diff options
Diffstat (limited to 'tests/src')
12 files changed, 356 insertions, 110 deletions
diff --git a/tests/src/com/android/inputmethod/latin/BinaryDictionaryDecayingTests.java b/tests/src/com/android/inputmethod/latin/BinaryDictionaryDecayingTests.java index f3bbe4ad4..8d5d6ccec 100644 --- a/tests/src/com/android/inputmethod/latin/BinaryDictionaryDecayingTests.java +++ b/tests/src/com/android/inputmethod/latin/BinaryDictionaryDecayingTests.java @@ -20,7 +20,7 @@ import android.test.AndroidTestCase; import android.test.suitebuilder.annotation.LargeTest; import android.util.Pair; -import com.android.inputmethod.latin.PrevWordsInfo.WordInfo; +import com.android.inputmethod.latin.NgramContext.WordInfo; import com.android.inputmethod.latin.makedict.BinaryDictIOUtils; import com.android.inputmethod.latin.makedict.CodePointUtils; import com.android.inputmethod.latin.makedict.DictDecoder; @@ -78,13 +78,13 @@ public class BinaryDictionaryDecayingTests extends AndroidTestCase { private void addBigramWords(final BinaryDictionary binaryDictionary, final String word0, final String word1, final int probability) { - binaryDictionary.addNgramEntry(new PrevWordsInfo(new WordInfo(word0)), word1, probability, + binaryDictionary.addNgramEntry(new NgramContext(new WordInfo(word0)), word1, probability, mCurrentTime /* timestamp */); } private static boolean isValidBigram(final BinaryDictionary binaryDictionary, final String word0, final String word1) { - return binaryDictionary.isValidNgram(new PrevWordsInfo(new WordInfo(word0)), word1); + return binaryDictionary.isValidNgram(new NgramContext(new WordInfo(word0)), word1); } private void forcePassingShortTime(final BinaryDictionary binaryDictionary) { @@ -661,31 +661,31 @@ public class BinaryDictionaryDecayingTests extends AndroidTestCase { BinaryDictionary.NOT_A_PROBABILITY /* shortcutProbability */, true /* isBeginningOfSentence */, true /* isNotAWord */, false /* isBlacklisted */, mCurrentTime); - final PrevWordsInfo prevWordsInfoStartOfSentence = PrevWordsInfo.BEGINNING_OF_SENTENCE; + final NgramContext beginningOfSentenceContext = NgramContext.BEGINNING_OF_SENTENCE; addUnigramWord(binaryDictionary, "aaa", DUMMY_PROBABILITY); - binaryDictionary.addNgramEntry(prevWordsInfoStartOfSentence, "aaa", DUMMY_PROBABILITY, + binaryDictionary.addNgramEntry(beginningOfSentenceContext, "aaa", DUMMY_PROBABILITY, mCurrentTime); - assertTrue(binaryDictionary.isValidNgram(prevWordsInfoStartOfSentence, "aaa")); - binaryDictionary.addNgramEntry(prevWordsInfoStartOfSentence, "aaa", DUMMY_PROBABILITY, + assertTrue(binaryDictionary.isValidNgram(beginningOfSentenceContext, "aaa")); + binaryDictionary.addNgramEntry(beginningOfSentenceContext, "aaa", DUMMY_PROBABILITY, mCurrentTime); addUnigramWord(binaryDictionary, "bbb", DUMMY_PROBABILITY); - binaryDictionary.addNgramEntry(prevWordsInfoStartOfSentence, "bbb", DUMMY_PROBABILITY, + binaryDictionary.addNgramEntry(beginningOfSentenceContext, "bbb", DUMMY_PROBABILITY, mCurrentTime); - assertTrue(binaryDictionary.isValidNgram(prevWordsInfoStartOfSentence, "aaa")); - assertTrue(binaryDictionary.isValidNgram(prevWordsInfoStartOfSentence, "bbb")); + assertTrue(binaryDictionary.isValidNgram(beginningOfSentenceContext, "aaa")); + assertTrue(binaryDictionary.isValidNgram(beginningOfSentenceContext, "bbb")); forcePassingLongTime(binaryDictionary); - assertFalse(binaryDictionary.isValidNgram(prevWordsInfoStartOfSentence, "aaa")); - assertFalse(binaryDictionary.isValidNgram(prevWordsInfoStartOfSentence, "bbb")); + assertFalse(binaryDictionary.isValidNgram(beginningOfSentenceContext, "aaa")); + assertFalse(binaryDictionary.isValidNgram(beginningOfSentenceContext, "bbb")); addUnigramWord(binaryDictionary, "aaa", DUMMY_PROBABILITY); - binaryDictionary.addNgramEntry(prevWordsInfoStartOfSentence, "aaa", DUMMY_PROBABILITY, + binaryDictionary.addNgramEntry(beginningOfSentenceContext, "aaa", DUMMY_PROBABILITY, mCurrentTime); addUnigramWord(binaryDictionary, "bbb", DUMMY_PROBABILITY); - binaryDictionary.addNgramEntry(prevWordsInfoStartOfSentence, "bbb", DUMMY_PROBABILITY, + binaryDictionary.addNgramEntry(beginningOfSentenceContext, "bbb", DUMMY_PROBABILITY, mCurrentTime); - assertTrue(binaryDictionary.isValidNgram(prevWordsInfoStartOfSentence, "aaa")); - assertTrue(binaryDictionary.isValidNgram(prevWordsInfoStartOfSentence, "bbb")); + assertTrue(binaryDictionary.isValidNgram(beginningOfSentenceContext, "aaa")); + assertTrue(binaryDictionary.isValidNgram(beginningOfSentenceContext, "bbb")); binaryDictionary.close(); dictFile.delete(); } diff --git a/tests/src/com/android/inputmethod/latin/BinaryDictionaryTests.java b/tests/src/com/android/inputmethod/latin/BinaryDictionaryTests.java index 4025744f8..e6f00b668 100644 --- a/tests/src/com/android/inputmethod/latin/BinaryDictionaryTests.java +++ b/tests/src/com/android/inputmethod/latin/BinaryDictionaryTests.java @@ -21,7 +21,7 @@ import android.test.suitebuilder.annotation.LargeTest; import android.text.TextUtils; import android.util.Pair; -import com.android.inputmethod.latin.PrevWordsInfo.WordInfo; +import com.android.inputmethod.latin.NgramContext.WordInfo; import com.android.inputmethod.latin.makedict.CodePointUtils; import com.android.inputmethod.latin.makedict.FormatSpec; import com.android.inputmethod.latin.makedict.WeightedString; @@ -208,45 +208,45 @@ public class BinaryDictionaryTests extends AndroidTestCase { private static void addBigramWords(final BinaryDictionary binaryDictionary, final String word0, final String word1, final int probability) { - binaryDictionary.addNgramEntry(new PrevWordsInfo(new WordInfo(word0)), word1, probability, + binaryDictionary.addNgramEntry(new NgramContext(new WordInfo(word0)), word1, probability, BinaryDictionary.NOT_A_VALID_TIMESTAMP /* timestamp */); } private static void addTrigramEntry(final BinaryDictionary binaryDictionary, final String word0, final String word1, final String word2, final int probability) { - final PrevWordsInfo prevWordsInfo = - new PrevWordsInfo(new WordInfo[] { new WordInfo(word1), new WordInfo(word0) } ); - binaryDictionary.addNgramEntry(prevWordsInfo, word2, probability, + final NgramContext ngramContext = + new NgramContext(new WordInfo[] { new WordInfo(word1), new WordInfo(word0) } ); + binaryDictionary.addNgramEntry(ngramContext, word2, probability, BinaryDictionary.NOT_A_VALID_TIMESTAMP /* timestamp */); } private static boolean isValidBigram(final BinaryDictionary binaryDictionary, final String word0, final String word1) { - return binaryDictionary.isValidNgram(new PrevWordsInfo(new WordInfo(word0)), word1); + return binaryDictionary.isValidNgram(new NgramContext(new WordInfo(word0)), word1); } private static void removeBigramEntry(final BinaryDictionary binaryDictionary, final String word0, final String word1) { - binaryDictionary.removeNgramEntry(new PrevWordsInfo(new WordInfo(word0)), word1); + binaryDictionary.removeNgramEntry(new NgramContext(new WordInfo(word0)), word1); } private static void removeTrigramEntry(final BinaryDictionary binaryDictionary, final String word0, final String word1, final String word2) { - final PrevWordsInfo prevWordsInfo = - new PrevWordsInfo(new WordInfo[] { new WordInfo(word1), new WordInfo(word0) } ); - binaryDictionary.removeNgramEntry(prevWordsInfo, word2); + final NgramContext ngramContext = + new NgramContext(new WordInfo[] { new WordInfo(word1), new WordInfo(word0) } ); + binaryDictionary.removeNgramEntry(ngramContext, word2); } private static int getBigramProbability(final BinaryDictionary binaryDictionary, final String word0, final String word1) { - return binaryDictionary.getNgramProbability(new PrevWordsInfo(new WordInfo(word0)), word1); + return binaryDictionary.getNgramProbability(new NgramContext(new WordInfo(word0)), word1); } private static int getTrigramProbability(final BinaryDictionary binaryDictionary, final String word0, final String word1, final String word2) { - final PrevWordsInfo prevWordsInfo = - new PrevWordsInfo(new WordInfo[] { new WordInfo(word1), new WordInfo(word0) } ); - return binaryDictionary.getNgramProbability(prevWordsInfo, word2); + final NgramContext ngramContext = + new NgramContext(new WordInfo[] { new WordInfo(word1), new WordInfo(word0) } ); + return binaryDictionary.getNgramProbability(ngramContext, word2); } public void testAddUnigramWord() { @@ -1422,7 +1422,7 @@ public class BinaryDictionaryTests extends AndroidTestCase { binaryDictionary.addUnigramEntry("ddd", unigramProbability, null /* shortcutTarget */, Dictionary.NOT_A_PROBABILITY, false /* isBeginningOfSentence */, true /* isNotAWord */, true /* isBlacklisted */, 0 /* timestamp */); - binaryDictionary.addNgramEntry(PrevWordsInfo.BEGINNING_OF_SENTENCE, + binaryDictionary.addNgramEntry(NgramContext.BEGINNING_OF_SENTENCE, "aaa", bigramProbability, 0 /* timestamp */); assertEquals(unigramProbability, binaryDictionary.getFrequency("aaa")); assertEquals(unigramProbability, binaryDictionary.getFrequency("bbb")); @@ -1436,7 +1436,7 @@ public class BinaryDictionaryTests extends AndroidTestCase { if (canCheckBigramProbability(toFormatVersion)) { assertEquals(bigramProbability, getBigramProbability(binaryDictionary, "aaa", "bbb")); assertEquals(bigramProbability, binaryDictionary.getNgramProbability( - PrevWordsInfo.BEGINNING_OF_SENTENCE, "aaa")); + NgramContext.BEGINNING_OF_SENTENCE, "aaa")); } assertTrue(isValidBigram(binaryDictionary, "aaa", "bbb")); WordProperty wordProperty = binaryDictionary.getWordProperty("ccc", @@ -1546,23 +1546,23 @@ public class BinaryDictionaryTests extends AndroidTestCase { 0 /* offset */, dictFile.length(), true /* useFullEditDistance */, Locale.getDefault(), TEST_LOCALE, true /* isUpdatable */); final int dummyProbability = 0; - final PrevWordsInfo prevWordsInfoBeginningOfSentence = PrevWordsInfo.BEGINNING_OF_SENTENCE; + final NgramContext beginningOfSentenceContext = NgramContext.BEGINNING_OF_SENTENCE; final int bigramProbability = 200; addUnigramWord(binaryDictionary, "aaa", dummyProbability); - binaryDictionary.addNgramEntry(prevWordsInfoBeginningOfSentence, "aaa", bigramProbability, + binaryDictionary.addNgramEntry(beginningOfSentenceContext, "aaa", bigramProbability, BinaryDictionary.NOT_A_VALID_TIMESTAMP /* timestamp */); assertEquals(bigramProbability, - binaryDictionary.getNgramProbability(prevWordsInfoBeginningOfSentence, "aaa")); - binaryDictionary.addNgramEntry(prevWordsInfoBeginningOfSentence, "aaa", bigramProbability, + binaryDictionary.getNgramProbability(beginningOfSentenceContext, "aaa")); + binaryDictionary.addNgramEntry(beginningOfSentenceContext, "aaa", bigramProbability, BinaryDictionary.NOT_A_VALID_TIMESTAMP /* timestamp */); addUnigramWord(binaryDictionary, "bbb", dummyProbability); - binaryDictionary.addNgramEntry(prevWordsInfoBeginningOfSentence, "bbb", bigramProbability, + binaryDictionary.addNgramEntry(beginningOfSentenceContext, "bbb", bigramProbability, BinaryDictionary.NOT_A_VALID_TIMESTAMP /* timestamp */); binaryDictionary.flushWithGC(); assertEquals(bigramProbability, - binaryDictionary.getNgramProbability(prevWordsInfoBeginningOfSentence, "aaa")); + binaryDictionary.getNgramProbability(beginningOfSentenceContext, "aaa")); assertEquals(bigramProbability, - binaryDictionary.getNgramProbability(prevWordsInfoBeginningOfSentence, "bbb")); + binaryDictionary.getNgramProbability(beginningOfSentenceContext, "bbb")); } public void testGetMaxFrequencyOfExactMatches() { diff --git a/tests/src/com/android/inputmethod/latin/BlueUnderlineTests.java b/tests/src/com/android/inputmethod/latin/BlueUnderlineTests.java index 6e894decf..30b088137 100644 --- a/tests/src/com/android/inputmethod/latin/BlueUnderlineTests.java +++ b/tests/src/com/android/inputmethod/latin/BlueUnderlineTests.java @@ -63,7 +63,7 @@ public class BlueUnderlineTests extends InputTestsBase { final int typedLength = STRING_TO_TYPE.length(); final int EXPECTED_SUGGESTION_SPAN_START = -1; final int EXPECTED_UNDERLINE_SPAN_START = 0; - final int EXPECTED_UNDERLINE_SPAN_END = 4; + final int EXPECTED_UNDERLINE_SPAN_END = 3; type(STRING_TO_TYPE); sleep(DELAY_TO_WAIT_FOR_UNDERLINE); runMessages(); diff --git a/tests/src/com/android/inputmethod/latin/InputLogicTests.java b/tests/src/com/android/inputmethod/latin/InputLogicTests.java index 59b858dbd..ec249dab3 100644 --- a/tests/src/com/android/inputmethod/latin/InputLogicTests.java +++ b/tests/src/com/android/inputmethod/latin/InputLogicTests.java @@ -159,8 +159,11 @@ public class InputLogicTests extends InputTestsBase { } public void testAutoCorrectWithSpaceThenRevert() { + // Backspacing to cancel the "tgis"->"this" autocorrection should result in + // a "phantom space": if the user presses space immediately after, + // only one space will be inserted in total. final String STRING_TO_TYPE = "tgis "; - final String EXPECTED_RESULT = "tgis "; + final String EXPECTED_RESULT = "tgis"; type(STRING_TO_TYPE); mLatinIME.onUpdateSelection(0, 0, STRING_TO_TYPE.length(), STRING_TO_TYPE.length(), -1, -1); type(Constants.CODE_DELETE); @@ -168,6 +171,24 @@ public class InputLogicTests extends InputTestsBase { mEditText.getText().toString()); } + public void testAutoCorrectWithSpaceThenRevertThenTypeMore() { + final String STRING_TO_TYPE_FIRST = "tgis "; + final String STRING_TO_TYPE_SECOND = "a"; + final String EXPECTED_RESULT = "tgis a"; + type(STRING_TO_TYPE_FIRST); + mLatinIME.onUpdateSelection(0, 0, + STRING_TO_TYPE_FIRST.length(), STRING_TO_TYPE_FIRST.length(), -1, -1); + type(Constants.CODE_DELETE); + + type(STRING_TO_TYPE_SECOND); + mLatinIME.onUpdateSelection(STRING_TO_TYPE_FIRST.length(), STRING_TO_TYPE_FIRST.length(), + STRING_TO_TYPE_FIRST.length() - 1 + STRING_TO_TYPE_SECOND.length(), + STRING_TO_TYPE_FIRST.length() - 1 + STRING_TO_TYPE_SECOND.length(), + -1, -1); + assertEquals("auto-correct with space then revert then type more", EXPECTED_RESULT, + mEditText.getText().toString()); + } + public void testAutoCorrectToSelfDoesNotRevert() { final String STRING_TO_TYPE = "this "; final String EXPECTED_RESULT = "this"; diff --git a/tests/src/com/android/inputmethod/latin/NgramContextTests.java b/tests/src/com/android/inputmethod/latin/NgramContextTests.java new file mode 100644 index 000000000..ecc2c634d --- /dev/null +++ b/tests/src/com/android/inputmethod/latin/NgramContextTests.java @@ -0,0 +1,66 @@ +/* + * Copyright (C) 2014 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; + +import com.android.inputmethod.latin.NgramContext.WordInfo; + +import android.test.AndroidTestCase; +import android.test.suitebuilder.annotation.SmallTest; + +@SmallTest +public class NgramContextTests extends AndroidTestCase { + public void testConstruct() { + assertEquals(new NgramContext(new WordInfo("a")), new NgramContext(new WordInfo("a"))); + assertEquals(new NgramContext(WordInfo.BEGINNING_OF_SENTENCE), + new NgramContext(WordInfo.BEGINNING_OF_SENTENCE)); + assertEquals(new NgramContext(WordInfo.EMPTY_WORD_INFO), + new NgramContext(WordInfo.EMPTY_WORD_INFO)); + assertEquals(new NgramContext(WordInfo.EMPTY_WORD_INFO), + new NgramContext(WordInfo.EMPTY_WORD_INFO)); + } + + public void testIsBeginningOfSentenceContext() { + assertFalse(new NgramContext().isBeginningOfSentenceContext()); + assertTrue(new NgramContext(WordInfo.BEGINNING_OF_SENTENCE) + .isBeginningOfSentenceContext()); + assertTrue(NgramContext.BEGINNING_OF_SENTENCE.isBeginningOfSentenceContext()); + assertFalse(new NgramContext(new WordInfo("a")).isBeginningOfSentenceContext()); + assertFalse(new NgramContext(new WordInfo("")).isBeginningOfSentenceContext()); + assertFalse(new NgramContext(WordInfo.EMPTY_WORD_INFO).isBeginningOfSentenceContext()); + assertTrue(new NgramContext(WordInfo.BEGINNING_OF_SENTENCE, new WordInfo("a")) + .isBeginningOfSentenceContext()); + assertFalse(new NgramContext(new WordInfo("a"), WordInfo.BEGINNING_OF_SENTENCE) + .isBeginningOfSentenceContext()); + assertFalse(new NgramContext(WordInfo.EMPTY_WORD_INFO, WordInfo.BEGINNING_OF_SENTENCE) + .isBeginningOfSentenceContext()); + } + + public void testGetNextNgramContext() { + final NgramContext ngramContext_a = new NgramContext(new WordInfo("a")); + final NgramContext ngramContext_b_a = + ngramContext_a.getNextNgramContext(new WordInfo("b")); + assertEquals("b", ngramContext_b_a.getNthPrevWord(1)); + assertEquals("a", ngramContext_b_a.getNthPrevWord(2)); + final NgramContext ngramContext_bos_b = + ngramContext_b_a.getNextNgramContext(WordInfo.BEGINNING_OF_SENTENCE); + assertTrue(ngramContext_bos_b.isBeginningOfSentenceContext()); + assertEquals("b", ngramContext_bos_b.getNthPrevWord(2)); + final NgramContext ngramContext_c_bos = + ngramContext_b_a.getNextNgramContext(new WordInfo("c")); + assertEquals("c", ngramContext_c_bos.getNthPrevWord(1)); + } +} diff --git a/tests/src/com/android/inputmethod/latin/RichInputConnectionAndTextRangeTests.java b/tests/src/com/android/inputmethod/latin/RichInputConnectionAndTextRangeTests.java index f9d72269e..7a3233625 100644 --- a/tests/src/com/android/inputmethod/latin/RichInputConnectionAndTextRangeTests.java +++ b/tests/src/com/android/inputmethod/latin/RichInputConnectionAndTextRangeTests.java @@ -30,9 +30,8 @@ import android.view.inputmethod.ExtractedTextRequest; import android.view.inputmethod.InputConnection; import android.view.inputmethod.InputConnectionWrapper; -import com.android.inputmethod.latin.PrevWordsInfo.WordInfo; import com.android.inputmethod.latin.settings.SpacingAndPunctuations; -import com.android.inputmethod.latin.utils.PrevWordsInfoUtils; +import com.android.inputmethod.latin.utils.NgramContextUtils; import com.android.inputmethod.latin.utils.RunInLocale; import com.android.inputmethod.latin.utils.ScriptUtils; import com.android.inputmethod.latin.utils.StringUtils; @@ -158,26 +157,25 @@ public class RichInputConnectionAndTextRangeTests extends AndroidTestCase { */ public void testGetPreviousWord() { // If one of the following cases breaks, the bigram suggestions won't work. - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc def", mSpacingAndPunctuations, 2).mPrevWordsInfo[0].mWord, "abc"); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc", mSpacingAndPunctuations, 2), PrevWordsInfo.BEGINNING_OF_SENTENCE); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc. def", mSpacingAndPunctuations, 2), PrevWordsInfo.BEGINNING_OF_SENTENCE); - - assertFalse(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc def", mSpacingAndPunctuations, 2).mPrevWordsInfo[0].mIsBeginningOfSentence); - assertTrue(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc", mSpacingAndPunctuations, 2).mPrevWordsInfo[0].mIsBeginningOfSentence); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc def", mSpacingAndPunctuations, 2).getNthPrevWord(1), "abc"); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc", mSpacingAndPunctuations, 2), NgramContext.BEGINNING_OF_SENTENCE); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc. def", mSpacingAndPunctuations, 2), NgramContext.BEGINNING_OF_SENTENCE); + + assertFalse(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc def", mSpacingAndPunctuations, 2).isBeginningOfSentenceContext()); + assertTrue(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc", mSpacingAndPunctuations, 2).isBeginningOfSentenceContext()); // For n-gram - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc def", mSpacingAndPunctuations, 1).mPrevWordsInfo[0].mWord, "def"); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc def", mSpacingAndPunctuations, 1).mPrevWordsInfo[1].mWord, "abc"); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc def", mSpacingAndPunctuations, 2).mPrevWordsInfo[1], - WordInfo.BEGINNING_OF_SENTENCE); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc def", mSpacingAndPunctuations, 1).getNthPrevWord(1), "def"); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc def", mSpacingAndPunctuations, 1).getNthPrevWord(2), "abc"); + assertTrue(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc def", mSpacingAndPunctuations, 2).isNthPrevWordBeginningOfSontence(2)); // The following tests reflect the current behavior of the function // RichInputConnection#getNthPreviousWord. @@ -186,33 +184,33 @@ public class RichInputConnectionAndTextRangeTests extends AndroidTestCase { // this function if needed - especially since it does not seem very // logical. These tests are just there to catch any unintentional // changes in the behavior of the RichInputConnection#getPreviousWord method. - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc def ", mSpacingAndPunctuations, 2).mPrevWordsInfo[0].mWord, "abc"); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc def.", mSpacingAndPunctuations, 2).mPrevWordsInfo[0].mWord, "abc"); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc def .", mSpacingAndPunctuations, 2).mPrevWordsInfo[0].mWord, "def"); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc ", mSpacingAndPunctuations, 2), PrevWordsInfo.BEGINNING_OF_SENTENCE); - - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc def", mSpacingAndPunctuations, 1).mPrevWordsInfo[0].mWord, "def"); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc def ", mSpacingAndPunctuations, 1).mPrevWordsInfo[0].mWord, "def"); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc 'def", mSpacingAndPunctuations, 1).mPrevWordsInfo[0].mWord, "'def"); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc def.", mSpacingAndPunctuations, 1), PrevWordsInfo.BEGINNING_OF_SENTENCE); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc def .", mSpacingAndPunctuations, 1), PrevWordsInfo.BEGINNING_OF_SENTENCE); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc, def", mSpacingAndPunctuations, 2), PrevWordsInfo.EMPTY_PREV_WORDS_INFO); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc? def", mSpacingAndPunctuations, 2), PrevWordsInfo.EMPTY_PREV_WORDS_INFO); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc! def", mSpacingAndPunctuations, 2), PrevWordsInfo.EMPTY_PREV_WORDS_INFO); - assertEquals(PrevWordsInfoUtils.getPrevWordsInfoFromNthPreviousWord( - "abc 'def", mSpacingAndPunctuations, 2), PrevWordsInfo.EMPTY_PREV_WORDS_INFO); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc def ", mSpacingAndPunctuations, 2).getNthPrevWord(1), "abc"); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc def.", mSpacingAndPunctuations, 2).getNthPrevWord(1), "abc"); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc def .", mSpacingAndPunctuations, 2).getNthPrevWord(1), "def"); + assertTrue(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc ", mSpacingAndPunctuations, 2).isBeginningOfSentenceContext()); + + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc def", mSpacingAndPunctuations, 1).getNthPrevWord(1), "def"); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc def ", mSpacingAndPunctuations, 1).getNthPrevWord(1), "def"); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc 'def", mSpacingAndPunctuations, 1).getNthPrevWord(1), "'def"); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc def.", mSpacingAndPunctuations, 1), NgramContext.BEGINNING_OF_SENTENCE); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc def .", mSpacingAndPunctuations, 1), NgramContext.BEGINNING_OF_SENTENCE); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc, def", mSpacingAndPunctuations, 2), NgramContext.EMPTY_PREV_WORDS_INFO); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc? def", mSpacingAndPunctuations, 2), NgramContext.EMPTY_PREV_WORDS_INFO); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc! def", mSpacingAndPunctuations, 2), NgramContext.EMPTY_PREV_WORDS_INFO); + assertEquals(NgramContextUtils.getNgramContextFromNthPreviousWord( + "abc 'def", mSpacingAndPunctuations, 2), NgramContext.EMPTY_PREV_WORDS_INFO); } public void testGetWordRangeAtCursor() { diff --git a/tests/src/com/android/inputmethod/latin/makedict/BinaryDictEncoderUtils.java b/tests/src/com/android/inputmethod/latin/makedict/BinaryDictEncoderUtils.java index 2b3fd892a..12290e6aa 100644 --- a/tests/src/com/android/inputmethod/latin/makedict/BinaryDictEncoderUtils.java +++ b/tests/src/com/android/inputmethod/latin/makedict/BinaryDictEncoderUtils.java @@ -27,6 +27,8 @@ import java.io.ByteArrayOutputStream; import java.io.IOException; import java.io.OutputStream; import java.util.ArrayList; +import java.util.HashMap; +import java.util.Map.Entry; /** * Encodes binary files for a FusionDictionary. @@ -791,10 +793,12 @@ public class BinaryDictEncoderUtils { * @param destination the stream to write the file header to. * @param dict the dictionary to write. * @param formatOptions file format options. + * @param codePointOccurrenceArray code points ordered by occurrence count. * @return the size of the header. */ /* package */ static int writeDictionaryHeader(final OutputStream destination, - final FusionDictionary dict, final FormatOptions formatOptions) + final FusionDictionary dict, final FormatOptions formatOptions, + final ArrayList<Entry<Integer, Integer>> codePointOccurrenceArray) throws IOException, UnsupportedFormatException { final int version = formatOptions.mVersion; if (version < FormatSpec.MINIMUM_SUPPORTED_VERSION @@ -833,6 +837,9 @@ public class BinaryDictEncoderUtils { CharEncoding.writeString(headerBuffer, key); CharEncoding.writeString(headerBuffer, value); } + + // TODO: Write out the code point table. + final int size = headerBuffer.size(); final byte[] bytes = headerBuffer.toByteArray(); // Write out the header size. @@ -845,4 +852,15 @@ public class BinaryDictEncoderUtils { headerBuffer.close(); return size; } + + static final class CodePointTable { + final HashMap<Integer, Integer> mCodePointToOneByteCodeMap; + final ArrayList<Entry<Integer, Integer>> mCodePointOccurrenceArray; + + CodePointTable(final HashMap<Integer, Integer> codePointToOneByteCodeMap, + final ArrayList<Entry<Integer, Integer>> codePointOccurrenceArray) { + mCodePointToOneByteCodeMap = codePointToOneByteCodeMap; + mCodePointOccurrenceArray = codePointOccurrenceArray; + } + } } diff --git a/tests/src/com/android/inputmethod/latin/makedict/Ver2DictEncoder.java b/tests/src/com/android/inputmethod/latin/makedict/Ver2DictEncoder.java index 0fa75e8ee..c47190190 100644 --- a/tests/src/com/android/inputmethod/latin/makedict/Ver2DictEncoder.java +++ b/tests/src/com/android/inputmethod/latin/makedict/Ver2DictEncoder.java @@ -18,6 +18,7 @@ package com.android.inputmethod.latin.makedict; import com.android.inputmethod.annotations.UsedForTesting; import com.android.inputmethod.latin.makedict.BinaryDictDecoderUtils.CharEncoding; +import com.android.inputmethod.latin.makedict.BinaryDictEncoderUtils.CodePointTable; import com.android.inputmethod.latin.makedict.FormatSpec.FormatOptions; import com.android.inputmethod.latin.makedict.FusionDictionary.PtNode; import com.android.inputmethod.latin.makedict.FusionDictionary.PtNodeArray; @@ -28,7 +29,11 @@ import java.io.FileOutputStream; import java.io.IOException; import java.io.OutputStream; import java.util.ArrayList; +import java.util.Collections; +import java.util.Comparator; +import java.util.HashMap; import java.util.Iterator; +import java.util.Map.Entry; /** * An implementation of DictEncoder for version 2 binary dictionary. @@ -73,6 +78,49 @@ public class Ver2DictEncoder implements DictEncoder { } } + // Package for testing + static CodePointTable makeCodePointTable(final FusionDictionary dict) { + final HashMap<Integer, Integer> codePointOccurrenceCounts = new HashMap<>(); + for (final WordProperty word : dict) { + // Store per code point occurrence + final String wordString = word.mWord; + for (int i = 0; i < wordString.length(); ++i) { + final int codePoint = Character.codePointAt(wordString, i); + if (codePointOccurrenceCounts.containsKey(codePoint)) { + codePointOccurrenceCounts.put(codePoint, + codePointOccurrenceCounts.get(codePoint) + 1); + } else { + codePointOccurrenceCounts.put(codePoint, 1); + } + } + } + final ArrayList<Entry<Integer, Integer>> codePointOccurrenceArray = + new ArrayList<>(codePointOccurrenceCounts.entrySet()); + // Descending order sort by occurrence (value side) + Collections.sort(codePointOccurrenceArray, new Comparator<Entry<Integer, Integer>>() { + @Override + public int compare(final Entry<Integer, Integer> a, final Entry<Integer, Integer> b) { + if (a.getValue() != b.getValue()) { + return b.getValue().compareTo(a.getValue()); + } + return b.getKey().compareTo(a.getKey()); + } + }); + int currentCodePointTableIndex = FormatSpec.MINIMAL_ONE_BYTE_CHARACTER_VALUE; + // Temporary map for writing of nodes + final HashMap<Integer, Integer> codePointToOneByteCodeMap = new HashMap<>(); + for (final Entry<Integer, Integer> entry : codePointOccurrenceArray) { + // Put a relation from the original code point to the one byte code. + codePointToOneByteCodeMap.put(entry.getKey(), currentCodePointTableIndex); + if (FormatSpec.MAXIMAL_ONE_BYTE_CHARACTER_VALUE < ++currentCodePointTableIndex) { + break; + } + } + // codePointToOneByteCodeMap for writing the trie + // codePointOccurrenceArray for writing the header + return new CodePointTable(codePointToOneByteCodeMap, codePointOccurrenceArray); + } + @Override public void writeDictionary(final FusionDictionary dict, final FormatOptions formatOptions) throws IOException, UnsupportedFormatException { @@ -85,7 +133,12 @@ public class Ver2DictEncoder implements DictEncoder { if (mOutStream == null) { openStream(); } - BinaryDictEncoderUtils.writeDictionaryHeader(mOutStream, dict, formatOptions); + + // Make code point conversion table ordered by occurrence of code points + final CodePointTable codePointTable = makeCodePointTable(dict); + + BinaryDictEncoderUtils.writeDictionaryHeader(mOutStream, dict, formatOptions, + codePointTable.mCodePointOccurrenceArray); // Addresses are limited to 3 bytes, but since addresses can be relative to each node // array, the structure itself is not limited to 16MB. However, if it is over 16MB deciding diff --git a/tests/src/com/android/inputmethod/latin/makedict/Ver2DictEncoderTests.java b/tests/src/com/android/inputmethod/latin/makedict/Ver2DictEncoderTests.java new file mode 100644 index 000000000..9104c2fcb --- /dev/null +++ b/tests/src/com/android/inputmethod/latin/makedict/Ver2DictEncoderTests.java @@ -0,0 +1,91 @@ +/* + * Copyright (C) 2014 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.makedict; + +import java.util.ArrayList; +import java.util.Arrays; +import java.util.HashMap; +import java.util.List; +import java.util.Map.Entry; + +import com.android.inputmethod.latin.makedict.BinaryDictEncoderUtils.CodePointTable; +import com.android.inputmethod.latin.makedict.FusionDictionary.PtNodeArray; + +import android.test.AndroidTestCase; +import android.test.suitebuilder.annotation.LargeTest; +import android.util.Log; + +/** + * Unit tests for Ver2DictEncoder + */ +@LargeTest +public class Ver2DictEncoderTests extends AndroidTestCase { + private static final String TAG = Ver2DictEncoderTests.class.getSimpleName(); + private static final int UNIGRAM_FREQ = 10; + + public void testCodePointTable() { + final String[] wordSource = {"words", "used", "for", "testing", "a", "code point", "table"}; + final List<String> words = Arrays.asList(wordSource); + final String correctCodePointTable = "eotdsanirfg bclwup"; + final String correctCodePointOccurrenceArrayString = + "11641114101411531003114211021052972119111711121108110311021991981321"; + final String correctCodePointExpectedMapString = "343332363540383937464549484744414243"; + final String dictName = "codePointTableTest"; + final String dictVersion = Long.toString(System.currentTimeMillis()); + + final FormatSpec.FormatOptions formatOptions = + new FormatSpec.FormatOptions(FormatSpec.VERSION2); + final FusionDictionary sourcedict = new FusionDictionary(new PtNodeArray(), + BinaryDictUtils.makeDictionaryOptions(dictName, dictVersion, formatOptions)); + addUnigrams(sourcedict, words, null /* shortcutMap */); + final CodePointTable codePointTable = Ver2DictEncoder.makeCodePointTable(sourcedict); + + // Check if mCodePointOccurrenceArray is correct + final StringBuilder codePointOccurrenceArrayString = new StringBuilder(); + for (Entry<Integer, Integer> entry : codePointTable.mCodePointOccurrenceArray) { + codePointOccurrenceArrayString.append(entry.getKey()); + codePointOccurrenceArrayString.append(entry.getValue()); + } + assertEquals(correctCodePointOccurrenceArrayString, + codePointOccurrenceArrayString.toString()); + + // Check if mCodePointToOneByteCodeMap is correct + final StringBuilder codePointExpectedMapString = new StringBuilder(); + for (int i = 0; i < correctCodePointTable.length(); ++i) { + codePointExpectedMapString.append(codePointTable.mCodePointToOneByteCodeMap.get( + correctCodePointTable.codePointAt(i))); + } + assertEquals(correctCodePointExpectedMapString, codePointExpectedMapString.toString()); + } + + /** + * Adds unigrams to the dictionary. + */ + private void addUnigrams(final FusionDictionary dict, final List<String> words, + final HashMap<String, List<String>> shortcutMap) { + for (final String word : words) { + final ArrayList<WeightedString> shortcuts = new ArrayList<>(); + if (shortcutMap != null && shortcutMap.containsKey(word)) { + for (final String shortcut : shortcutMap.get(word)) { + shortcuts.add(new WeightedString(shortcut, UNIGRAM_FREQ)); + } + } + dict.add(word, new ProbabilityInfo(UNIGRAM_FREQ), + (shortcutMap == null) ? null : shortcuts, false /* isNotAWord */); + } + } +} diff --git a/tests/src/com/android/inputmethod/latin/makedict/Ver4DictEncoder.java b/tests/src/com/android/inputmethod/latin/makedict/Ver4DictEncoder.java index 2e5435159..74da93766 100644 --- a/tests/src/com/android/inputmethod/latin/makedict/Ver4DictEncoder.java +++ b/tests/src/com/android/inputmethod/latin/makedict/Ver4DictEncoder.java @@ -19,7 +19,7 @@ package com.android.inputmethod.latin.makedict; import com.android.inputmethod.annotations.UsedForTesting; import com.android.inputmethod.latin.BinaryDictionary; import com.android.inputmethod.latin.Dictionary; -import com.android.inputmethod.latin.PrevWordsInfo; +import com.android.inputmethod.latin.NgramContext; import com.android.inputmethod.latin.makedict.FormatSpec.FormatOptions; import com.android.inputmethod.latin.makedict.FusionDictionary.PtNode; import com.android.inputmethod.latin.utils.BinaryDictionaryUtils; @@ -104,12 +104,12 @@ public class Ver4DictEncoder implements DictEncoder { for (final WordProperty word0Property : dict) { if (null == word0Property.mBigrams) continue; for (final WeightedString word1 : word0Property.mBigrams) { - final PrevWordsInfo prevWordsInfo = - new PrevWordsInfo(new PrevWordsInfo.WordInfo(word0Property.mWord)); - if (!binaryDict.addNgramEntry(prevWordsInfo, word1.mWord, + final NgramContext ngramContext = + new NgramContext(new NgramContext.WordInfo(word0Property.mWord)); + if (!binaryDict.addNgramEntry(ngramContext, word1.mWord, word1.getProbability(), 0 /* timestamp */)) { MakedictLog.e("Cannot add n-gram entry for " - + prevWordsInfo + " -> " + word1.mWord); + + ngramContext + " -> " + word1.mWord); return; } if (binaryDict.needsToRunGC(true /* mindsBlockByGC */)) { diff --git a/tests/src/com/android/inputmethod/latin/personalization/UserHistoryDictionaryTests.java b/tests/src/com/android/inputmethod/latin/personalization/UserHistoryDictionaryTests.java index 616209682..766627334 100644 --- a/tests/src/com/android/inputmethod/latin/personalization/UserHistoryDictionaryTests.java +++ b/tests/src/com/android/inputmethod/latin/personalization/UserHistoryDictionaryTests.java @@ -21,8 +21,8 @@ import android.test.suitebuilder.annotation.LargeTest; import android.util.Log; import com.android.inputmethod.latin.ExpandableBinaryDictionary; -import com.android.inputmethod.latin.PrevWordsInfo; -import com.android.inputmethod.latin.PrevWordsInfo.WordInfo; +import com.android.inputmethod.latin.NgramContext; +import com.android.inputmethod.latin.NgramContext.WordInfo; import com.android.inputmethod.latin.utils.BinaryDictionaryUtils; import com.android.inputmethod.latin.utils.DistracterFilter; import com.android.inputmethod.latin.utils.FileUtils; @@ -159,13 +159,13 @@ public class UserHistoryDictionaryTests extends AndroidTestCase { return new ArrayList<>(wordSet); } - private static void addToDict(final UserHistoryDictionary dict, final List<String> words) { - PrevWordsInfo prevWordsInfo = PrevWordsInfo.EMPTY_PREV_WORDS_INFO; + private static void addToDict(final UserHistoryDictionary dict, final List<String> words, + final int timestamp) { + NgramContext ngramContext = NgramContext.EMPTY_PREV_WORDS_INFO; for (String word : words) { - UserHistoryDictionary.addToDictionary(dict, prevWordsInfo, word, true, - (int)TimeUnit.MILLISECONDS.toSeconds(System.currentTimeMillis()), + UserHistoryDictionary.addToDictionary(dict, ngramContext, word, true, timestamp, DistracterFilter.EMPTY_DISTRACTER_FILTER); - prevWordsInfo = prevWordsInfo.getNextPrevWordsInfo(new WordInfo(word)); + ngramContext = ngramContext.getNextNgramContext(new WordInfo(word)); } } @@ -177,7 +177,7 @@ public class UserHistoryDictionaryTests extends AndroidTestCase { final int numberOfWords, final Random random, final boolean checkContents) { final List<String> words = generateWords(numberOfWords, random); // Add random words to the user history dictionary. - addToDict(dict, words); + addToDict(dict, words, mCurrentTime); if (checkContents) { dict.waitAllTasksForTests(); for (int i = 0; i < numberOfWords; ++i) { @@ -288,11 +288,11 @@ public class UserHistoryDictionaryTests extends AndroidTestCase { clearHistory(dict); final List<String> words = generateWords(numberOfWords, random); dict.waitAllTasksForTests(); - PrevWordsInfo prevWordsInfo = PrevWordsInfo.EMPTY_PREV_WORDS_INFO; + NgramContext ngramContext = NgramContext.EMPTY_PREV_WORDS_INFO; for (final String word : words) { - UserHistoryDictionary.addToDictionary(dict, prevWordsInfo, word, true, mCurrentTime, + UserHistoryDictionary.addToDictionary(dict, ngramContext, word, true, mCurrentTime, DistracterFilter.EMPTY_DISTRACTER_FILTER); - prevWordsInfo = prevWordsInfo.getNextPrevWordsInfo(new WordInfo(word)); + ngramContext = ngramContext.getNextNgramContext(new WordInfo(word)); dict.waitAllTasksForTests(); assertTrue(dict.isInDictionary(word)); } @@ -308,6 +308,5 @@ public class UserHistoryDictionaryTests extends AndroidTestCase { for (final String word : words) { assertFalse(dict.isInDictionary(word)); } - stopTestModeInNativeCode(); } } diff --git a/tests/src/com/android/inputmethod/latin/utils/DistracterFilterTest.java b/tests/src/com/android/inputmethod/latin/utils/DistracterFilterTest.java index 6ed912088..8360d53fb 100644 --- a/tests/src/com/android/inputmethod/latin/utils/DistracterFilterTest.java +++ b/tests/src/com/android/inputmethod/latin/utils/DistracterFilterTest.java @@ -24,7 +24,7 @@ import android.test.AndroidTestCase; import android.test.suitebuilder.annotation.LargeTest; import android.view.inputmethod.InputMethodSubtype; -import com.android.inputmethod.latin.PrevWordsInfo; +import com.android.inputmethod.latin.NgramContext; import com.android.inputmethod.latin.RichInputMethodManager; import com.android.inputmethod.latin.utils.DistracterFilter.HandlingType; @@ -58,7 +58,7 @@ public class DistracterFilterTest extends AndroidTestCase { } public void testIsDistracterToWordsInDictionaries() { - final PrevWordsInfo EMPTY_PREV_WORDS_INFO = PrevWordsInfo.EMPTY_PREV_WORDS_INFO; + final NgramContext EMPTY_PREV_WORDS_INFO = NgramContext.EMPTY_PREV_WORDS_INFO; final Locale localeEnUs = new Locale("en", "US"); String typedWord; @@ -204,7 +204,7 @@ public class DistracterFilterTest extends AndroidTestCase { public void testGetWordHandlingType() { final Locale localeEnUs = new Locale("en", "US"); - final PrevWordsInfo EMPTY_PREV_WORDS_INFO = PrevWordsInfo.EMPTY_PREV_WORDS_INFO; + final NgramContext EMPTY_PREV_WORDS_INFO = NgramContext.EMPTY_PREV_WORDS_INFO; int handlingType = 0; handlingType = mDistracterFilter.getWordHandlingType(EMPTY_PREV_WORDS_INFO, |