am 748f32fe: Merge "Rename UnigramProperty to WordProperty."
* commit '748f32feba044fe4d9c3ae1ab74ed732873d0a83': Rename UnigramProperty to WordProperty.main
commit
1484054c6b
|
@ -27,7 +27,7 @@ import com.android.inputmethod.latin.utils.CollectionUtils;
|
||||||
import com.android.inputmethod.latin.utils.JniUtils;
|
import com.android.inputmethod.latin.utils.JniUtils;
|
||||||
import com.android.inputmethod.latin.utils.LanguageModelParam;
|
import com.android.inputmethod.latin.utils.LanguageModelParam;
|
||||||
import com.android.inputmethod.latin.utils.StringUtils;
|
import com.android.inputmethod.latin.utils.StringUtils;
|
||||||
import com.android.inputmethod.latin.utils.UnigramProperty;
|
import com.android.inputmethod.latin.utils.WordProperty;
|
||||||
|
|
||||||
import java.io.File;
|
import java.io.File;
|
||||||
import java.util.ArrayList;
|
import java.util.ArrayList;
|
||||||
|
@ -61,18 +61,18 @@ public final class BinaryDictionary extends Dictionary {
|
||||||
|
|
||||||
public static final int NOT_A_VALID_TIMESTAMP = -1;
|
public static final int NOT_A_VALID_TIMESTAMP = -1;
|
||||||
|
|
||||||
// Format to get unigram flags from native side via getUnigramPropertyNative().
|
// Format to get unigram flags from native side via getWordPropertyNative().
|
||||||
private static final int FORMAT_UNIGRAM_PROPERTY_OUTPUT_FLAG_COUNT = 4;
|
private static final int FORMAT_WORD_PROPERTY_OUTPUT_FLAG_COUNT = 4;
|
||||||
private static final int FORMAT_UNIGRAM_PROPERTY_IS_NOT_A_WORD_INDEX = 0;
|
private static final int FORMAT_WORD_PROPERTY_IS_NOT_A_WORD_INDEX = 0;
|
||||||
private static final int FORMAT_UNIGRAM_PROPERTY_IS_BLACKLISTED_INDEX = 1;
|
private static final int FORMAT_WORD_PROPERTY_IS_BLACKLISTED_INDEX = 1;
|
||||||
private static final int FORMAT_UNIGRAM_PROPERTY_HAS_BIGRAMS_INDEX = 2;
|
private static final int FORMAT_WORD_PROPERTY_HAS_BIGRAMS_INDEX = 2;
|
||||||
private static final int FORMAT_UNIGRAM_PROPERTY_HAS_SHORTCUTS_INDEX = 3;
|
private static final int FORMAT_WORD_PROPERTY_HAS_SHORTCUTS_INDEX = 3;
|
||||||
|
|
||||||
// Format to get unigram historical info from native side via getUnigramPropertyNative().
|
// Format to get unigram historical info from native side via getWordPropertyNative().
|
||||||
private static final int FORMAT_UNIGRAM_PROPERTY_OUTPUT_HISTORICAL_INFO_COUNT = 3;
|
private static final int FORMAT_WORD_PROPERTY_OUTPUT_HISTORICAL_INFO_COUNT = 3;
|
||||||
private static final int FORMAT_UNIGRAM_PROPERTY_TIMESTAMP_INDEX = 0;
|
private static final int FORMAT_WORD_PROPERTY_TIMESTAMP_INDEX = 0;
|
||||||
private static final int FORMAT_UNIGRAM_PROPERTY_LEVEL_INDEX = 1;
|
private static final int FORMAT_WORD_PROPERTY_LEVEL_INDEX = 1;
|
||||||
private static final int FORMAT_UNIGRAM_PROPERTY_COUNT_INDEX = 2;
|
private static final int FORMAT_WORD_PROPERTY_COUNT_INDEX = 2;
|
||||||
|
|
||||||
private long mNativeDict;
|
private long mNativeDict;
|
||||||
private final Locale mLocale;
|
private final Locale mLocale;
|
||||||
|
@ -143,7 +143,7 @@ public final class BinaryDictionary extends Dictionary {
|
||||||
private static native int getFormatVersionNative(long dict);
|
private static native int getFormatVersionNative(long dict);
|
||||||
private static native int getProbabilityNative(long dict, int[] word);
|
private static native int getProbabilityNative(long dict, int[] word);
|
||||||
private static native int getBigramProbabilityNative(long dict, int[] word0, int[] word1);
|
private static native int getBigramProbabilityNative(long dict, int[] word0, int[] word1);
|
||||||
private static native void getUnigramPropertyNative(long dict, int[] word,
|
private static native void getWordPropertyNative(long dict, int[] word,
|
||||||
int[] outCodePoints, boolean[] outFlags, int[] outProbability,
|
int[] outCodePoints, boolean[] outFlags, int[] outProbability,
|
||||||
int[] outHistoricalInfo, ArrayList<int[]> outShortcutTargets,
|
int[] outHistoricalInfo, ArrayList<int[]> outShortcutTargets,
|
||||||
ArrayList<Integer> outShortcutProbabilities);
|
ArrayList<Integer> outShortcutProbabilities);
|
||||||
|
@ -306,28 +306,28 @@ public final class BinaryDictionary extends Dictionary {
|
||||||
}
|
}
|
||||||
|
|
||||||
@UsedForTesting
|
@UsedForTesting
|
||||||
public UnigramProperty getUnigramProperty(final String word) {
|
public WordProperty getWordProperty(final String word) {
|
||||||
if (TextUtils.isEmpty(word)) {
|
if (TextUtils.isEmpty(word)) {
|
||||||
return null;
|
return null;
|
||||||
}
|
}
|
||||||
final int[] codePoints = StringUtils.toCodePointArray(word);
|
final int[] codePoints = StringUtils.toCodePointArray(word);
|
||||||
final int[] outCodePoints = new int[MAX_WORD_LENGTH];
|
final int[] outCodePoints = new int[MAX_WORD_LENGTH];
|
||||||
final boolean[] outFlags = new boolean[FORMAT_UNIGRAM_PROPERTY_OUTPUT_FLAG_COUNT];
|
final boolean[] outFlags = new boolean[FORMAT_WORD_PROPERTY_OUTPUT_FLAG_COUNT];
|
||||||
final int[] outProbability = new int[1];
|
final int[] outProbability = new int[1];
|
||||||
final int[] outHistoricalInfo =
|
final int[] outHistoricalInfo =
|
||||||
new int[FORMAT_UNIGRAM_PROPERTY_OUTPUT_HISTORICAL_INFO_COUNT];
|
new int[FORMAT_WORD_PROPERTY_OUTPUT_HISTORICAL_INFO_COUNT];
|
||||||
final ArrayList<int[]> outShortcutTargets = CollectionUtils.newArrayList();
|
final ArrayList<int[]> outShortcutTargets = CollectionUtils.newArrayList();
|
||||||
final ArrayList<Integer> outShortcutProbabilities = CollectionUtils.newArrayList();
|
final ArrayList<Integer> outShortcutProbabilities = CollectionUtils.newArrayList();
|
||||||
getUnigramPropertyNative(mNativeDict, codePoints, outCodePoints, outFlags, outProbability,
|
getWordPropertyNative(mNativeDict, codePoints, outCodePoints, outFlags, outProbability,
|
||||||
outHistoricalInfo, outShortcutTargets, outShortcutProbabilities);
|
outHistoricalInfo, outShortcutTargets, outShortcutProbabilities);
|
||||||
return new UnigramProperty(codePoints,
|
return new WordProperty(codePoints,
|
||||||
outFlags[FORMAT_UNIGRAM_PROPERTY_IS_NOT_A_WORD_INDEX],
|
outFlags[FORMAT_WORD_PROPERTY_IS_NOT_A_WORD_INDEX],
|
||||||
outFlags[FORMAT_UNIGRAM_PROPERTY_IS_BLACKLISTED_INDEX],
|
outFlags[FORMAT_WORD_PROPERTY_IS_BLACKLISTED_INDEX],
|
||||||
outFlags[FORMAT_UNIGRAM_PROPERTY_HAS_BIGRAMS_INDEX],
|
outFlags[FORMAT_WORD_PROPERTY_HAS_BIGRAMS_INDEX],
|
||||||
outFlags[FORMAT_UNIGRAM_PROPERTY_HAS_SHORTCUTS_INDEX], outProbability[0],
|
outFlags[FORMAT_WORD_PROPERTY_HAS_SHORTCUTS_INDEX], outProbability[0],
|
||||||
outHistoricalInfo[FORMAT_UNIGRAM_PROPERTY_TIMESTAMP_INDEX],
|
outHistoricalInfo[FORMAT_WORD_PROPERTY_TIMESTAMP_INDEX],
|
||||||
outHistoricalInfo[FORMAT_UNIGRAM_PROPERTY_LEVEL_INDEX],
|
outHistoricalInfo[FORMAT_WORD_PROPERTY_LEVEL_INDEX],
|
||||||
outHistoricalInfo[FORMAT_UNIGRAM_PROPERTY_COUNT_INDEX],
|
outHistoricalInfo[FORMAT_WORD_PROPERTY_COUNT_INDEX],
|
||||||
outShortcutTargets, outShortcutProbabilities);
|
outShortcutTargets, outShortcutProbabilities);
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
|
@ -26,7 +26,7 @@ import java.util.ArrayList;
|
||||||
// This has information that belong to a unigram. This class has some detailed attributes such as
|
// This has information that belong to a unigram. This class has some detailed attributes such as
|
||||||
// historical information but they have to be checked only for testing purpose.
|
// historical information but they have to be checked only for testing purpose.
|
||||||
@UsedForTesting
|
@UsedForTesting
|
||||||
public class UnigramProperty {
|
public class WordProperty {
|
||||||
public final String mCodePoints;
|
public final String mCodePoints;
|
||||||
public final boolean mIsNotAWord;
|
public final boolean mIsNotAWord;
|
||||||
public final boolean mIsBlacklisted;
|
public final boolean mIsBlacklisted;
|
||||||
|
@ -50,8 +50,8 @@ public class UnigramProperty {
|
||||||
return codePoints.length;
|
return codePoints.length;
|
||||||
}
|
}
|
||||||
|
|
||||||
// This represents invalid unigram when the probability is BinaryDictionary.NOT_A_PROBABILITY.
|
// This represents invalid word when the probability is BinaryDictionary.NOT_A_PROBABILITY.
|
||||||
public UnigramProperty(final int[] codePoints, final boolean isNotAWord,
|
public WordProperty(final int[] codePoints, final boolean isNotAWord,
|
||||||
final boolean isBlacklisted, final boolean hasBigram,
|
final boolean isBlacklisted, final boolean hasBigram,
|
||||||
final boolean hasShortcuts, final int probability, final int timestamp,
|
final boolean hasShortcuts, final int probability, final int timestamp,
|
||||||
final int level, final int count, final ArrayList<int[]> shortcutTargets,
|
final int level, final int count, final ArrayList<int[]> shortcutTargets,
|
|
@ -32,7 +32,7 @@ LATIN_IME_CORE_SRC_FILES := \
|
||||||
error_type_utils.cpp \
|
error_type_utils.cpp \
|
||||||
multi_bigram_map.cpp \
|
multi_bigram_map.cpp \
|
||||||
suggestions_output_utils.cpp \
|
suggestions_output_utils.cpp \
|
||||||
unigram_property.cpp) \
|
word_property.cpp) \
|
||||||
$(addprefix suggest/core/layout/, \
|
$(addprefix suggest/core/layout/, \
|
||||||
additional_proximity_chars.cpp \
|
additional_proximity_chars.cpp \
|
||||||
proximity_info.cpp \
|
proximity_info.cpp \
|
||||||
|
|
|
@ -24,7 +24,7 @@
|
||||||
#include "jni.h"
|
#include "jni.h"
|
||||||
#include "jni_common.h"
|
#include "jni_common.h"
|
||||||
#include "suggest/core/dictionary/dictionary.h"
|
#include "suggest/core/dictionary/dictionary.h"
|
||||||
#include "suggest/core/dictionary/unigram_property.h"
|
#include "suggest/core/dictionary/word_property.h"
|
||||||
#include "suggest/core/suggest_options.h"
|
#include "suggest/core/suggest_options.h"
|
||||||
#include "suggest/policyimpl/dictionary/structure/dictionary_structure_with_buffer_policy_factory.h"
|
#include "suggest/policyimpl/dictionary/structure/dictionary_structure_with_buffer_policy_factory.h"
|
||||||
#include "suggest/policyimpl/dictionary/utils/dict_file_writing_utils.h"
|
#include "suggest/policyimpl/dictionary/utils/dict_file_writing_utils.h"
|
||||||
|
@ -260,7 +260,7 @@ static jint latinime_BinaryDictionary_getBigramProbability(JNIEnv *env, jclass c
|
||||||
word1Length);
|
word1Length);
|
||||||
}
|
}
|
||||||
|
|
||||||
static void latinime_BinaryDictionary_getUnigramProperty(JNIEnv *env, jclass clazz,
|
static void latinime_BinaryDictionary_getWordProperty(JNIEnv *env, jclass clazz,
|
||||||
jlong dict, jintArray word, jintArray outCodePoints, jbooleanArray outFlags,
|
jlong dict, jintArray word, jintArray outCodePoints, jbooleanArray outFlags,
|
||||||
jintArray outProbability, jintArray outHistoricalInfo, jobject outShortcutTargets,
|
jintArray outProbability, jintArray outHistoricalInfo, jobject outShortcutTargets,
|
||||||
jobject outShortcutProbabilities) {
|
jobject outShortcutProbabilities) {
|
||||||
|
@ -269,9 +269,8 @@ static void latinime_BinaryDictionary_getUnigramProperty(JNIEnv *env, jclass cla
|
||||||
const jsize wordLength = env->GetArrayLength(word);
|
const jsize wordLength = env->GetArrayLength(word);
|
||||||
int wordCodePoints[wordLength];
|
int wordCodePoints[wordLength];
|
||||||
env->GetIntArrayRegion(word, 0, wordLength, wordCodePoints);
|
env->GetIntArrayRegion(word, 0, wordLength, wordCodePoints);
|
||||||
const UnigramProperty unigramProperty = dictionary->getUnigramProperty(
|
const WordProperty wordProperty = dictionary->getWordProperty(wordCodePoints, wordLength);
|
||||||
wordCodePoints, wordLength);
|
wordProperty.outputProperties(env, outCodePoints, outFlags, outProbability,
|
||||||
unigramProperty.outputProperties(env, outCodePoints, outFlags, outProbability,
|
|
||||||
outHistoricalInfo, outShortcutTargets, outShortcutProbabilities);
|
outHistoricalInfo, outShortcutTargets, outShortcutProbabilities);
|
||||||
}
|
}
|
||||||
|
|
||||||
|
@ -521,9 +520,9 @@ static const JNINativeMethod sMethods[] = {
|
||||||
reinterpret_cast<void *>(latinime_BinaryDictionary_getBigramProbability)
|
reinterpret_cast<void *>(latinime_BinaryDictionary_getBigramProbability)
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
const_cast<char *>("getUnigramPropertyNative"),
|
const_cast<char *>("getWordPropertyNative"),
|
||||||
const_cast<char *>("(J[I[I[Z[I[ILjava/util/ArrayList;Ljava/util/ArrayList;)V"),
|
const_cast<char *>("(J[I[I[Z[I[ILjava/util/ArrayList;Ljava/util/ArrayList;)V"),
|
||||||
reinterpret_cast<void *>(latinime_BinaryDictionary_getUnigramProperty)
|
reinterpret_cast<void *>(latinime_BinaryDictionary_getWordProperty)
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
const_cast<char *>("calcNormalizedScoreNative"),
|
const_cast<char *>("calcNormalizedScoreNative"),
|
||||||
|
|
|
@ -143,10 +143,10 @@ void Dictionary::getProperty(const char *const query, const int queryLength, cha
|
||||||
maxResultLength);
|
maxResultLength);
|
||||||
}
|
}
|
||||||
|
|
||||||
const UnigramProperty Dictionary::getUnigramProperty(const int *const codePoints,
|
const WordProperty Dictionary::getWordProperty(const int *const codePoints,
|
||||||
const int codePointCount) {
|
const int codePointCount) {
|
||||||
TimeKeeper::setCurrentTime();
|
TimeKeeper::setCurrentTime();
|
||||||
return mDictionaryStructureWithBufferPolicy.get()->getUnigramProperty(
|
return mDictionaryStructureWithBufferPolicy.get()->getWordProperty(
|
||||||
codePoints, codePointCount);
|
codePoints, codePointCount);
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
|
@ -22,7 +22,7 @@
|
||||||
#include "defines.h"
|
#include "defines.h"
|
||||||
#include "jni.h"
|
#include "jni.h"
|
||||||
#include "suggest/core/dictionary/bigram_dictionary.h"
|
#include "suggest/core/dictionary/bigram_dictionary.h"
|
||||||
#include "suggest/core/dictionary/unigram_property.h"
|
#include "suggest/core/dictionary/word_property.h"
|
||||||
#include "suggest/core/policy/dictionary_header_structure_policy.h"
|
#include "suggest/core/policy/dictionary_header_structure_policy.h"
|
||||||
#include "suggest/core/policy/dictionary_structure_with_buffer_policy.h"
|
#include "suggest/core/policy/dictionary_structure_with_buffer_policy.h"
|
||||||
#include "suggest/core/suggest_interface.h"
|
#include "suggest/core/suggest_interface.h"
|
||||||
|
@ -34,7 +34,7 @@ class DictionaryStructureWithBufferPolicy;
|
||||||
class DicTraverseSession;
|
class DicTraverseSession;
|
||||||
class ProximityInfo;
|
class ProximityInfo;
|
||||||
class SuggestOptions;
|
class SuggestOptions;
|
||||||
class UnigramProperty;
|
class WordProperty;
|
||||||
|
|
||||||
class Dictionary {
|
class Dictionary {
|
||||||
public:
|
public:
|
||||||
|
@ -94,7 +94,7 @@ class Dictionary {
|
||||||
void getProperty(const char *const query, const int queryLength, char *const outResult,
|
void getProperty(const char *const query, const int queryLength, char *const outResult,
|
||||||
const int maxResultLength);
|
const int maxResultLength);
|
||||||
|
|
||||||
const UnigramProperty getUnigramProperty(const int *const codePoints, const int codePointCount);
|
const WordProperty getWordProperty(const int *const codePoints, const int codePointCount);
|
||||||
|
|
||||||
const DictionaryStructureWithBufferPolicy *getDictionaryStructurePolicy() const {
|
const DictionaryStructureWithBufferPolicy *getDictionaryStructurePolicy() const {
|
||||||
return mDictionaryStructureWithBufferPolicy.get();
|
return mDictionaryStructureWithBufferPolicy.get();
|
||||||
|
|
|
@ -14,11 +14,11 @@
|
||||||
* limitations under the License.
|
* limitations under the License.
|
||||||
*/
|
*/
|
||||||
|
|
||||||
#include "suggest/core/dictionary/unigram_property.h"
|
#include "suggest/core/dictionary/word_property.h"
|
||||||
|
|
||||||
namespace latinime {
|
namespace latinime {
|
||||||
|
|
||||||
void UnigramProperty::outputProperties(JNIEnv *const env, jintArray outCodePoints,
|
void WordProperty::outputProperties(JNIEnv *const env, jintArray outCodePoints,
|
||||||
jbooleanArray outFlags, jintArray outProbability, jintArray outHistoricalInfo,
|
jbooleanArray outFlags, jintArray outProbability, jintArray outHistoricalInfo,
|
||||||
jobject outShortcutTargets, jobject outShortcutProbabilities) const {
|
jobject outShortcutTargets, jobject outShortcutProbabilities) const {
|
||||||
env->SetIntArrayRegion(outCodePoints, 0 /* start */, mCodePoints.size(), &mCodePoints[0]);
|
env->SetIntArrayRegion(outCodePoints, 0 /* start */, mCodePoints.size(), &mCodePoints[0]);
|
|
@ -14,8 +14,8 @@
|
||||||
* limitations under the License.
|
* limitations under the License.
|
||||||
*/
|
*/
|
||||||
|
|
||||||
#ifndef LATINIME_UNIGRAM_PROPERTY_H
|
#ifndef LATINIME_WORD_PROPERTY_H
|
||||||
#define LATINIME_UNIGRAM_PROPERTY_H
|
#define LATINIME_WORD_PROPERTY_H
|
||||||
|
|
||||||
#include <cstring>
|
#include <cstring>
|
||||||
#include <vector>
|
#include <vector>
|
||||||
|
@ -25,16 +25,17 @@
|
||||||
|
|
||||||
namespace latinime {
|
namespace latinime {
|
||||||
|
|
||||||
// This class is used for returning information belonging to a unigram to java side.
|
// This class is used for returning information belonging to a word to java side.
|
||||||
class UnigramProperty {
|
class WordProperty {
|
||||||
public:
|
public:
|
||||||
// Invalid unigram.
|
// TODO: Add bigram information.
|
||||||
UnigramProperty()
|
// Invalid word.
|
||||||
|
WordProperty()
|
||||||
: mCodePoints(), mIsNotAWord(false), mIsBlacklisted(false),
|
: mCodePoints(), mIsNotAWord(false), mIsBlacklisted(false),
|
||||||
mHasBigrams(false), mHasShortcuts(false), mProbability(NOT_A_PROBABILITY),
|
mHasBigrams(false), mHasShortcuts(false), mProbability(NOT_A_PROBABILITY),
|
||||||
mTimestamp(0), mLevel(0), mCount(0), mShortcutTargets(), mShortcutProbabilities() {}
|
mTimestamp(0), mLevel(0), mCount(0), mShortcutTargets(), mShortcutProbabilities() {}
|
||||||
|
|
||||||
UnigramProperty(const std::vector<int> *const codePoints,
|
WordProperty(const std::vector<int> *const codePoints,
|
||||||
const bool isNotAWord, const bool isBlacklisted, const bool hasBigrams,
|
const bool isNotAWord, const bool isBlacklisted, const bool hasBigrams,
|
||||||
const bool hasShortcuts, const int probability, const int timestamp,
|
const bool hasShortcuts, const int probability, const int timestamp,
|
||||||
const int level, const int count,
|
const int level, const int count,
|
||||||
|
@ -50,7 +51,7 @@ class UnigramProperty {
|
||||||
jobject outShortcutProbabilities) const;
|
jobject outShortcutProbabilities) const;
|
||||||
|
|
||||||
private:
|
private:
|
||||||
DISALLOW_ASSIGNMENT_OPERATOR(UnigramProperty);
|
DISALLOW_ASSIGNMENT_OPERATOR(WordProperty);
|
||||||
|
|
||||||
std::vector<int> mCodePoints;
|
std::vector<int> mCodePoints;
|
||||||
bool mIsNotAWord;
|
bool mIsNotAWord;
|
||||||
|
@ -67,4 +68,4 @@ class UnigramProperty {
|
||||||
std::vector<int> mShortcutProbabilities;
|
std::vector<int> mShortcutProbabilities;
|
||||||
};
|
};
|
||||||
} // namespace latinime
|
} // namespace latinime
|
||||||
#endif // LATINIME_UNIGRAM_PROPERTY_H
|
#endif // LATINIME_WORD_PROPERTY_H
|
|
@ -18,7 +18,7 @@
|
||||||
#define LATINIME_DICTIONARY_STRUCTURE_POLICY_H
|
#define LATINIME_DICTIONARY_STRUCTURE_POLICY_H
|
||||||
|
|
||||||
#include "defines.h"
|
#include "defines.h"
|
||||||
#include "suggest/core/dictionary/unigram_property.h"
|
#include "suggest/core/dictionary/word_property.h"
|
||||||
#include "utils/exclusive_ownership_pointer.h"
|
#include "utils/exclusive_ownership_pointer.h"
|
||||||
|
|
||||||
namespace latinime {
|
namespace latinime {
|
||||||
|
@ -92,7 +92,7 @@ class DictionaryStructureWithBufferPolicy {
|
||||||
const int maxResultLength) = 0;
|
const int maxResultLength) = 0;
|
||||||
|
|
||||||
// Used for testing.
|
// Used for testing.
|
||||||
virtual const UnigramProperty getUnigramProperty(const int *const codePonts,
|
virtual const WordProperty getWordProperty(const int *const codePonts,
|
||||||
const int codePointCount) const = 0;
|
const int codePointCount) const = 0;
|
||||||
|
|
||||||
protected:
|
protected:
|
||||||
|
|
|
@ -123,10 +123,10 @@ class PatriciaTriePolicy : public DictionaryStructureWithBufferPolicy {
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
const UnigramProperty getUnigramProperty(const int *const codePoints,
|
const WordProperty getWordProperty(const int *const codePoints,
|
||||||
const int codePointCount) const {
|
const int codePointCount) const {
|
||||||
// getUnigramProperty is not supported.
|
// getWordProperty is not supported.
|
||||||
return UnigramProperty();
|
return WordProperty();
|
||||||
}
|
}
|
||||||
|
|
||||||
private:
|
private:
|
||||||
|
|
|
@ -20,7 +20,7 @@
|
||||||
|
|
||||||
#include "suggest/core/dicnode/dic_node.h"
|
#include "suggest/core/dicnode/dic_node.h"
|
||||||
#include "suggest/core/dicnode/dic_node_vector.h"
|
#include "suggest/core/dicnode/dic_node_vector.h"
|
||||||
#include "suggest/core/dictionary/unigram_property.h"
|
#include "suggest/core/dictionary/word_property.h"
|
||||||
#include "suggest/policyimpl/dictionary/structure/pt_common/dynamic_pt_reading_helper.h"
|
#include "suggest/policyimpl/dictionary/structure/pt_common/dynamic_pt_reading_helper.h"
|
||||||
#include "suggest/policyimpl/dictionary/structure/v4/ver4_patricia_trie_node_reader.h"
|
#include "suggest/policyimpl/dictionary/structure/v4/ver4_patricia_trie_node_reader.h"
|
||||||
#include "suggest/policyimpl/dictionary/utils/forgetting_curve_utils.h"
|
#include "suggest/policyimpl/dictionary/utils/forgetting_curve_utils.h"
|
||||||
|
@ -317,13 +317,13 @@ void Ver4PatriciaTriePolicy::getProperty(const char *const query, const int quer
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
const UnigramProperty Ver4PatriciaTriePolicy::getUnigramProperty(const int *const codePoints,
|
const WordProperty Ver4PatriciaTriePolicy::getWordProperty(const int *const codePoints,
|
||||||
const int codePointCount) const {
|
const int codePointCount) const {
|
||||||
const int ptNodePos = getTerminalPtNodePositionOfWord(codePoints, codePointCount,
|
const int ptNodePos = getTerminalPtNodePositionOfWord(codePoints, codePointCount,
|
||||||
false /* forceLowerCaseSearch */);
|
false /* forceLowerCaseSearch */);
|
||||||
if (ptNodePos == NOT_A_DICT_POS) {
|
if (ptNodePos == NOT_A_DICT_POS) {
|
||||||
AKLOGE("fetchUnigramProperty is called for invalid word.");
|
AKLOGE("getWordProperty is called for invalid word.");
|
||||||
return UnigramProperty();
|
return WordProperty();
|
||||||
}
|
}
|
||||||
const PtNodeParams ptNodeParams = mNodeReader.fetchNodeInfoInBufferFromPtNodePos(ptNodePos);
|
const PtNodeParams ptNodeParams = mNodeReader.fetchNodeInfoInBufferFromPtNodePos(ptNodePos);
|
||||||
std::vector<int> codePointVector(ptNodeParams.getCodePoints(),
|
std::vector<int> codePointVector(ptNodeParams.getCodePoints(),
|
||||||
|
@ -351,7 +351,7 @@ const UnigramProperty Ver4PatriciaTriePolicy::getUnigramProperty(const int *cons
|
||||||
shortcutProbabilities.push_back(shortcutProbability);
|
shortcutProbabilities.push_back(shortcutProbability);
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
return UnigramProperty(&codePointVector, ptNodeParams.isNotAWord(),
|
return WordProperty(&codePointVector, ptNodeParams.isNotAWord(),
|
||||||
ptNodeParams.isBlacklisted(), ptNodeParams.hasBigrams(),
|
ptNodeParams.isBlacklisted(), ptNodeParams.hasBigrams(),
|
||||||
ptNodeParams.hasShortcutTargets(), ptNodeParams.getProbability(),
|
ptNodeParams.hasShortcutTargets(), ptNodeParams.getProbability(),
|
||||||
historicalInfo->getTimeStamp(), historicalInfo->getLevel(),
|
historicalInfo->getTimeStamp(), historicalInfo->getLevel(),
|
||||||
|
|
|
@ -106,7 +106,7 @@ class Ver4PatriciaTriePolicy : public DictionaryStructureWithBufferPolicy {
|
||||||
void getProperty(const char *const query, const int queryLength, char *const outResult,
|
void getProperty(const char *const query, const int queryLength, char *const outResult,
|
||||||
const int maxResultLength);
|
const int maxResultLength);
|
||||||
|
|
||||||
const UnigramProperty getUnigramProperty(const int *const codePoints,
|
const WordProperty getWordProperty(const int *const codePoints,
|
||||||
const int codePointCount) const;
|
const int codePointCount) const;
|
||||||
|
|
||||||
private:
|
private:
|
||||||
|
|
|
@ -26,7 +26,7 @@ import com.android.inputmethod.latin.makedict.FormatSpec;
|
||||||
import com.android.inputmethod.latin.makedict.FusionDictionary.WeightedString;
|
import com.android.inputmethod.latin.makedict.FusionDictionary.WeightedString;
|
||||||
import com.android.inputmethod.latin.utils.FileUtils;
|
import com.android.inputmethod.latin.utils.FileUtils;
|
||||||
import com.android.inputmethod.latin.utils.LanguageModelParam;
|
import com.android.inputmethod.latin.utils.LanguageModelParam;
|
||||||
import com.android.inputmethod.latin.utils.UnigramProperty;
|
import com.android.inputmethod.latin.utils.WordProperty;
|
||||||
|
|
||||||
import java.io.File;
|
import java.io.File;
|
||||||
import java.io.IOException;
|
import java.io.IOException;
|
||||||
|
@ -892,8 +892,7 @@ public class BinaryDictionaryTests extends AndroidTestCase {
|
||||||
0 /* offset */, dictFile.length(), true /* useFullEditDistance */,
|
0 /* offset */, dictFile.length(), true /* useFullEditDistance */,
|
||||||
Locale.getDefault(), TEST_LOCALE, true /* isUpdatable */);
|
Locale.getDefault(), TEST_LOCALE, true /* isUpdatable */);
|
||||||
|
|
||||||
final UnigramProperty invalidUnigramProperty =
|
final WordProperty invalidUnigramProperty = binaryDictionary.getWordProperty("dummyWord");
|
||||||
binaryDictionary.getUnigramProperty("dummyWord");
|
|
||||||
assertFalse(invalidUnigramProperty.isValid());
|
assertFalse(invalidUnigramProperty.isValid());
|
||||||
|
|
||||||
for (int i = 0; i < ITERATION_COUNT; i++) {
|
for (int i = 0; i < ITERATION_COUNT; i++) {
|
||||||
|
@ -905,8 +904,7 @@ public class BinaryDictionaryTests extends AndroidTestCase {
|
||||||
binaryDictionary.addUnigramWord(word, unigramProbability,
|
binaryDictionary.addUnigramWord(word, unigramProbability,
|
||||||
null /* shortcutTarget */, BinaryDictionary.NOT_A_PROBABILITY,
|
null /* shortcutTarget */, BinaryDictionary.NOT_A_PROBABILITY,
|
||||||
isNotAWord, isBlacklisted, BinaryDictionary.NOT_A_VALID_TIMESTAMP);
|
isNotAWord, isBlacklisted, BinaryDictionary.NOT_A_VALID_TIMESTAMP);
|
||||||
final UnigramProperty unigramProperty =
|
final WordProperty unigramProperty = binaryDictionary.getWordProperty(word);
|
||||||
binaryDictionary.getUnigramProperty(word);
|
|
||||||
assertEquals(word, unigramProperty.mCodePoints);
|
assertEquals(word, unigramProperty.mCodePoints);
|
||||||
assertTrue(unigramProperty.isValid());
|
assertTrue(unigramProperty.isValid());
|
||||||
assertEquals(isNotAWord, unigramProperty.mIsNotAWord);
|
assertEquals(isNotAWord, unigramProperty.mIsNotAWord);
|
||||||
|
@ -938,7 +936,7 @@ public class BinaryDictionaryTests extends AndroidTestCase {
|
||||||
binaryDictionary.addUnigramWord("aaa", unigramProbability, "zzz",
|
binaryDictionary.addUnigramWord("aaa", unigramProbability, "zzz",
|
||||||
shortcutProbability, false /* isNotAWord */, false /* isBlacklisted */,
|
shortcutProbability, false /* isNotAWord */, false /* isBlacklisted */,
|
||||||
0 /* timestamp */);
|
0 /* timestamp */);
|
||||||
UnigramProperty unigramProperty = binaryDictionary.getUnigramProperty("aaa");
|
WordProperty unigramProperty = binaryDictionary.getWordProperty("aaa");
|
||||||
assertEquals(1, unigramProperty.mShortcutTargets.size());
|
assertEquals(1, unigramProperty.mShortcutTargets.size());
|
||||||
assertEquals("zzz", unigramProperty.mShortcutTargets.get(0).mWord);
|
assertEquals("zzz", unigramProperty.mShortcutTargets.get(0).mWord);
|
||||||
assertEquals(shortcutProbability, unigramProperty.mShortcutTargets.get(0).mFrequency);
|
assertEquals(shortcutProbability, unigramProperty.mShortcutTargets.get(0).mFrequency);
|
||||||
|
@ -946,7 +944,7 @@ public class BinaryDictionaryTests extends AndroidTestCase {
|
||||||
binaryDictionary.addUnigramWord("aaa", unigramProbability, "zzz",
|
binaryDictionary.addUnigramWord("aaa", unigramProbability, "zzz",
|
||||||
updatedShortcutProbability, false /* isNotAWord */, false /* isBlacklisted */,
|
updatedShortcutProbability, false /* isNotAWord */, false /* isBlacklisted */,
|
||||||
0 /* timestamp */);
|
0 /* timestamp */);
|
||||||
unigramProperty = binaryDictionary.getUnigramProperty("aaa");
|
unigramProperty = binaryDictionary.getWordProperty("aaa");
|
||||||
assertEquals(1, unigramProperty.mShortcutTargets.size());
|
assertEquals(1, unigramProperty.mShortcutTargets.size());
|
||||||
assertEquals("zzz", unigramProperty.mShortcutTargets.get(0).mWord);
|
assertEquals("zzz", unigramProperty.mShortcutTargets.get(0).mWord);
|
||||||
assertEquals(updatedShortcutProbability,
|
assertEquals(updatedShortcutProbability,
|
||||||
|
@ -957,7 +955,7 @@ public class BinaryDictionaryTests extends AndroidTestCase {
|
||||||
final HashMap<String, Integer> shortcutTargets = new HashMap<String, Integer>();
|
final HashMap<String, Integer> shortcutTargets = new HashMap<String, Integer>();
|
||||||
shortcutTargets.put("zzz", updatedShortcutProbability);
|
shortcutTargets.put("zzz", updatedShortcutProbability);
|
||||||
shortcutTargets.put("yyy", shortcutProbability);
|
shortcutTargets.put("yyy", shortcutProbability);
|
||||||
unigramProperty = binaryDictionary.getUnigramProperty("aaa");
|
unigramProperty = binaryDictionary.getWordProperty("aaa");
|
||||||
assertEquals(2, unigramProperty.mShortcutTargets.size());
|
assertEquals(2, unigramProperty.mShortcutTargets.size());
|
||||||
for (WeightedString shortcutTarget : unigramProperty.mShortcutTargets) {
|
for (WeightedString shortcutTarget : unigramProperty.mShortcutTargets) {
|
||||||
assertTrue(shortcutTargets.containsKey(shortcutTarget.mWord));
|
assertTrue(shortcutTargets.containsKey(shortcutTarget.mWord));
|
||||||
|
@ -967,7 +965,7 @@ public class BinaryDictionaryTests extends AndroidTestCase {
|
||||||
shortcutTargets.put("zzz", updatedShortcutProbability);
|
shortcutTargets.put("zzz", updatedShortcutProbability);
|
||||||
shortcutTargets.put("yyy", shortcutProbability);
|
shortcutTargets.put("yyy", shortcutProbability);
|
||||||
binaryDictionary.flushWithGC();
|
binaryDictionary.flushWithGC();
|
||||||
unigramProperty = binaryDictionary.getUnigramProperty("aaa");
|
unigramProperty = binaryDictionary.getWordProperty("aaa");
|
||||||
assertEquals(2, unigramProperty.mShortcutTargets.size());
|
assertEquals(2, unigramProperty.mShortcutTargets.size());
|
||||||
for (WeightedString shortcutTarget : unigramProperty.mShortcutTargets) {
|
for (WeightedString shortcutTarget : unigramProperty.mShortcutTargets) {
|
||||||
assertTrue(shortcutTargets.containsKey(shortcutTarget.mWord));
|
assertTrue(shortcutTargets.containsKey(shortcutTarget.mWord));
|
||||||
|
@ -1036,7 +1034,7 @@ public class BinaryDictionaryTests extends AndroidTestCase {
|
||||||
}
|
}
|
||||||
|
|
||||||
for (final String word : words) {
|
for (final String word : words) {
|
||||||
final UnigramProperty unigramProperty = binaryDictionary.getUnigramProperty(word);
|
final WordProperty unigramProperty = binaryDictionary.getWordProperty(word);
|
||||||
assertEquals((int)unigramProbabilities.get(word), unigramProperty.mProbability);
|
assertEquals((int)unigramProbabilities.get(word), unigramProperty.mProbability);
|
||||||
if (!shortcutTargets.containsKey(word)) {
|
if (!shortcutTargets.containsKey(word)) {
|
||||||
// The word does not have shortcut targets.
|
// The word does not have shortcut targets.
|
||||||
|
|
Loading…
Reference in New Issue