From f4befcf1258ecd418834ceab644b8154aca04cab Mon Sep 17 00:00:00 2001
From: Hieu Hoang <hieuhoang@gmail.com>
Date: Wed, 19 Mar 2014 12:59:55 +0000
Subject: [PATCH] probing pt source vocab

---
 contrib/other-builds/moses/.cproject          | 13 ++++++-----
 .../TranslationModel/ProbingPT/ProbingPT.cpp  | 22 +++++++++++++++----
 moses/TranslationModel/ProbingPT/ProbingPT.h  |  7 ++++--
 3 files changed, 31 insertions(+), 11 deletions(-)

diff --git a/contrib/other-builds/moses/.cproject b/contrib/other-builds/moses/.cproject
index 8840f82bb..8a4410c16 100644
--- a/contrib/other-builds/moses/.cproject
+++ b/contrib/other-builds/moses/.cproject
@@ -11,12 +11,12 @@
 					</externalSetting>
 				</externalSettings>
 				<extensions>
-					<extension id="org.eclipse.cdt.core.ELF" point="org.eclipse.cdt.core.BinaryParser"/>
-					<extension id="org.eclipse.cdt.core.MachO64" point="org.eclipse.cdt.core.BinaryParser"/>
 					<extension id="org.eclipse.cdt.core.GmakeErrorParser" point="org.eclipse.cdt.core.ErrorParser"/>
 					<extension id="org.eclipse.cdt.core.CWDLocator" point="org.eclipse.cdt.core.ErrorParser"/>
 					<extension id="org.eclipse.cdt.core.GCCErrorParser" point="org.eclipse.cdt.core.ErrorParser"/>
 					<extension id="org.eclipse.cdt.core.GASErrorParser" point="org.eclipse.cdt.core.ErrorParser"/>
+					<extension id="org.eclipse.cdt.core.ELF" point="org.eclipse.cdt.core.BinaryParser"/>
+					<extension id="org.eclipse.cdt.core.MachO64" point="org.eclipse.cdt.core.BinaryParser"/>
 				</extensions>
 			</storageModule>
 			<storageModule moduleId="cdtBuildSystem" version="4.0.0">
@@ -79,8 +79,11 @@
 					</folderInfo>
 					<fileInfo id="cdt.managedbuild.config.gnu.exe.debug.656913512.511477442" name="Rand.h" rcbsApplicability="disable" resourcePath="LM/Rand.h" toolsToInvoke=""/>
 					<fileInfo id="cdt.managedbuild.config.gnu.exe.debug.656913512.790052015" name="IRST.h" rcbsApplicability="disable" resourcePath="LM/IRST.h" toolsToInvoke=""/>
+					<fileInfo id="cdt.managedbuild.config.gnu.exe.debug.656913512.1228569547" name="FuzzyMatchWrapper.cpp" rcbsApplicability="disable" resourcePath="TranslationModel/fuzzy-match/FuzzyMatchWrapper.cpp" toolsToInvoke="cdt.managedbuild.tool.gnu.cpp.compiler.exe.debug.1774992327.200046295">
+						<tool id="cdt.managedbuild.tool.gnu.cpp.compiler.exe.debug.1774992327.200046295" name="GCC C++ Compiler" superClass="cdt.managedbuild.tool.gnu.cpp.compiler.exe.debug.1774992327"/>
+					</fileInfo>
 					<sourceEntries>
-						<entry excluding="TranslationModel/CompactPT|LM/NeuralLMWrapper.cpp|FF/PhraseLengthFeatureTest.cpp|PhraseLengthFeatureTest.cpp|LM/BackwardTest.cpp|LM/BackwardLMState.h|LM/BackwardLMState.cpp|LM/Backward.h|LM/Backward.cpp|FeatureVectorTest.cpp|LM/ParallelBackoff.h|LM/ParallelBackoff.cpp|src/SyntacticLanguageModelState.h|src/SyntacticLanguageModelFiles.h|src/SyntacticLanguageModel.h|src/SyntacticLanguageModel.cpp|src/LM/SRI.h|src/LM/SRI.cpp|src/LM/Rand.h|src/LM/Rand.cpp|src/LM/LDHT.h|src/LM/LDHT.cpp|SyntacticLanguageModelState.h|SyntacticLanguageModelFiles.h|SyntacticLanguageModel.h|SyntacticLanguageModel.cpp|LM/LDHT.h|LM/LDHT.cpp" flags="VALUE_WORKSPACE_PATH|RESOLVED" kind="sourcePath" name=""/>
+						<entry excluding="TranslationModel/fuzzy-match/FuzzyMatchWrapper.cpp|TranslationModel/CompactPT|LM/NeuralLMWrapper.cpp|FF/PhraseLengthFeatureTest.cpp|PhraseLengthFeatureTest.cpp|LM/BackwardTest.cpp|LM/BackwardLMState.h|LM/BackwardLMState.cpp|LM/Backward.h|LM/Backward.cpp|FeatureVectorTest.cpp|LM/ParallelBackoff.h|LM/ParallelBackoff.cpp|src/SyntacticLanguageModelState.h|src/SyntacticLanguageModelFiles.h|src/SyntacticLanguageModel.h|src/SyntacticLanguageModel.cpp|src/LM/SRI.h|src/LM/SRI.cpp|src/LM/Rand.h|src/LM/Rand.cpp|src/LM/LDHT.h|src/LM/LDHT.cpp|SyntacticLanguageModelState.h|SyntacticLanguageModelFiles.h|SyntacticLanguageModel.h|SyntacticLanguageModel.cpp|LM/LDHT.h|LM/LDHT.cpp" flags="VALUE_WORKSPACE_PATH|RESOLVED" kind="sourcePath" name=""/>
 					</sourceEntries>
 				</configuration>
 			</storageModule>
@@ -90,13 +93,13 @@
 			<storageModule buildSystemId="org.eclipse.cdt.managedbuilder.core.configurationDataProvider" id="cdt.managedbuild.config.gnu.exe.release.401150096" moduleId="org.eclipse.cdt.core.settings" name="Release">
 				<externalSettings/>
 				<extensions>
-					<extension id="org.eclipse.cdt.core.ELF" point="org.eclipse.cdt.core.BinaryParser"/>
-					<extension id="org.eclipse.cdt.core.MachO64" point="org.eclipse.cdt.core.BinaryParser"/>
 					<extension id="org.eclipse.cdt.core.GmakeErrorParser" point="org.eclipse.cdt.core.ErrorParser"/>
 					<extension id="org.eclipse.cdt.core.CWDLocator" point="org.eclipse.cdt.core.ErrorParser"/>
 					<extension id="org.eclipse.cdt.core.GCCErrorParser" point="org.eclipse.cdt.core.ErrorParser"/>
 					<extension id="org.eclipse.cdt.core.GASErrorParser" point="org.eclipse.cdt.core.ErrorParser"/>
 					<extension id="org.eclipse.cdt.core.GLDErrorParser" point="org.eclipse.cdt.core.ErrorParser"/>
+					<extension id="org.eclipse.cdt.core.ELF" point="org.eclipse.cdt.core.BinaryParser"/>
+					<extension id="org.eclipse.cdt.core.MachO64" point="org.eclipse.cdt.core.BinaryParser"/>
 				</extensions>
 			</storageModule>
 			<storageModule moduleId="cdtBuildSystem" version="4.0.0">
diff --git a/moses/TranslationModel/ProbingPT/ProbingPT.cpp b/moses/TranslationModel/ProbingPT/ProbingPT.cpp
index e6ba3498a..5174e99ff 100644
--- a/moses/TranslationModel/ProbingPT/ProbingPT.cpp
+++ b/moses/TranslationModel/ProbingPT/ProbingPT.cpp
@@ -32,7 +32,21 @@ void ProbingPT::Load()
 
 	m_unkId = 456456546456;
 
-	// vocab
+	// source vocab
+	const std::map<uint64_t, std::string> &sourceVocab = m_engine->getSourceVocab();
+	std::map<uint64_t, std::string>::const_iterator iter;
+	for (iter = sourceVocab.begin(); iter != sourceVocab.end(); ++iter) {
+	  const string &wordStr = iter->second;
+	  const Factor *factor = FactorCollection::Instance().AddFactor(wordStr);
+
+	  uint64_t probingId = iter->first;
+
+	  SourceVocabMap::value_type entry(factor, probingId);
+	  m_sourceVocabMap.insert(entry);
+
+	}
+
+	// target vocab
 	const std::map<unsigned int, std::string> &probingVocab = m_engine->getVocab();
 	std::map<unsigned int, std::string>::const_iterator iter;
 	for (iter = probingVocab.begin(); iter != probingVocab.end(); ++iter) {
@@ -41,7 +55,7 @@ void ProbingPT::Load()
 
 	  unsigned int probingId = iter->first;
 
-	  VocabMap::value_type entry(factor, probingId);
+	  TargetVocabMap::value_type entry(factor, probingId);
 	  m_vocabMap.insert(entry);
 
 	}
@@ -173,7 +187,7 @@ TargetPhrase *ProbingPT::CreateTargetPhrase(const Phrase &sourcePhrase, const ta
 
 const Factor *ProbingPT::GetFactor(uint64_t probingId) const
 {
-	VocabMap::right_map::const_iterator iter;
+	TargetVocabMap::right_map::const_iterator iter;
 	iter = m_vocabMap.right.find(probingId);
 	if (iter != m_vocabMap.right.end()) {
 		return iter->second;
@@ -186,7 +200,7 @@ const Factor *ProbingPT::GetFactor(uint64_t probingId) const
 
 uint64_t ProbingPT::GetProbingId(const Factor *factor) const
 {
-	VocabMap::left_map::const_iterator iter;
+	TargetVocabMap::left_map::const_iterator iter;
 	iter = m_vocabMap.left.find(factor);
 	if (iter != m_vocabMap.left.end()) {
 		return iter->second;
diff --git a/moses/TranslationModel/ProbingPT/ProbingPT.h b/moses/TranslationModel/ProbingPT/ProbingPT.h
index b1f7ff7de..1746f236c 100644
--- a/moses/TranslationModel/ProbingPT/ProbingPT.h
+++ b/moses/TranslationModel/ProbingPT/ProbingPT.h
@@ -37,8 +37,11 @@ public:
 protected:
   QueryEngine *m_engine;
 
-  typedef boost::bimap<const Factor *, unsigned int> VocabMap;
-  mutable VocabMap m_vocabMap;
+  typedef boost::bimap<const Factor *, uint64_t> SourceVocabMap;
+  mutable SourceVocabMap m_sourceVocabMap;
+
+  typedef boost::bimap<const Factor *, unsigned int> TargetVocabMap;
+  mutable TargetVocabMap m_vocabMap;
 
   TargetPhraseCollection *CreateTargetPhrase(const Phrase &sourcePhrase) const;
   TargetPhrase *CreateTargetPhrase(const Phrase &sourcePhrase, const target_text &probingTargetPhrase) const;