|
+
subl->pp_info[j].domain_name = (const char **) exalloc(sizeof(const char *)*k);
|
1865
|
+
}
|
1866
|
+
k = 0;
|
1867
|
+
for (d = pp->d_type_array[j]; d != NULL; d = d->next)
|
1868
|
+
{
|
1869
|
+
char buff[5];
|
1870
|
+
sprintf(buff, "%c", d->type);
|
1871
|
+
subl->pp_info[j].domain_name[k] = string_set_add (buff, sent->string_set);
|
1872
|
+
|
1873
|
+
k++;
|
1874
|
+
}
|
1875
|
+
}
|
1876
|
+
subl->pp_data = postprocessor->pp_data;
|
1877
|
+
if (pp->violation != NULL)
|
1878
|
+
{
|
1879
|
+
char * s = (char *) exalloc(sizeof(char)*(strlen(pp->violation)+1));
|
1880
|
+
strcpy(s, pp->violation);
|
1881
|
+
subl->violation = s;
|
1882
|
+
}
|
1883
|
+
}
|
1884
|
+
}
|
1885
|
+
post_process_close_sentence(postprocessor);
|
1886
|
+
}
|
1887
|
+
#endif
|