From 340bbcacb00fc7c6cca5ec343d3b4805334b4e0f Mon Sep 17 00:00:00 2001
From: Adam Wardynski <award@.(win7-laptop)>
Date: Thu, 9 Dec 2010 21:03:00 +0100
Subject: [PATCH] Rework strong agreement operator. Clearer and better
performing version, also more likely to actually work as expected :)
---
.../bool/predicates/strongagreement.cpp | 54 +++++++++++--------
1 file changed, 33 insertions(+), 21 deletions(-)
diff --git a/libwccl/ops/functions/bool/predicates/strongagreement.cpp b/libwccl/ops/functions/bool/predicates/strongagreement.cpp
index 4081562..a29be8d 100644
--- a/libwccl/ops/functions/bool/predicates/strongagreement.cpp
+++ b/libwccl/ops/functions/bool/predicates/strongagreement.cpp
@@ -52,34 +52,46 @@ StrongAgreement::BaseRetValPtr StrongAgreement::apply_internal(const FunExecCont
int min_card = attribs->categories_count(tagset_);
- for(int i = abs_left; i <= abs_right; ++i) {
- bool i_has_matched_tag = false;
- foreach (const Corpus2::Lexeme& i_lex, sc.at(i)->lexemes()) {
- const Corpus2::Tag& i_tag = i_lex.tag();
- if (attribs->matching_categories(i_tag) >= min_card) {
- i_has_matched_tag = true;
- for(int j = abs_right; j > i; --j) {
- bool i_agrees_with_j = false;
- foreach(const Corpus2::Lexeme& j_lex, sc.at(j)->lexemes()) {
- Corpus2::Tag intersection = i_tag.get_masked(j_lex.tag());
- // if the intersection matches enough categories we have agreement
- if (attribs->matching_categories(intersection) >= min_card) {
- i_agrees_with_j = true;
- break;
+ const Corpus2::Token* t1 = sc.at(abs_left);
+ const Corpus2::Token* t2 = sc.at(abs_right);
+ // to optimize a bit, make sure t1 is the one with less lexemes
+ if (t1->lexemes().size() > t2->lexemes().size()) {
+ std::swap(t1, t2);
+ }
+ // check strong agreement between range endpoints
+ // for each possible agreement between them,
+ // check if remaining tokens meet that agreement too
+ // return true if an agreement met by all tokens is found
+ foreach (const Corpus2::Lexeme& t1_lex, t1->lexemes()) {
+ const Corpus2::Tag& t1_tag = t1_lex.tag();
+ // don't bother checking t2 unless current t1_tag matches enough categories
+ if (attribs->matching_categories(t1_tag) >= min_card) {
+ foreach (const Corpus2::Lexeme& t2_lex, t2->lexemes()) {
+ Corpus2::Tag inter = t1_tag.get_masked(t2_lex.tag());
+ // if the intersection matches enough categories we have agreement
+ if (attribs->matching_categories(inter) >= min_card) {
+ // check if selected agreement is met by all remaining tokens
+ bool agreement_met = true;
+ for(int i = abs_left + 1; agreement_met && (i < abs_right); ++i) {
+ agreement_met = false;
+ foreach(const Corpus2::Lexeme& i_lex, sc.at(i)->lexemes()) {
+ Corpus2::Tag i_inter = i_lex.tag().get_masked(inter);
+ if (attribs->matching_categories(i_inter) < min_card) {
+ agreement_met = true;
+ break;
+ }
}
}
- if (!i_agrees_with_j) {
- return Predicate::False(context);
+ if (agreement_met) {
+ return Predicate::True(context);
}
}
}
}
- if (!i_has_matched_tag) {
- return Predicate::False(context);
- }
}
-
- return Predicate::True(context);
+ // None of possible agreements between endpoints carried over to all tokens
+ // or there was no such possible agreement at all, so return False.
+ return Predicate::False(context);
}
} /* end ns Wccl */
--
GitLab