diff options
17 files changed, 567 insertions, 135 deletions
diff --git a/java/res/values-fr-rCA/config-spacing-and-punctuations.xml b/java/res/values-fr-rCA/config-spacing-and-punctuations.xml index 06254800b..c3a1a0a01 100644 --- a/java/res/values-fr-rCA/config-spacing-and-punctuations.xml +++ b/java/res/values-fr-rCA/config-spacing-and-punctuations.xml @@ -25,7 +25,7 @@ <string name="symbols_followed_by_space" translatable="false">.,;:!?)]}&</string> <!-- Symbols that separate words --> <!-- Don't remove the enclosing double quotes, they protect whitespace (not just U+0020) --> - <string name="symbols_word_separators" translatable="false">"	 \n"()[]{}*&<>+=|.,;:!?/_\"</string> + <string name="symbols_word_separators" translatable="false">"	 
"()[]{}*&<>+=|.,;:!?/_\"</string> <!-- Word connectors --> <string name="symbols_word_connectors" translatable="false">\'-</string> </resources> diff --git a/java/res/values-fr/config-spacing-and-punctuations.xml b/java/res/values-fr/config-spacing-and-punctuations.xml index 33e0236b0..8b53fb3e8 100644 --- a/java/res/values-fr/config-spacing-and-punctuations.xml +++ b/java/res/values-fr/config-spacing-and-punctuations.xml @@ -24,7 +24,7 @@ <string name="symbols_followed_by_space" translatable="false">.,;:!?)]}&</string> <!-- Symbols that separate words --> <!-- Don't remove the enclosing double quotes, they protect whitespace (not just U+0020) --> - <string name="symbols_word_separators" translatable="false">"	 \n"()[]{}*&<>+=|.,;:!?/_\"</string> + <string name="symbols_word_separators" translatable="false">"	 
"()[]{}*&<>+=|.,;:!?/_\"</string> <!-- Word connectors --> <string name="symbols_word_connectors" translatable="false">\'-</string> </resources> diff --git a/java/res/values-hy-rAM/config-spacing-and-punctuations.xml b/java/res/values-hy-rAM/config-spacing-and-punctuations.xml index f26a30def..8bc1b85a1 100644 --- a/java/res/values-hy-rAM/config-spacing-and-punctuations.xml +++ b/java/res/values-hy-rAM/config-spacing-and-punctuations.xml @@ -25,7 +25,7 @@ <string name="symbols_followed_by_space" translatable="false">.,;:!?)]}&։՝</string> <!-- Symbols that separate words. Adding armenian period and comma. --> <!-- Don't remove the enclosing double quotes, they protect whitespace (not just U+0020) --> - <string name="symbols_word_separators" translatable="false">"	 \n"()[]{}*&<>+=|.,;:!?/_\"։՝</string> + <string name="symbols_word_separators" translatable="false">"	 
"()[]{}*&<>+=|.,;:!?/_\"։՝</string> <!-- The sentence separator code point, for capitalization --> <!-- U+0589: "։" ARMENIAN FULL STOP ; 589h = 1417d --> <integer name="sentence_separator" translatable="false">1417</integer> diff --git a/java/res/values/config-spacing-and-punctuations.xml b/java/res/values/config-spacing-and-punctuations.xml index f10f810dc..2f52edd4a 100644 --- a/java/res/values/config-spacing-and-punctuations.xml +++ b/java/res/values/config-spacing-and-punctuations.xml @@ -28,7 +28,7 @@ <string name="symbols_followed_by_space" translatable="false">.,;:!?)]}&</string> <!-- Symbols that separate words --> <!-- Don't remove the enclosing double quotes, they protect whitespace (not just U+0020) --> - <string name="symbols_word_separators" translatable="false">"	 \n"()[]{}*&<>+=|.,;:!?/_\"</string> + <string name="symbols_word_separators" translatable="false">"	 
"()[]{}*&<>+=|.,;:!?/_\"</string> <!-- Word connectors --> <string name="symbols_word_connectors" translatable="false">\'-</string> <!-- The sentence separator code point, for capitalization --> diff --git a/java/src/com/android/inputmethod/latin/LatinIME.java b/java/src/com/android/inputmethod/latin/LatinIME.java index 3fca4fd19..d3e6a1bc2 100644 --- a/java/src/com/android/inputmethod/latin/LatinIME.java +++ b/java/src/com/android/inputmethod/latin/LatinIME.java @@ -1407,7 +1407,7 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen // TODO[IL]: Move this to InputLogic public SuggestedWords maybeRetrieveOlderSuggestions(final String typedWord, - final SuggestedWords suggestedWords) { + final SuggestedWords suggestedWords, final SuggestedWords previousSuggestedWords) { // TODO: consolidate this into getSuggestedWords // We update the suggestion strip only when we have some suggestions to show, i.e. when // the suggestion count is > 1; else, we leave the old suggestions, with the typed word @@ -1420,28 +1420,22 @@ public class LatinIME extends InputMethodService implements KeyboardActionListen || mSuggestionStripView.isShowingAddToDictionaryHint()) { return suggestedWords; } else { - return getOlderSuggestions(typedWord); - } - } - - private SuggestedWords getOlderSuggestions(final String typedWord) { - SuggestedWords previousSuggestedWords = mInputLogic.mSuggestedWords; - if (previousSuggestedWords - == mSettings.getCurrent().mSpacingAndPunctuations.mSuggestPuncList) { - previousSuggestedWords = SuggestedWords.EMPTY; - } - if (typedWord == null) { - return previousSuggestedWords; + final SuggestedWords punctuationList = + mSettings.getCurrent().mSpacingAndPunctuations.mSuggestPuncList; + final SuggestedWords oldSuggestedWords = previousSuggestedWords == punctuationList + ? SuggestedWords.EMPTY : previousSuggestedWords; + if (TextUtils.isEmpty(typedWord)) { + return oldSuggestedWords; + } + final ArrayList<SuggestedWords.SuggestedWordInfo> typedWordAndPreviousSuggestions = + SuggestedWords.getTypedWordAndPreviousSuggestions(typedWord, oldSuggestedWords); + return new SuggestedWords(typedWordAndPreviousSuggestions, + false /* typedWordValid */, + false /* hasAutoCorrectionCandidate */, + false /* isPunctuationSuggestions */, + true /* isObsoleteSuggestions */, + false /* isPrediction */); } - final ArrayList<SuggestedWords.SuggestedWordInfo> typedWordAndPreviousSuggestions = - SuggestedWords.getTypedWordAndPreviousSuggestions(typedWord, - previousSuggestedWords); - return new SuggestedWords(typedWordAndPreviousSuggestions, - false /* typedWordValid */, - false /* hasAutoCorrectionCandidate */, - false /* isPunctuationSuggestions */, - true /* isObsoleteSuggestions */, - false /* isPrediction */); } private void showSuggestionStripWithTypedWord(final SuggestedWords suggestedWords, diff --git a/java/src/com/android/inputmethod/latin/RichInputConnection.java b/java/src/com/android/inputmethod/latin/RichInputConnection.java index a5b147aae..325a0d981 100644 --- a/java/src/com/android/inputmethod/latin/RichInputConnection.java +++ b/java/src/com/android/inputmethod/latin/RichInputConnection.java @@ -35,6 +35,7 @@ import com.android.inputmethod.latin.utils.StringUtils; import com.android.inputmethod.latin.utils.TextRange; import com.android.inputmethod.research.ResearchLogger; +import java.util.Arrays; import java.util.regex.Pattern; /** @@ -557,8 +558,8 @@ public final class RichInputConnection { return getNthPreviousWord(prev, spacingAndPunctuations, n); } - private static boolean isSeparator(int code, String sep) { - return sep.indexOf(code) != -1; + private static boolean isSeparator(final int code, final int[] sortedSeparators) { + return Arrays.binarySearch(sortedSeparators, code) >= 0; } // Get the nth word before cursor. n = 1 retrieves the word immediately before the cursor, @@ -597,29 +598,29 @@ public final class RichInputConnection { } /** - * @param separators characters which may separate words + * @param sortedSeparators a sorted array of code points which may separate words * @return the word that surrounds the cursor, including up to one trailing * separator. For example, if the field contains "he|llo world", where | * represents the cursor, then "hello " will be returned. */ - public CharSequence getWordAtCursor(String separators) { + public CharSequence getWordAtCursor(final int[] sortedSeparators) { // getWordRangeAtCursor returns null if the connection is null - TextRange r = getWordRangeAtCursor(separators, 0); + final TextRange r = getWordRangeAtCursor(sortedSeparators, 0); return (r == null) ? null : r.mWord; } /** * Returns the text surrounding the cursor. * - * @param sep a string of characters that split words. + * @param sortedSeparators a sorted array of code points that split words. * @param additionalPrecedingWordsCount the number of words before the current word that should * be included in the returned range * @return a range containing the text surrounding the cursor */ - public TextRange getWordRangeAtCursor(final String sep, + public TextRange getWordRangeAtCursor(final int[] sortedSeparators, final int additionalPrecedingWordsCount) { mIC = mParent.getCurrentInputConnection(); - if (mIC == null || sep == null) { + if (mIC == null) { return null; } final CharSequence before = mIC.getTextBeforeCursor(Constants.EDITOR_CONTENTS_CACHE_SIZE, @@ -638,7 +639,7 @@ public final class RichInputConnection { while (true) { // see comments below for why this is guaranteed to halt while (startIndexInBefore > 0) { final int codePoint = Character.codePointBefore(before, startIndexInBefore); - if (isStoppingAtWhitespace == isSeparator(codePoint, sep)) { + if (isStoppingAtWhitespace == isSeparator(codePoint, sortedSeparators)) { break; // inner loop } --startIndexInBefore; @@ -659,7 +660,7 @@ public final class RichInputConnection { int endIndexInAfter = -1; while (++endIndexInAfter < after.length()) { final int codePoint = Character.codePointAt(after, endIndexInAfter); - if (isSeparator(codePoint, sep)) { + if (isSeparator(codePoint, sortedSeparators)) { break; } if (Character.isSupplementaryCodePoint(codePoint)) { diff --git a/java/src/com/android/inputmethod/latin/inputlogic/InputLogic.java b/java/src/com/android/inputmethod/latin/inputlogic/InputLogic.java index 3ff20791f..375a42e3b 100644 --- a/java/src/com/android/inputmethod/latin/inputlogic/InputLogic.java +++ b/java/src/com/android/inputmethod/latin/inputlogic/InputLogic.java @@ -970,7 +970,8 @@ public final class InputLogic { if (TextUtils.isEmpty(selectedText)) return; // Race condition with the input connection mRecapitalizeStatus.initialize(mConnection.getExpectedSelectionStart(), mConnection.getExpectedSelectionEnd(), selectedText.toString(), - settingsValues.mLocale, settingsValues.mSpacingAndPunctuations.mWordSeparators); + settingsValues.mLocale, + settingsValues.mSpacingAndPunctuations.mSortedWordSeparators); // We trim leading and trailing whitespace. mRecapitalizeStatus.trim(); } @@ -1031,7 +1032,8 @@ public final class InputLogic { public void onGetSuggestedWords(final SuggestedWords suggestedWords) { final SuggestedWords suggestedWordsWithMaybeOlderSuggestions = mLatinIME.maybeRetrieveOlderSuggestions( - mWordComposer.getTypedWord(), suggestedWords); + mWordComposer.getTypedWord(), suggestedWords, + mSuggestedWords); holder.set(suggestedWordsWithMaybeOlderSuggestions); } } @@ -1072,7 +1074,7 @@ public final class InputLogic { final int expectedCursorPosition = mConnection.getExpectedSelectionStart(); if (!mConnection.isCursorTouchingWord(settingsValues.mSpacingAndPunctuations)) return; final TextRange range = mConnection.getWordRangeAtCursor( - settingsValues.mSpacingAndPunctuations.mWordSeparators, + settingsValues.mSpacingAndPunctuations.mSortedWordSeparators, 0 /* additionalPrecedingWordsCount */); if (null == range) return; // Happens if we don't have an input connection at all if (range.length() <= 0) return; // Race condition. No text to resume on, so bail out. diff --git a/java/src/com/android/inputmethod/latin/makedict/Ver2DictEncoder.java b/java/src/com/android/inputmethod/latin/makedict/Ver2DictEncoder.java index e5430423d..a3a6c2c34 100644 --- a/java/src/com/android/inputmethod/latin/makedict/Ver2DictEncoder.java +++ b/java/src/com/android/inputmethod/latin/makedict/Ver2DictEncoder.java @@ -16,6 +16,7 @@ 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.FormatSpec.FormatOptions; import com.android.inputmethod.latin.makedict.FusionDictionary.PtNode; @@ -33,6 +34,7 @@ import java.util.Iterator; /** * An implementation of DictEncoder for version 2 binary dictionary. */ +@UsedForTesting public class Ver2DictEncoder implements DictEncoder { private final File mDictFile; @@ -40,6 +42,7 @@ public class Ver2DictEncoder implements DictEncoder { private byte[] mBuffer; private int mPosition; + @UsedForTesting public Ver2DictEncoder(final File dictFile) { mDictFile = dictFile; mOutStream = null; @@ -49,6 +52,7 @@ public class Ver2DictEncoder implements DictEncoder { // This constructor is used only by BinaryDictOffdeviceUtilsTests. // If you want to use this in the production code, you should consider keeping consistency of // the interface of Ver3DictDecoder by using factory. + @UsedForTesting public Ver2DictEncoder(final OutputStream outStream) { mDictFile = null; mOutStream = outStream; diff --git a/java/src/com/android/inputmethod/latin/settings/Settings.java b/java/src/com/android/inputmethod/latin/settings/Settings.java index 7db10714a..9bf269b6e 100644 --- a/java/src/com/android/inputmethod/latin/settings/Settings.java +++ b/java/src/com/android/inputmethod/latin/settings/Settings.java @@ -181,10 +181,6 @@ public final class Settings implements SharedPreferences.OnSharedPreferenceChang return mSettingsValues.mIsInternal; } - public String getWordSeparators() { - return mSettingsValues.mSpacingAndPunctuations.mWordSeparators; - } - public boolean isWordSeparator(final int code) { return mSettingsValues.isWordSeparator(code); } diff --git a/java/src/com/android/inputmethod/latin/settings/SpacingAndPunctuations.java b/java/src/com/android/inputmethod/latin/settings/SpacingAndPunctuations.java index 29bd3e7b3..8ba32ff76 100644 --- a/java/src/com/android/inputmethod/latin/settings/SpacingAndPunctuations.java +++ b/java/src/com/android/inputmethod/latin/settings/SpacingAndPunctuations.java @@ -32,11 +32,11 @@ import java.util.Arrays; import java.util.Locale; public final class SpacingAndPunctuations { - private final int[] mSymbolsPrecededBySpace; - private final int[] mSymbolsFollowedBySpace; - private final int[] mWordConnectors; + private final int[] mSortedSymbolsPrecededBySpace; + private final int[] mSortedSymbolsFollowedBySpace; + private final int[] mSortedWordConnectors; + public final int[] mSortedWordSeparators; public final SuggestedWords mSuggestPuncList; - public final String mWordSeparators; private final int mSentenceSeparator; public final String mSentenceSeparatorAndSpace; public final boolean mCurrentLanguageHasSpaces; @@ -44,19 +44,20 @@ public final class SpacingAndPunctuations { public final boolean mUsesGermanRules; public SpacingAndPunctuations(final Resources res) { - mSymbolsPrecededBySpace = - StringUtils.toCodePointArray(res.getString(R.string.symbols_preceded_by_space)); - Arrays.sort(mSymbolsPrecededBySpace); - mSymbolsFollowedBySpace = - StringUtils.toCodePointArray(res.getString(R.string.symbols_followed_by_space)); - Arrays.sort(mSymbolsFollowedBySpace); - mWordConnectors = - StringUtils.toCodePointArray(res.getString(R.string.symbols_word_connectors)); - Arrays.sort(mWordConnectors); + // To be able to binary search the code point. See {@link #isUsuallyPrecededBySpace(int)}. + mSortedSymbolsPrecededBySpace = StringUtils.toSortedCodePointArray( + res.getString(R.string.symbols_preceded_by_space)); + // To be able to binary search the code point. See {@link #isUsuallyFollowedBySpace(int)}. + mSortedSymbolsFollowedBySpace = StringUtils.toSortedCodePointArray( + res.getString(R.string.symbols_followed_by_space)); + // To be able to binary search the code point. See {@link #isWordConnector(int)}. + mSortedWordConnectors = StringUtils.toSortedCodePointArray( + res.getString(R.string.symbols_word_connectors)); + mSortedWordSeparators = StringUtils.toSortedCodePointArray( + res.getString(R.string.symbols_word_separators)); final String[] suggestPuncsSpec = KeySpecParser.splitKeySpecs(res.getString( R.string.suggested_punctuations)); mSuggestPuncList = createSuggestPuncList(suggestPuncsSpec); - mWordSeparators = res.getString(R.string.symbols_word_separators); mSentenceSeparator = res.getInteger(R.integer.sentence_separator); mSentenceSeparatorAndSpace = new String(new int[] { mSentenceSeparator, Constants.CODE_SPACE }, 0, 2); @@ -74,6 +75,7 @@ public final class SpacingAndPunctuations { if (puncs != null) { for (final String puncSpec : puncs) { // TODO: Stop using KeySpceParser.getLabel(). + // TODO: Punctuation suggestions should honor RTL languages. puncList.add(new SuggestedWordInfo(KeySpecParser.getLabel(puncSpec), SuggestedWordInfo.MAX_SCORE, SuggestedWordInfo.KIND_HARDCODED, Dictionary.DICTIONARY_HARDCODED, @@ -90,11 +92,11 @@ public final class SpacingAndPunctuations { } public boolean isWordSeparator(final int code) { - return mWordSeparators.contains(String.valueOf((char)code)); + return Arrays.binarySearch(mSortedWordSeparators, code) >= 0; } public boolean isWordConnector(final int code) { - return Arrays.binarySearch(mWordConnectors, code) >= 0; + return Arrays.binarySearch(mSortedWordConnectors, code) >= 0; } public boolean isWordCodePoint(final int code) { @@ -102,11 +104,11 @@ public final class SpacingAndPunctuations { } public boolean isUsuallyPrecededBySpace(final int code) { - return Arrays.binarySearch(mSymbolsPrecededBySpace, code) >= 0; + return Arrays.binarySearch(mSortedSymbolsPrecededBySpace, code) >= 0; } public boolean isUsuallyFollowedBySpace(final int code) { - return Arrays.binarySearch(mSymbolsFollowedBySpace, code) >= 0; + return Arrays.binarySearch(mSortedSymbolsFollowedBySpace, code) >= 0; } public boolean isSentenceSeparator(final int code) { diff --git a/java/src/com/android/inputmethod/latin/utils/RecapitalizeStatus.java b/java/src/com/android/inputmethod/latin/utils/RecapitalizeStatus.java index 0f5cd80db..4521ec531 100644 --- a/java/src/com/android/inputmethod/latin/utils/RecapitalizeStatus.java +++ b/java/src/com/android/inputmethod/latin/utils/RecapitalizeStatus.java @@ -37,12 +37,12 @@ public class RecapitalizeStatus { CAPS_MODE_ALL_UPPER }; - private static final int getStringMode(final String string, final String separators) { + private static final int getStringMode(final String string, final int[] sortedSeparators) { if (StringUtils.isIdenticalAfterUpcase(string)) { return CAPS_MODE_ALL_UPPER; } else if (StringUtils.isIdenticalAfterDowncase(string)) { return CAPS_MODE_ALL_LOWER; - } else if (StringUtils.isIdenticalAfterCapitalizeEachWord(string, separators)) { + } else if (StringUtils.isIdenticalAfterCapitalizeEachWord(string, sortedSeparators)) { return CAPS_MODE_FIRST_WORD_UPPER; } else { return CAPS_MODE_ORIGINAL_MIXED_CASE; @@ -60,26 +60,28 @@ public class RecapitalizeStatus { private int mRotationStyleCurrentIndex; private boolean mSkipOriginalMixedCaseMode; private Locale mLocale; - private String mSeparators; + private int[] mSortedSeparators; private String mStringAfter; private boolean mIsActive; + private static final int[] EMPTY_STORTED_SEPARATORS = {}; + public RecapitalizeStatus() { // By default, initialize with dummy values that won't match any real recapitalize. - initialize(-1, -1, "", Locale.getDefault(), ""); + initialize(-1, -1, "", Locale.getDefault(), EMPTY_STORTED_SEPARATORS); deactivate(); } public void initialize(final int cursorStart, final int cursorEnd, final String string, - final Locale locale, final String separators) { + final Locale locale, final int[] sortedSeparators) { mCursorStartBefore = cursorStart; mStringBefore = string; mCursorStartAfter = cursorStart; mCursorEndAfter = cursorEnd; mStringAfter = string; - final int initialMode = getStringMode(mStringBefore, separators); + final int initialMode = getStringMode(mStringBefore, sortedSeparators); mLocale = locale; - mSeparators = separators; + mSortedSeparators = sortedSeparators; if (CAPS_MODE_ORIGINAL_MIXED_CASE == initialMode) { mRotationStyleCurrentIndex = 0; mSkipOriginalMixedCaseMode = false; @@ -131,7 +133,7 @@ public class RecapitalizeStatus { mStringAfter = mStringBefore.toLowerCase(mLocale); break; case CAPS_MODE_FIRST_WORD_UPPER: - mStringAfter = StringUtils.capitalizeEachWord(mStringBefore, mSeparators, + mStringAfter = StringUtils.capitalizeEachWord(mStringBefore, mSortedSeparators, mLocale); break; case CAPS_MODE_ALL_UPPER: diff --git a/java/src/com/android/inputmethod/latin/utils/StringUtils.java b/java/src/com/android/inputmethod/latin/utils/StringUtils.java index c5ed39310..5920c68f1 100644 --- a/java/src/com/android/inputmethod/latin/utils/StringUtils.java +++ b/java/src/com/android/inputmethod/latin/utils/StringUtils.java @@ -22,6 +22,7 @@ import com.android.inputmethod.annotations.UsedForTesting; import com.android.inputmethod.latin.Constants; import java.util.ArrayList; +import java.util.Arrays; import java.util.Locale; public final class StringUtils { @@ -183,6 +184,12 @@ public final class StringUtils { return codePoints; } + public static int[] toSortedCodePointArray(final String string) { + final int[] codePoints = toCodePointArray(string); + Arrays.sort(codePoints); + return codePoints; + } + // This method assumes the text is not null. For the empty string, it returns CAPITALIZE_NONE. public static int getCapitalizationType(final String text) { // If the first char is not uppercase, then the word is either all lower case or @@ -265,39 +272,39 @@ public final class StringUtils { } public static boolean isIdenticalAfterCapitalizeEachWord(final String text, - final String separators) { - boolean needCapsNext = true; + final int[] sortedSeparators) { + boolean needsCapsNext = true; final int len = text.length(); for (int i = 0; i < len; i = text.offsetByCodePoints(i, 1)) { final int codePoint = text.codePointAt(i); if (Character.isLetter(codePoint)) { - if ((needCapsNext && !Character.isUpperCase(codePoint)) - || (!needCapsNext && !Character.isLowerCase(codePoint))) { + if ((needsCapsNext && !Character.isUpperCase(codePoint)) + || (!needsCapsNext && !Character.isLowerCase(codePoint))) { return false; } } // We need a capital letter next if this is a separator. - needCapsNext = (-1 != separators.indexOf(codePoint)); + needsCapsNext = (Arrays.binarySearch(sortedSeparators, codePoint) >= 0); } return true; } // TODO: like capitalizeFirst*, this does not work perfectly for Dutch because of the IJ digraph // which should be capitalized together in *some* cases. - public static String capitalizeEachWord(final String text, final String separators, + public static String capitalizeEachWord(final String text, final int[] sortedSeparators, final Locale locale) { final StringBuilder builder = new StringBuilder(); - boolean needCapsNext = true; + boolean needsCapsNext = true; final int len = text.length(); for (int i = 0; i < len; i = text.offsetByCodePoints(i, 1)) { final String nextChar = text.substring(i, text.offsetByCodePoints(i, 1)); - if (needCapsNext) { + if (needsCapsNext) { builder.append(nextChar.toUpperCase(locale)); } else { builder.append(nextChar.toLowerCase(locale)); } // We need a capital letter next if this is a separator. - needCapsNext = (-1 != separators.indexOf(nextChar.codePointAt(0))); + needsCapsNext = (Arrays.binarySearch(sortedSeparators, nextChar.codePointAt(0)) >= 0); } return builder.toString(); } diff --git a/java/src/com/android/inputmethod/research/ResearchLogger.java b/java/src/com/android/inputmethod/research/ResearchLogger.java index e7f49a605..11fb3a156 100644 --- a/java/src/com/android/inputmethod/research/ResearchLogger.java +++ b/java/src/com/android/inputmethod/research/ResearchLogger.java @@ -59,6 +59,7 @@ import com.android.inputmethod.latin.RichInputConnection; import com.android.inputmethod.latin.SuggestedWords; import com.android.inputmethod.latin.define.ProductionFlag; import com.android.inputmethod.latin.utils.InputTypeUtils; +import com.android.inputmethod.latin.utils.StringUtils; import com.android.inputmethod.latin.utils.TextRange; import com.android.inputmethod.research.MotionEventReader.ReplayData; import com.android.inputmethod.research.ui.SplashScreen; @@ -131,7 +132,8 @@ public class ResearchLogger implements SharedPreferences.OnSharedPreferenceChang public static final String RESEARCH_KEY_OUTPUT_TEXT = ".research."; // constants related to specific log points - private static final String WHITESPACE_SEPARATORS = " \t\n\r"; + private static final int[] WHITESPACE_SEPARATORS = + StringUtils.toSortedCodePointArray(" \t\n\r"); private static final int MAX_INPUTVIEW_LENGTH_TO_CAPTURE = 8192; // must be >=1 private static final String PREF_RESEARCH_SAVED_CHANNEL = "pref_research_saved_channel"; diff --git a/tests/src/com/android/inputmethod/latin/RichInputConnectionAndTextRangeTests.java b/tests/src/com/android/inputmethod/latin/RichInputConnectionAndTextRangeTests.java index f19d1853e..7f0743543 100644 --- a/tests/src/com/android/inputmethod/latin/RichInputConnectionAndTextRangeTests.java +++ b/tests/src/com/android/inputmethod/latin/RichInputConnectionAndTextRangeTests.java @@ -32,6 +32,7 @@ import android.view.inputmethod.InputConnectionWrapper; import com.android.inputmethod.latin.settings.SpacingAndPunctuations; import com.android.inputmethod.latin.utils.RunInLocale; +import com.android.inputmethod.latin.utils.StringUtils; import com.android.inputmethod.latin.utils.TextRange; import java.util.Locale; @@ -183,6 +184,12 @@ public class RichInputConnectionAndTextRangeTests extends AndroidTestCase { /** * Test logic in getting the word range at the cursor. */ + private static final int[] SPACE = { Constants.CODE_SPACE }; + static final int[] TAB = { Constants.CODE_TAB }; + private static final int[] SPACE_TAB = StringUtils.toSortedCodePointArray(" \t"); + // A character that needs surrogate pair to represent its code point (U+2008A). + private static final String SUPPLEMENTARY_CHAR = "\uD840\uDC8A"; + public void testGetWordRangeAtCursor() { ExtractedText et = new ExtractedText(); final MockInputMethodService mockInputMethodService = new MockInputMethodService(); @@ -194,48 +201,47 @@ public class RichInputConnectionAndTextRangeTests extends AndroidTestCase { ic.beginBatchEdit(); // basic case - r = ic.getWordRangeAtCursor(" ", 0); + r = ic.getWordRangeAtCursor(SPACE, 0); assertTrue(TextUtils.equals("word", r.mWord)); // more than one word - r = ic.getWordRangeAtCursor(" ", 1); + r = ic.getWordRangeAtCursor(SPACE, 1); assertTrue(TextUtils.equals("word word", r.mWord)); ic.endBatchEdit(); // tab character instead of space mockInputMethodService.setInputConnection(new MockConnection("one\tword\two", "rd", et)); ic.beginBatchEdit(); - r = ic.getWordRangeAtCursor("\t", 1); + r = ic.getWordRangeAtCursor(TAB, 1); ic.endBatchEdit(); assertTrue(TextUtils.equals("word\tword", r.mWord)); // only one word doesn't go too far mockInputMethodService.setInputConnection(new MockConnection("one\tword\two", "rd", et)); ic.beginBatchEdit(); - r = ic.getWordRangeAtCursor("\t", 1); + r = ic.getWordRangeAtCursor(TAB, 1); ic.endBatchEdit(); assertTrue(TextUtils.equals("word\tword", r.mWord)); // tab or space mockInputMethodService.setInputConnection(new MockConnection("one word\two", "rd", et)); ic.beginBatchEdit(); - r = ic.getWordRangeAtCursor(" \t", 1); + r = ic.getWordRangeAtCursor(SPACE_TAB, 1); ic.endBatchEdit(); assertTrue(TextUtils.equals("word\tword", r.mWord)); // tab or space multiword mockInputMethodService.setInputConnection(new MockConnection("one word\two", "rd", et)); ic.beginBatchEdit(); - r = ic.getWordRangeAtCursor(" \t", 2); + r = ic.getWordRangeAtCursor(SPACE_TAB, 2); ic.endBatchEdit(); assertTrue(TextUtils.equals("one word\tword", r.mWord)); // splitting on supplementary character - final String supplementaryChar = "\uD840\uDC8A"; mockInputMethodService.setInputConnection( - new MockConnection("one word" + supplementaryChar + "wo", "rd", et)); + new MockConnection("one word" + SUPPLEMENTARY_CHAR + "wo", "rd", et)); ic.beginBatchEdit(); - r = ic.getWordRangeAtCursor(supplementaryChar, 0); + r = ic.getWordRangeAtCursor(StringUtils.toSortedCodePointArray(SUPPLEMENTARY_CHAR), 0); ic.endBatchEdit(); assertTrue(TextUtils.equals("word", r.mWord)); } @@ -265,7 +271,7 @@ public class RichInputConnectionAndTextRangeTests extends AndroidTestCase { TextRange r; SuggestionSpan[] suggestions; - r = ic.getWordRangeAtCursor(" ", 0); + r = ic.getWordRangeAtCursor(SPACE, 0); suggestions = r.getSuggestionSpansAtWord(); assertEquals(suggestions.length, 1); MoreAsserts.assertEquals(suggestions[0].getSuggestions(), SUGGESTIONS1); @@ -277,7 +283,7 @@ public class RichInputConnectionAndTextRangeTests extends AndroidTestCase { text.setSpan(new SuggestionSpan(Locale.ENGLISH, SUGGESTIONS2, 0 /* flags */), 10 /* start */, 16 /* end */, 0 /* flags */); mockInputMethodService.setInputConnection(new MockConnection(text, cursorPos)); - r = ic.getWordRangeAtCursor(" ", 0); + r = ic.getWordRangeAtCursor(SPACE, 0); suggestions = r.getSuggestionSpansAtWord(); assertEquals(suggestions.length, 2); MoreAsserts.assertEquals(suggestions[0].getSuggestions(), SUGGESTIONS1); @@ -290,7 +296,7 @@ public class RichInputConnectionAndTextRangeTests extends AndroidTestCase { text.setSpan(new SuggestionSpan(Locale.ENGLISH, SUGGESTIONS2, 0 /* flags */), 5 /* start */, 16 /* end */, 0 /* flags */); mockInputMethodService.setInputConnection(new MockConnection(text, cursorPos)); - r = ic.getWordRangeAtCursor(" ", 0); + r = ic.getWordRangeAtCursor(SPACE, 0); suggestions = r.getSuggestionSpansAtWord(); assertEquals(suggestions.length, 1); MoreAsserts.assertEquals(suggestions[0].getSuggestions(), SUGGESTIONS1); @@ -302,7 +308,7 @@ public class RichInputConnectionAndTextRangeTests extends AndroidTestCase { text.setSpan(new SuggestionSpan(Locale.ENGLISH, SUGGESTIONS2, 0 /* flags */), 10 /* start */, 20 /* end */, 0 /* flags */); mockInputMethodService.setInputConnection(new MockConnection(text, cursorPos)); - r = ic.getWordRangeAtCursor(" ", 0); + r = ic.getWordRangeAtCursor(SPACE, 0); suggestions = r.getSuggestionSpansAtWord(); assertEquals(suggestions.length, 1); MoreAsserts.assertEquals(suggestions[0].getSuggestions(), SUGGESTIONS1); @@ -314,7 +320,7 @@ public class RichInputConnectionAndTextRangeTests extends AndroidTestCase { text.setSpan(new SuggestionSpan(Locale.ENGLISH, SUGGESTIONS2, 0 /* flags */), 5 /* start */, 20 /* end */, 0 /* flags */); mockInputMethodService.setInputConnection(new MockConnection(text, cursorPos)); - r = ic.getWordRangeAtCursor(" ", 0); + r = ic.getWordRangeAtCursor(SPACE, 0); suggestions = r.getSuggestionSpansAtWord(); assertEquals(suggestions.length, 1); MoreAsserts.assertEquals(suggestions[0].getSuggestions(), SUGGESTIONS1); @@ -326,7 +332,7 @@ public class RichInputConnectionAndTextRangeTests extends AndroidTestCase { text.setSpan(new SuggestionSpan(Locale.ENGLISH, SUGGESTIONS2, 0 /* flags */), 5 /* start */, 20 /* end */, 0 /* flags */); mockInputMethodService.setInputConnection(new MockConnection(text, cursorPos)); - r = ic.getWordRangeAtCursor(" ", 0); + r = ic.getWordRangeAtCursor(SPACE, 0); suggestions = r.getSuggestionSpansAtWord(); assertEquals(suggestions.length, 0); } diff --git a/tests/src/com/android/inputmethod/latin/settings/SpacingAndPunctuationsTests.java b/tests/src/com/android/inputmethod/latin/settings/SpacingAndPunctuationsTests.java new file mode 100644 index 000000000..424e7ff1b --- /dev/null +++ b/tests/src/com/android/inputmethod/latin/settings/SpacingAndPunctuationsTests.java @@ -0,0 +1,401 @@ +/* + * 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.settings; + +import android.content.res.Resources; +import android.test.AndroidTestCase; +import android.test.suitebuilder.annotation.SmallTest; + +import com.android.inputmethod.latin.SuggestedWords; +import com.android.inputmethod.latin.utils.RunInLocale; + +import junit.framework.AssertionFailedError; + +import java.util.Locale; + +@SmallTest +public class SpacingAndPunctuationsTests extends AndroidTestCase { + private static final int ARMENIAN_FULL_STOP = '\u0589'; + private static final int ARMENIAN_COMMA = '\u055D'; + + private SpacingAndPunctuations ENGLISH; + private SpacingAndPunctuations FRENCH; + private SpacingAndPunctuations GERMAN; + private SpacingAndPunctuations ARMENIAN; + private SpacingAndPunctuations THAI; + private SpacingAndPunctuations KHMER; + private SpacingAndPunctuations LAO; + private SpacingAndPunctuations ARABIC; + private SpacingAndPunctuations PERSIAN; + private SpacingAndPunctuations HEBREW; + + private SpacingAndPunctuations UNITED_STATES; + private SpacingAndPunctuations UNITED_KINGDOM; + private SpacingAndPunctuations CANADA_FRENCH; + private SpacingAndPunctuations SWISS_GERMAN; + private SpacingAndPunctuations INDIA_ENGLISH; + private SpacingAndPunctuations ARMENIA_ARMENIAN; + private SpacingAndPunctuations CAMBODIA_KHMER; + private SpacingAndPunctuations LAOS_LAO; + + @Override + protected void setUp() throws Exception { + super.setUp(); + + // Language only + ENGLISH = getSpacingAndPunctuations(Locale.ENGLISH); + FRENCH = getSpacingAndPunctuations(Locale.FRENCH); + GERMAN = getSpacingAndPunctuations(Locale.GERMAN); + THAI = getSpacingAndPunctuations(new Locale("th")); + ARMENIAN = getSpacingAndPunctuations(new Locale("hy")); + KHMER = getSpacingAndPunctuations(new Locale("km")); + LAO = getSpacingAndPunctuations(new Locale("lo")); + ARABIC = getSpacingAndPunctuations(new Locale("ar")); + PERSIAN = getSpacingAndPunctuations(new Locale("fa")); + HEBREW = getSpacingAndPunctuations(new Locale("iw")); + + // Language and Country + UNITED_STATES = getSpacingAndPunctuations(Locale.US); + UNITED_KINGDOM = getSpacingAndPunctuations(Locale.UK); + CANADA_FRENCH = getSpacingAndPunctuations(Locale.CANADA_FRENCH); + SWISS_GERMAN = getSpacingAndPunctuations(new Locale("de", "CH")); + INDIA_ENGLISH = getSpacingAndPunctuations(new Locale("en", "IN")); + ARMENIA_ARMENIAN = getSpacingAndPunctuations(new Locale("hy", "AM")); + CAMBODIA_KHMER = getSpacingAndPunctuations(new Locale("km", "KH")); + LAOS_LAO = getSpacingAndPunctuations(new Locale("lo", "LA")); + } + + private SpacingAndPunctuations getSpacingAndPunctuations(final Locale locale) { + final RunInLocale<SpacingAndPunctuations> job = new RunInLocale<SpacingAndPunctuations>() { + @Override + protected SpacingAndPunctuations job(Resources res) { + return new SpacingAndPunctuations(res); + } + }; + return job.runInLocale(getContext().getResources(), locale); + } + + private static void testingStandardWordSeparator(final SpacingAndPunctuations sp) { + assertTrue("Tab", sp.isWordSeparator('\t')); + assertTrue("Newline", sp.isWordSeparator('\n')); + assertTrue("Space", sp.isWordSeparator(' ')); + assertTrue("Exclamation", sp.isWordSeparator('!')); + assertTrue("Quotation", sp.isWordSeparator('"')); + assertFalse("Number", sp.isWordSeparator('#')); + assertFalse("Dollar", sp.isWordSeparator('$')); + assertFalse("Percent", sp.isWordSeparator('%')); + assertTrue("Ampersand", sp.isWordSeparator('&')); + assertFalse("Apostrophe", sp.isWordSeparator('\'')); + assertTrue("L Paren", sp.isWordSeparator('(')); + assertTrue("R Paren", sp.isWordSeparator(')')); + assertTrue("Asterisk", sp.isWordSeparator('*')); + assertTrue("Plus", sp.isWordSeparator('+')); + assertTrue("Comma", sp.isWordSeparator(',')); + assertFalse("Minus", sp.isWordSeparator('-')); + assertTrue("Period", sp.isWordSeparator('.')); + assertTrue("Slash", sp.isWordSeparator('/')); + assertTrue("Colon", sp.isWordSeparator(':')); + assertTrue("Semicolon", sp.isWordSeparator(';')); + assertTrue("L Angle", sp.isWordSeparator('<')); + assertTrue("Equal", sp.isWordSeparator('=')); + assertTrue("R Angle", sp.isWordSeparator('>')); + assertTrue("Question", sp.isWordSeparator('?')); + assertFalse("Atmark", sp.isWordSeparator('@')); + assertTrue("L S Bracket", sp.isWordSeparator('[')); + assertFalse("B Slash", sp.isWordSeparator('\\')); + assertTrue("R S Bracket", sp.isWordSeparator(']')); + assertFalse("Circumflex", sp.isWordSeparator('^')); + assertTrue("Underscore", sp.isWordSeparator('_')); + assertFalse("Grave", sp.isWordSeparator('`')); + assertTrue("L C Brace", sp.isWordSeparator('{')); + assertTrue("V Line", sp.isWordSeparator('|')); + assertTrue("R C Brace", sp.isWordSeparator('}')); + assertFalse("Tilde", sp.isWordSeparator('~')); + } + + public void testWordSeparator() { + testingStandardWordSeparator(ENGLISH); + testingStandardWordSeparator(FRENCH); + testingStandardWordSeparator(CANADA_FRENCH); + testingStandardWordSeparator(ARMENIA_ARMENIAN); + assertTrue(ARMENIA_ARMENIAN.isWordSeparator(ARMENIAN_FULL_STOP)); + assertTrue(ARMENIA_ARMENIAN.isWordSeparator(ARMENIAN_COMMA)); + // TODO: We should fix these. + testingStandardWordSeparator(ARMENIAN); + assertFalse(ARMENIAN.isWordSeparator(ARMENIAN_FULL_STOP)); + assertFalse(ARMENIAN.isWordSeparator(ARMENIAN_COMMA)); + } + + private static void testingStandardWordConnector(final SpacingAndPunctuations sp) { + assertFalse("Tab", sp.isWordConnector('\t')); + assertFalse("Newline", sp.isWordConnector('\n')); + assertFalse("Space", sp.isWordConnector(' ')); + assertFalse("Exclamation", sp.isWordConnector('!')); + assertFalse("Quotation", sp.isWordConnector('"')); + assertFalse("Number", sp.isWordConnector('#')); + assertFalse("Dollar", sp.isWordConnector('$')); + assertFalse("Percent", sp.isWordConnector('%')); + assertFalse("Ampersand", sp.isWordConnector('&')); + assertTrue("Apostrophe", sp.isWordConnector('\'')); + assertFalse("L Paren", sp.isWordConnector('(')); + assertFalse("R Paren", sp.isWordConnector(')')); + assertFalse("Asterisk", sp.isWordConnector('*')); + assertFalse("Plus", sp.isWordConnector('+')); + assertFalse("Comma", sp.isWordConnector(',')); + assertTrue("Minus", sp.isWordConnector('-')); + assertFalse("Period", sp.isWordConnector('.')); + assertFalse("Slash", sp.isWordConnector('/')); + assertFalse("Colon", sp.isWordConnector(':')); + assertFalse("Semicolon", sp.isWordConnector(';')); + assertFalse("L Angle", sp.isWordConnector('<')); + assertFalse("Equal", sp.isWordConnector('=')); + assertFalse("R Angle", sp.isWordConnector('>')); + assertFalse("Question", sp.isWordConnector('?')); + assertFalse("Atmark", sp.isWordConnector('@')); + assertFalse("L S Bracket", sp.isWordConnector('[')); + assertFalse("B Slash", sp.isWordConnector('\\')); + assertFalse("R S Bracket", sp.isWordConnector(']')); + assertFalse("Circumflex", sp.isWordConnector('^')); + assertFalse("Underscore", sp.isWordConnector('_')); + assertFalse("Grave", sp.isWordConnector('`')); + assertFalse("L C Brace", sp.isWordConnector('{')); + assertFalse("V Line", sp.isWordConnector('|')); + assertFalse("R C Brace", sp.isWordConnector('}')); + assertFalse("Tilde", sp.isWordConnector('~')); + + } + + public void testWordConnector() { + testingStandardWordConnector(ENGLISH); + testingStandardWordConnector(FRENCH); + testingStandardWordConnector(CANADA_FRENCH); + testingStandardWordConnector(ARMENIA_ARMENIAN); + } + + private static void testingCommonPrecededBySpace(final SpacingAndPunctuations sp) { + assertFalse("Tab", sp.isUsuallyPrecededBySpace('\t')); + assertFalse("Newline", sp.isUsuallyPrecededBySpace('\n')); + assertFalse("Space", sp.isUsuallyPrecededBySpace(' ')); + //assertFalse("Exclamation", sp.isUsuallyPrecededBySpace('!')); + assertFalse("Quotation", sp.isUsuallyPrecededBySpace('"')); + assertFalse("Number", sp.isUsuallyPrecededBySpace('#')); + assertFalse("Dollar", sp.isUsuallyPrecededBySpace('$')); + assertFalse("Percent", sp.isUsuallyPrecededBySpace('%')); + assertTrue("Ampersand", sp.isUsuallyPrecededBySpace('&')); + assertFalse("Apostrophe", sp.isUsuallyPrecededBySpace('\'')); + assertTrue("L Paren", sp.isUsuallyPrecededBySpace('(')); + assertFalse("R Paren", sp.isUsuallyPrecededBySpace(')')); + assertFalse("Asterisk", sp.isUsuallyPrecededBySpace('*')); + assertFalse("Plus", sp.isUsuallyPrecededBySpace('+')); + assertFalse("Comma", sp.isUsuallyPrecededBySpace(',')); + assertFalse("Minus", sp.isUsuallyPrecededBySpace('-')); + assertFalse("Period", sp.isUsuallyPrecededBySpace('.')); + assertFalse("Slash", sp.isUsuallyPrecededBySpace('/')); + //assertFalse("Colon", sp.isUsuallyPrecededBySpace(':')); + //assertFalse("Semicolon", sp.isUsuallyPrecededBySpace(';')); + assertFalse("L Angle", sp.isUsuallyPrecededBySpace('<')); + assertFalse("Equal", sp.isUsuallyPrecededBySpace('=')); + assertFalse("R Angle", sp.isUsuallyPrecededBySpace('>')); + //assertFalse("Question", sp.isUsuallyPrecededBySpace('?')); + assertFalse("Atmark", sp.isUsuallyPrecededBySpace('@')); + assertTrue("L S Bracket", sp.isUsuallyPrecededBySpace('[')); + assertFalse("B Slash", sp.isUsuallyPrecededBySpace('\\')); + assertFalse("R S Bracket", sp.isUsuallyPrecededBySpace(']')); + assertFalse("Circumflex", sp.isUsuallyPrecededBySpace('^')); + assertFalse("Underscore", sp.isUsuallyPrecededBySpace('_')); + assertFalse("Grave", sp.isUsuallyPrecededBySpace('`')); + assertTrue("L C Brace", sp.isUsuallyPrecededBySpace('{')); + assertFalse("V Line", sp.isUsuallyPrecededBySpace('|')); + assertFalse("R C Brace", sp.isUsuallyPrecededBySpace('}')); + assertFalse("Tilde", sp.isUsuallyPrecededBySpace('~')); + } + + private static void testingStandardPrecededBySpace(final SpacingAndPunctuations sp) { + testingCommonPrecededBySpace(sp); + assertFalse("Exclamation", sp.isUsuallyPrecededBySpace('!')); + assertFalse("Colon", sp.isUsuallyPrecededBySpace(':')); + assertFalse("Semicolon", sp.isUsuallyPrecededBySpace(';')); + assertFalse("Question", sp.isUsuallyPrecededBySpace('?')); + } + + public void testIsUsuallyPrecededBySpace() { + testingStandardPrecededBySpace(ENGLISH); + testingCommonPrecededBySpace(FRENCH); + assertTrue("Exclamation", FRENCH.isUsuallyPrecededBySpace('!')); + assertTrue("Colon", FRENCH.isUsuallyPrecededBySpace(':')); + assertTrue("Semicolon", FRENCH.isUsuallyPrecededBySpace(';')); + assertTrue("Question", FRENCH.isUsuallyPrecededBySpace('?')); + testingCommonPrecededBySpace(CANADA_FRENCH); + assertFalse("Exclamation", CANADA_FRENCH.isUsuallyPrecededBySpace('!')); + assertTrue("Colon", CANADA_FRENCH.isUsuallyPrecededBySpace(':')); + assertFalse("Semicolon", CANADA_FRENCH.isUsuallyPrecededBySpace(';')); + assertFalse("Question", CANADA_FRENCH.isUsuallyPrecededBySpace('?')); + testingStandardPrecededBySpace(ARMENIA_ARMENIAN); + } + + private static void testingStandardFollowedBySpace(final SpacingAndPunctuations sp) { + assertFalse("Tab", sp.isUsuallyFollowedBySpace('\t')); + assertFalse("Newline", sp.isUsuallyFollowedBySpace('\n')); + assertFalse("Space", sp.isUsuallyFollowedBySpace(' ')); + assertTrue("Exclamation", sp.isUsuallyFollowedBySpace('!')); + assertFalse("Quotation", sp.isUsuallyFollowedBySpace('"')); + assertFalse("Number", sp.isUsuallyFollowedBySpace('#')); + assertFalse("Dollar", sp.isUsuallyFollowedBySpace('$')); + assertFalse("Percent", sp.isUsuallyFollowedBySpace('%')); + assertTrue("Ampersand", sp.isUsuallyFollowedBySpace('&')); + assertFalse("Apostrophe", sp.isUsuallyFollowedBySpace('\'')); + assertFalse("L Paren", sp.isUsuallyFollowedBySpace('(')); + assertTrue("R Paren", sp.isUsuallyFollowedBySpace(')')); + assertFalse("Asterisk", sp.isUsuallyFollowedBySpace('*')); + assertFalse("Plus", sp.isUsuallyFollowedBySpace('+')); + assertTrue("Comma", sp.isUsuallyFollowedBySpace(',')); + assertFalse("Minus", sp.isUsuallyFollowedBySpace('-')); + assertTrue("Period", sp.isUsuallyFollowedBySpace('.')); + assertFalse("Slash", sp.isUsuallyFollowedBySpace('/')); + assertTrue("Colon", sp.isUsuallyFollowedBySpace(':')); + assertTrue("Semicolon", sp.isUsuallyFollowedBySpace(';')); + assertFalse("L Angle", sp.isUsuallyFollowedBySpace('<')); + assertFalse("Equal", sp.isUsuallyFollowedBySpace('=')); + assertFalse("R Angle", sp.isUsuallyFollowedBySpace('>')); + assertTrue("Question", sp.isUsuallyFollowedBySpace('?')); + assertFalse("Atmark", sp.isUsuallyFollowedBySpace('@')); + assertFalse("L S Bracket", sp.isUsuallyFollowedBySpace('[')); + assertFalse("B Slash", sp.isUsuallyFollowedBySpace('\\')); + assertTrue("R S Bracket", sp.isUsuallyFollowedBySpace(']')); + assertFalse("Circumflex", sp.isUsuallyFollowedBySpace('^')); + assertFalse("Underscore", sp.isUsuallyFollowedBySpace('_')); + assertFalse("Grave", sp.isUsuallyFollowedBySpace('`')); + assertFalse("L C Brace", sp.isUsuallyFollowedBySpace('{')); + assertFalse("V Line", sp.isUsuallyFollowedBySpace('|')); + assertTrue("R C Brace", sp.isUsuallyFollowedBySpace('}')); + assertFalse("Tilde", sp.isUsuallyFollowedBySpace('~')); + } + + public void testIsUsuallyFollowedBySpace() { + testingStandardFollowedBySpace(ENGLISH); + testingStandardFollowedBySpace(FRENCH); + testingStandardFollowedBySpace(CANADA_FRENCH); + testingStandardFollowedBySpace(ARMENIA_ARMENIAN); + assertTrue(ARMENIA_ARMENIAN.isUsuallyFollowedBySpace(ARMENIAN_FULL_STOP)); + assertTrue(ARMENIA_ARMENIAN.isUsuallyFollowedBySpace(ARMENIAN_COMMA)); + } + + private static void testingStandardSentenceSeparator(final SpacingAndPunctuations sp) { + assertFalse("Tab", sp.isUsuallyFollowedBySpace('\t')); + assertFalse("Newline", sp.isUsuallyFollowedBySpace('\n')); + assertFalse("Space", sp.isUsuallyFollowedBySpace(' ')); + assertFalse("Exclamation", sp.isUsuallyFollowedBySpace('!')); + assertFalse("Quotation", sp.isUsuallyFollowedBySpace('"')); + assertFalse("Number", sp.isUsuallyFollowedBySpace('#')); + assertFalse("Dollar", sp.isUsuallyFollowedBySpace('$')); + assertFalse("Percent", sp.isUsuallyFollowedBySpace('%')); + assertFalse("Ampersand", sp.isUsuallyFollowedBySpace('&')); + assertFalse("Apostrophe", sp.isUsuallyFollowedBySpace('\'')); + assertFalse("L Paren", sp.isUsuallyFollowedBySpace('(')); + assertFalse("R Paren", sp.isUsuallyFollowedBySpace(')')); + assertFalse("Asterisk", sp.isUsuallyFollowedBySpace('*')); + assertFalse("Plus", sp.isUsuallyFollowedBySpace('+')); + assertFalse("Comma", sp.isUsuallyFollowedBySpace(',')); + assertFalse("Minus", sp.isUsuallyFollowedBySpace('-')); + assertTrue("Period", sp.isUsuallyFollowedBySpace('.')); + assertFalse("Slash", sp.isUsuallyFollowedBySpace('/')); + assertFalse("Colon", sp.isUsuallyFollowedBySpace(':')); + assertFalse("Semicolon", sp.isUsuallyFollowedBySpace(';')); + assertFalse("L Angle", sp.isUsuallyFollowedBySpace('<')); + assertFalse("Equal", sp.isUsuallyFollowedBySpace('=')); + assertFalse("R Angle", sp.isUsuallyFollowedBySpace('>')); + assertFalse("Question", sp.isUsuallyFollowedBySpace('?')); + assertFalse("Atmark", sp.isUsuallyFollowedBySpace('@')); + assertFalse("L S Bracket", sp.isUsuallyFollowedBySpace('[')); + assertFalse("B Slash", sp.isUsuallyFollowedBySpace('\\')); + assertFalse("R S Bracket", sp.isUsuallyFollowedBySpace(']')); + assertFalse("Circumflex", sp.isUsuallyFollowedBySpace('^')); + assertFalse("Underscore", sp.isUsuallyFollowedBySpace('_')); + assertFalse("Grave", sp.isUsuallyFollowedBySpace('`')); + assertFalse("L C Brace", sp.isUsuallyFollowedBySpace('{')); + assertFalse("V Line", sp.isUsuallyFollowedBySpace('|')); + assertFalse("R C Brace", sp.isUsuallyFollowedBySpace('}')); + assertFalse("Tilde", sp.isUsuallyFollowedBySpace('~')); + } + + public void isSentenceSeparator() { + testingStandardSentenceSeparator(ENGLISH); + try { + testingStandardSentenceSeparator(ARMENIA_ARMENIAN); + fail("Armenian Sentence Separator"); + } catch (final AssertionFailedError e) { + assertEquals("Period", e.getMessage()); + } + assertTrue(ARMENIA_ARMENIAN.isSentenceSeparator(ARMENIAN_FULL_STOP)); + assertFalse(ARMENIA_ARMENIAN.isSentenceSeparator(ARMENIAN_COMMA)); + } + + public void testLanguageHasSpace() { + assertTrue(ENGLISH.mCurrentLanguageHasSpaces); + assertTrue(FRENCH.mCurrentLanguageHasSpaces); + assertTrue(GERMAN.mCurrentLanguageHasSpaces); + assertFalse(THAI.mCurrentLanguageHasSpaces); + assertFalse(CAMBODIA_KHMER.mCurrentLanguageHasSpaces); + assertFalse(LAOS_LAO.mCurrentLanguageHasSpaces); + // TODO: We should fix these. + assertTrue(KHMER.mCurrentLanguageHasSpaces); + assertTrue(LAO.mCurrentLanguageHasSpaces); + } + + public void testUsesAmericanTypography() { + assertTrue(ENGLISH.mUsesAmericanTypography); + assertTrue(UNITED_STATES.mUsesAmericanTypography); + assertTrue(UNITED_KINGDOM.mUsesAmericanTypography); + assertTrue(INDIA_ENGLISH.mUsesAmericanTypography); + assertFalse(FRENCH.mUsesAmericanTypography); + assertFalse(GERMAN.mUsesAmericanTypography); + assertFalse(SWISS_GERMAN.mUsesAmericanTypography); + } + + public void testUsesGermanRules() { + assertFalse(ENGLISH.mUsesGermanRules); + assertFalse(FRENCH.mUsesGermanRules); + assertTrue(GERMAN.mUsesGermanRules); + assertTrue(SWISS_GERMAN.mUsesGermanRules); + } + + private static void testingStandardPunctuationSuggestions(final SpacingAndPunctuations sp) { + final SuggestedWords suggestedWords = sp.mSuggestPuncList; + assertFalse("typedWordValid", suggestedWords.mTypedWordValid); + assertFalse("willAutoCorrect", suggestedWords.mWillAutoCorrect); + assertTrue("isPunctuationSuggestions", suggestedWords.mIsPunctuationSuggestions); + assertFalse("isObsoleteSuggestions", suggestedWords.mIsObsoleteSuggestions); + assertFalse("isPrediction", suggestedWords.mIsPrediction); + final String[] punctuations = { + "!", "?", ",", ":", ";", "\"", "(", ")", "'", "-", "/", "@", "_" + }; + assertEquals("size", punctuations.length, suggestedWords.size()); + for (int index = 0; index < punctuations.length; index++) { + assertEquals("punctuation at " + index, + punctuations[index], suggestedWords.getWord(index)); + } + } + public void testPunctuationSuggestions() { + testingStandardPunctuationSuggestions(ENGLISH); + testingStandardPunctuationSuggestions(FRENCH); + testingStandardPunctuationSuggestions(GERMAN); + // TODO: Should fix these RTL languages + testingStandardPunctuationSuggestions(ARABIC); + testingStandardPunctuationSuggestions(PERSIAN); + testingStandardPunctuationSuggestions(HEBREW); + } +} diff --git a/tests/src/com/android/inputmethod/latin/utils/RecapitalizeStatusTests.java b/tests/src/com/android/inputmethod/latin/utils/RecapitalizeStatusTests.java index a52041264..ada80c3fa 100644 --- a/tests/src/com/android/inputmethod/latin/utils/RecapitalizeStatusTests.java +++ b/tests/src/com/android/inputmethod/latin/utils/RecapitalizeStatusTests.java @@ -19,31 +19,35 @@ package com.android.inputmethod.latin.utils; import android.test.AndroidTestCase; import android.test.suitebuilder.annotation.SmallTest; +import com.android.inputmethod.latin.Constants; + import java.util.Locale; @SmallTest public class RecapitalizeStatusTests extends AndroidTestCase { + private static final int[] SPACE = { Constants.CODE_SPACE }; + public void testTrim() { final RecapitalizeStatus status = new RecapitalizeStatus(); - status.initialize(30, 40, "abcdefghij", Locale.ENGLISH, " "); + status.initialize(30, 40, "abcdefghij", Locale.ENGLISH, SPACE); status.trim(); assertEquals("abcdefghij", status.getRecapitalizedString()); assertEquals(30, status.getNewCursorStart()); assertEquals(40, status.getNewCursorEnd()); - status.initialize(30, 44, " abcdefghij", Locale.ENGLISH, " "); + status.initialize(30, 44, " abcdefghij", Locale.ENGLISH, SPACE); status.trim(); assertEquals("abcdefghij", status.getRecapitalizedString()); assertEquals(34, status.getNewCursorStart()); assertEquals(44, status.getNewCursorEnd()); - status.initialize(30, 40, "abcdefgh ", Locale.ENGLISH, " "); + status.initialize(30, 40, "abcdefgh ", Locale.ENGLISH, SPACE); status.trim(); assertEquals("abcdefgh", status.getRecapitalizedString()); assertEquals(30, status.getNewCursorStart()); assertEquals(38, status.getNewCursorEnd()); - status.initialize(30, 45, " abcdefghij ", Locale.ENGLISH, " "); + status.initialize(30, 45, " abcdefghij ", Locale.ENGLISH, SPACE); status.trim(); assertEquals("abcdefghij", status.getRecapitalizedString()); assertEquals(33, status.getNewCursorStart()); @@ -52,7 +56,7 @@ public class RecapitalizeStatusTests extends AndroidTestCase { public void testRotate() { final RecapitalizeStatus status = new RecapitalizeStatus(); - status.initialize(29, 40, "abcd efghij", Locale.ENGLISH, " "); + status.initialize(29, 40, "abcd efghij", Locale.ENGLISH, SPACE); status.rotate(); assertEquals("Abcd Efghij", status.getRecapitalizedString()); assertEquals(29, status.getNewCursorStart()); @@ -64,7 +68,7 @@ public class RecapitalizeStatusTests extends AndroidTestCase { status.rotate(); assertEquals("Abcd Efghij", status.getRecapitalizedString()); - status.initialize(29, 40, "Abcd Efghij", Locale.ENGLISH, " "); + status.initialize(29, 40, "Abcd Efghij", Locale.ENGLISH, SPACE); status.rotate(); assertEquals("ABCD EFGHIJ", status.getRecapitalizedString()); assertEquals(29, status.getNewCursorStart()); @@ -76,7 +80,7 @@ public class RecapitalizeStatusTests extends AndroidTestCase { status.rotate(); assertEquals("ABCD EFGHIJ", status.getRecapitalizedString()); - status.initialize(29, 40, "ABCD EFGHIJ", Locale.ENGLISH, " "); + status.initialize(29, 40, "ABCD EFGHIJ", Locale.ENGLISH, SPACE); status.rotate(); assertEquals("abcd efghij", status.getRecapitalizedString()); assertEquals(29, status.getNewCursorStart()); @@ -88,7 +92,7 @@ public class RecapitalizeStatusTests extends AndroidTestCase { status.rotate(); assertEquals("abcd efghij", status.getRecapitalizedString()); - status.initialize(29, 39, "AbCDefghij", Locale.ENGLISH, " "); + status.initialize(29, 39, "AbCDefghij", Locale.ENGLISH, SPACE); status.rotate(); assertEquals("abcdefghij", status.getRecapitalizedString()); assertEquals(29, status.getNewCursorStart()); @@ -102,7 +106,7 @@ public class RecapitalizeStatusTests extends AndroidTestCase { status.rotate(); assertEquals("abcdefghij", status.getRecapitalizedString()); - status.initialize(29, 40, "Abcd efghij", Locale.ENGLISH, " "); + status.initialize(29, 40, "Abcd efghij", Locale.ENGLISH, SPACE); status.rotate(); assertEquals("abcd efghij", status.getRecapitalizedString()); assertEquals(29, status.getNewCursorStart()); @@ -116,7 +120,8 @@ public class RecapitalizeStatusTests extends AndroidTestCase { status.rotate(); assertEquals("abcd efghij", status.getRecapitalizedString()); - status.initialize(30, 34, "grüß", Locale.GERMAN, " "); status.rotate(); + status.initialize(30, 34, "grüß", Locale.GERMAN, SPACE); + status.rotate(); assertEquals("Grüß", status.getRecapitalizedString()); assertEquals(30, status.getNewCursorStart()); assertEquals(34, status.getNewCursorEnd()); @@ -133,7 +138,8 @@ public class RecapitalizeStatusTests extends AndroidTestCase { assertEquals(30, status.getNewCursorStart()); assertEquals(34, status.getNewCursorEnd()); - status.initialize(30, 33, "œuf", Locale.FRENCH, " "); status.rotate(); + status.initialize(30, 33, "œuf", Locale.FRENCH, SPACE); + status.rotate(); assertEquals("Œuf", status.getRecapitalizedString()); assertEquals(30, status.getNewCursorStart()); assertEquals(33, status.getNewCursorEnd()); @@ -150,7 +156,8 @@ public class RecapitalizeStatusTests extends AndroidTestCase { assertEquals(30, status.getNewCursorStart()); assertEquals(33, status.getNewCursorEnd()); - status.initialize(30, 33, "œUf", Locale.FRENCH, " "); status.rotate(); + status.initialize(30, 33, "œUf", Locale.FRENCH, SPACE); + status.rotate(); assertEquals("œuf", status.getRecapitalizedString()); assertEquals(30, status.getNewCursorStart()); assertEquals(33, status.getNewCursorEnd()); @@ -171,7 +178,8 @@ public class RecapitalizeStatusTests extends AndroidTestCase { assertEquals(30, status.getNewCursorStart()); assertEquals(33, status.getNewCursorEnd()); - status.initialize(30, 35, "école", Locale.FRENCH, " "); status.rotate(); + status.initialize(30, 35, "école", Locale.FRENCH, SPACE); + status.rotate(); assertEquals("École", status.getRecapitalizedString()); assertEquals(30, status.getNewCursorStart()); assertEquals(35, status.getNewCursorEnd()); diff --git a/tests/src/com/android/inputmethod/latin/utils/StringAndJsonUtilsTests.java b/tests/src/com/android/inputmethod/latin/utils/StringAndJsonUtilsTests.java index 71966124f..e55c32bd0 100644 --- a/tests/src/com/android/inputmethod/latin/utils/StringAndJsonUtilsTests.java +++ b/tests/src/com/android/inputmethod/latin/utils/StringAndJsonUtilsTests.java @@ -19,6 +19,8 @@ package com.android.inputmethod.latin.utils; import android.test.AndroidTestCase; import android.test.suitebuilder.annotation.SmallTest; +import com.android.inputmethod.latin.Constants; + import java.util.Arrays; import java.util.List; import java.util.Locale; @@ -206,42 +208,47 @@ public class StringAndJsonUtilsTests extends AndroidTestCase { assertTrue(StringUtils.isIdenticalAfterDowncase("")); } - private static void checkCapitalize(final String src, final String dst, final String separators, - final Locale locale) { - assertEquals(dst, StringUtils.capitalizeEachWord(src, separators, locale)); + private static void checkCapitalize(final String src, final String dst, + final int[] sortedSeparators, final Locale locale) { + assertEquals(dst, StringUtils.capitalizeEachWord(src, sortedSeparators, locale)); assert(src.equals(dst) - == StringUtils.isIdenticalAfterCapitalizeEachWord(src, separators)); + == StringUtils.isIdenticalAfterCapitalizeEachWord(src, sortedSeparators)); } + private static final int[] SPACE = { Constants.CODE_SPACE }; + private static final int[] SPACE_PERIOD = StringUtils.toSortedCodePointArray(" ."); + private static final int[] SENTENCE_SEPARATORS = + StringUtils.toSortedCodePointArray(" \n.!?*()&"); + private static final int[] WORD_SEPARATORS = StringUtils.toSortedCodePointArray(" \n.!?*,();&"); + public void testCapitalizeEachWord() { - checkCapitalize("", "", " ", Locale.ENGLISH); - checkCapitalize("test", "Test", " ", Locale.ENGLISH); - checkCapitalize(" test", " Test", " ", Locale.ENGLISH); - checkCapitalize("Test", "Test", " ", Locale.ENGLISH); - checkCapitalize(" Test", " Test", " ", Locale.ENGLISH); - checkCapitalize(".Test", ".test", " ", Locale.ENGLISH); - checkCapitalize(".Test", ".Test", " .", Locale.ENGLISH); - checkCapitalize(".Test", ".Test", ". ", Locale.ENGLISH); - checkCapitalize("test and retest", "Test And Retest", " .", Locale.ENGLISH); - checkCapitalize("Test and retest", "Test And Retest", " .", Locale.ENGLISH); - checkCapitalize("Test And Retest", "Test And Retest", " .", Locale.ENGLISH); - checkCapitalize("Test And.Retest ", "Test And.Retest ", " .", Locale.ENGLISH); - checkCapitalize("Test And.retest ", "Test And.Retest ", " .", Locale.ENGLISH); - checkCapitalize("Test And.retest ", "Test And.retest ", " ", Locale.ENGLISH); - checkCapitalize("Test And.Retest ", "Test And.retest ", " ", Locale.ENGLISH); - checkCapitalize("test and ietest", "Test And İetest", " .", new Locale("tr")); - checkCapitalize("test and ietest", "Test And Ietest", " .", Locale.ENGLISH); - checkCapitalize("Test&Retest", "Test&Retest", " \n.!?*()&", Locale.ENGLISH); - checkCapitalize("Test&retest", "Test&Retest", " \n.!?*()&", Locale.ENGLISH); - checkCapitalize("test&Retest", "Test&Retest", " \n.!?*()&", Locale.ENGLISH); + checkCapitalize("", "", SPACE, Locale.ENGLISH); + checkCapitalize("test", "Test", SPACE, Locale.ENGLISH); + checkCapitalize(" test", " Test", SPACE, Locale.ENGLISH); + checkCapitalize("Test", "Test", SPACE, Locale.ENGLISH); + checkCapitalize(" Test", " Test", SPACE, Locale.ENGLISH); + checkCapitalize(".Test", ".test", SPACE, Locale.ENGLISH); + checkCapitalize(".Test", ".Test", SPACE_PERIOD, Locale.ENGLISH); + checkCapitalize("test and retest", "Test And Retest", SPACE_PERIOD, Locale.ENGLISH); + checkCapitalize("Test and retest", "Test And Retest", SPACE_PERIOD, Locale.ENGLISH); + checkCapitalize("Test And Retest", "Test And Retest", SPACE_PERIOD, Locale.ENGLISH); + checkCapitalize("Test And.Retest ", "Test And.Retest ", SPACE_PERIOD, Locale.ENGLISH); + checkCapitalize("Test And.retest ", "Test And.Retest ", SPACE_PERIOD, Locale.ENGLISH); + checkCapitalize("Test And.retest ", "Test And.retest ", SPACE, Locale.ENGLISH); + checkCapitalize("Test And.Retest ", "Test And.retest ", SPACE, Locale.ENGLISH); + checkCapitalize("test and ietest", "Test And İetest", SPACE_PERIOD, new Locale("tr")); + checkCapitalize("test and ietest", "Test And Ietest", SPACE_PERIOD, Locale.ENGLISH); + checkCapitalize("Test&Retest", "Test&Retest", SENTENCE_SEPARATORS, Locale.ENGLISH); + checkCapitalize("Test&retest", "Test&Retest", SENTENCE_SEPARATORS, Locale.ENGLISH); + checkCapitalize("test&Retest", "Test&Retest", SENTENCE_SEPARATORS, Locale.ENGLISH); checkCapitalize("rest\nrecreation! And in the end...", - "Rest\nRecreation! And In The End...", " \n.!?*,();&", Locale.ENGLISH); + "Rest\nRecreation! And In The End...", WORD_SEPARATORS, Locale.ENGLISH); checkCapitalize("lorem ipsum dolor sit amet", "Lorem Ipsum Dolor Sit Amet", - " \n.,!?*()&;", Locale.ENGLISH); + WORD_SEPARATORS, Locale.ENGLISH); checkCapitalize("Lorem!Ipsum (Dolor) Sit * Amet", "Lorem!Ipsum (Dolor) Sit * Amet", - " \n,.;!?*()&", Locale.ENGLISH); + WORD_SEPARATORS, Locale.ENGLISH); checkCapitalize("Lorem!Ipsum (dolor) Sit * Amet", "Lorem!Ipsum (Dolor) Sit * Amet", - " \n,.;!?*()&", Locale.ENGLISH); + WORD_SEPARATORS, Locale.ENGLISH); } public void testLooksLikeURL() { |