diff options
6 files changed, 40 insertions, 29 deletions
diff --git a/native/jni/src/suggest/core/dictionary/error_type_utils.cpp b/native/jni/src/suggest/core/dictionary/error_type_utils.cpp index b6bf7a98c..1e2494e92 100644 --- a/native/jni/src/suggest/core/dictionary/error_type_utils.cpp +++ b/native/jni/src/suggest/core/dictionary/error_type_utils.cpp @@ -19,17 +19,18 @@ namespace latinime { const ErrorTypeUtils::ErrorType ErrorTypeUtils::NOT_AN_ERROR = 0x0; -const ErrorTypeUtils::ErrorType ErrorTypeUtils::MATCH_WITH_CASE_ERROR = 0x1; -const ErrorTypeUtils::ErrorType ErrorTypeUtils::MATCH_WITH_ACCENT_ERROR = 0x2; -const ErrorTypeUtils::ErrorType ErrorTypeUtils::MATCH_WITH_DIGRAPH = 0x4; -const ErrorTypeUtils::ErrorType ErrorTypeUtils::INTENTIONAL_OMISSION = 0x8; -const ErrorTypeUtils::ErrorType ErrorTypeUtils::EDIT_CORRECTION = 0x10; -const ErrorTypeUtils::ErrorType ErrorTypeUtils::PROXIMITY_CORRECTION = 0x20; -const ErrorTypeUtils::ErrorType ErrorTypeUtils::COMPLETION = 0x40; -const ErrorTypeUtils::ErrorType ErrorTypeUtils::NEW_WORD = 0x80; +const ErrorTypeUtils::ErrorType ErrorTypeUtils::MATCH_WITH_WRONG_CASE = 0x1; +const ErrorTypeUtils::ErrorType ErrorTypeUtils::MATCH_WITH_MISSING_ACCENT = 0x2; +const ErrorTypeUtils::ErrorType ErrorTypeUtils::MATCH_WITH_WRONG_ACCENT = 0x4; +const ErrorTypeUtils::ErrorType ErrorTypeUtils::MATCH_WITH_DIGRAPH = 0x8; +const ErrorTypeUtils::ErrorType ErrorTypeUtils::INTENTIONAL_OMISSION = 0x10; +const ErrorTypeUtils::ErrorType ErrorTypeUtils::EDIT_CORRECTION = 0x20; +const ErrorTypeUtils::ErrorType ErrorTypeUtils::PROXIMITY_CORRECTION = 0x40; +const ErrorTypeUtils::ErrorType ErrorTypeUtils::COMPLETION = 0x80; +const ErrorTypeUtils::ErrorType ErrorTypeUtils::NEW_WORD = 0x100; const ErrorTypeUtils::ErrorType ErrorTypeUtils::ERRORS_TREATED_AS_AN_EXACT_MATCH = - NOT_AN_ERROR | MATCH_WITH_CASE_ERROR | MATCH_WITH_ACCENT_ERROR | MATCH_WITH_DIGRAPH; + NOT_AN_ERROR | MATCH_WITH_WRONG_CASE | MATCH_WITH_MISSING_ACCENT | MATCH_WITH_DIGRAPH; const ErrorTypeUtils::ErrorType ErrorTypeUtils::ERRORS_TREATED_AS_AN_EXACT_MATCH_WITH_INTENTIONAL_OMISSION = diff --git a/native/jni/src/suggest/core/dictionary/error_type_utils.h b/native/jni/src/suggest/core/dictionary/error_type_utils.h index e3e76b238..fd1d5fcff 100644 --- a/native/jni/src/suggest/core/dictionary/error_type_utils.h +++ b/native/jni/src/suggest/core/dictionary/error_type_utils.h @@ -30,8 +30,9 @@ class ErrorTypeUtils { typedef uint32_t ErrorType; static const ErrorType NOT_AN_ERROR; - static const ErrorType MATCH_WITH_CASE_ERROR; - static const ErrorType MATCH_WITH_ACCENT_ERROR; + static const ErrorType MATCH_WITH_WRONG_CASE; + static const ErrorType MATCH_WITH_MISSING_ACCENT; + static const ErrorType MATCH_WITH_WRONG_ACCENT; static const ErrorType MATCH_WITH_DIGRAPH; // Treat error as an intentional omission when the CorrectionType is omission and the node can // be intentional omission. diff --git a/native/jni/src/suggest/policyimpl/dictionary/structure/backward/v402/ver4_patricia_trie_policy.cpp b/native/jni/src/suggest/policyimpl/dictionary/structure/backward/v402/ver4_patricia_trie_policy.cpp index e3f14e687..9c6452e40 100644 --- a/native/jni/src/suggest/policyimpl/dictionary/structure/backward/v402/ver4_patricia_trie_policy.cpp +++ b/native/jni/src/suggest/policyimpl/dictionary/structure/backward/v402/ver4_patricia_trie_policy.cpp @@ -210,7 +210,7 @@ bool Ver4PatriciaTriePolicy::addUnigramEntry(const int *const word, const int le } for (const auto &shortcut : unigramProperty->getShortcuts()) { if (shortcut.getTargetCodePoints()->size() > MAX_WORD_LENGTH) { - AKLOGE("One of shortcut targets is too long to insert to the dictionary, length: %d", + AKLOGE("One of shortcut targets is too long to insert to the dictionary, length: %zd", shortcut.getTargetCodePoints()->size()); return false; } @@ -245,7 +245,7 @@ bool Ver4PatriciaTriePolicy::addUnigramEntry(const int *const word, const int le if (!mUpdatingHelper.addShortcutTarget(wordPos, shortcut.getTargetCodePoints()->data(), shortcut.getTargetCodePoints()->size(), shortcut.getProbability())) { - AKLOGE("Cannot add new shortcut target. PtNodePos: %d, length: %d, " + AKLOGE("Cannot add new shortcut target. PtNodePos: %d, length: %zd, " "probability: %d", wordPos, shortcut.getTargetCodePoints()->size(), shortcut.getProbability()); return false; @@ -289,7 +289,7 @@ bool Ver4PatriciaTriePolicy::addNgramEntry(const PrevWordsInfo *const prevWordsI } if (bigramProperty->getTargetCodePoints()->size() > MAX_WORD_LENGTH) { AKLOGE("The word is too long to insert the ngram to the dictionary. " - "length: %d", bigramProperty->getTargetCodePoints()->size()); + "length: %zd", bigramProperty->getTargetCodePoints()->size()); return false; } int prevWordsPtNodePos[MAX_PREV_WORD_COUNT_FOR_N_GRAM]; diff --git a/native/jni/src/suggest/policyimpl/dictionary/structure/v4/ver4_patricia_trie_policy.cpp b/native/jni/src/suggest/policyimpl/dictionary/structure/v4/ver4_patricia_trie_policy.cpp index 723808399..d8f4595ff 100644 --- a/native/jni/src/suggest/policyimpl/dictionary/structure/v4/ver4_patricia_trie_policy.cpp +++ b/native/jni/src/suggest/policyimpl/dictionary/structure/v4/ver4_patricia_trie_policy.cpp @@ -200,7 +200,7 @@ bool Ver4PatriciaTriePolicy::addUnigramEntry(const int *const word, const int le } for (const auto &shortcut : unigramProperty->getShortcuts()) { if (shortcut.getTargetCodePoints()->size() > MAX_WORD_LENGTH) { - AKLOGE("One of shortcut targets is too long to insert to the dictionary, length: %d", + AKLOGE("One of shortcut targets is too long to insert to the dictionary, length: %zd", shortcut.getTargetCodePoints()->size()); return false; } @@ -235,7 +235,7 @@ bool Ver4PatriciaTriePolicy::addUnigramEntry(const int *const word, const int le if (!mUpdatingHelper.addShortcutTarget(wordPos, shortcut.getTargetCodePoints()->data(), shortcut.getTargetCodePoints()->size(), shortcut.getProbability())) { - AKLOGE("Cannot add new shortcut target. PtNodePos: %d, length: %d, " + AKLOGE("Cannot add new shortcut target. PtNodePos: %d, length: %zd, " "probability: %d", wordPos, shortcut.getTargetCodePoints()->size(), shortcut.getProbability()); return false; @@ -286,7 +286,7 @@ bool Ver4PatriciaTriePolicy::addNgramEntry(const PrevWordsInfo *const prevWordsI } if (bigramProperty->getTargetCodePoints()->size() > MAX_WORD_LENGTH) { AKLOGE("The word is too long to insert the ngram to the dictionary. " - "length: %d", bigramProperty->getTargetCodePoints()->size()); + "length: %zd", bigramProperty->getTargetCodePoints()->size()); return false; } int prevWordsPtNodePos[MAX_PREV_WORD_COUNT_FOR_N_GRAM]; diff --git a/native/jni/src/suggest/policyimpl/typing/typing_scoring.h b/native/jni/src/suggest/policyimpl/typing/typing_scoring.h index 04cb6603a..52c4251f0 100644 --- a/native/jni/src/suggest/policyimpl/typing/typing_scoring.h +++ b/native/jni/src/suggest/policyimpl/typing/typing_scoring.h @@ -51,10 +51,10 @@ class TypingScoring : public Scoring { } if (boostExactMatches && ErrorTypeUtils::isExactMatch(containedErrorTypes)) { score += ScoringParams::EXACT_MATCH_PROMOTION; - if ((ErrorTypeUtils::MATCH_WITH_CASE_ERROR & containedErrorTypes) != 0) { + if ((ErrorTypeUtils::MATCH_WITH_WRONG_CASE & containedErrorTypes) != 0) { score -= ScoringParams::CASE_ERROR_PENALTY_FOR_EXACT_MATCH; } - if ((ErrorTypeUtils::MATCH_WITH_ACCENT_ERROR & containedErrorTypes) != 0) { + if ((ErrorTypeUtils::MATCH_WITH_MISSING_ACCENT & containedErrorTypes) != 0) { score -= ScoringParams::ACCENT_ERROR_PENALTY_FOR_EXACT_MATCH; } if ((ErrorTypeUtils::MATCH_WITH_DIGRAPH & containedErrorTypes) != 0) { diff --git a/native/jni/src/suggest/policyimpl/typing/typing_weighting.cpp b/native/jni/src/suggest/policyimpl/typing/typing_weighting.cpp index 54f65c786..1d590c353 100644 --- a/native/jni/src/suggest/policyimpl/typing/typing_weighting.cpp +++ b/native/jni/src/suggest/policyimpl/typing/typing_weighting.cpp @@ -36,25 +36,34 @@ ErrorTypeUtils::ErrorType TypingWeighting::getErrorType(const CorrectionType cor // Compare the node code point with original primary code point on the keyboard. const ProximityInfoState *const pInfoState = traverseSession->getProximityInfoState(0); - const int primaryOriginalCodePoint = pInfoState->getPrimaryOriginalCodePointAt( + const int primaryCodePoint = pInfoState->getPrimaryCodePointAt( dicNode->getInputIndex(0)); const int nodeCodePoint = dicNode->getNodeCodePoint(); - if (primaryOriginalCodePoint == nodeCodePoint) { + // TODO: Check whether the input code point is on the keyboard. + if (primaryCodePoint == nodeCodePoint) { // Node code point is same as original code point on the keyboard. return ErrorTypeUtils::NOT_AN_ERROR; - } else if (CharUtils::toLowerCase(primaryOriginalCodePoint) == + } else if (CharUtils::toLowerCase(primaryCodePoint) == CharUtils::toLowerCase(nodeCodePoint)) { // Only cases of the code points are different. - return ErrorTypeUtils::MATCH_WITH_CASE_ERROR; - } else if (CharUtils::toBaseCodePoint(primaryOriginalCodePoint) == - CharUtils::toBaseCodePoint(nodeCodePoint)) { + return ErrorTypeUtils::MATCH_WITH_WRONG_CASE; + } else if (primaryCodePoint == CharUtils::toBaseCodePoint(nodeCodePoint)) { // Node code point is a variant of original code point. - return ErrorTypeUtils::MATCH_WITH_ACCENT_ERROR; - } else { + return ErrorTypeUtils::MATCH_WITH_MISSING_ACCENT; + } else if (CharUtils::toBaseCodePoint(primaryCodePoint) + == CharUtils::toBaseCodePoint(nodeCodePoint)) { + // Base code points are the same but the code point is intentionally input. + return ErrorTypeUtils::MATCH_WITH_WRONG_ACCENT; + } else if (CharUtils::toLowerCase(primaryCodePoint) + == CharUtils::toBaseLowerCase(nodeCodePoint)) { // Node code point is a variant of original code point and the cases are also // different. - return ErrorTypeUtils::MATCH_WITH_ACCENT_ERROR - | ErrorTypeUtils::MATCH_WITH_CASE_ERROR; + return ErrorTypeUtils::MATCH_WITH_MISSING_ACCENT + | ErrorTypeUtils::MATCH_WITH_WRONG_CASE; + } else { + // Base code points are the same and the cases are different. + return ErrorTypeUtils::MATCH_WITH_WRONG_ACCENT + | ErrorTypeUtils::MATCH_WITH_WRONG_CASE; } } break; |