core/java/android/service/wallpaper/IWallpaperService.aidl \
core/java/android/service/chooser/IChooserTargetService.aidl \
core/java/android/service/chooser/IChooserTargetResult.aidl \
+ core/java/android/service/resolver/IResolverRankerService.aidl \
+ core/java/android/service/resolver/IResolverRankerResult.aidl \
core/java/android/text/ITextClassificationService.aidl \
core/java/android/view/accessibility/IAccessibilityInteractionConnection.aidl\
core/java/android/view/accessibility/IAccessibilityInteractionConnectionCallback.aidl\
frameworks/base/core/java/android/service/notification/SnoozeCriterion.aidl \
frameworks/base/core/java/android/service/notification/StatusBarNotification.aidl \
frameworks/base/core/java/android/service/chooser/ChooserTarget.aidl \
+ frameworks/base/core/java/android/service/resolver/ResolverTarget.aidl \
frameworks/base/core/java/android/speech/tts/Voice.aidl \
frameworks/base/core/java/android/app/usage/CacheQuotaHint.aidl \
frameworks/base/core/java/android/app/usage/ExternalStorageStats.aidl \
field public static final java.lang.String BIND_PRINT_SERVICE = "android.permission.BIND_PRINT_SERVICE";
field public static final java.lang.String BIND_QUICK_SETTINGS_TILE = "android.permission.BIND_QUICK_SETTINGS_TILE";
field public static final java.lang.String BIND_REMOTEVIEWS = "android.permission.BIND_REMOTEVIEWS";
+ field public static final java.lang.String BIND_RESOLVER_RANKER_SERVICE = "android.permission.BIND_RESOLVER_RANKER_SERVICE";
field public static final java.lang.String BIND_RUNTIME_PERMISSION_PRESENTER_SERVICE = "android.permission.BIND_RUNTIME_PERMISSION_PRESENTER_SERVICE";
field public static final java.lang.String BIND_SCREENING_SERVICE = "android.permission.BIND_SCREENING_SERVICE";
field public static final java.lang.String BIND_TELECOM_CONNECTION_SERVICE = "android.permission.BIND_TELECOM_CONNECTION_SERVICE";
}
+package android.service.resolver {
+
+ public abstract class ResolverRankerService extends android.app.Service {
+ ctor public ResolverRankerService();
+ method public android.os.IBinder onBind(android.content.Intent);
+ method public void onPredictSharingProbabilities(java.util.List<android.service.resolver.ResolverTarget>);
+ method public void onTrainRankingModel(java.util.List<android.service.resolver.ResolverTarget>, int);
+ field public static final java.lang.String BIND_PERMISSION = "android.permission.BIND_RESOLVER_RANKER_SERVICE";
+ field public static final java.lang.String SERVICE_INTERFACE = "android.service.resolver.ResolverRankerService";
+ }
+
+ public final class ResolverTarget implements android.os.Parcelable {
+ ctor public ResolverTarget();
+ method public int describeContents();
+ method public float getChooserScore();
+ method public float getLaunchScore();
+ method public float getRecencyScore();
+ method public float getSelectProbability();
+ method public float getTimeSpentScore();
+ method public void setChooserScore(float);
+ method public void setLaunchScore(float);
+ method public void setRecencyScore(float);
+ method public void setSelectProbability(float);
+ method public void setTimeSpentScore(float);
+ method public void writeToParcel(android.os.Parcel, int);
+ field public static final android.os.Parcelable.Creator<android.service.resolver.ResolverTarget> CREATOR;
+ }
+
+}
+
package android.service.restrictions {
public abstract class RestrictionsReceiver extends android.content.BroadcastReceiver {
--- /dev/null
+/*
+ * Copyright (C) 2017 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 android.service.resolver;
+
+import android.service.resolver.ResolverTarget;
+
+/**
+ * @hide
+ */
+oneway interface IResolverRankerResult
+{
+ void sendResult(in List<ResolverTarget> results);
+}
\ No newline at end of file
--- /dev/null
+/*
+ * Copyright (C) 2017 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 android.service.resolver;
+
+import android.service.resolver.IResolverRankerResult;
+import android.service.resolver.ResolverTarget;
+
+/**
+ * @hide
+ */
+oneway interface IResolverRankerService
+{
+ void predict(in List<ResolverTarget> targets, IResolverRankerResult result);
+ void train(in List<ResolverTarget> targets, int selectedPosition);
+}
\ No newline at end of file
--- /dev/null
+/*
+ * Copyright (C) 2017 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 android.service.resolver;
+
+import android.annotation.SdkConstant;
+import android.annotation.SystemApi;
+import android.app.Service;
+import android.content.ComponentName;
+import android.content.Intent;
+import android.os.IBinder;
+import android.os.Handler;
+import android.os.HandlerThread;
+import android.os.RemoteException;
+import android.service.resolver.ResolverTarget;
+import android.util.Log;
+
+import java.util.List;
+import java.util.Map;
+
+/**
+ * A service to rank apps according to usage stats of apps, when the system is resolving targets for
+ * an Intent.
+ *
+ * <p>To extend this class, you must declare the service in your manifest file with the
+ * {@link android.Manifest.permission#BIND_RESOLVER_RANKER_SERVICE} permission, and include an
+ * intent filter with the {@link #SERVICE_INTERFACE} action. For example:</p>
+ * <pre>
+ * <service android:name=".MyResolverRankerService"
+ * android:exported="true"
+ * android:priority="100"
+ * android:permission="android.permission.BIND_RESOLVER_RANKER_SERVICE">
+ * <intent-filter>
+ * <action android:name="android.service.resolver.ResolverRankerService" />
+ * </intent-filter>
+ * </service>
+ * </pre>
+ * @hide
+ */
+@SystemApi
+public abstract class ResolverRankerService extends Service {
+
+ private static final String TAG = "ResolverRankerService";
+
+ private static final boolean DEBUG = false;
+
+ /**
+ * The Intent action that a service must respond to. Add it to the intent filter of the service
+ * in its manifest.
+ */
+ @SdkConstant(SdkConstant.SdkConstantType.SERVICE_ACTION)
+ public static final String SERVICE_INTERFACE = "android.service.resolver.ResolverRankerService";
+
+ /**
+ * The permission that a service must require to ensure that only Android system can bind to it.
+ * If this permission is not enforced in the AndroidManifest of the service, the system will
+ * skip that service.
+ */
+ public static final String BIND_PERMISSION = "android.permission.BIND_RESOLVER_RANKER_SERVICE";
+
+ private ResolverRankerServiceWrapper mWrapper = null;
+
+ /**
+ * Called by the system to retrieve a list of probabilities to rank apps/options. To implement
+ * it, set selectProbability of each input {@link ResolverTarget}. The higher the
+ * selectProbability is, the more likely the {@link ResolverTarget} will be selected by the
+ * user. Override this function to provide prediction results.
+ *
+ * @param targets a list of {@link ResolverTarget}, for the list of apps to be ranked.
+ *
+ * @throws Exception when the prediction task fails.
+ */
+ public void onPredictSharingProbabilities(final List<ResolverTarget> targets) {}
+
+ /**
+ * Called by the system to train/update a ranking service, after the user makes a selection from
+ * the ranked list of apps. Override this function to enable model updates.
+ *
+ * @param targets a list of {@link ResolverTarget}, for the list of apps to be ranked.
+ * @param selectedPosition the position of the selected app in the list.
+ *
+ * @throws Exception when the training task fails.
+ */
+ public void onTrainRankingModel(
+ final List<ResolverTarget> targets, final int selectedPosition) {}
+
+ private static final String HANDLER_THREAD_NAME = "RESOLVER_RANKER_SERVICE";
+ private volatile Handler mHandler;
+ private HandlerThread mHandlerThread;
+
+ @Override
+ public IBinder onBind(Intent intent) {
+ if (DEBUG) Log.d(TAG, "onBind " + intent);
+ if (!SERVICE_INTERFACE.equals(intent.getAction())) {
+ if (DEBUG) Log.d(TAG, "bad intent action " + intent.getAction() + "; returning null");
+ return null;
+ }
+ if (mHandlerThread == null) {
+ mHandlerThread = new HandlerThread(HANDLER_THREAD_NAME);
+ mHandlerThread.start();
+ mHandler = new Handler(mHandlerThread.getLooper());
+ }
+ if (mWrapper == null) {
+ mWrapper = new ResolverRankerServiceWrapper();
+ }
+ return mWrapper;
+ }
+
+ @Override
+ public void onDestroy() {
+ mHandler = null;
+ if (mHandlerThread != null) {
+ mHandlerThread.quitSafely();
+ }
+ super.onDestroy();
+ }
+
+ private static void sendResult(List<ResolverTarget> targets, IResolverRankerResult result) {
+ try {
+ result.sendResult(targets);
+ } catch (Exception e) {
+ Log.e(TAG, "failed to send results: " + e);
+ }
+ }
+
+ private class ResolverRankerServiceWrapper extends IResolverRankerService.Stub {
+
+ @Override
+ public void predict(final List<ResolverTarget> targets, final IResolverRankerResult result)
+ throws RemoteException {
+ Runnable predictRunnable = new Runnable() {
+ @Override
+ public void run() {
+ try {
+ if (DEBUG) {
+ Log.d(TAG, "predict calls onPredictSharingProbabilities.");
+ }
+ onPredictSharingProbabilities(targets);
+ sendResult(targets, result);
+ } catch (Exception e) {
+ Log.e(TAG, "onPredictSharingProbabilities failed; send null results: " + e);
+ sendResult(null, result);
+ }
+ }
+ };
+ final Handler h = mHandler;
+ if (h != null) {
+ h.post(predictRunnable);
+ }
+ }
+
+ @Override
+ public void train(final List<ResolverTarget> targets, final int selectedPosition)
+ throws RemoteException {
+ Runnable trainRunnable = new Runnable() {
+ @Override
+ public void run() {
+ try {
+ if (DEBUG) {
+ Log.d(TAG, "train calls onTranRankingModel");
+ }
+ onTrainRankingModel(targets, selectedPosition);
+ } catch (Exception e) {
+ Log.e(TAG, "onTrainRankingModel failed; skip train: " + e);
+ }
+ }
+ };
+ final Handler h = mHandler;
+ if (h != null) {
+ h.post(trainRunnable);
+ }
+ }
+ }
+}
--- /dev/null
+/*
+ * Copyright (C) 2017 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 android.service.resolver;
+
+/**
+ * @hide
+ */
+parcelable ResolverTarget;
--- /dev/null
+/*
+ * Copyright (C) 2017 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 android.service.resolver;
+
+import android.annotation.SystemApi;
+import android.os.Bundle;
+import android.os.Parcel;
+import android.os.Parcelable;
+import android.util.ArrayMap;
+
+import java.util.Map;
+
+/**
+ * A ResolverTarget contains features by which an app or option will be ranked, in
+ * {@link ResolverRankerService}.
+ * @hide
+ */
+@SystemApi
+public final class ResolverTarget implements Parcelable {
+ private static final String TAG = "ResolverTarget";
+
+ /**
+ * a float score for recency of last use.
+ */
+ private float mRecencyScore;
+
+ /**
+ * a float score for total time spent.
+ */
+ private float mTimeSpentScore;
+
+ /**
+ * a float score for number of launches.
+ */
+ private float mLaunchScore;
+
+ /**
+ * a float score for number of selected.
+ */
+ private float mChooserScore;
+
+ /**
+ * a float score for the probability to be selected.
+ */
+ private float mSelectProbability;
+
+ // constructor for the class.
+ public ResolverTarget() {}
+
+ ResolverTarget(Parcel in) {
+ mRecencyScore = in.readFloat();
+ mTimeSpentScore = in.readFloat();
+ mLaunchScore = in.readFloat();
+ mChooserScore = in.readFloat();
+ mSelectProbability = in.readFloat();
+ }
+
+ /**
+ * Gets the score for how recently the target was used in the foreground.
+ *
+ * @return a float score whose range is [0, 1]. The higher the score is, the more recently the
+ * target was used.
+ */
+ public float getRecencyScore() {
+ return mRecencyScore;
+ }
+
+ /**
+ * Sets the score for how recently the target was used in the foreground.
+ *
+ * @param recencyScore a float score whose range is [0, 1]. The higher the score is, the more
+ * recently the target was used.
+ */
+ public void setRecencyScore(float recencyScore) {
+ this.mRecencyScore = recencyScore;
+ }
+
+ /**
+ * Gets the score for how long the target has been used in the foreground.
+ *
+ * @return a float score whose range is [0, 1]. The higher the score is, the longer the target
+ * has been used for.
+ */
+ public float getTimeSpentScore() {
+ return mTimeSpentScore;
+ }
+
+ /**
+ * Sets the score for how long the target has been used in the foreground.
+ *
+ * @param timeSpentScore a float score whose range is [0, 1]. The higher the score is, the
+ * longer the target has been used for.
+ */
+ public void setTimeSpentScore(float timeSpentScore) {
+ this.mTimeSpentScore = timeSpentScore;
+ }
+
+ /**
+ * Gets the score for how many times the target has been launched to the foreground.
+ *
+ * @return a float score whose range is [0, 1]. The higher the score is, the more times the
+ * target has been launched.
+ */
+ public float getLaunchScore() {
+ return mLaunchScore;
+ }
+
+ /**
+ * Sets the score for how many times the target has been launched to the foreground.
+ *
+ * @param launchScore a float score whose range is [0, 1]. The higher the score is, the more
+ * times the target has been launched.
+ */
+ public void setLaunchScore(float launchScore) {
+ this.mLaunchScore = launchScore;
+ }
+
+ /**
+ * Gets the score for how many times the target has been selected by the user to share the same
+ * types of content.
+ *
+ * @return a float score whose range is [0, 1]. The higher the score is, the
+ * more times the target has been selected by the user to share the same types of content for.
+ */
+ public float getChooserScore() {
+ return mChooserScore;
+ }
+
+ /**
+ * Sets the score for how many times the target has been selected by the user to share the same
+ * types of content.
+ *
+ * @param chooserScore a float score whose range is [0, 1]. The higher the score is, the more
+ * times the target has been selected by the user to share the same types
+ * of content for.
+ */
+ public void setChooserScore(float chooserScore) {
+ this.mChooserScore = chooserScore;
+ }
+
+ /**
+ * Gets the probability of how likely this target will be selected by the user.
+ *
+ * @return a float score whose range is [0, 1]. The higher the score is, the more likely the
+ * user is going to select this target.
+ */
+ public float getSelectProbability() {
+ return mSelectProbability;
+ }
+
+ /**
+ * Sets the probability for how like this target will be selected by the user.
+ *
+ * @param selectProbability a float score whose range is [0, 1]. The higher the score is, the
+ * more likely tht user is going to select this target.
+ */
+ public void setSelectProbability(float selectProbability) {
+ this.mSelectProbability = selectProbability;
+ }
+
+ // serialize the class to a string.
+ @Override
+ public String toString() {
+ return "ResolverTarget{"
+ + mRecencyScore + ", "
+ + mTimeSpentScore + ", "
+ + mLaunchScore + ", "
+ + mChooserScore + ", "
+ + mSelectProbability + "}";
+ }
+
+ // describes the kinds of special objects contained in this Parcelable instance's marshaled
+ // representation.
+ @Override
+ public int describeContents() {
+ return 0;
+ }
+
+ // flattens this object in to a Parcel.
+ @Override
+ public void writeToParcel(Parcel dest, int flags) {
+ dest.writeFloat(mRecencyScore);
+ dest.writeFloat(mTimeSpentScore);
+ dest.writeFloat(mLaunchScore);
+ dest.writeFloat(mChooserScore);
+ dest.writeFloat(mSelectProbability);
+ }
+
+ // creator definition for the class.
+ public static final Creator<ResolverTarget> CREATOR
+ = new Creator<ResolverTarget>() {
+ @Override
+ public ResolverTarget createFromParcel(Parcel source) {
+ return new ResolverTarget(source);
+ }
+
+ @Override
+ public ResolverTarget[] newArray(int size) {
+ return new ResolverTarget[size];
+ }
+ };
+}
--- /dev/null
+/*
+ * Copyright (C) 2017 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.internal.app;
+
+import android.content.Context;
+import android.content.Intent;
+import android.content.SharedPreferences;
+import android.os.Environment;
+import android.os.IBinder;
+import android.os.storage.StorageManager;
+import android.service.resolver.ResolverRankerService;
+import android.service.resolver.ResolverTarget;
+import android.util.ArrayMap;
+import android.util.Log;
+
+import java.io.File;
+import java.util.Collection;
+import java.util.List;
+import java.util.Map;
+
+/**
+ * A Logistic Regression based {@link android.service.resolver.ResolverRankerService}, to be used
+ * in {@link ResolverComparator}.
+ */
+public final class LRResolverRankerService extends ResolverRankerService {
+ private static final String TAG = "LRResolverRankerService";
+
+ private static final boolean DEBUG = false;
+
+ private static final String PARAM_SHARED_PREF_NAME = "resolver_ranker_params";
+ private static final String BIAS_PREF_KEY = "bias";
+ private static final String VERSION_PREF_KEY = "version";
+
+ private static final String LAUNCH_SCORE = "launch";
+ private static final String TIME_SPENT_SCORE = "timeSpent";
+ private static final String RECENCY_SCORE = "recency";
+ private static final String CHOOSER_SCORE = "chooser";
+
+ // parameters for a pre-trained model, to initialize the app ranker. When updating the
+ // pre-trained model, please update these params, as well as initModel().
+ private static final int CURRENT_VERSION = 1;
+ private static final float LEARNING_RATE = 0.0001f;
+ private static final float REGULARIZER_PARAM = 0.0001f;
+
+ private SharedPreferences mParamSharedPref;
+ private ArrayMap<String, Float> mFeatureWeights;
+ private float mBias;
+
+ @Override
+ public IBinder onBind(Intent intent) {
+ initModel();
+ return super.onBind(intent);
+ }
+
+ @Override
+ public void onPredictSharingProbabilities(List<ResolverTarget> targets) {
+ final int size = targets.size();
+ for (int i = 0; i < size; ++i) {
+ ResolverTarget target = targets.get(i);
+ ArrayMap<String, Float> features = getFeatures(target);
+ target.setSelectProbability(predict(features));
+ }
+ }
+
+ @Override
+ public void onTrainRankingModel(List<ResolverTarget> targets, int selectedPosition) {
+ final int size = targets.size();
+ if (selectedPosition < 0 || selectedPosition >= size) {
+ if (DEBUG) {
+ Log.d(TAG, "Invalid Position of Selected App " + selectedPosition);
+ }
+ return;
+ }
+ final ArrayMap<String, Float> positive = getFeatures(targets.get(selectedPosition));
+ final float positiveProbability = targets.get(selectedPosition).getSelectProbability();
+ final int targetSize = targets.size();
+ for (int i = 0; i < targetSize; ++i) {
+ if (i == selectedPosition) {
+ continue;
+ }
+ final ArrayMap<String, Float> negative = getFeatures(targets.get(i));
+ final float negativeProbability = targets.get(i).getSelectProbability();
+ if (negativeProbability > positiveProbability) {
+ update(negative, negativeProbability, false);
+ update(positive, positiveProbability, true);
+ }
+ }
+ commitUpdate();
+ }
+
+ private void initModel() {
+ mParamSharedPref = getParamSharedPref();
+ mFeatureWeights = new ArrayMap<>(4);
+ if (mParamSharedPref == null ||
+ mParamSharedPref.getInt(VERSION_PREF_KEY, 0) < CURRENT_VERSION) {
+ // Initializing the app ranker to a pre-trained model. When updating the pre-trained
+ // model, please increment CURRENT_VERSION, and update LEARNING_RATE and
+ // REGULARIZER_PARAM.
+ mBias = -1.6568f;
+ mFeatureWeights.put(LAUNCH_SCORE, 2.5543f);
+ mFeatureWeights.put(TIME_SPENT_SCORE, 2.8412f);
+ mFeatureWeights.put(RECENCY_SCORE, 0.269f);
+ mFeatureWeights.put(CHOOSER_SCORE, 4.2222f);
+ } else {
+ mBias = mParamSharedPref.getFloat(BIAS_PREF_KEY, 0.0f);
+ mFeatureWeights.put(LAUNCH_SCORE, mParamSharedPref.getFloat(LAUNCH_SCORE, 0.0f));
+ mFeatureWeights.put(
+ TIME_SPENT_SCORE, mParamSharedPref.getFloat(TIME_SPENT_SCORE, 0.0f));
+ mFeatureWeights.put(RECENCY_SCORE, mParamSharedPref.getFloat(RECENCY_SCORE, 0.0f));
+ mFeatureWeights.put(CHOOSER_SCORE, mParamSharedPref.getFloat(CHOOSER_SCORE, 0.0f));
+ }
+ }
+
+ private ArrayMap<String, Float> getFeatures(ResolverTarget target) {
+ ArrayMap<String, Float> features = new ArrayMap<>(4);
+ features.put(RECENCY_SCORE, target.getRecencyScore());
+ features.put(TIME_SPENT_SCORE, target.getTimeSpentScore());
+ features.put(LAUNCH_SCORE, target.getLaunchScore());
+ features.put(CHOOSER_SCORE, target.getChooserScore());
+ return features;
+ }
+
+ private float predict(ArrayMap<String, Float> target) {
+ if (target == null) {
+ return 0.0f;
+ }
+ final int featureSize = target.size();
+ float sum = 0.0f;
+ for (int i = 0; i < featureSize; i++) {
+ String featureName = target.keyAt(i);
+ float weight = mFeatureWeights.getOrDefault(featureName, 0.0f);
+ sum += weight * target.valueAt(i);
+ }
+ return (float) (1.0 / (1.0 + Math.exp(-mBias - sum)));
+ }
+
+ private void update(ArrayMap<String, Float> target, float predict, boolean isSelected) {
+ if (target == null) {
+ return;
+ }
+ final int featureSize = target.size();
+ float error = isSelected ? 1.0f - predict : -predict;
+ for (int i = 0; i < featureSize; i++) {
+ String featureName = target.keyAt(i);
+ float currentWeight = mFeatureWeights.getOrDefault(featureName, 0.0f);
+ mBias += LEARNING_RATE * error;
+ currentWeight = currentWeight - LEARNING_RATE * REGULARIZER_PARAM * currentWeight +
+ LEARNING_RATE * error * target.valueAt(i);
+ mFeatureWeights.put(featureName, currentWeight);
+ }
+ if (DEBUG) {
+ Log.d(TAG, "Weights: " + mFeatureWeights + " Bias: " + mBias);
+ }
+ }
+
+ private void commitUpdate() {
+ try {
+ SharedPreferences.Editor editor = mParamSharedPref.edit();
+ editor.putFloat(BIAS_PREF_KEY, mBias);
+ final int size = mFeatureWeights.size();
+ for (int i = 0; i < size; i++) {
+ editor.putFloat(mFeatureWeights.keyAt(i), mFeatureWeights.valueAt(i));
+ }
+ editor.putInt(VERSION_PREF_KEY, CURRENT_VERSION);
+ editor.apply();
+ } catch (Exception e) {
+ Log.e(TAG, "Failed to commit update" + e);
+ }
+ }
+
+ private SharedPreferences getParamSharedPref() {
+ // The package info in the context isn't initialized in the way it is for normal apps,
+ // so the standard, name-based context.getSharedPreferences doesn't work. Instead, we
+ // build the path manually below using the same policy that appears in ContextImpl.
+ if (DEBUG) {
+ Log.d(TAG, "Context Package Name: " + getPackageName());
+ }
+ final File prefsFile = new File(new File(
+ Environment.getDataUserCePackageDirectory(
+ StorageManager.UUID_PRIVATE_INTERNAL, getUserId(), getPackageName()),
+ "shared_prefs"),
+ PARAM_SHARED_PREF_NAME + ".xml");
+ return getSharedPreferences(prefsFile, Context.MODE_PRIVATE);
+ }
+}
\ No newline at end of file
getMainThreadHandler().removeCallbacks(mPostListReadyRunnable);
mPostListReadyRunnable = null;
}
+ if (mAdapter != null && mAdapter.mResolverListController != null) {
+ mAdapter.mResolverListController.destroy();
+ }
}
@Override
import android.content.pm.ApplicationInfo;
import android.content.pm.ComponentInfo;
import android.content.pm.PackageManager;
+import android.content.pm.PackageManager.NameNotFoundException;
import android.content.pm.ResolveInfo;
import android.content.SharedPreferences;
+import android.content.ServiceConnection;
import android.os.Environment;
+import android.os.Handler;
+import android.os.IBinder;
+import android.os.Looper;
+import android.os.Message;
+import android.os.RemoteException;
import android.os.storage.StorageManager;
import android.os.UserHandle;
+import android.service.resolver.IResolverRankerService;
+import android.service.resolver.IResolverRankerResult;
+import android.service.resolver.ResolverRankerService;
+import android.service.resolver.ResolverTarget;
import android.text.TextUtils;
import android.util.ArrayMap;
import android.util.Log;
import com.android.internal.app.ResolverActivity.ResolvedComponentInfo;
import java.io.File;
+import java.lang.InterruptedException;
import java.text.Collator;
import java.util.ArrayList;
import java.util.Comparator;
+import java.util.concurrent.CountDownLatch;
+import java.util.concurrent.TimeUnit;
import java.util.LinkedHashMap;
import java.util.List;
import java.util.Map;
private static final float RECENCY_MULTIPLIER = 2.f;
- // feature names used in ranking.
- private static final String LAUNCH_SCORE = "launch";
- private static final String TIME_SPENT_SCORE = "timeSpent";
- private static final String RECENCY_SCORE = "recency";
- private static final String CHOOSER_SCORE = "chooser";
+ // message types
+ private static final int RESOLVER_RANKER_SERVICE_RESULT = 0;
+ private static final int RESOLVER_RANKER_RESULT_TIMEOUT = 1;
+
+ // timeout for establishing connections with a ResolverRankerService.
+ private static final int CONNECTION_COST_TIMEOUT_MILLIS = 200;
+ // timeout for establishing connections with a ResolverRankerService, collecting features and
+ // predicting ranking scores.
+ private static final int WATCHDOG_TIMEOUT_MILLIS = 500;
private final Collator mCollator;
private final boolean mHttp;
private final Map<String, UsageStats> mStats;
private final long mCurrentTime;
private final long mSinceTime;
- private final LinkedHashMap<ComponentName, ScoredTarget> mScoredTargets = new LinkedHashMap<>();
+ private final LinkedHashMap<ComponentName, ResolverTarget> mTargetsDict = new LinkedHashMap<>();
private final String mReferrerPackage;
+ private final Object mLock = new Object();
+ private ArrayList<ResolverTarget> mTargets;
private String mContentType;
private String[] mAnnotations;
private String mAction;
- private LogisticRegressionAppRanker mRanker;
+ private IResolverRankerService mRanker;
+ private ResolverRankerServiceConnection mConnection;
+ private AfterCompute mAfterCompute;
+ private Context mContext;
+ private CountDownLatch mConnectSignal;
+
+ private final Handler mHandler = new Handler(Looper.getMainLooper()) {
+ public void handleMessage(Message msg) {
+ switch (msg.what) {
+ case RESOLVER_RANKER_SERVICE_RESULT:
+ if (DEBUG) {
+ Log.d(TAG, "RESOLVER_RANKER_SERVICE_RESULT");
+ }
+ if (mHandler.hasMessages(RESOLVER_RANKER_RESULT_TIMEOUT)) {
+ if (msg.obj != null) {
+ final List<ResolverTarget> receivedTargets =
+ (List<ResolverTarget>) msg.obj;
+ if (receivedTargets != null && mTargets != null
+ && receivedTargets.size() == mTargets.size()) {
+ final int size = mTargets.size();
+ for (int i = 0; i < size; ++i) {
+ mTargets.get(i).setSelectProbability(
+ receivedTargets.get(i).getSelectProbability());
+ }
+ } else {
+ Log.e(TAG, "Sizes of sent and received ResolverTargets diff.");
+ }
+ } else {
+ Log.e(TAG, "Receiving null prediction results.");
+ }
+ mHandler.removeMessages(RESOLVER_RANKER_RESULT_TIMEOUT);
+ mAfterCompute.afterCompute();
+ }
+ break;
+
+ case RESOLVER_RANKER_RESULT_TIMEOUT:
+ if (DEBUG) {
+ Log.d(TAG, "RESOLVER_RANKER_RESULT_TIMEOUT; unbinding services");
+ }
+ mHandler.removeMessages(RESOLVER_RANKER_SERVICE_RESULT);
+ mAfterCompute.afterCompute();
+ break;
- public ResolverComparator(Context context, Intent intent, String referrerPackage) {
+ default:
+ super.handleMessage(msg);
+ }
+ }
+ };
+
+ public interface AfterCompute {
+ public void afterCompute ();
+ }
+
+ public ResolverComparator(Context context, Intent intent, String referrerPackage,
+ AfterCompute afterCompute) {
mCollator = Collator.getInstance(context.getResources().getConfiguration().locale);
String scheme = intent.getScheme();
mHttp = "http".equals(scheme) || "https".equals(scheme);
mReferrerPackage = referrerPackage;
+ mAfterCompute = afterCompute;
+ mContext = context;
mPm = context.getPackageManager();
mUsm = (UsageStatsManager) context.getSystemService(Context.USAGE_STATS_SERVICE);
mContentType = intent.getType();
getContentAnnotations(intent);
mAction = intent.getAction();
- mRanker = new LogisticRegressionAppRanker(context);
}
+ // get annotations of content from intent.
public void getContentAnnotations(Intent intent) {
ArrayList<String> annotations = intent.getStringArrayListExtra(
Intent.EXTRA_CONTENT_ANNOTATIONS);
}
}
+ public void setCallBack(AfterCompute afterCompute) {
+ mAfterCompute = afterCompute;
+ }
+
+ // compute features for each target according to usage stats of targets.
public void compute(List<ResolvedComponentInfo> targets) {
- mScoredTargets.clear();
+ reset();
final long recentSinceTime = mCurrentTime - RECENCY_TIME_PERIOD;
- long mostRecentlyUsedTime = recentSinceTime + 1;
- long mostTimeSpent = 1;
- int mostLaunched = 1;
- int mostSelected = 1;
+ float mostRecencyScore = 1.0f;
+ float mostTimeSpentScore = 1.0f;
+ float mostLaunchScore = 1.0f;
+ float mostChooserScore = 1.0f;
for (ResolvedComponentInfo target : targets) {
- final ScoredTarget scoredTarget
- = new ScoredTarget(target.getResolveInfoAt(0).activityInfo);
- mScoredTargets.put(target.name, scoredTarget);
+ final ResolverTarget resolverTarget = new ResolverTarget();
+ mTargetsDict.put(target.name, resolverTarget);
final UsageStats pkStats = mStats.get(target.name.getPackageName());
if (pkStats != null) {
// Only count recency for apps that weren't the caller
// Persistent processes muck this up, so omit them too.
if (!target.name.getPackageName().equals(mReferrerPackage)
&& !isPersistentProcess(target)) {
- final long lastTimeUsed = pkStats.getLastTimeUsed();
- scoredTarget.lastTimeUsed = lastTimeUsed;
- if (lastTimeUsed > mostRecentlyUsedTime) {
- mostRecentlyUsedTime = lastTimeUsed;
+ final float recencyScore =
+ (float) Math.max(pkStats.getLastTimeUsed() - recentSinceTime, 0);
+ resolverTarget.setRecencyScore(recencyScore);
+ if (recencyScore > mostRecencyScore) {
+ mostRecencyScore = recencyScore;
}
}
- final long timeSpent = pkStats.getTotalTimeInForeground();
- scoredTarget.timeSpent = timeSpent;
- if (timeSpent > mostTimeSpent) {
- mostTimeSpent = timeSpent;
+ final float timeSpentScore = (float) pkStats.getTotalTimeInForeground();
+ resolverTarget.setTimeSpentScore(timeSpentScore);
+ if (timeSpentScore > mostTimeSpentScore) {
+ mostTimeSpentScore = timeSpentScore;
}
- final int launched = pkStats.mLaunchCount;
- scoredTarget.launchCount = launched;
- if (launched > mostLaunched) {
- mostLaunched = launched;
+ final float launchScore = (float) pkStats.mLaunchCount;
+ resolverTarget.setLaunchScore(launchScore);
+ if (launchScore > mostLaunchScore) {
+ mostLaunchScore = launchScore;
}
- int selected = 0;
+ float chooserScore = 0.0f;
if (pkStats.mChooserCounts != null && mAction != null
&& pkStats.mChooserCounts.get(mAction) != null) {
- selected = pkStats.mChooserCounts.get(mAction).getOrDefault(mContentType, 0);
+ chooserScore = (float) pkStats.mChooserCounts.get(mAction)
+ .getOrDefault(mContentType, 0);
if (mAnnotations != null) {
final int size = mAnnotations.length;
for (int i = 0; i < size; i++) {
- selected += pkStats.mChooserCounts.get(mAction)
+ chooserScore += (float) pkStats.mChooserCounts.get(mAction)
.getOrDefault(mAnnotations[i], 0);
}
}
Log.d(TAG, "Action type is null");
} else {
Log.d(TAG, "Chooser Count of " + mAction + ":" +
- target.name.getPackageName() + " is " + Integer.toString(selected));
+ target.name.getPackageName() + " is " +
+ Float.toString(chooserScore));
}
}
- scoredTarget.chooserCount = selected;
- if (selected > mostSelected) {
- mostSelected = selected;
+ resolverTarget.setChooserScore(chooserScore);
+ if (chooserScore > mostChooserScore) {
+ mostChooserScore = chooserScore;
}
}
}
-
if (DEBUG) {
- Log.d(TAG, "compute - mostRecentlyUsedTime: " + mostRecentlyUsedTime
- + " mostTimeSpent: " + mostTimeSpent
- + " recentSinceTime: " + recentSinceTime
- + " mostLaunched: " + mostLaunched);
+ Log.d(TAG, "compute - mostRecencyScore: " + mostRecencyScore
+ + " mostTimeSpentScore: " + mostTimeSpentScore
+ + " mostLaunchScore: " + mostLaunchScore
+ + " mostChooserScore: " + mostChooserScore);
}
- for (ScoredTarget target : mScoredTargets.values()) {
- final float recency = (float) Math.max(target.lastTimeUsed - recentSinceTime, 0)
- / (mostRecentlyUsedTime - recentSinceTime);
- target.setFeatures((float) target.launchCount / mostLaunched,
- (float) target.timeSpent / mostTimeSpent,
- recency * recency * RECENCY_MULTIPLIER,
- (float) target.chooserCount / mostSelected);
- target.selectProb = mRanker.predict(target.getFeatures());
+ mTargets = new ArrayList<>(mTargetsDict.values());
+ for (ResolverTarget target : mTargets) {
+ final float recency = target.getRecencyScore() / mostRecencyScore;
+ setFeatures(target, recency * recency * RECENCY_MULTIPLIER,
+ target.getLaunchScore() / mostLaunchScore,
+ target.getTimeSpentScore() / mostTimeSpentScore,
+ target.getChooserScore() / mostChooserScore);
+ addDefaultSelectProbability(target);
if (DEBUG) {
Log.d(TAG, "Scores: " + target);
}
}
- }
-
- static boolean isPersistentProcess(ResolvedComponentInfo rci) {
- if (rci != null && rci.getCount() > 0) {
- return (rci.getResolveInfoAt(0).activityInfo.applicationInfo.flags &
- ApplicationInfo.FLAG_PERSISTENT) != 0;
- }
- return false;
+ predictSelectProbabilities(mTargets);
}
@Override
// Pinned items stay stable within a normal lexical sort and ignore scoring.
if (!lPinned && !rPinned) {
if (mStats != null) {
- final ScoredTarget lhsTarget = mScoredTargets.get(new ComponentName(
+ final ResolverTarget lhsTarget = mTargetsDict.get(new ComponentName(
lhs.activityInfo.packageName, lhs.activityInfo.name));
- final ScoredTarget rhsTarget = mScoredTargets.get(new ComponentName(
+ final ResolverTarget rhsTarget = mTargetsDict.get(new ComponentName(
rhs.activityInfo.packageName, rhs.activityInfo.name));
- final int selectProbDiff = Float.compare(
- rhsTarget.selectProb, lhsTarget.selectProb);
+ final int selectProbabilityDiff = Float.compare(
+ rhsTarget.getSelectProbability(), lhsTarget.getSelectProbability());
- if (selectProbDiff != 0) {
- return selectProbDiff > 0 ? 1 : -1;
+ if (selectProbabilityDiff != 0) {
+ return selectProbabilityDiff > 0 ? 1 : -1;
}
}
}
}
public float getScore(ComponentName name) {
- final ScoredTarget target = mScoredTargets.get(name);
+ final ResolverTarget target = mTargetsDict.get(name);
if (target != null) {
- return target.selectProb;
+ return target.getSelectProbability();
}
return 0;
}
- static class ScoredTarget {
- public final ComponentInfo componentInfo;
- public long lastTimeUsed;
- public long timeSpent;
- public long launchCount;
- public long chooserCount;
- public ArrayMap<String, Float> features;
- public float selectProb;
-
- public ScoredTarget(ComponentInfo ci) {
- componentInfo = ci;
- features = new ArrayMap<>(5);
+ public void updateChooserCounts(String packageName, int userId, String action) {
+ if (mUsm != null) {
+ mUsm.reportChooserSelection(packageName, userId, mContentType, mAnnotations, action);
}
+ }
- @Override
- public String toString() {
- return "ScoredTarget{" + componentInfo
- + " lastTimeUsed: " + lastTimeUsed
- + " timeSpent: " + timeSpent
- + " launchCount: " + launchCount
- + " chooserCount: " + chooserCount
- + " selectProb: " + selectProb
- + "}";
+ // update ranking model when the connection to it is valid.
+ public void updateModel(ComponentName componentName) {
+ synchronized (mLock) {
+ if (mRanker != null) {
+ try {
+ int selectedPos = new ArrayList<ComponentName>(mTargetsDict.keySet())
+ .indexOf(componentName);
+ if (selectedPos > 0) {
+ mRanker.train(mTargets, selectedPos);
+ } else {
+ if (DEBUG) {
+ Log.d(TAG, "Selected a unknown component: " + componentName);
+ }
+ }
+ } catch (RemoteException e) {
+ Log.e(TAG, "Error in Train: " + e);
+ }
+ } else {
+ if (DEBUG) {
+ Log.d(TAG, "Ranker is null; skip updateModel.");
+ }
+ }
}
+ }
- public void setFeatures(float launchCountScore, float usageTimeScore, float recencyScore,
- float chooserCountScore) {
- features.put(LAUNCH_SCORE, launchCountScore);
- features.put(TIME_SPENT_SCORE, usageTimeScore);
- features.put(RECENCY_SCORE, recencyScore);
- features.put(CHOOSER_SCORE, chooserCountScore);
+ // unbind the service and clear unhandled messges.
+ public void destroy() {
+ mHandler.removeMessages(RESOLVER_RANKER_SERVICE_RESULT);
+ mHandler.removeMessages(RESOLVER_RANKER_RESULT_TIMEOUT);
+ if (mConnection != null) {
+ mContext.unbindService(mConnection);
+ mConnection.destroy();
}
-
- public ArrayMap<String, Float> getFeatures() {
- return features;
+ if (DEBUG) {
+ Log.d(TAG, "Unbinded Resolver Ranker.");
}
}
- public void updateChooserCounts(String packageName, int userId, String action) {
- if (mUsm != null) {
- mUsm.reportChooserSelection(packageName, userId, mContentType, mAnnotations, action);
+ // connect to a ranking service.
+ private void initRanker(Context context) {
+ synchronized (mLock) {
+ if (mConnection != null && mRanker != null) {
+ if (DEBUG) {
+ Log.d(TAG, "Ranker still exists; reusing the existing one.");
+ }
+ return;
+ }
}
- }
-
- public void updateModel(ComponentName componentName) {
- if (mScoredTargets == null || componentName == null ||
- !mScoredTargets.containsKey(componentName)) {
+ Intent intent = resolveRankerService();
+ if (intent == null) {
return;
}
- ScoredTarget selected = mScoredTargets.get(componentName);
- for (ComponentName targetComponent : mScoredTargets.keySet()) {
- if (targetComponent.equals(componentName)) {
+ mConnectSignal = new CountDownLatch(1);
+ mConnection = new ResolverRankerServiceConnection(mConnectSignal);
+ context.bindServiceAsUser(intent, mConnection, Context.BIND_AUTO_CREATE, UserHandle.SYSTEM);
+ }
+
+ // resolve the service for ranking.
+ private Intent resolveRankerService() {
+ Intent intent = new Intent(ResolverRankerService.SERVICE_INTERFACE);
+ final List<ResolveInfo> resolveInfos = mPm.queryIntentServices(intent, 0);
+ for (ResolveInfo resolveInfo : resolveInfos) {
+ if (resolveInfo == null || resolveInfo.serviceInfo == null
+ || resolveInfo.serviceInfo.applicationInfo == null) {
+ if (DEBUG) {
+ Log.d(TAG, "Failed to retrieve a ranker: " + resolveInfo);
+ }
continue;
}
- ScoredTarget target = mScoredTargets.get(targetComponent);
- // A potential point of optimization. Save updates or derive a closed form for the
- // positive case, to avoid calculating them repeatedly.
- if (target.selectProb >= selected.selectProb) {
- mRanker.update(target.getFeatures(), target.selectProb, false);
- mRanker.update(selected.getFeatures(), selected.selectProb, true);
+ ComponentName componentName = new ComponentName(
+ resolveInfo.serviceInfo.applicationInfo.packageName,
+ resolveInfo.serviceInfo.name);
+ try {
+ final String perm = mPm.getServiceInfo(componentName, 0).permission;
+ if (!ResolverRankerService.BIND_PERMISSION.equals(perm)) {
+ Log.w(TAG, "ResolverRankerService " + componentName + " does not require"
+ + " permission " + ResolverRankerService.BIND_PERMISSION
+ + " - this service will not be queried for ResolverComparator."
+ + " add android:permission=\""
+ + ResolverRankerService.BIND_PERMISSION + "\""
+ + " to the <service> tag for " + componentName
+ + " in the manifest.");
+ continue;
+ }
+ } catch (NameNotFoundException e) {
+ Log.e(TAG, "Could not look up service " + componentName
+ + "; component name not found");
+ continue;
+ }
+ if (DEBUG) {
+ Log.d(TAG, "Succeeded to retrieve a ranker: " + componentName);
}
+ intent.setComponent(componentName);
+ return intent;
}
- mRanker.commitUpdate();
+ return null;
}
- class LogisticRegressionAppRanker {
- private static final String PARAM_SHARED_PREF_NAME = "resolver_ranker_params";
- private static final String BIAS_PREF_KEY = "bias";
- private static final String VERSION_PREF_KEY = "version";
-
- // parameters for a pre-trained model, to initialize the app ranker. When updating the
- // pre-trained model, please update these params, as well as initModel().
- private static final int CURRENT_VERSION = 1;
- private static final float LEARNING_RATE = 0.0001f;
- private static final float REGULARIZER_PARAM = 0.0001f;
+ // set a watchdog, to avoid waiting for ranking service for too long.
+ private void startWatchDog(int timeOutLimit) {
+ if (DEBUG) Log.d(TAG, "Setting watchdog timer for " + timeOutLimit + "ms");
+ if (mHandler == null) {
+ Log.d(TAG, "Error: Handler is Null; Needs to be initialized.");
+ }
+ mHandler.sendEmptyMessageDelayed(RESOLVER_RANKER_RESULT_TIMEOUT, timeOutLimit);
+ }
- private SharedPreferences mParamSharedPref;
- private ArrayMap<String, Float> mFeatureWeights;
- private float mBias;
+ private class ResolverRankerServiceConnection implements ServiceConnection {
+ private final CountDownLatch mConnectSignal;
- public LogisticRegressionAppRanker(Context context) {
- mParamSharedPref = getParamSharedPref(context);
- initModel();
+ public ResolverRankerServiceConnection(CountDownLatch connectSignal) {
+ mConnectSignal = connectSignal;
}
- public float predict(ArrayMap<String, Float> target) {
- if (target == null) {
- return 0.0f;
- }
- final int featureSize = target.size();
- float sum = 0.0f;
- for (int i = 0; i < featureSize; i++) {
- String featureName = target.keyAt(i);
- float weight = mFeatureWeights.getOrDefault(featureName, 0.0f);
- sum += weight * target.valueAt(i);
+ public final IResolverRankerResult resolverRankerResult =
+ new IResolverRankerResult.Stub() {
+ @Override
+ public void sendResult(List<ResolverTarget> targets) throws RemoteException {
+ if (DEBUG) {
+ Log.d(TAG, "Sending Result back to Resolver: " + targets);
+ }
+ synchronized (mLock) {
+ final Message msg = Message.obtain();
+ msg.what = RESOLVER_RANKER_SERVICE_RESULT;
+ msg.obj = targets;
+ mHandler.sendMessage(msg);
+ }
}
- return (float) (1.0 / (1.0 + Math.exp(-mBias - sum)));
- }
+ };
- public void update(ArrayMap<String, Float> target, float predict, boolean isSelected) {
- if (target == null) {
- return;
+ @Override
+ public void onServiceConnected(ComponentName name, IBinder service) {
+ if (DEBUG) {
+ Log.d(TAG, "onServiceConnected: " + name);
}
- final int featureSize = target.size();
- float error = isSelected ? 1.0f - predict : -predict;
- for (int i = 0; i < featureSize; i++) {
- String featureName = target.keyAt(i);
- float currentWeight = mFeatureWeights.getOrDefault(featureName, 0.0f);
- mBias += LEARNING_RATE * error;
- currentWeight = currentWeight - LEARNING_RATE * REGULARIZER_PARAM * currentWeight +
- LEARNING_RATE * error * target.valueAt(i);
- mFeatureWeights.put(featureName, currentWeight);
+ synchronized (mLock) {
+ mRanker = IResolverRankerService.Stub.asInterface(service);
+ mConnectSignal.countDown();
}
+ }
+
+ @Override
+ public void onServiceDisconnected(ComponentName name) {
if (DEBUG) {
- Log.d(TAG, "Weights: " + mFeatureWeights + " Bias: " + mBias);
+ Log.d(TAG, "onServiceDisconnected: " + name);
+ }
+ synchronized (mLock) {
+ destroy();
}
}
- public void commitUpdate() {
- SharedPreferences.Editor editor = mParamSharedPref.edit();
- editor.putFloat(BIAS_PREF_KEY, mBias);
- final int size = mFeatureWeights.size();
- for (int i = 0; i < size; i++) {
- editor.putFloat(mFeatureWeights.keyAt(i), mFeatureWeights.valueAt(i));
+ public void destroy() {
+ synchronized (mLock) {
+ mRanker = null;
}
- editor.putInt(VERSION_PREF_KEY, CURRENT_VERSION);
- editor.apply();
}
+ }
- private SharedPreferences getParamSharedPref(Context context) {
- // The package info in the context isn't initialized in the way it is for normal apps,
- // so the standard, name-based context.getSharedPreferences doesn't work. Instead, we
- // build the path manually below using the same policy that appears in ContextImpl.
+ private void reset() {
+ mTargetsDict.clear();
+ mTargets = null;
+ startWatchDog(WATCHDOG_TIMEOUT_MILLIS);
+ initRanker(mContext);
+ }
+
+ // predict select probabilities if ranking service is valid.
+ private void predictSelectProbabilities(List<ResolverTarget> targets) {
+ if (mConnection == null) {
if (DEBUG) {
- Log.d(TAG, "Context Package Name: " + context.getPackageName());
+ Log.d(TAG, "Has not found valid ResolverRankerService; Skip Prediction");
+ }
+ return;
+ } else {
+ try {
+ mConnectSignal.await(CONNECTION_COST_TIMEOUT_MILLIS, TimeUnit.MILLISECONDS);
+ synchronized (mLock) {
+ if (mRanker != null) {
+ mRanker.predict(targets, mConnection.resolverRankerResult);
+ return;
+ } else {
+ if (DEBUG) {
+ Log.d(TAG, "Ranker has not been initialized; skip predict.");
+ }
+ }
+ }
+ } catch (InterruptedException e) {
+ Log.e(TAG, "Error in Wait for Service Connection.");
+ } catch (RemoteException e) {
+ Log.e(TAG, "Error in Predict: " + e);
}
- final File prefsFile = new File(new File(
- Environment.getDataUserCePackageDirectory(StorageManager.UUID_PRIVATE_INTERNAL,
- context.getUserId(), context.getPackageName()),
- "shared_prefs"),
- PARAM_SHARED_PREF_NAME + ".xml");
- return context.getSharedPreferences(prefsFile, Context.MODE_PRIVATE);
}
+ mAfterCompute.afterCompute();
+ }
- private void initModel() {
- mFeatureWeights = new ArrayMap<>(4);
- if (mParamSharedPref == null ||
- mParamSharedPref.getInt(VERSION_PREF_KEY, 0) < CURRENT_VERSION) {
- // Initializing the app ranker to a pre-trained model. When updating the pre-trained
- // model, please increment CURRENT_VERSION, and update LEARNING_RATE and
- // REGULARIZER_PARAM.
- mBias = -1.6568f;
- mFeatureWeights.put(LAUNCH_SCORE, 2.5543f);
- mFeatureWeights.put(TIME_SPENT_SCORE, 2.8412f);
- mFeatureWeights.put(RECENCY_SCORE, 0.269f);
- mFeatureWeights.put(CHOOSER_SCORE, 4.2222f);
- } else {
- mBias = mParamSharedPref.getFloat(BIAS_PREF_KEY, 0.0f);
- mFeatureWeights.put(LAUNCH_SCORE, mParamSharedPref.getFloat(LAUNCH_SCORE, 0.0f));
- mFeatureWeights.put(
- TIME_SPENT_SCORE, mParamSharedPref.getFloat(TIME_SPENT_SCORE, 0.0f));
- mFeatureWeights.put(RECENCY_SCORE, mParamSharedPref.getFloat(RECENCY_SCORE, 0.0f));
- mFeatureWeights.put(CHOOSER_SCORE, mParamSharedPref.getFloat(CHOOSER_SCORE, 0.0f));
- }
+ // adds select prob as the default values, according to a pre-trained Logistic Regression model.
+ private void addDefaultSelectProbability(ResolverTarget target) {
+ float sum = 2.5543f * target.getLaunchScore() + 2.8412f * target.getTimeSpentScore() +
+ 0.269f * target.getRecencyScore() + 4.2222f * target.getChooserScore();
+ target.setSelectProbability((float) (1.0 / (1.0 + Math.exp(1.6568f - sum))));
+ }
+
+ // sets features for each target
+ private void setFeatures(ResolverTarget target, float recencyScore, float launchScore,
+ float timeSpentScore, float chooserScore) {
+ target.setRecencyScore(recencyScore);
+ target.setLaunchScore(launchScore);
+ target.setTimeSpentScore(timeSpentScore);
+ target.setChooserScore(chooserScore);
+ }
+
+ static boolean isPersistentProcess(ResolvedComponentInfo rci) {
+ if (rci != null && rci.getCount() > 0) {
+ return (rci.getResolveInfoAt(0).activityInfo.applicationInfo.flags &
+ ApplicationInfo.FLAG_PERSISTENT) != 0;
}
+ return false;
}
}
import android.util.Log;
import com.android.internal.annotations.VisibleForTesting;
+import java.lang.InterruptedException;
import java.util.ArrayList;
import java.util.Collections;
+import java.util.concurrent.CountDownLatch;
import java.util.List;
/**
return listToReturn;
}
+ private class ComputeCallback implements ResolverComparator.AfterCompute {
+
+ private CountDownLatch mFinishComputeSignal;
+
+ public ComputeCallback(CountDownLatch finishComputeSignal) {
+ mFinishComputeSignal = finishComputeSignal;
+ }
+
+ public void afterCompute () {
+ mFinishComputeSignal.countDown();
+ }
+ }
+
@VisibleForTesting
@WorkerThread
public void sort(List<ResolverActivity.ResolvedComponentInfo> inputList) {
+ final CountDownLatch finishComputeSignal = new CountDownLatch(1);
+ ComputeCallback callback = new ComputeCallback(finishComputeSignal);
if (mResolverComparator == null) {
- mResolverComparator = new ResolverComparator(mContext, mTargetIntent, mReferrerPackage);
+ mResolverComparator =
+ new ResolverComparator(mContext, mTargetIntent, mReferrerPackage, callback);
+ } else {
+ mResolverComparator.setCallBack(callback);
+ }
+ try {
+ long beforeRank = System.currentTimeMillis();
+ mResolverComparator.compute(inputList);
+ finishComputeSignal.await();
+ Collections.sort(inputList, mResolverComparator);
+ long afterRank = System.currentTimeMillis();
+ if (DEBUG) {
+ Log.d(TAG, "Time Cost: " + Long.toString(afterRank - beforeRank));
+ }
+ } catch (InterruptedException e) {
+ Log.e(TAG, "Compute & Sort was interrupted: " + e);
}
- mResolverComparator.compute(inputList);
- Collections.sort(inputList, mResolverComparator);
}
private static boolean isSameResolvedComponent(ResolveInfo a,
@VisibleForTesting
public float getScore(ResolverActivity.DisplayResolveInfo target) {
if (mResolverComparator == null) {
- mResolverComparator = new ResolverComparator(mContext, mTargetIntent, mReferrerPackage);
+ return 0.0f;
}
return mResolverComparator.getScore(target.getResolvedComponentName());
}
mResolverComparator.updateChooserCounts(packageName, userId, action);
}
}
+
+ public void destroy() {
+ if (mResolverComparator != null) {
+ mResolverComparator.destroy();
+ }
+ }
}
<permission android:name="android.permission.BIND_CHOOSER_TARGET_SERVICE"
android:protectionLevel="signature" />
+ <!-- @SystemApi Must be required by services that extend
+ {@link android.service.resolver.ResolverRankerService}, to ensure that only the system can
+ bind to them.
+ <p>Protection level: signature
+ @hide
+ -->
+ <permission android:name="android.permission.BIND_RESOLVER_RANKER_SERVICE"
+ android:protectionLevel="signature" />
+
<!-- Must be required by a {@link
android.service.notification.ConditionProviderService},
to ensure that only the system can bind to it.
android:permission="android.permission.BIND_JOB_SERVICE" >
</service>
+ <service android:name="com.android.internal.app.LRResolverRankerService"
+ android:permission="android.permission.BIND_RESOLVER_RANKER_SERVICE"
+ android:exported="false"
+ android:priority="-1" >
+ <intent-filter>
+ <action android:name="android.service.resolver.ResolverRankerService" />
+ </intent-filter>
+ </service>
</application>
</manifest>
OSDN Git Service
