nyancrimew
/
LatinIME
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Some refactoring 

Getting the frequency of a terminal is not very useful, however
getting its position will be very useful for retrieving bigrams
later.
Moreover, from the position it's easy to find out the frequency.

Bug: 5046459
Change-Id: Ica53472c2038c7e407dbd1399d336511c731087f
main
Jean Chalard 2011-07-22 18:18:15 +09:00
parent 7139335dcd
commit 848b69a5f9
1 changed files with 6 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
native/src/unigram_dictionary.cpp
View File

@ -1055,9 +1055,9 @@ int UnigramDictionary::getMostFrequentWordLikeInner(const uint16_t * const inWor
return maxFreq; return maxFreq;
} }
// This function gets the frequency of the exact matching word in the dictionary. // This function gets the byte position of the last chargroup of the exact matching word in the
// If no match is found, it returns NOT_VALID_WORD. // dictionary. If no match is found, it returns NOT_VALID_WORD.
static inline int getFrequency(const uint8_t* const root, const uint16_t* const inWord, static inline int getTerminalPosition(const uint8_t* const root, const uint16_t* const inWord,
const int length) { const int length) {
int pos = 0; int pos = 0;
int wordPos = 0; int wordPos = 0;
@ -1072,6 +1072,7 @@ static inline int getFrequency(const uint8_t* const root, const uint16_t* const
// If there are no more character groups in this node, it means we could not // If there are no more character groups in this node, it means we could not
// find a matching character for this depth, therefore there is no match. // find a matching character for this depth, therefore there is no match.
if (0 >= charGroupCount) return NOT_VALID_WORD; if (0 >= charGroupCount) return NOT_VALID_WORD;
const int charGroupPos = pos;
const uint8_t flags = BinaryFormat::getFlagsAndForwardPointer(root, &pos); const uint8_t flags = BinaryFormat::getFlagsAndForwardPointer(root, &pos);
int32_t character = BinaryFormat::getCharCodeAndForwardPointer(root, &pos); int32_t character = BinaryFormat::getCharCodeAndForwardPointer(root, &pos);
if (character == wChar) { if (character == wChar) {
@ -1099,7 +1100,7 @@ static inline int getFrequency(const uint8_t* const root, const uint16_t* const
++wordPos; ++wordPos;
if (UnigramDictionary::FLAG_IS_TERMINAL & flags) { if (UnigramDictionary::FLAG_IS_TERMINAL & flags) {
if (wordPos == length) { if (wordPos == length) {
return BinaryFormat::readFrequencyWithoutMovingPointer(root, pos); return charGroupPos;
} }
pos = BinaryFormat::skipFrequency(UnigramDictionary::FLAG_IS_TERMINAL, pos); pos = BinaryFormat::skipFrequency(UnigramDictionary::FLAG_IS_TERMINAL, pos);
} }
@ -1126,7 +1127,7 @@ static inline int getFrequency(const uint8_t* const root, const uint16_t* const
} }
bool UnigramDictionary::isValidWord(const uint16_t* const inWord, const int length) const { bool UnigramDictionary::isValidWord(const uint16_t* const inWord, const int length) const {
return NOT_VALID_WORD != getFrequency(DICT_ROOT, inWord, length); return NOT_VALID_WORD != getTerminalPosition(DICT_ROOT, inWord, length);
} }
int UnigramDictionary::getBigrams(unsigned short *word, int length, int *codes, int codesSize, int UnigramDictionary::getBigrams(unsigned short *word, int length, int *codes, int codesSize,