From 1270ffe8237eff041bd8745f503e3e90982402c3 Mon Sep 17 00:00:00 2001
From: Adam Wardynski <award@.(win7-laptop)>
Date: Tue, 26 Apr 2011 16:50:25 +0200
Subject: [PATCH] Fix MatchVector::first_token to ignore negative values. It
was looking for minimum value and Position::Nowhere is the smallest you can
get value-wise (min_int). So if there was an empty Match in a vector,
first_token for the vector would always end up returning Position::Nowhere,
which was wrong (empty submatches should've been ignored and
Position::Nowhere returned only if there were no non-empty alternatives).
---
libwccl/values/matchvector.cpp | 13 +++++++++++--
tests/match.cpp | 12 ++++++++++++
2 files changed, 23 insertions(+), 2 deletions(-)
diff --git a/libwccl/values/matchvector.cpp b/libwccl/values/matchvector.cpp
index bf80136..6415ed0 100644
--- a/libwccl/values/matchvector.cpp
+++ b/libwccl/values/matchvector.cpp
@@ -27,12 +27,21 @@ Position MatchVector::first_token(const boost::shared_ptr<Corpus2::AnnotatedSent
if (matches_.empty()) {
return Position(Position::Nowhere);
} else {
+ // Negative positions are invalid, including specials like Nowhere,
+ // so we can't just find minimum value but minimum *non-negative* value.
+ // Note: yes, the code assumes the special values like Nowhere are indeed negative.
Position p = matches_.front()->first_token(s);
- for (size_t i = 1; i < matches_.size(); ++i) {
+ size_t i = 1;
+ while ((p.get_value() < 0) && (i < matches_.size())) {
+ p = matches_[i]->first_token(s);
+ ++i;
+ }
+ while (i < matches_.size()) {
Position c = matches_[i]->first_token(s);
- if (c.get_value() < p.get_value()) {
+ if ((c.get_value() >= 0) && (c.get_value() < p.get_value())) {
p = c;
}
+ ++i;
}
return p;
}
diff --git a/tests/match.cpp b/tests/match.cpp
index 72ea47d..9b3ab09 100644
--- a/tests/match.cpp
+++ b/tests/match.cpp
@@ -98,6 +98,18 @@ BOOST_AUTO_TEST_CASE(matchvector_first_last)
m1->append(boost::make_shared<TokenMatch>(1));
BOOST_CHECK_EQUAL(m2.first_token(ptr).get_value(), 1);
BOOST_CHECK_EQUAL(m2.last_token(ptr).get_value(), 7);
+ Wccl::MatchVector m3;
+ BOOST_CHECK_EQUAL(m3.to_raw_string(), "MATCH()");
+ BOOST_CHECK_EQUAL(m3.first_token(ptr).get_value(), Wccl::Position::Nowhere);
+ BOOST_CHECK_EQUAL(m3.last_token(ptr).get_value(), Wccl::Position::Nowhere);
+ m3.append(boost::make_shared<MatchVector>());
+ BOOST_CHECK_EQUAL(m3.to_raw_string(), "MATCH(MATCH())");
+ BOOST_CHECK_EQUAL(m3.first_token(ptr).get_value(), Wccl::Position::Nowhere);
+ BOOST_CHECK_EQUAL(m3.last_token(ptr).get_value(), Wccl::Position::Nowhere);
+ m3.append(boost::make_shared<TokenMatch>(1));
+ BOOST_CHECK_EQUAL(m3.to_raw_string(), "MATCH(MATCH(),TOK[1])");
+ BOOST_CHECK_EQUAL(m3.first_token(ptr).get_value(), 1);
+ BOOST_CHECK_EQUAL(m3.last_token(ptr).get_value(), 1);
}
BOOST_AUTO_TEST_CASE(varmatch)
--
GitLab