fix: 修复因copy_to导致的offset错误

KennFalcon · Nov 29, 2019 · a88a696 · a88a696
1 parent 985b703
commit a88a696
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 1 deletion.
diff --git a/src/main/java/com/hankcs/lucene/HanLPTokenizer.java b/src/main/java/com/hankcs/lucene/HanLPTokenizer.java
@@ -73,6 +73,7 @@ final public boolean incrementToken() throws IOException {
         do {
             term = segment.next();
             if (term == null) {
+                totalOffset += segment.offset;
                 return false;
             }
             if (TextUtility.isBlank(term.word)) {

diff --git a/src/main/java/com/hankcs/lucene/SegmentWrapper.java b/src/main/java/com/hankcs/lucene/SegmentWrapper.java
@@ -42,7 +42,7 @@ public class SegmentWrapper {
     /**
      * term的偏移量，由于wrapper是按行读取的，必须对term.offset做一个校正
      */
-    private int offset;
+    int offset;
 
     Configuration configuration;