public static Attribute Create(String natureWithFrequency)
{
try {
string[] param = natureWithFrequency.Split(' ');
int natureCount = param.Length / 2;
Attribute attribute = new Attribute(natureCount);
for (int i = 0; i < natureCount; ++i)
{
attribute.nature[i] = LexiconUtility.convertStringToNature(param[2 * i], null);
attribute.frequency[i] = int.Parse(param[1 + 2 * i]);
attribute.totalFrequency += attribute.frequency[i];
}
return(attribute);
}
catch (Exception e) {
Predefine.logger.Warn("使用字符串" + natureWithFrequency + "创建词条属性失败！" + e.Message);
return(null);
}
}
/**
* 获取本词语在HanLP词库中的频次
* @return 频次，0代表这是个OOV
*/
public int getFrequency()
{
return(LexiconUtility.getFrequency(word));
}
private static bool loadMainDictionary(String mainPath)
{
logger.info("自定义词典开始加载:" + mainPath);
if (loadDat(mainPath))
{
return(true);
}
Dictionary <String, CoreDictionary.Attribute> map = new Dictionary <string, CoreDictionary.Attribute>();
HashSet <Nature> customNatureCollector = new HashSet <Nature>();
try
{
foreach (String p in path)
{
Nature defaultNature = Nature.n;
int cut = p.IndexOf(' ');
if (cut > 0)
{
// 有默认词性
String nature = p.Substring(cut + 1);
p = p.Substring(0, cut);
try
{
defaultNature = LexiconUtility.convertStringToNature(nature, customNatureCollector);
}
catch (Exception e)
{
logger.severe("配置文件【" + p + "】写错了！" + e);
continue;
}
}
logger.info("以默认词性[" + defaultNature + "]加载自定义词典" + p + "中……");
bool success = load(p, defaultNature, map, customNatureCollector);
if (!success)
{
logger.warning("失败：" + p);
}
}
if (map.Count == 0)
{
logger.warning("没有加载到任何词条");
map.put(Predefine.TAG_OTHER, null); // 当作空白占位符
}
logger.info("正在构建DoubleArrayTrie……");
dat.build(map);
// 缓存成dat文件，下次加载会快很多
logger.info("正在缓存词典为dat文件……");
// 缓存值文件
List <CoreDictionary.Attribute> attributeList = new List <CoreDictionary.Attribute>();
foreach (KeyValuePair <String, CoreDictionary.Attribute> entry in map)
{
attributeList.Add(entry.Value);
}
DataOutputStream out = new DataOutputStream(new FileOutputStream(mainPath + Predefine.BIN_EXT));
// 缓存用户词性
IOUtil.writeCustomNature(out, customNatureCollector);
// 缓存正文
out.writeInt(attributeList.size());
for (CoreDictionary.Attribute attribute : attributeList)
{
out.writeInt(attribute.totalFrequency);
out.writeInt(attribute.nature.length);
