词法分析器seg接口支持自定义词性覆盖统计词性 fix #1156

hankcs · Apr 20, 2019 · 259825c · 259825c
1 parent 6888837
commit 259825c
Show file tree

Hide file tree

Showing 2 changed files with 21 additions and 0 deletions.
diff --git a/src/main/java/com/hankcs/hanlp/tokenizer/lexical/AbstractLexicalAnalyzer.java b/src/main/java/com/hankcs/hanlp/tokenizer/lexical/AbstractLexicalAnalyzer.java
@@ -402,6 +402,16 @@ protected List<Term> segSentence(char[] sentence)
                     {
                         childrenList.add(iterator.next());
                     }
+                    if (attributeList != null)
+                    {
+                        attributeIterator = attributeList.iterator();
+                        for (int i = 0; i < wordArray.length; i++)
+                        {
+                            CoreDictionary.Attribute attribute = attributeIterator.next();
+                            if (attribute != null)
+                                posArray[i] = attribute.nature[0].toString();
+                        }
+                    }
                     String prePos = posArray[0];
                     offset = 0;
 

diff --git a/src/test/java/com/hankcs/hanlp/tokenizer/lexical/AbstractLexicalAnalyzerTest.java b/src/test/java/com/hankcs/hanlp/tokenizer/lexical/AbstractLexicalAnalyzerTest.java
@@ -2,11 +2,13 @@
 
 import com.hankcs.hanlp.HanLP;
 import com.hankcs.hanlp.dictionary.CustomDictionary;
+import com.hankcs.hanlp.model.crf.CRFLexicalAnalyzer;
 import com.hankcs.hanlp.model.perceptron.PerceptronLexicalAnalyzer;
 import com.hankcs.hanlp.seg.Segment;
 import com.hankcs.hanlp.seg.common.Term;
 import junit.framework.TestCase;
 
+import java.io.IOException;
 import java.util.List;
 
 public class AbstractLexicalAnalyzerTest extends TestCase
@@ -34,4 +36,13 @@ public void testCustomDictionary() throws Exception
         CustomDictionary.add("攻城狮");
         System.out.println(analyzer.segment(text));
     }
+
+    public void testOverwriteTag() throws IOException
+    {
+        CRFLexicalAnalyzer analyzer = new CRFLexicalAnalyzer();
+        String text = "强行修改词性";
+        System.out.println(analyzer.seg(text));
+        CustomDictionary.add("修改", "自定义词性");
+        System.out.println(analyzer.seg(text));
+    }
 }