Github user jimczi commented on a diff in the pull request:
https://github.com/apache/lucene-solr/pull/384#discussion_r191572617
--- Diff:
lucene/analysis/common/src/java/org/apache/lucene/analysis/miscellaneous/ConcatenateGraphFilter.java
---
@@ -119,27 +147,19 @@ public boolean incrementToken() throws IOException {
@Override
public void end() throws IOException {
- super.end();
- if (finiteStrings == null) {
- inputTokenStream.end();
- }
- }
-
- @Override
- public void close() throws IOException {
- if (finiteStrings == null) {
- inputTokenStream.close();
- }
+ restoreState(endState);
}
+ //nocommit move method to before incrementToken
@Override
public void reset() throws IOException {
- super.reset();
- if (hasAttribute(CharTermAttribute.class)) {
- // we only create this if we really need it to safe the UTF-8 to
UTF-16 conversion
- charTermAttribute = getAttribute(CharTermAttribute.class);
--- End diff --
I am referring to this comment:
````
//nocommit move method to before incrementToken
````
but maybe I misunderstood what you meant by move. Let me ask you this
differently, why do you need to build the automaton on reset ? Unless there is
a good reason to do it we shouldn't change the behavior here which is to build
the automaton on the first call to incrementToken.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
