private void checkDictionary(
final FusionDictionary dict,
final List<String> words,
final SparseArray<List<Integer>> bigrams,
final HashMap<String, List<String>> shortcutMap) {
assertNotNull(dict);
// check unigram
for (final String word : words) {
final PtNode ptNode = FusionDictionary.findWordInTree(dict.mRootNodeArray, word);
assertNotNull(ptNode);
}
// check bigram
for (int i = 0; i < bigrams.size(); ++i) {
final int w1 = bigrams.keyAt(i);
for (final int w2 : bigrams.valueAt(i)) {
final PtNode ptNode = FusionDictionary.findWordInTree(dict.mRootNodeArray, words.get(w1));
assertNotNull(words.get(w1) + "," + words.get(w2), ptNode.getBigram(words.get(w2)));
}
}
// check shortcut
if (shortcutMap != null) {
for (final Entry<String, List<String>> entry : shortcutMap.entrySet()) {
assertTrue(words.contains(entry.getKey()));
final PtNode ptNode = FusionDictionary.findWordInTree(dict.mRootNodeArray, entry.getKey());
for (final String word : entry.getValue()) {
assertNotNull(
"shortcut not found: " + entry.getKey() + ", " + word, ptNode.getShortcut(word));
}
}
}
}
private void addBigrams(
final FusionDictionary dict,
final List<String> words,
final SparseArray<List<Integer>> bigrams) {
for (int i = 0; i < bigrams.size(); ++i) {
final int w1 = bigrams.keyAt(i);
for (int w2 : bigrams.valueAt(i)) {
dict.setBigram(words.get(w1), words.get(w2), new ProbabilityInfo(BIGRAM_FREQ));
}
}
}
/** Adds unigrams to the dictionary. */
private void addUnigrams(
final int number,
final FusionDictionary dict,
final List<String> words,
final HashMap<String, List<String>> shortcutMap) {
for (int i = 0; i < number; ++i) {
final String word = words.get(i);
final ArrayList<WeightedString> shortcuts = new ArrayList<>();
if (shortcutMap != null && shortcutMap.containsKey(word)) {
for (final String shortcut : shortcutMap.get(word)) {
shortcuts.add(new WeightedString(shortcut, UNIGRAM_FREQ));
}
}
dict.add(
word,
new ProbabilityInfo(UNIGRAM_FREQ),
(shortcutMap == null) ? null : shortcuts,
false /* isNotAWord */);
}
}