Merge "Add a jni method to fetch unigram information."

2 files changed, 126 insertions, 0 deletions

diff --git a/java/src/com/android/inputmethod/latin/BinaryDictionary.java b/java/src/com/android/inputmethod/latin/BinaryDictionary.java
index a41cc6a82..db4234c63 100644
--- a/java/src/com/android/inputmethod/latin/BinaryDictionary.java
+++ b/java/src/com/android/inputmethod/latin/BinaryDictionary.java
@@ -26,6 +26,7 @@ import com.android.inputmethod.latin.settings.NativeSuggestOptions;
 import com.android.inputmethod.latin.utils.CollectionUtils;
 import com.android.inputmethod.latin.utils.JniUtils;
 import com.android.inputmethod.latin.utils.StringUtils;
+import com.android.inputmethod.latin.utils.UnigramProperty;
 
 import java.io.File;
 import java.util.ArrayList;
@@ -59,6 +60,19 @@ public final class BinaryDictionary extends Dictionary {
 
     public static final int NOT_A_VALID_TIMESTAMP = -1;
 
+    // Format to get unigram flags from native side via getUnigramPropertyNative().
+    private static final int FORMAT_UNIGRAM_PROPERTY_OUTPUT_FLAG_COUNT = 4;
+    private static final int FORMAT_UNIGRAM_PROPERTY_IS_NOT_A_WORD_INDEX = 0;
+    private static final int FORMAT_UNIGRAM_PROPERTY_IS_BLACKLISTED_INDEX = 1;
+    private static final int FORMAT_UNIGRAM_PROPERTY_HAS_BIGRAMS_INDEX = 2;
+    private static final int FORMAT_UNIGRAM_PROPERTY_HAS_SHORTCUTS_INDEX = 3;
+
+    // Format to get unigram historical info from native side via getUnigramPropertyNative().
+    private static final int FORMAT_UNIGRAM_PROPERTY_OUTPUT_HISTORICAL_INFO_COUNT = 3;
+    private static final int FORMAT_UNIGRAM_PROPERTY_TIMESTAMP_INDEX = 0;
+    private static final int FORMAT_UNIGRAM_PROPERTY_LEVEL_INDEX = 1;
+    private static final int FORMAT_UNIGRAM_PROPERTY_COUNT_INDEX = 2;
+
     private long mNativeDict;
     private final Locale mLocale;
     private final long mDictSize;
@@ -128,6 +142,10 @@ public final class BinaryDictionary extends Dictionary {
     private static native int getFormatVersionNative(long dict);
     private static native int getProbabilityNative(long dict, int[] word);
     private static native int getBigramProbabilityNative(long dict, int[] word0, int[] word1);
+    private static native void getUnigramPropertyNative(long dict, int[] word,
+            int[] outCodePoints, boolean[] outFlags, int[] outProbability,
+            int[] outHistoricalInfo, ArrayList<int[]> outShortcutTargets,
+            ArrayList<Integer> outShortcutProbabilities);
     private static native int getSuggestionsNative(long dict, long proximityInfo,
             long traverseSession, int[] xCoordinates, int[] yCoordinates, int[] times,
             int[] pointerIds, int[] inputCodePoints, int inputSize, int commitPoint,
@@ -285,6 +303,32 @@ public final class BinaryDictionary extends Dictionary {
         return getBigramProbabilityNative(mNativeDict, codePoints0, codePoints1);
     }
 
+    @UsedForTesting
+    public UnigramProperty getUnigramProperty(final String word) {
+        if (TextUtils.isEmpty(word)) {
+            return null;
+        }
+        final int[] codePoints = StringUtils.toCodePointArray(word);
+        final int[] outCodePoints = new int[MAX_WORD_LENGTH];
+        final boolean[] outFlags = new boolean[FORMAT_UNIGRAM_PROPERTY_OUTPUT_FLAG_COUNT];
+        final int[] outProbability = new int[1];
+        final int[] outHistoricalInfo =
+                new int[FORMAT_UNIGRAM_PROPERTY_OUTPUT_HISTORICAL_INFO_COUNT];
+        final ArrayList<int[]> outShortcutTargets = CollectionUtils.newArrayList();
+        final ArrayList<Integer> outShortcutProbabilities = CollectionUtils.newArrayList();
+        getUnigramPropertyNative(mNativeDict, codePoints, outCodePoints, outFlags, outProbability,
+                outHistoricalInfo, outShortcutTargets, outShortcutProbabilities);
+        return new UnigramProperty(codePoints,
+                outFlags[FORMAT_UNIGRAM_PROPERTY_IS_NOT_A_WORD_INDEX],
+                outFlags[FORMAT_UNIGRAM_PROPERTY_IS_BLACKLISTED_INDEX],
+                outFlags[FORMAT_UNIGRAM_PROPERTY_HAS_BIGRAMS_INDEX],
+                outFlags[FORMAT_UNIGRAM_PROPERTY_HAS_SHORTCUTS_INDEX], outProbability[0],
+                outHistoricalInfo[FORMAT_UNIGRAM_PROPERTY_TIMESTAMP_INDEX],
+                outHistoricalInfo[FORMAT_UNIGRAM_PROPERTY_LEVEL_INDEX],
+                outHistoricalInfo[FORMAT_UNIGRAM_PROPERTY_COUNT_INDEX],
+                outShortcutTargets, outShortcutProbabilities);
+    }
+
     // Add a unigram entry to binary dictionary with unigram attributes in native code.
     public void addUnigramWord(final String word, final int probability,
             final String shortcutTarget, final int shortcutProbability, final boolean isNotAWord,
diff --git a/java/src/com/android/inputmethod/latin/utils/UnigramProperty.java b/java/src/com/android/inputmethod/latin/utils/UnigramProperty.java
new file mode 100644
index 000000000..4feee4393
--- /dev/null
+++ b/java/src/com/android/inputmethod/latin/utils/UnigramProperty.java
@@ -0,0 +1,82 @@
+/*
+ * Copyright (C) 2013 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+
+package com.android.inputmethod.latin.utils;
+
+import com.android.inputmethod.annotations.UsedForTesting;
+import com.android.inputmethod.latin.BinaryDictionary;
+import com.android.inputmethod.latin.makedict.FusionDictionary.WeightedString;
+
+import java.util.ArrayList;
+
+// 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.
+@UsedForTesting
+public class UnigramProperty {
+    public final String mCodePoints;
+    public final boolean mIsNotAWord;
+    public final boolean mIsBlacklisted;
+    public final boolean mHasBigrams;
+    public final boolean mHasShortcuts;
+    public final int mProbability;
+    // mTimestamp, mLevel and mCount are historical info. These values are depend on the
+    // implementation in native code; thus, we must not use them and have any assumptions about
+    // them except for tests.
+    public final int mTimestamp;
+    public final int mLevel;
+    public final int mCount;
+    public final ArrayList<WeightedString> mShortcutTargets = CollectionUtils.newArrayList();
+
+    private static int getCodePointCount(final int[] codePoints) {
+        for (int i = 0; i < codePoints.length; i++) {
+            if (codePoints[i] == 0) {
+                return i;
+            }
+        }
+        return codePoints.length;
+    }
+
+    // This represents invalid unigram when the probability is BinaryDictionary.NOT_A_PROBABILITY.
+    public UnigramProperty(final int[] codePoints, final boolean isNotAWord,
+            final boolean isBlacklisted, final boolean hasBigram,
+            final boolean hasShortcuts, final int probability, final int timestamp,
+            final int level, final int count, final ArrayList<int[]> shortcutTargets,
+            final ArrayList<Integer> shortcutProbabilities) {
+        mCodePoints = new String(codePoints, 0 /* offset */, getCodePointCount(codePoints));
+        mIsNotAWord = isNotAWord;
+        mIsBlacklisted = isBlacklisted;
+        mHasBigrams = hasBigram;
+        mHasShortcuts = hasShortcuts;
+        mProbability = probability;
+        mTimestamp = timestamp;
+        mLevel = level;
+        mCount = count;
+        final int shortcutTargetCount = shortcutTargets.size();
+        for (int i = 0; i < shortcutTargetCount; i++) {
+            final int[] shortcutTargetCodePointArray = shortcutTargets.get(i);
+            final String shortcutTargetString = new String(shortcutTargetCodePointArray,
+                    0 /* offset */, getCodePointCount(shortcutTargetCodePointArray));
+            mShortcutTargets.add(
+                    new WeightedString(shortcutTargetString, shortcutProbabilities.get(i)));
+        }
+    }
+
+    @UsedForTesting
+    public boolean isValid() {
+        return mProbability != BinaryDictionary.NOT_A_PROBABILITY;
+    }
+}
\ No newline at end of file