Merge remote-tracking branch 'origin/branch_6x' into branch_6x

lucene/CHANGES.txt

@@ -63,6 +63,10 @@ Improvements
 
 * LUCENE-7496: Better toString for SweetSpotSimilarity (janhoy)
 
+* LUCENE-7520: Highlighter's WeightedSpanTermExtractor shouldn't attempt to expand a MultiTermQuery
+  when its field doesn't match the field the extraction is scoped to.
+  (Cao Manh Dat via David Smiley)
+
 Optimizations
 
 * LUCENE-7501: BKDReader should not store the split dimension explicitly in the

lucene/highlighter/src/java/org/apache/lucene/search/highlight/WeightedSpanTermExtractor.java

@@ -217,13 +217,14 @@ public class WeightedSpanTermExtractor {
     } else if (isQueryUnsupported(query.getClass())) {
       // nothing
     } else {
+      if (query instanceof MultiTermQuery &&
+          (!expandMultiTermQuery || !fieldNameComparator(((MultiTermQuery)query).getField()))) {
+        return;
+      }
       Query origQuery = query;
       final IndexReader reader = getLeafContext().reader();
       Query rewritten;
       if (query instanceof MultiTermQuery) {
-        if (!expandMultiTermQuery) {
-          return;
-        }
         rewritten = MultiTermQuery.SCORING_BOOLEAN_REWRITE.rewrite(reader, (MultiTermQuery) query);
       } else {
         rewritten = origQuery.rewrite(reader);
@@ -513,11 +514,7 @@ public class WeightedSpanTermExtractor {
    */
   public Map<String,WeightedSpanTerm> getWeightedSpanTerms(Query query, float boost, TokenStream tokenStream,
       String fieldName) throws IOException {
-    if (fieldName != null) {
     this.fieldName = fieldName;
-    } else {
-      this.fieldName = null;
-    }
 
     Map<String,WeightedSpanTerm> terms = new PositionCheckingMap<>();
     this.tokenStream = tokenStream;

lucene/highlighter/src/test/org/apache/lucene/search/highlight/HighlighterTest.java

@@ -33,6 +33,7 @@ import java.util.StringTokenizer;
 import org.apache.lucene.analysis.Analyzer;
 import org.apache.lucene.analysis.BaseTokenStreamTestCase;
 import org.apache.lucene.analysis.CachingTokenFilter;
+import org.apache.lucene.analysis.CannedTokenStream;
 import org.apache.lucene.analysis.MockAnalyzer;
 import org.apache.lucene.analysis.MockPayloadAnalyzer;
 import org.apache.lucene.analysis.MockTokenFilter;
@@ -1339,6 +1340,22 @@ public class HighlighterTest extends BaseTokenStreamTestCase implements Formatte
 
   }
 
+  public void testNotRewriteMultiTermQuery() throws IOException {
+    // field "bar": (not the field we ultimately want to extract)
+    MultiTermQuery mtq = new TermRangeQuery("bar", new BytesRef("aa"), new BytesRef("zz"), true, true) ;
+    WeightedSpanTermExtractor extractor = new WeightedSpanTermExtractor() {
+      @Override
+      protected void extract(Query query, float boost, Map<String, WeightedSpanTerm> terms) throws IOException {
+        assertEquals(mtq, query);
+        super.extract(query, boost, terms);
+      }
+    };
+    extractor.setExpandMultiTermQuery(true);
+    extractor.setMaxDocCharsToAnalyze(51200);
+    extractor.getWeightedSpanTerms(
+        mtq, 3, new CannedTokenStream(new Token("aa",0,2), new Token("bb", 2,4)), "foo"); // field "foo"
+  }
+
   public void testGetBestSingleFragmentWithWeights() throws Exception {
 
     TestHighlightRunner helper = new TestHighlightRunner() {

solr/CHANGES.txt

@@ -195,6 +195,7 @@ Optimizations
 
 * SOLR-9506: cache IndexFingerprint for each segment (Pushkar Raste, yonik, noble)
 
+* SOLR-7506: Roll over GC logs by default via bin/solr scripts (shalin, janhoy)
 
 Other Changes
 ----------------------

solr/bin/solr

@@ -1407,13 +1407,14 @@ if [ -z ${GC_LOG_OPTS+x} ]; then
 else
   GC_LOG_OPTS=($GC_LOG_OPTS)
 fi
-# if verbose gc logging enabled, setup the location of the log file
+
+# if verbose gc logging enabled, setup the location of the log file and rotation
 if [ "$GC_LOG_OPTS" != "" ]; then
   gc_log_flag="-Xloggc"
   if [ "$JAVA_VENDOR" == "IBM J9" ]; then
     gc_log_flag="-Xverbosegclog"
   fi
-  GC_LOG_OPTS+=("$gc_log_flag:$SOLR_LOGS_DIR/solr_gc.log")
+  GC_LOG_OPTS+=("$gc_log_flag:$SOLR_LOGS_DIR/solr_gc.log" -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=9 -XX:GCLogFileSize=20M)
 fi
 
 # If ZK_HOST is defined, the assume SolrCloud mode

solr/bin/solr.cmd

@@ -1013,23 +1013,23 @@ IF NOT EXIST "%SOLR_SERVER_DIR%\tmp" (
 )
 
 IF "%JAVA_VENDOR%" == "IBM J9" (
-  set "GCLOG_OPT=-Xverbosegclog"
+  set GCLOG_OPT="-Xverbosegclog:!SOLR_LOGS_DIR!\solr_gc.log" -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=9 -XX:GCLogFileSize=20M
 ) else (
-  set "GCLOG_OPT=-Xloggc"
+  set GCLOG_OPT="-Xloggc:!SOLR_LOGS_DIR!\solr_gc.log" -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=9 -XX:GCLogFileSize=20M
 )
 
 IF "%FG%"=="1" (
   REM run solr in the foreground
   title "Solr-%SOLR_PORT%"
   echo %SOLR_PORT%>"%SOLR_TIP%"\bin\solr-%SOLR_PORT%.port
-  "%JAVA%" %SERVEROPT% %SOLR_JAVA_MEM% %START_OPTS% %GCLOG_OPT%:"!SOLR_LOGS_DIR!/solr_gc.log" ^
+  "%JAVA%" %SERVEROPT% %SOLR_JAVA_MEM% %START_OPTS% %GCLOG_OPT% ^
     -Dlog4j.configuration="%LOG4J_CONFIG%" -DSTOP.PORT=!STOP_PORT! -DSTOP.KEY=%STOP_KEY% ^
     -Dsolr.solr.home="%SOLR_HOME%" -Dsolr.install.dir="%SOLR_TIP%" ^
     -Djetty.host=%SOLR_JETTY_HOST% -Djetty.port=%SOLR_PORT% -Djetty.home="%SOLR_SERVER_DIR%" ^
     -Djava.io.tmpdir="%SOLR_SERVER_DIR%\tmp" -jar start.jar "%SOLR_JETTY_CONFIG%"
 ) ELSE (
   START /B "Solr-%SOLR_PORT%" /D "%SOLR_SERVER_DIR%" ^
-    "%JAVA%" %SERVEROPT% %SOLR_JAVA_MEM% %START_OPTS% %GCLOG_OPT%:"!SOLR_LOGS_DIR!/solr_gc.log" ^
+    "%JAVA%" %SERVEROPT% %SOLR_JAVA_MEM% %START_OPTS% %GCLOG_OPT% ^
     -Dlog4j.configuration="%LOG4J_CONFIG%" -DSTOP.PORT=!STOP_PORT! -DSTOP.KEY=%STOP_KEY% ^
     -Dsolr.log.muteconsole ^
     -Dsolr.solr.home="%SOLR_HOME%" -Dsolr.install.dir="%SOLR_TIP%" ^

solr/core/src/java/org/apache/solr/util/SolrCLI.java

@@ -3445,13 +3445,13 @@ public class SolrCLI {
         Files.createDirectories(archivePath);
       }
       List<Path> archived = Files.find(archivePath, 1, (f, a) 
-          -> a.isRegularFile() && String.valueOf(f.getFileName()).startsWith("solr_gc_"))
+          -> a.isRegularFile() && String.valueOf(f.getFileName()).matches("^solr_gc[_.].+"))
           .collect(Collectors.toList());
       for (Path p : archived) {
         Files.delete(p);
       }
       List<Path> files = Files.find(logsPath, 1, (f, a) 
-          -> a.isRegularFile() && String.valueOf(f.getFileName()).startsWith("solr_gc_"))
+          -> a.isRegularFile() && String.valueOf(f.getFileName()).matches("^solr_gc[_.].+"))
           .collect(Collectors.toList());
       if (files.size() > 0) {
         out("Archiving " + files.size() + " old GC log files to " + archivePath);

solr/core/src/test/org/apache/solr/util/UtilsToolTest.java

@@ -55,6 +55,10 @@ public class UtilsToolTest extends SolrTestCaseJ4 {
       "solr_log_20160304", 
       "solr-8983-console.log",
       "solr_gc_log_20160102", 
+      "solr_gcnotremove", 
+      "solr_gc.log", 
+      "solr_gc.log.0", 
+      "solr_gc.log.0.current", 
       "solr_gc_log_2");
   
   @Before
@@ -136,7 +140,7 @@ public class UtilsToolTest extends SolrTestCaseJ4 {
     String[] args = {"utils", "-archive_gc_logs", "-l", dir.toString()};
     assertEquals(files.size(), fileCount());
     assertEquals(0, runTool(args));
-    assertEquals(files.size()-2, fileCount());
+    assertEquals(files.size()-5, fileCount());
     assertFalse(listFiles().contains("solr_gc_log_2"));
     assertTrue(Files.exists(dir.resolve("archived").resolve("solr_gc_log_2")));
     assertEquals(0, runTool(args));