Improve magic space handling.
Split symbols into: - Symbols that should swap themselves with a magic space. - Symbols that should remove the magic space. - Symbols that should transform the magic space into a real space. Also, have symbols inserted from the suggestion strip always consider the space before them to be magic. Also, do minor refactoring to remove useless variables or make clearer how to carry some symbol-related information. Bug: 4391347 Change-Id: Ie2308a62c8fef55e3610897c0f58943acb7aa8c9main
parent
aac2d87dcf
commit
863f95b605
|
@ -18,8 +18,12 @@
|
|||
*/
|
||||
-->
|
||||
<resources xmlns:xliff="urn:oasis:names:tc:xliff:document:1.2">
|
||||
<!-- Symbols that are commonly considered word separators in this language -->
|
||||
<string name="word_separators">.\u0009\u0020,;:!?\n()[]*&@{}/<>_+=|\u0022</string>
|
||||
<!-- Symbols that are sentence separators, for purposes of making it hug the last sentence. -->
|
||||
<string name="sentence_separators">.,</string>
|
||||
<!-- Symbols that should be swapped with a magic space -->
|
||||
<string name="magic_space_swapping_symbols">.,\u0022)]}</string>
|
||||
<!-- Symbols that should strip a magic space -->
|
||||
<string name="magic_space_stripping_symbols">\u0009\u0020\u0027\n-/_</string>
|
||||
<!-- Symbols that should promote magic spaces into real space -->
|
||||
<string name="magic_space_promoting_symbols">;:!?([*&@{<>+=|</string>
|
||||
<!-- Symbols that do NOT separate words -->
|
||||
<string name="non_word_separator_symbols">\u0027</string>
|
||||
</resources>
|
||||
|
|
|
@ -18,6 +18,6 @@
|
|||
*/
|
||||
-->
|
||||
<resources xmlns:xliff="urn:oasis:names:tc:xliff:document:1.2">
|
||||
<!-- Symbols that are commonly considered word separators in this language -->
|
||||
<string name="word_separators">.\u0009\u0020,;:!?\'\n()[]*&@{}/<>_+=|\u0022</string>
|
||||
<!-- Symbols that do NOT separate words -->
|
||||
<string name="non_word_separator_symbols"></string>
|
||||
</resources>
|
||||
|
|
|
@ -18,12 +18,19 @@
|
|||
*/
|
||||
-->
|
||||
<resources xmlns:xliff="urn:oasis:names:tc:xliff:document:1.2">
|
||||
<!-- Symbols that are commonly considered word separators in this language -->
|
||||
<string name="word_separators">.\u0009\u0020,;:!?\n()[]*&@{}/<>_+=|\u0022</string>
|
||||
<!-- Symbols that are sentence separators, for purposes of making it hug the last sentence. -->
|
||||
<string name="sentence_separators">.,!?)</string>
|
||||
<!-- Symbols that are suggested between words -->
|
||||
<string name="suggested_punctuations">!?,\u0022\u0027:();-/@_</string>
|
||||
<!-- Symbols that should be swapped with a magic space -->
|
||||
<string name="magic_space_swapping_symbols">.,;:!?)]}\u0022</string>
|
||||
<!-- Symbols that should strip a magic space -->
|
||||
<string name="magic_space_stripping_symbols">\u0009\u0020\n/_\u0027-</string>
|
||||
<!-- Symbols that should convert magic spaces into real space -->
|
||||
<string name="magic_space_promoting_symbols">([*&@{<>+=|</string>
|
||||
<!-- Symbols that do NOT separate words -->
|
||||
<string name="non_word_separator_symbols">\u0027-</string>
|
||||
<!-- Word separator list is the union of all symbols except those that are not separators:
|
||||
magic_space_swapping_symbols | magic_space_stripping_symbols |
|
||||
magic_space_neutral_symbols \ non_word_separator_symbols -->
|
||||
|
||||
<!-- Label for ALT modifier key. Must be short to fit on key! -->
|
||||
<string name="label_alt_key">ALT</string>
|
||||
|
|
|
@ -216,7 +216,8 @@ public class LatinIME extends InputMethodServiceCompatWrapper implements Keyboar
|
|||
private boolean mSilentMode;
|
||||
|
||||
/* package */ String mWordSeparators;
|
||||
private String mSentenceSeparators;
|
||||
private String mMagicSpaceStrippers;
|
||||
private String mMagicSpaceSwappers;
|
||||
private String mSuggestPuncs;
|
||||
// TODO: Move this flag to VoiceProxy
|
||||
private boolean mConfigurationChanging;
|
||||
|
@ -505,8 +506,14 @@ public class LatinIME extends InputMethodServiceCompatWrapper implements Keyboar
|
|||
mSuggest.setUserBigramDictionary(mUserBigramDictionary);
|
||||
|
||||
updateCorrectionMode();
|
||||
mWordSeparators = res.getString(R.string.word_separators);
|
||||
mSentenceSeparators = res.getString(R.string.sentence_separators);
|
||||
mMagicSpaceStrippers = res.getString(R.string.magic_space_stripping_symbols);
|
||||
mMagicSpaceSwappers = res.getString(R.string.magic_space_swapping_symbols);
|
||||
String wordSeparators = mMagicSpaceStrippers + mMagicSpaceSwappers
|
||||
+ res.getString(R.string.magic_space_promoting_symbols);
|
||||
final String notWordSeparators = res.getString(R.string.non_word_separator_symbols);
|
||||
for (int i = notWordSeparators.length() - 1; i >= 0; --i)
|
||||
wordSeparators = wordSeparators.replace(notWordSeparators.substring(i, i + 1), "");
|
||||
mWordSeparators = wordSeparators;
|
||||
|
||||
Utils.setSystemLocale(res, savedLocale);
|
||||
}
|
||||
|
@ -1039,13 +1046,13 @@ public class LatinIME extends InputMethodServiceCompatWrapper implements Keyboar
|
|||
return false;
|
||||
}
|
||||
|
||||
private void swapPunctuationAndSpace() {
|
||||
private void swapSwapperAndSpace() {
|
||||
final InputConnection ic = getCurrentInputConnection();
|
||||
if (ic == null) return;
|
||||
CharSequence lastTwo = ic.getTextBeforeCursor(2, 0);
|
||||
// It is guaranteed lastTwo.charAt(1) is a swapper - else this method is not called.
|
||||
if (lastTwo != null && lastTwo.length() == 2
|
||||
&& lastTwo.charAt(0) == Keyboard.CODE_SPACE
|
||||
&& isSentenceSeparator(lastTwo.charAt(1))) {
|
||||
&& lastTwo.charAt(0) == Keyboard.CODE_SPACE) {
|
||||
ic.beginBatchEdit();
|
||||
ic.deleteSurroundingText(2, 0);
|
||||
ic.commitText(lastTwo.charAt(1) + " ", 1);
|
||||
|
@ -1054,7 +1061,7 @@ public class LatinIME extends InputMethodServiceCompatWrapper implements Keyboar
|
|||
}
|
||||
}
|
||||
|
||||
private void doubleSpace() {
|
||||
private void maybeDoubleSpace() {
|
||||
if (mCorrectionMode == Suggest.CORRECTION_NONE) return;
|
||||
final InputConnection ic = getCurrentInputConnection();
|
||||
if (ic == null) return;
|
||||
|
@ -1330,14 +1337,11 @@ public class LatinIME extends InputMethodServiceCompatWrapper implements Keyboar
|
|||
private void handleCharacter(int primaryCode, int[] keyCodes, int x, int y) {
|
||||
mVoiceProxy.handleCharacter();
|
||||
|
||||
if (mJustAddedMagicSpace && primaryCode == Keyboard.CODE_SINGLE_QUOTE) {
|
||||
if (mJustAddedMagicSpace && isMagicSpaceStripper(primaryCode)) {
|
||||
removeTrailingSpace();
|
||||
}
|
||||
if (primaryCode != Keyboard.CODE_ENTER) {
|
||||
mJustAddedMagicSpace = false;
|
||||
}
|
||||
|
||||
if (mLastSelectionStart == mLastSelectionEnd && TextEntryState.isRecorrecting()) {
|
||||
if (mLastSelectionStart == mLastSelectionEnd) {
|
||||
abortRecorrection(false);
|
||||
}
|
||||
|
||||
|
@ -1389,6 +1393,12 @@ public class LatinIME extends InputMethodServiceCompatWrapper implements Keyboar
|
|||
} else {
|
||||
sendKeyChar((char)code);
|
||||
}
|
||||
if (mJustAddedMagicSpace && isMagicSpaceSwapper(primaryCode)) {
|
||||
swapSwapperAndSpace();
|
||||
} else {
|
||||
mJustAddedMagicSpace = false;
|
||||
}
|
||||
|
||||
switcher.updateShiftState();
|
||||
if (LatinIME.PERF_DEBUG) measureCps();
|
||||
TextEntryState.typedCharacter((char) code, isWordSeparator(code), x, y);
|
||||
|
@ -1422,27 +1432,25 @@ public class LatinIME extends InputMethodServiceCompatWrapper implements Keyboar
|
|||
}
|
||||
}
|
||||
|
||||
if (mJustAddedMagicSpace && primaryCode == Keyboard.CODE_SPACE) {
|
||||
mJustAddedMagicSpace = false;
|
||||
} else if (mJustAddedMagicSpace && primaryCode == Keyboard.CODE_ENTER) {
|
||||
removeTrailingSpace();
|
||||
mJustAddedMagicSpace = false;
|
||||
sendKeyChar((char)primaryCode);
|
||||
if (mJustAddedMagicSpace) {
|
||||
if (isMagicSpaceSwapper(primaryCode)) {
|
||||
sendKeyChar((char)primaryCode);
|
||||
swapSwapperAndSpace();
|
||||
} else {
|
||||
if (isMagicSpaceStripper(primaryCode)) removeTrailingSpace();
|
||||
sendKeyChar((char)primaryCode);
|
||||
mJustAddedMagicSpace = false;
|
||||
}
|
||||
} else {
|
||||
sendKeyChar((char)primaryCode);
|
||||
}
|
||||
|
||||
if (isSuggestionsRequested() && primaryCode == Keyboard.CODE_SPACE) {
|
||||
maybeDoubleSpace();
|
||||
}
|
||||
|
||||
TextEntryState.typedCharacter((char) primaryCode, true, x, y);
|
||||
|
||||
if (TextEntryState.isPunctuationAfterAccepted() && primaryCode != Keyboard.CODE_ENTER
|
||||
&& mJustAddedMagicSpace) {
|
||||
swapPunctuationAndSpace();
|
||||
} else if (isSuggestionsRequested() && primaryCode == Keyboard.CODE_SPACE) {
|
||||
doubleSpace();
|
||||
mJustAddedMagicSpace = false;
|
||||
} else {
|
||||
mJustAddedMagicSpace = false;
|
||||
}
|
||||
if (pickedDefault) {
|
||||
CharSequence typedWord = mWord.getTypedWord();
|
||||
TextEntryState.backToAcceptedDefault(typedWord);
|
||||
|
@ -1687,10 +1695,17 @@ public class LatinIME extends InputMethodServiceCompatWrapper implements Keyboar
|
|||
// So, LatinImeLogger logs "" as a user's input.
|
||||
LatinImeLogger.logOnManualSuggestion(
|
||||
"", suggestion.toString(), index, suggestions.mWords);
|
||||
// Find out whether the previous character is a space. If it is, as a special case
|
||||
// for punctuation entered through the suggestion strip, it should be considered
|
||||
// a magic space even if it was a normal space. This is meant to help in case the user
|
||||
// pressed space on purpose of displaying the suggestion strip punctuation.
|
||||
final char primaryCode = suggestion.charAt(0);
|
||||
final int toLeft = (ic == null) ? 0 : ic.getTextBeforeCursor(1, 0).charAt(0);
|
||||
if (Keyboard.CODE_SPACE == toLeft) mJustAddedMagicSpace = true;
|
||||
onCodeInput(primaryCode, new int[] { primaryCode },
|
||||
KeyboardActionListener.NOT_A_TOUCH_COORDINATE,
|
||||
KeyboardActionListener.NOT_A_TOUCH_COORDINATE);
|
||||
mJustAddedMagicSpace = false;
|
||||
if (ic != null) {
|
||||
ic.endBatchEdit();
|
||||
}
|
||||
|
@ -1917,8 +1932,11 @@ public class LatinIME extends InputMethodServiceCompatWrapper implements Keyboar
|
|||
}
|
||||
|
||||
if (mUserBigramDictionary != null) {
|
||||
// We don't want to register as bigrams words separated by a separator.
|
||||
// For example "I will, and you too" : we don't want the pair ("will" "and") to be
|
||||
// a bigram.
|
||||
CharSequence prevWord = EditingUtils.getPreviousWord(getCurrentInputConnection(),
|
||||
mSentenceSeparators);
|
||||
mWordSeparators);
|
||||
if (!TextUtils.isEmpty(prevWord)) {
|
||||
mUserBigramDictionary.addBigrams(prevWord.toString(), suggestion.toString());
|
||||
}
|
||||
|
@ -1993,8 +2011,12 @@ public class LatinIME extends InputMethodServiceCompatWrapper implements Keyboar
|
|||
return separators.contains(String.valueOf((char)code));
|
||||
}
|
||||
|
||||
private boolean isSentenceSeparator(int code) {
|
||||
return mSentenceSeparators.contains(String.valueOf((char)code));
|
||||
private boolean isMagicSpaceStripper(int code) {
|
||||
return mMagicSpaceStrippers.contains(String.valueOf((char)code));
|
||||
}
|
||||
|
||||
private boolean isMagicSpaceSwapper(int code) {
|
||||
return mMagicSpaceSwappers.contains(String.valueOf((char)code));
|
||||
}
|
||||
|
||||
private void sendMagicSpace() {
|
||||
|
|
Loading…
Reference in New Issue