From 20edcc90470b4e719da1d544dbe49b410efa5983 Mon Sep 17 00:00:00 2001
From: omekr <roman.kurc@pwr.wroc.pl>
Date: Fri, 5 Aug 2011 13:10:39 +0200
Subject: [PATCH] test against converted multiword units
---
libmwereader/mwe.cpp | 15 +++++++++++----
libmwereader/mweparser.cpp | 17 +++++++++++------
libmwereader/mwereader.cpp | 3 +++
libmwereader/tests/mwefunctional.cpp | 2 ++
4 files changed, 27 insertions(+), 10 deletions(-)
diff --git a/libmwereader/mwe.cpp b/libmwereader/mwe.cpp
index 4fcc72b..9397a4f 100644
--- a/libmwereader/mwe.cpp
+++ b/libmwereader/mwe.cpp
@@ -1,6 +1,7 @@
#include "mwe.h"
#include <boost/algorithm/string.hpp>
#include <libwccl/values/strset.h>
+#include <boost/algorithm/string/predicate.hpp>
namespace Corpus2{
@@ -27,11 +28,14 @@ LexicalUnit::LexicalUnit(const std::string &base,
bool LexicalUnit::IsHere(const Wccl::SentenceContext &sc,
std::set<int> &out_position, int &head_pos) const
{
- // set variables
+ // set variables, skip vars with names starting with '!'
for(variables_map::const_iterator ivars = variables_.begin();
- ivars != variables_.end(); ++ivars){
- condition_->set<Wccl::StrSet>(ivars->first, ivars->second);
- }
+ ivars != variables_.end(); ++ivars){
+ if(!boost::starts_with(ivars->first, "!")){
+ std::cout << ivars->first << " " << std::endl;
+ condition_->set<Wccl::StrSet>(ivars->first, ivars->second);
+ }
+ }
// fire up the operator
boost::shared_ptr<const Wccl::Bool> pResult = condition_->apply(sc);
@@ -102,6 +106,7 @@ void MWEIndex::add_lexicalunit(LexicalUnit::Ptr lu)
{
foreach(const std::string& base, lu->get_potential_bases()){
value_type::iterator find = index_.find(base);
+ std::cout << "b:"<<base<<std::endl;
if(find == index_.end()){ // not found -> create new one
luvec v;
v.push_back(lu);
@@ -113,6 +118,8 @@ void MWEIndex::add_lexicalunit(LexicalUnit::Ptr lu)
}
const MWEIndex::luvec& MWEIndex::get_potential_lu(const std::string &base){
+ std::cout << "index " << index_.size()<< std::endl;
+ std::cout << "sb:"<<base<<std::endl;
value_type::iterator find = index_.find(base);
if(find == index_.end()){ // not found -> return empty
return empty_;
diff --git a/libmwereader/mweparser.cpp b/libmwereader/mweparser.cpp
index 1edb9c6..5d7534f 100644
--- a/libmwereader/mweparser.cpp
+++ b/libmwereader/mweparser.cpp
@@ -34,9 +34,9 @@ namespace Corpus2 {
if(search != where.end())
return search->second;
-
+ //std::cout << " dddddddddddddd "<< cond << std::endl;
BoolOpPtr op = parser_.parseBoolOperator(cond);
-
+ //std::cout << " dddddddddddddd $$$$" << cond << std::endl;
where[cond] = op;
return op;
@@ -57,24 +57,28 @@ namespace Corpus2 {
void MWEParser::create_mwe()
{
print_current_mwe(true);
+ //std::cout << " kupa cond" << std::endl;
MWEBuilder::BoolOpPtr main = mwe_builder_->get_mwe_condition(
wccl_operator_);
+ //std::cout << " kupa head" << std::endl;
MWEBuilder::BoolOpPtr head = mwe_builder_->get_head_condition(
head_cond_);
-
+ //std::cout << " kupa " << std::endl;
if(group_type_ == "fix"){ // group_name_ -> lower case
-
+ //std::cout << " kupa fix" << std::endl;
mwe_index_.add_lexicalunit( LexicalUnit::Ptr(new FixedLU(mwe_base_, main, head,
variables_)));
} else if(group_type_ == "flex"){
+ //std::cout << " kupa flex" << std::endl;
mwe_index_.add_lexicalunit(LexicalUnit::Ptr(new FlexLU(mwe_base_, main, head,
variables_)));
} else {
throw Wccl::WcclError("Unknown type of lexical unit:"
+ group_type_);
}
-
+ //std::cout << " kupa clear" << std::endl;
variables_.clear();
+ //std::cout << "po kupie " << std::endl;
}
std::string MWEParser::get_attribute(const AttributeList& attributes,
@@ -109,8 +113,9 @@ namespace Corpus2 {
void MWEParser::on_start_element(const Glib::ustring &name,
const AttributeList& attributes)
{
+ std::cout << "about to check" << std::endl;
std::cout << state_ << ": " << name << std::endl;
-
+ std::cout << "done with check" << std::endl;
if(state_ == NONE && name == "units_description"){
tagset_ = get_attribute(attributes, "tagset");
mwe_builder_ = boost::shared_ptr<MWEBuilder>(new MWEBuilder(Corpus2::get_named_tagset(tagset_)));
diff --git a/libmwereader/mwereader.cpp b/libmwereader/mwereader.cpp
index 83739ee..90b3bcd 100644
--- a/libmwereader/mwereader.cpp
+++ b/libmwereader/mwereader.cpp
@@ -47,10 +47,13 @@ bool MWEReader::registered = TokenReader::register_path_reader<MWEReader>(
if(lex.is_disamb()){
std::string base = lex.lemma_utf8();
const MWEIndex::luvec& potential = mwe_index_.get_potential_lu(base);
+ std::cout << "potential " << potential.size() << std::endl;
foreach(LexicalUnit::Ptr pLU, potential){
std::set<int> positions;
int head;
+ //std::cout << " is " << std::endl;
bool is_here = pLU->IsHere(sc, positions, head);
+ //std::cout << " is out" << std::endl;
if(is_here){
std::string new_orth_utf8;
Corpus2::Sentence::Ptr sent = sc.get_sentence_ptr();
diff --git a/libmwereader/tests/mwefunctional.cpp b/libmwereader/tests/mwefunctional.cpp
index 141d3fc..5e0b082 100644
--- a/libmwereader/tests/mwefunctional.cpp
+++ b/libmwereader/tests/mwefunctional.cpp
@@ -31,6 +31,8 @@ struct Fixture{
};
+
+
BOOST_FIXTURE_TEST_CASE( preferred_lexeme, Fixture)
{
BOOST_MESSAGE("test: finding preferred lexeme");
--
GitLab