diff options
Diffstat (limited to 'tests/src')
| -rw-r--r-- | tests/src/com/android/inputmethod/latin/BinaryDictionaryTests.java | 170 | ||||
| -rw-r--r-- | tests/src/com/android/inputmethod/latin/utils/UserHistoryDictIOUtilsTests.java | 244 |
2 files changed, 159 insertions, 255 deletions
diff --git a/tests/src/com/android/inputmethod/latin/BinaryDictionaryTests.java b/tests/src/com/android/inputmethod/latin/BinaryDictionaryTests.java index 5294bb006..bab86e546 100644 --- a/tests/src/com/android/inputmethod/latin/BinaryDictionaryTests.java +++ b/tests/src/com/android/inputmethod/latin/BinaryDictionaryTests.java @@ -878,7 +878,8 @@ public class BinaryDictionaryTests extends AndroidTestCase { private void testGetWordProperties(final int formatVersion) { final long seed = System.currentTimeMillis(); final Random random = new Random(seed); - final int ITERATION_COUNT = 1000; + final int UNIGRAM_COUNT = 1000; + final int BIGRAM_COUNT = 1000; final int codePointSetSize = 20; final int[] codePointSet = CodePointUtils.generateCodePointSet(codePointSetSize, random); @@ -895,7 +896,13 @@ public class BinaryDictionaryTests extends AndroidTestCase { final WordProperty invalidWordProperty = binaryDictionary.getWordProperty("dummyWord"); assertFalse(invalidWordProperty.isValid()); - for (int i = 0; i < ITERATION_COUNT; i++) { + final ArrayList<String> words = new ArrayList<String>(); + final HashMap<String, Integer> wordProbabilities = new HashMap<String, Integer>(); + final HashMap<String, HashSet<String>> bigrams = new HashMap<String, HashSet<String>>(); + final HashMap<Pair<String, String>, Integer> bigramProbabilities = + new HashMap<Pair<String, String>, Integer>(); + + for (int i = 0; i < UNIGRAM_COUNT; i++) { final String word = CodePointUtils.generateWord(random, codePointSet); final int unigramProbability = random.nextInt(0xFF); final boolean isNotAWord = random.nextBoolean(); @@ -904,16 +911,157 @@ public class BinaryDictionaryTests extends AndroidTestCase { binaryDictionary.addUnigramWord(word, unigramProbability, null /* shortcutTarget */, BinaryDictionary.NOT_A_PROBABILITY, isNotAWord, isBlacklisted, BinaryDictionary.NOT_A_VALID_TIMESTAMP); - final WordProperty wordProperty = binaryDictionary.getWordProperty(word); - assertEquals(word, wordProperty.mCodePoints); - assertTrue(wordProperty.isValid()); - assertEquals(isNotAWord, wordProperty.mIsNotAWord); - assertEquals(isBlacklisted, wordProperty.mIsBlacklisted); - assertEquals(false, wordProperty.mHasBigrams); - assertEquals(false, wordProperty.mHasShortcuts); - assertEquals(unigramProbability, wordProperty.mProbabilityInfo.mProbability); - assertTrue(wordProperty.mShortcutTargets.isEmpty()); + if (binaryDictionary.needsToRunGC(false /* mindsBlockByGC */)) { + binaryDictionary.flushWithGC(); + } + words.add(word); + wordProbabilities.put(word, unigramProbability); + final WordProperty unigramProperty = binaryDictionary.getWordProperty(word); + assertEquals(word, unigramProperty.mCodePoints); + assertTrue(unigramProperty.isValid()); + assertEquals(isNotAWord, unigramProperty.mIsNotAWord); + assertEquals(isBlacklisted, unigramProperty.mIsBlacklisted); + assertEquals(false, unigramProperty.mHasBigrams); + assertEquals(false, unigramProperty.mHasShortcuts); + assertEquals(unigramProbability, unigramProperty.mProbabilityInfo.mProbability); + assertTrue(unigramProperty.mShortcutTargets.isEmpty()); + } + + for (int i = 0; i < BIGRAM_COUNT; i++) { + final int word0Index = random.nextInt(wordProbabilities.size()); + final int word1Index = random.nextInt(wordProbabilities.size()); + if (word0Index == word1Index) { + continue; + } + final String word0 = words.get(word0Index); + final String word1 = words.get(word1Index); + final int bigramProbability = random.nextInt(0xF); + binaryDictionary.addBigramWords(word0, word1, bigramProbability, + BinaryDictionary.NOT_A_VALID_TIMESTAMP); + if (binaryDictionary.needsToRunGC(false /* mindsBlockByGC */)) { + binaryDictionary.flushWithGC(); + } + if (!bigrams.containsKey(word0)) { + final HashSet<String> bigramWord1s = new HashSet<String>(); + bigrams.put(word0, bigramWord1s); + } + bigrams.get(word0).add(word1); + bigramProbabilities.put(new Pair<String, String>(word0, word1), bigramProbability); + } + + for (int i = 0; i < words.size(); i++) { + final String word0 = words.get(i); + if (!bigrams.containsKey(word0)) { + continue; + } + final HashSet<String> bigramWord1s = bigrams.get(word0); + final WordProperty unigramProperty = binaryDictionary.getWordProperty(word0); + assertEquals(bigramWord1s.size(), unigramProperty.mBigramTargets.size()); + assertEquals(unigramProperty.mBigramTargets.size(), + unigramProperty.mBigramProbabilityInfo.size()); + for (int j = 0; j < unigramProperty.mBigramTargets.size(); j++) { + final String word1 = unigramProperty.mBigramTargets.get(j).mWord; + assertTrue(bigramWord1s.contains(word1)); + final int probability = unigramProperty.mBigramTargets.get(j).mFrequency; + assertEquals((int)bigramProbabilities.get(new Pair<String, String>(word0, word1)), + probability); + assertEquals(unigramProperty.mBigramProbabilityInfo.get(j).mProbability, + probability); + } + } + } + + public void testIterateAllWords() { + testIterateAllWords(FormatSpec.VERSION4); + } + + private void testIterateAllWords(final int formatVersion) { + final long seed = System.currentTimeMillis(); + final Random random = new Random(seed); + final int UNIGRAM_COUNT = 1000; + final int BIGRAM_COUNT = 1000; + final int codePointSetSize = 20; + final int[] codePointSet = CodePointUtils.generateCodePointSet(codePointSetSize, random); + + File dictFile = null; + try { + dictFile = createEmptyDictionaryAndGetFile("TestBinaryDictionary", formatVersion); + } catch (IOException e) { + fail("IOException while writing an initial dictionary : " + e); + } + final BinaryDictionary binaryDictionary = new BinaryDictionary(dictFile.getAbsolutePath(), + 0 /* offset */, dictFile.length(), true /* useFullEditDistance */, + Locale.getDefault(), TEST_LOCALE, true /* isUpdatable */); + + final WordProperty invalidWordProperty = binaryDictionary.getWordProperty("dummyWord"); + assertFalse(invalidWordProperty.isValid()); + + final ArrayList<String> words = new ArrayList<String>(); + final HashMap<String, Integer> wordProbabilitiesToCheckLater = + new HashMap<String, Integer>(); + final HashMap<String, HashSet<String>> bigrams = new HashMap<String, HashSet<String>>(); + final HashMap<Pair<String, String>, Integer> bigramProbabilitiesToCheckLater = + new HashMap<Pair<String, String>, Integer>(); + + for (int i = 0; i < UNIGRAM_COUNT; i++) { + final String word = CodePointUtils.generateWord(random, codePointSet); + final int unigramProbability = random.nextInt(0xFF); + addUnigramWord(binaryDictionary, word, unigramProbability); + if (binaryDictionary.needsToRunGC(false /* mindsBlockByGC */)) { + binaryDictionary.flushWithGC(); + } + words.add(word); + wordProbabilitiesToCheckLater.put(word, unigramProbability); } + + for (int i = 0; i < BIGRAM_COUNT; i++) { + final int word0Index = random.nextInt(wordProbabilitiesToCheckLater.size()); + final int word1Index = random.nextInt(wordProbabilitiesToCheckLater.size()); + if (word0Index == word1Index) { + continue; + } + final String word0 = words.get(word0Index); + final String word1 = words.get(word1Index); + final int bigramProbability = random.nextInt(0xF); + binaryDictionary.addBigramWords(word0, word1, bigramProbability, + BinaryDictionary.NOT_A_VALID_TIMESTAMP); + if (binaryDictionary.needsToRunGC(false /* mindsBlockByGC */)) { + binaryDictionary.flushWithGC(); + } + if (!bigrams.containsKey(word0)) { + final HashSet<String> bigramWord1s = new HashSet<String>(); + bigrams.put(word0, bigramWord1s); + } + bigrams.get(word0).add(word1); + bigramProbabilitiesToCheckLater.put( + new Pair<String, String>(word0, word1), bigramProbability); + } + + final HashSet<String> wordSet = new HashSet<String>(words); + final HashSet<Pair<String, String>> bigramSet = + new HashSet<Pair<String,String>>(bigramProbabilitiesToCheckLater.keySet()); + int token = 0; + do { + final BinaryDictionary.GetNextWordPropertyResult result = + binaryDictionary.getNextWordProperty(token); + final WordProperty wordProperty = result.mWordProperty; + final String word0 = wordProperty.mCodePoints; + assertEquals((int)wordProbabilitiesToCheckLater.get(word0), + wordProperty.mProbabilityInfo.mProbability); + wordSet.remove(word0); + final HashSet<String> bigramWord1s = bigrams.get(word0); + for (int j = 0; j < wordProperty.mBigramTargets.size(); j++) { + final String word1 = wordProperty.mBigramTargets.get(j).mWord; + assertTrue(bigramWord1s.contains(word1)); + final int probability = wordProperty.mBigramTargets.get(j).mFrequency; + final Pair<String, String> bigram = new Pair<String, String>(word0, word1); + assertEquals((int)bigramProbabilitiesToCheckLater.get(bigram), probability); + bigramSet.remove(bigram); + } + token = result.mNextToken; + } while (token != 0); + assertTrue(wordSet.isEmpty()); + assertTrue(bigramSet.isEmpty()); } public void testAddShortcuts() { diff --git a/tests/src/com/android/inputmethod/latin/utils/UserHistoryDictIOUtilsTests.java b/tests/src/com/android/inputmethod/latin/utils/UserHistoryDictIOUtilsTests.java deleted file mode 100644 index 93731b3cd..000000000 --- a/tests/src/com/android/inputmethod/latin/utils/UserHistoryDictIOUtilsTests.java +++ /dev/null @@ -1,244 +0,0 @@ -/* - * Copyright (C) 2012 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 android.content.Context; -import android.test.AndroidTestCase; -import android.test.suitebuilder.annotation.LargeTest; -import android.util.Log; - -import com.android.inputmethod.latin.makedict.DictDecoder; -import com.android.inputmethod.latin.makedict.DictEncoder; -import com.android.inputmethod.latin.makedict.FormatSpec; -import com.android.inputmethod.latin.makedict.FormatSpec.FileHeader; -import com.android.inputmethod.latin.makedict.FusionDictionary; -import com.android.inputmethod.latin.makedict.FusionDictionary.PtNode; -import com.android.inputmethod.latin.makedict.UnsupportedFormatException; -import com.android.inputmethod.latin.makedict.Ver2DictDecoder; -import com.android.inputmethod.latin.makedict.Ver2DictEncoder; -import com.android.inputmethod.latin.personalization.UserHistoryDictionaryBigramList; -import com.android.inputmethod.latin.utils.UserHistoryDictIOUtils.BigramDictionaryInterface; -import com.android.inputmethod.latin.utils.UserHistoryDictIOUtils.OnAddWordListener; - -import java.io.File; -import java.io.FileNotFoundException; -import java.io.IOException; -import java.util.ArrayList; -import java.util.Collections; -import java.util.HashMap; - -/** - * Unit tests for UserHistoryDictIOUtils - */ -@LargeTest -public class UserHistoryDictIOUtilsTests extends AndroidTestCase - implements BigramDictionaryInterface { - - private static final String TAG = UserHistoryDictIOUtilsTests.class.getSimpleName(); - private static final int UNIGRAM_FREQUENCY = 50; - private static final int BIGRAM_FREQUENCY = 100; - private static final ArrayList<String> NOT_HAVE_BIGRAM = new ArrayList<String>(); - private static final FormatSpec.FormatOptions FORMAT_OPTIONS = new FormatSpec.FormatOptions(2); - private static final String TEST_DICT_FILE_EXTENSION = ".testDict"; - private static final HashMap<String, String> HEADER_OPTIONS = new HashMap<String, String>(); - static { - HEADER_OPTIONS.put(FileHeader.DICTIONARY_LOCALE_KEY, "en_US"); - HEADER_OPTIONS.put(FileHeader.DICTIONARY_ID_KEY, "test"); - HEADER_OPTIONS.put(FileHeader.DICTIONARY_VERSION_KEY, "1000"); - } - - /** - * Return same frequency for all words and bigrams - */ - @Override - public int getFrequency(String word1, String word2) { - if (word1 == null) return UNIGRAM_FREQUENCY; - return BIGRAM_FREQUENCY; - } - - // Utilities for Testing - - private void addWord(final String word, - final HashMap<String, ArrayList<String> > addedWords) { - if (!addedWords.containsKey(word)) { - addedWords.put(word, new ArrayList<String>()); - } - } - - private void addBigram(final String word1, final String word2, - final HashMap<String, ArrayList<String> > addedWords) { - addWord(word1, addedWords); - addWord(word2, addedWords); - addedWords.get(word1).add(word2); - } - - private void addBigramToBigramList(final String word1, final String word2, - final HashMap<String, ArrayList<String> > addedWords, - final UserHistoryDictionaryBigramList bigramList) { - bigramList.addBigram(null, word1); - bigramList.addBigram(word1, word2); - - addBigram(word1, word2, addedWords); - } - - private void checkWordInFusionDict(final FusionDictionary dict, final String word, - final ArrayList<String> expectedBigrams) { - final PtNode ptNode = FusionDictionary.findWordInTree(dict.mRootNodeArray, word); - assertNotNull(ptNode); - assertTrue(ptNode.isTerminal()); - - for (final String bigram : expectedBigrams) { - assertNotNull(ptNode.getBigram(bigram)); - } - } - - private void checkWordsInFusionDict(final FusionDictionary dict, - final HashMap<String, ArrayList<String> > bigrams) { - for (final String word : bigrams.keySet()) { - if (bigrams.containsKey(word)) { - checkWordInFusionDict(dict, word, bigrams.get(word)); - } else { - checkWordInFusionDict(dict, word, NOT_HAVE_BIGRAM); - } - } - } - - private void checkWordInBigramList( - final UserHistoryDictionaryBigramList bigramList, final String word, - final ArrayList<String> expectedBigrams) { - // check unigram - final HashMap<String,Byte> unigramMap = bigramList.getBigrams(null); - assertTrue(unigramMap.containsKey(word)); - - // check bigrams - final ArrayList<String> actualBigrams = new ArrayList<String>( - bigramList.getBigrams(word).keySet()); - - Collections.sort(expectedBigrams); - Collections.sort(actualBigrams); - assertEquals(expectedBigrams, actualBigrams); - } - - private void checkWordsInBigramList(final UserHistoryDictionaryBigramList bigramList, - final HashMap<String, ArrayList<String> > addedWords) { - for (final String word : addedWords.keySet()) { - if (addedWords.containsKey(word)) { - checkWordInBigramList(bigramList, word, addedWords.get(word)); - } else { - checkWordInBigramList(bigramList, word, NOT_HAVE_BIGRAM); - } - } - } - - private void writeDictToFile(final File file, - final UserHistoryDictionaryBigramList bigramList) { - final DictEncoder dictEncoder = new Ver2DictEncoder(file); - UserHistoryDictIOUtils.writeDictionary(dictEncoder, this, bigramList, FORMAT_OPTIONS, - HEADER_OPTIONS); - } - - private void readDictFromFile(final File file, final OnAddWordListener listener) - throws IOException, FileNotFoundException, UnsupportedFormatException { - final DictDecoder dictDecoder = FormatSpec.getDictDecoder(file, DictDecoder.USE_BYTEARRAY); - dictDecoder.openDictBuffer(); - UserHistoryDictIOUtils.readDictionaryBinary(dictDecoder, listener); - } - - public void testGenerateFusionDictionary() { - final UserHistoryDictionaryBigramList originalList = new UserHistoryDictionaryBigramList(); - - final HashMap<String, ArrayList<String> > addedWords = - new HashMap<String, ArrayList<String>>(); - addBigramToBigramList("this", "is", addedWords, originalList); - addBigramToBigramList("this", "was", addedWords, originalList); - addBigramToBigramList("hello", "world", addedWords, originalList); - - final FusionDictionary fusionDict = UserHistoryDictIOUtils.constructFusionDictionary( - this, originalList, HEADER_OPTIONS); - - checkWordsInFusionDict(fusionDict, addedWords); - } - - public void testReadAndWrite() throws IOException, FileNotFoundException, - UnsupportedFormatException { - final Context context = getContext(); - - File file = null; - try { - file = File.createTempFile("testReadAndWrite", TEST_DICT_FILE_EXTENSION, - getContext().getCacheDir()); - } catch (IOException e) { - Log.d(TAG, "IOException while creating a temporary file", e); - } - assertNotNull(file); - - // make original dictionary - final UserHistoryDictionaryBigramList originalList = new UserHistoryDictionaryBigramList(); - final HashMap<String, ArrayList<String>> addedWords = CollectionUtils.newHashMap(); - addBigramToBigramList("this" , "is" , addedWords, originalList); - addBigramToBigramList("this" , "was" , addedWords, originalList); - addBigramToBigramList("is" , "not" , addedWords, originalList); - addBigramToBigramList("hello", "world", addedWords, originalList); - - // write to file - writeDictToFile(file, originalList); - - // make result dict. - final UserHistoryDictionaryBigramList resultList = new UserHistoryDictionaryBigramList(); - final OnAddWordListener listener = new OnAddWordListener() { - @Override - public void setUnigram(final String word, final String shortcutTarget, - final int frequency, final int shortcutFreq) { - Log.d(TAG, "in: setUnigram: " + word + "," + frequency); - resultList.addBigram(null, word, (byte)frequency); - } - @Override - public void setBigram(final String word1, final String word2, final int frequency) { - Log.d(TAG, "in: setBigram: " + word1 + "," + word2 + "," + frequency); - resultList.addBigram(word1, word2, (byte)frequency); - } - }; - - // load from file - readDictFromFile(file, listener); - checkWordsInBigramList(resultList, addedWords); - - // add new bigram - addBigramToBigramList("hello", "java", addedWords, resultList); - - // rewrite - writeDictToFile(file, resultList); - final UserHistoryDictionaryBigramList resultList2 = new UserHistoryDictionaryBigramList(); - final OnAddWordListener listener2 = new OnAddWordListener() { - @Override - public void setUnigram(final String word, final String shortcutTarget, - final int frequency, final int shortcutFreq) { - Log.d(TAG, "in: setUnigram: " + word + "," + frequency); - resultList2.addBigram(null, word, (byte)frequency); - } - @Override - public void setBigram(final String word1, final String word2, final int frequency) { - Log.d(TAG, "in: setBigram: " + word1 + "," + word2 + "," + frequency); - resultList2.addBigram(word1, word2, (byte)frequency); - } - }; - - // load from file - readDictFromFile(file, listener2); - checkWordsInBigramList(resultList2, addedWords); - } -} |