/* 向词的cache中插入新的词。
* 参数：
* hz 词
* length 长度
* 返回：无
*/
void InsertCiToCache(HZ *hz, int length, int syllable_length, int set_fix_top)
{
int pos; //Cache中的位置
int item_length; //本词条的长度
int used_count; //词汇使用度
int i;
char item_save[WORDLIB_FEATURE_LENGTH + sizeof(HZ) * MAX_WORD_LENGTH]; //临时保存区
//如果以当前的位置确定词的顺序，则不能向词频中增加内容
if (pim_config->ci_option & CI_ADJUST_FREQ_NONE)
return;
if (length > MAX_WORD_LENGTH)
return;
//UCS4的词不加入Cache
if (length != syllable_length)
return;
item_length = length * sizeof(HZ) + WORDLIB_FEATURE_LENGTH;
//以下这个词是否出现在Cache中?
pos = GetCiCacheInfo(hz, length);
if (pos != -1) //找到
{
used_count = *(int*)&share_segment->ci_cache.cache[pos] >> 8; //找出词的使用度
used_count++;
}
/* 完整处理音节到汉字候选。
* 处理过程：
* 1. 检索汉字候选
* 2. 汉字候选排重
* 3. 候选排序
* 本函数将依照系统配置与IC配置中的当前模糊音、检索结果集以及输出集合
* 设置进行处理。
*
* 参数：
* syllable 音节
* *candidate_array 候选数组
* array_length 候选数组长度
* 返回值：
* 返回候选数目
*/
int ProcessZiCandidates(SYLLABLE syllable, CANDIDATE *candidate_array, int array_length, int zi_level)
{
int zi_count = 0, normal_zi_count = 0, small_count = 0, small_ci_count = 0, small_zi_count = 0;
SYLLABLE small_syllables[2];
CANDIDATE small_word_candidates[0x10];
//寻找类似xian输入西安的候选，双拼不用考虑
if (/*pim_config->special_parse_pin_yin && */pim_config->pinyin_mode != PINYIN_SHUANGPIN && GetSmallSyllables(syllable, small_syllables))
{
int i;
small_ci_count = ProcessCiCandidate(small_syllables,
2,
0,
small_word_candidates,
_SizeOf(small_word_candidates),
1);
for (i = 0; i < small_ci_count && i < array_length; i++)
{
candidate_array[i].type = CAND_TYPE_ZI;
candidate_array[i].hz.is_word = 1;
candidate_array[i].hz.word_item = small_word_candidates[i].word.item;
candidate_array[i].hz.origin_syllable = syllable;
candidate_array[i].hz.top_pos = DEFAULT_TOP_POS;
candidate_array[i].hz.cache_pos =
GetCiCacheInfo(GetItemHZPtr(candidate_array[i].hz.word_item), candidate_array[i].hz.word_item->ci_length);
}
//检索小音节拆分字
small_zi_count = GetZiCandidates(small_syllables[0],
candidate_array + small_ci_count,
array_length - small_ci_count,
pim_config->use_fuzzy ? pim_config->fuzzy_mode : 0,
zi_level,
pim_config->hz_output_mode);
//如果没有找到汉字，如eng，则必须扩大汉字的范围进行查找
if (!small_zi_count)
small_zi_count = GetZiCandidates(small_syllables[0],
candidate_array + small_ci_count,
array_length - small_ci_count,
pim_config->use_fuzzy ? pim_config->fuzzy_mode : 0,
HZ_ALL_USED,
HZ_OUTPUT_HANZI_ALL);
for (i = small_ci_count; i < small_ci_count + small_zi_count; i++)
{
candidate_array[i].hz.hz_type = ZI_TYPE_OTHER;
}
small_count = small_ci_count + small_zi_count;
}
//检索普通字
normal_zi_count = GetZiCandidates(syllable,
candidate_array + small_count,
array_length - small_count,
pim_config->use_fuzzy ? pim_config->fuzzy_mode : 0,
zi_level,
pim_config->hz_output_mode);
//如果没有找到汉字，如eng，则必须扩大汉字的范围进行查找
if (!normal_zi_count)
normal_zi_count = GetZiCandidates(syllable,
candidate_array + small_count,
array_length - small_count,
pim_config->use_fuzzy ? pim_config->fuzzy_mode : 0,
HZ_ALL_USED,
HZ_OUTPUT_HANZI_ALL);
zi_count = small_zi_count + normal_zi_count;
//去掉无法显示的汉字
zi_count = DeleteUnreadableZiCandidates(candidate_array + small_ci_count, zi_count);
//排重(只针对字)
zi_count = UnifyZiCandidates(candidate_array + small_ci_count, zi_count);
//排序(字和小音节词都要排)
SortZiCandidates(candidate_array, small_ci_count + zi_count);
return small_ci_count + zi_count;
}
