bugfix: 修复 bintrie 树全分词时 提前跳出循环 bug

src/main/java/com/hankcs/hanlp/collection/trie/bintrie/BinTrie.java

@@ -608,6 +608,14 @@ public void parseText(String text, AhoCorasickDoubleArrayTrie.IHit<V> processor)
                 {
                     processor.hit(begin, i + 1, value);
                 }
+
+                /*如果是最后一位，这里不能直接跳出循环， 要继续从下一个字符开始判断*/
+                if (i == length - 1)
+                {
+                    i = begin;
+                    ++begin;
+                    state = this;
+                }
             }
             else
             {
@@ -640,6 +648,14 @@ public void parseText(char[] text, AhoCorasickDoubleArrayTrie.IHit<V> processor)
                 {
                     processor.hit(begin, i + 1, value);
                 }
+
+                /*如果是最后一位，这里不能直接跳出循环， 要继续从下一个字符开始判断*/
+                if (i == length - 1)
+                {
+                    i = begin;
+                    ++begin;
+                    state = this;
+                }
             }
             else
             {

src/test/java/com/hankcs/hanlp/collection/trie/bintrie/BinTrieParseTextTest.java

@@ -0,0 +1,56 @@
+package com.hankcs.hanlp.collection.trie.bintrie;
+
+import com.hankcs.hanlp.collection.AhoCorasick.AhoCorasickDoubleArrayTrie;
+import org.junit.Assert;
+import org.junit.Before;
+import org.junit.Test;
+
+import java.util.HashSet;
+import java.util.Set;
+
+public class BinTrieParseTextTest {
+
+
+    private final String[] words = new String[]{"溜", "儿", "溜儿", "一溜儿", "一溜"};
+    private BinTrie<Integer> trie;
+
+    @Before
+    public void setup() {
+        this.trie = new BinTrie<Integer>();
+        /*构建一个简单的词典， 从 core dict 文件中扣出的一部分*/
+        for (int i = 0; i < words.length; i++) {
+            this.trie.put(words[i], i);
+        }
+    }
+
+
+    @Test
+    public void testFullParse() {
+        assertFullParse("一溜儿");
+        assertFullParse("一溜儿    ");
+        assertFullParse("一溜儿 ");
+    }
+
+    private void assertFullParse(String text) {
+        Set<String> result = parseText(text);
+        /*确保每个词都被分出来了*/
+        for (String word : words) {
+            Assert.assertTrue(result.contains(word));
+        }
+    }
+
+
+    private Set<String> parseText(final String text) {
+        final Set<String> result = new HashSet<String>(words.length);
+        trie.parseText(text, new AhoCorasickDoubleArrayTrie.IHit<Integer>() {
+            @Override
+            public void hit(int begin, int end, Integer value) {
+                result.add(text.substring(begin, end));
+            }
+        });
+
+        return result;
+    }
+
+
+}