1 files changed, 13 insertions, 10 deletions

diff --git a/native/jni/src/bigram_dictionary.cpp b/native/jni/src/bigram_dictionary.cpp
index c592542bd..b57227880 100644
--- a/native/jni/src/bigram_dictionary.cpp
+++ b/native/jni/src/bigram_dictionary.cpp
@@ -20,15 +20,18 @@
 
 #include "bigram_dictionary.h"
 
-#include "char_utils.h"
 #include "defines.h"
+#include "suggest/core/dictionary/binary_dictionary_info.h"
 #include "suggest/core/dictionary/binary_format.h"
 #include "suggest/core/dictionary/bloom_filter.h"
+#include "suggest/core/dictionary/char_utils.h"
 #include "suggest/core/dictionary/dictionary.h"
+#include "suggest/core/dictionary/probability_utils.h"
 
 namespace latinime {
 
-BigramDictionary::BigramDictionary(const uint8_t *const streamStart) : DICT_ROOT(streamStart) {
+BigramDictionary::BigramDictionary(const BinaryDictionaryInfo *const binaryDictionaryInfo)
+        : mBinaryDictionaryInfo(binaryDictionaryInfo) {
     if (DEBUG_DICT) {
         AKLOGI("BigramDictionary - constructor");
     }
@@ -52,7 +55,7 @@ void BigramDictionary::addWordBigram(int *word, int length, int probability, int
     int insertAt = 0;
     while (insertAt < MAX_RESULTS) {
         if (probability > bigramProbability[insertAt] || (bigramProbability[insertAt] == probability
-                && length < getCodePointCount(MAX_WORD_LENGTH,
+                && length < CharUtils::getCodePointCount(MAX_WORD_LENGTH,
                         bigramCodePoints + insertAt * MAX_WORD_LENGTH))) {
             break;
         }
@@ -103,7 +106,7 @@ int BigramDictionary::getBigrams(const int *prevWord, int prevWordLength, int *i
     // TODO: remove unused arguments, and refrain from storing stuff in members of this class
     // TODO: have "in" arguments before "out" ones, and make out args explicit in the name
 
-    const uint8_t *const root = DICT_ROOT;
+    const uint8_t *const root = mBinaryDictionaryInfo->getDictRoot();
     int pos = getBigramListPositionForWord(prevWord, prevWordLength,
             false /* forceLowerCaseSearch */);
     // getBigramListPositionForWord returns 0 if this word isn't in the dictionary or has no bigrams
@@ -134,7 +137,7 @@ int BigramDictionary::getBigrams(const int *prevWord, int prevWordLength, int *i
             // resulting probability is 8 - although in the practice it's never bigger than 3 or 4
             // in very bad cases. This means that sometimes, we'll see some bigrams interverted
             // here, but it can't get too bad.
-            const int probability = BinaryFormat::computeProbabilityForBigram(
+            const int probability = ProbabilityUtils::computeProbabilityForBigram(
                     unigramProbability, bigramProbabilityTemp);
             addWordBigram(bigramBuffer, length, probability, bigramProbability, bigramCodePoints,
                     outputTypes);
@@ -149,7 +152,7 @@ int BigramDictionary::getBigrams(const int *prevWord, int prevWordLength, int *i
 int BigramDictionary::getBigramListPositionForWord(const int *prevWord, const int prevWordLength,
         const bool forceLowerCaseSearch) const {
     if (0 >= prevWordLength) return 0;
-    const uint8_t *const root = DICT_ROOT;
+    const uint8_t *const root = mBinaryDictionaryInfo->getDictRoot();
     int pos = BinaryFormat::getTerminalPosition(root, prevWord, prevWordLength,
             forceLowerCaseSearch);
 
@@ -170,7 +173,7 @@ int BigramDictionary::getBigramListPositionForWord(const int *prevWord, const in
 void BigramDictionary::fillBigramAddressToProbabilityMapAndFilter(const int *prevWord,
         const int prevWordLength, std::map<int, int> *map, uint8_t *filter) const {
     memset(filter, 0, BIGRAM_FILTER_BYTE_SIZE);
-    const uint8_t *const root = DICT_ROOT;
+    const uint8_t *const root = mBinaryDictionaryInfo->getDictRoot();
     int pos = getBigramListPositionForWord(prevWord, prevWordLength,
             false /* forceLowerCaseSearch */);
     if (0 == pos) {
@@ -196,9 +199,9 @@ bool BigramDictionary::checkFirstCharacter(int *word, int *inputCodePoints) cons
     // what user typed.
 
     int maxAlt = MAX_ALTERNATIVES;
-    const int firstBaseLowerCodePoint = toBaseLowerCase(*word);
+    const int firstBaseLowerCodePoint = CharUtils::toBaseLowerCase(*word);
     while (maxAlt > 0) {
-        if (toBaseLowerCase(*inputCodePoints) == firstBaseLowerCodePoint) {
+        if (CharUtils::toBaseLowerCase(*inputCodePoints) == firstBaseLowerCodePoint) {
             return true;
         }
         inputCodePoints++;
@@ -209,7 +212,7 @@ bool BigramDictionary::checkFirstCharacter(int *word, int *inputCodePoints) cons
 
 bool BigramDictionary::isValidBigram(const int *word1, int length1, const int *word2,
         int length2) const {
-    const uint8_t *const root = DICT_ROOT;
+    const uint8_t *const root = mBinaryDictionaryInfo->getDictRoot();
     int pos = getBigramListPositionForWord(word1, length1, false /* forceLowerCaseSearch */);
     // getBigramListPositionForWord returns 0 if this word isn't in the dictionary or has no bigrams
     if (0 == pos) return false;