diff options
Diffstat (limited to 'java/src/org/kelar/inputmethod/latin/spellcheck/AndroidSpellCheckerSession.java')
-rw-r--r-- | java/src/org/kelar/inputmethod/latin/spellcheck/AndroidSpellCheckerSession.java | 225 |
1 files changed, 225 insertions, 0 deletions
diff --git a/java/src/org/kelar/inputmethod/latin/spellcheck/AndroidSpellCheckerSession.java b/java/src/org/kelar/inputmethod/latin/spellcheck/AndroidSpellCheckerSession.java new file mode 100644 index 000000000..3ab5138bf --- /dev/null +++ b/java/src/org/kelar/inputmethod/latin/spellcheck/AndroidSpellCheckerSession.java @@ -0,0 +1,225 @@ +/* + * Copyright (C) 2012 The Android Open Source Project + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.kelar.inputmethod.latin.spellcheck; + +import android.annotation.TargetApi; +import android.content.res.Resources; +import android.os.Binder; +import android.os.Build; +import android.text.TextUtils; +import android.util.Log; +import android.view.textservice.SentenceSuggestionsInfo; +import android.view.textservice.SuggestionsInfo; +import android.view.textservice.TextInfo; + +import org.kelar.inputmethod.compat.TextInfoCompatUtils; +import org.kelar.inputmethod.latin.NgramContext; +import org.kelar.inputmethod.latin.utils.SpannableStringUtils; + +import java.util.ArrayList; +import java.util.Locale; + +public final class AndroidSpellCheckerSession extends AndroidWordLevelSpellCheckerSession { + private static final String TAG = AndroidSpellCheckerSession.class.getSimpleName(); + private static final boolean DBG = false; + private final Resources mResources; + private SentenceLevelAdapter mSentenceLevelAdapter; + + public AndroidSpellCheckerSession(AndroidSpellCheckerService service) { + super(service); + mResources = service.getResources(); + } + + @TargetApi(Build.VERSION_CODES.JELLY_BEAN) + private SentenceSuggestionsInfo fixWronglyInvalidatedWordWithSingleQuote(TextInfo ti, + SentenceSuggestionsInfo ssi) { + final CharSequence typedText = TextInfoCompatUtils.getCharSequenceOrString(ti); + if (!typedText.toString().contains(AndroidSpellCheckerService.SINGLE_QUOTE)) { + return null; + } + final int N = ssi.getSuggestionsCount(); + final ArrayList<Integer> additionalOffsets = new ArrayList<>(); + final ArrayList<Integer> additionalLengths = new ArrayList<>(); + final ArrayList<SuggestionsInfo> additionalSuggestionsInfos = new ArrayList<>(); + CharSequence currentWord = null; + for (int i = 0; i < N; ++i) { + final SuggestionsInfo si = ssi.getSuggestionsInfoAt(i); + final int flags = si.getSuggestionsAttributes(); + if ((flags & SuggestionsInfo.RESULT_ATTR_IN_THE_DICTIONARY) == 0) { + continue; + } + final int offset = ssi.getOffsetAt(i); + final int length = ssi.getLengthAt(i); + final CharSequence subText = typedText.subSequence(offset, offset + length); + final NgramContext ngramContext = + new NgramContext(new NgramContext.WordInfo(currentWord)); + currentWord = subText; + if (!subText.toString().contains(AndroidSpellCheckerService.SINGLE_QUOTE)) { + continue; + } + // Split preserving spans. + final CharSequence[] splitTexts = SpannableStringUtils.split(subText, + AndroidSpellCheckerService.SINGLE_QUOTE, + true /* preserveTrailingEmptySegments */); + if (splitTexts == null || splitTexts.length <= 1) { + continue; + } + final int splitNum = splitTexts.length; + for (int j = 0; j < splitNum; ++j) { + final CharSequence splitText = splitTexts[j]; + if (TextUtils.isEmpty(splitText)) { + continue; + } + if (mSuggestionsCache.getSuggestionsFromCache(splitText.toString()) == null) { + continue; + } + final int newLength = splitText.length(); + // Neither RESULT_ATTR_IN_THE_DICTIONARY nor RESULT_ATTR_LOOKS_LIKE_TYPO + final int newFlags = 0; + final SuggestionsInfo newSi = new SuggestionsInfo(newFlags, EMPTY_STRING_ARRAY); + newSi.setCookieAndSequence(si.getCookie(), si.getSequence()); + if (DBG) { + Log.d(TAG, "Override and remove old span over: " + splitText + ", " + + offset + "," + newLength); + } + additionalOffsets.add(offset); + additionalLengths.add(newLength); + additionalSuggestionsInfos.add(newSi); + } + } + final int additionalSize = additionalOffsets.size(); + if (additionalSize <= 0) { + return null; + } + final int suggestionsSize = N + additionalSize; + final int[] newOffsets = new int[suggestionsSize]; + final int[] newLengths = new int[suggestionsSize]; + final SuggestionsInfo[] newSuggestionsInfos = new SuggestionsInfo[suggestionsSize]; + int i; + for (i = 0; i < N; ++i) { + newOffsets[i] = ssi.getOffsetAt(i); + newLengths[i] = ssi.getLengthAt(i); + newSuggestionsInfos[i] = ssi.getSuggestionsInfoAt(i); + } + for (; i < suggestionsSize; ++i) { + newOffsets[i] = additionalOffsets.get(i - N); + newLengths[i] = additionalLengths.get(i - N); + newSuggestionsInfos[i] = additionalSuggestionsInfos.get(i - N); + } + return new SentenceSuggestionsInfo(newSuggestionsInfos, newOffsets, newLengths); + } + + @Override + public SentenceSuggestionsInfo[] onGetSentenceSuggestionsMultiple(TextInfo[] textInfos, + int suggestionsLimit) { + final SentenceSuggestionsInfo[] retval = splitAndSuggest(textInfos, suggestionsLimit); + if (retval == null || retval.length != textInfos.length) { + return retval; + } + for (int i = 0; i < retval.length; ++i) { + final SentenceSuggestionsInfo tempSsi = + fixWronglyInvalidatedWordWithSingleQuote(textInfos[i], retval[i]); + if (tempSsi != null) { + retval[i] = tempSsi; + } + } + return retval; + } + + /** + * Get sentence suggestions for specified texts in an array of TextInfo. This is taken from + * SpellCheckerService#onGetSentenceSuggestionsMultiple that we can't use because it's + * using private variables. + * The default implementation splits the input text to words and returns + * {@link SentenceSuggestionsInfo} which contains suggestions for each word. + * This function will run on the incoming IPC thread. + * So, this is not called on the main thread, + * but will be called in series on another thread. + * @param textInfos an array of the text metadata + * @param suggestionsLimit the maximum number of suggestions to be returned + * @return an array of {@link SentenceSuggestionsInfo} returned by + * {@link android.service.textservice.SpellCheckerService.Session#onGetSuggestions(TextInfo, int)} + */ + private SentenceSuggestionsInfo[] splitAndSuggest(TextInfo[] textInfos, int suggestionsLimit) { + if (textInfos == null || textInfos.length == 0) { + return SentenceLevelAdapter.getEmptySentenceSuggestionsInfo(); + } + SentenceLevelAdapter sentenceLevelAdapter; + synchronized(this) { + sentenceLevelAdapter = mSentenceLevelAdapter; + if (sentenceLevelAdapter == null) { + final String localeStr = getLocale(); + if (!TextUtils.isEmpty(localeStr)) { + sentenceLevelAdapter = new SentenceLevelAdapter(mResources, + new Locale(localeStr)); + mSentenceLevelAdapter = sentenceLevelAdapter; + } + } + } + if (sentenceLevelAdapter == null) { + return SentenceLevelAdapter.getEmptySentenceSuggestionsInfo(); + } + final int infosSize = textInfos.length; + final SentenceSuggestionsInfo[] retval = new SentenceSuggestionsInfo[infosSize]; + for (int i = 0; i < infosSize; ++i) { + final SentenceLevelAdapter.SentenceTextInfoParams textInfoParams = + sentenceLevelAdapter.getSplitWords(textInfos[i]); + final ArrayList<SentenceLevelAdapter.SentenceWordItem> mItems = + textInfoParams.mItems; + final int itemsSize = mItems.size(); + final TextInfo[] splitTextInfos = new TextInfo[itemsSize]; + for (int j = 0; j < itemsSize; ++j) { + splitTextInfos[j] = mItems.get(j).mTextInfo; + } + retval[i] = SentenceLevelAdapter.reconstructSuggestions( + textInfoParams, onGetSuggestionsMultiple( + splitTextInfos, suggestionsLimit, true)); + } + return retval; + } + + @Override + public SuggestionsInfo[] onGetSuggestionsMultiple(TextInfo[] textInfos, + int suggestionsLimit, boolean sequentialWords) { + long ident = Binder.clearCallingIdentity(); + try { + final int length = textInfos.length; + final SuggestionsInfo[] retval = new SuggestionsInfo[length]; + for (int i = 0; i < length; ++i) { + final CharSequence prevWord; + if (sequentialWords && i > 0) { + final TextInfo prevTextInfo = textInfos[i - 1]; + final CharSequence prevWordCandidate = + TextInfoCompatUtils.getCharSequenceOrString(prevTextInfo); + // Note that an empty string would be used to indicate the initial word + // in the future. + prevWord = TextUtils.isEmpty(prevWordCandidate) ? null : prevWordCandidate; + } else { + prevWord = null; + } + final NgramContext ngramContext = + new NgramContext(new NgramContext.WordInfo(prevWord)); + final TextInfo textInfo = textInfos[i]; + retval[i] = onGetSuggestionsInternal(textInfo, ngramContext, suggestionsLimit); + retval[i].setCookieAndSequence(textInfo.getCookie(), textInfo.getSequence()); + } + return retval; + } finally { + Binder.restoreCallingIdentity(ident); + } + } +} |