tidy IndexCheck and improve its testing (#4442)

convert IndexCheck from static class

also fold the project paralell check into the class, use Mockito for testing
and rework exception propagation

fixes #4412

Author: vladak

opengrok-indexer/src/main/java/org/opengrok/indexer/configuration/Project.java

@@ -570,8 +570,9 @@ public static Project getProject(String path) {
     /**
      * Get the project for a specific file.
      *
-     * @param file the file to lookup
-     * @return the project that this file belongs to (or {@code null} if the file doesn't belong to a project)
+     * @param file file under source root
+     * @return the project that this file belongs to (or {@code null} if the file doesn't belong to a project,
+     * or it is a symbolic link that is not allowed)
      */
     @Nullable
     public static Project getProject(File file) {

opengrok-indexer/src/main/java/org/opengrok/indexer/index/IndexCheck.java

@@ -0,0 +1,51 @@
+/*
+ * CDDL HEADER START
+ *
+ * The contents of this file are subject to the terms of the
+ * Common Development and Distribution License (the "License").
+ * You may not use this file except in compliance with the License.
+ *
+ * See LICENSE.txt included in this distribution for the specific
+ * language governing permissions and limitations under the License.
+ *
+ * When distributing Covered Code, include this CDDL HEADER in each
+ * file and include the License file at LICENSE.txt.
+ * If applicable, add the following below this CDDL HEADER, with the
+ * fields enclosed by brackets "[]" replaced with your own identifying
+ * information: Portions Copyright [yyyy] [name of copyright owner]
+ *
+ * CDDL HEADER END
+ */
+
+/*
+ * Copyright (c) 2023, Oracle and/or its affiliates. All rights reserved.
+ */
+package org.opengrok.indexer.index;
+
+import java.nio.file.Path;
+import java.util.Collections;
+import java.util.HashSet;
+import java.util.Set;
+
+/**
+ * Common exception for all check modes.
+ */
+public class IndexCheckException extends Exception {
+    private static final long serialVersionUID = 5693446916108385595L;
+
+    private final Set<Path> failedPaths = new HashSet<>();
+
+    public IndexCheckException(String s, Path path) {
+        super(s);
+        failedPaths.add(path);
+    }
+
+    public IndexCheckException(String s, Set<Path> paths) {
+        super(s);
+        failedPaths.addAll(paths);
+    }
+
+    public Set<Path> getFailedPaths() {
+        return Collections.unmodifiableSet(failedPaths);
+    }
+}

opengrok-indexer/src/main/java/org/opengrok/indexer/index/IndexCheckException.java

@@ -0,0 +1,50 @@
+/*
+ * CDDL HEADER START
+ *
+ * The contents of this file are subject to the terms of the
+ * Common Development and Distribution License (the "License").
+ * You may not use this file except in compliance with the License.
+ *
+ * See LICENSE.txt included in this distribution for the specific
+ * language governing permissions and limitations under the License.
+ *
+ * When distributing Covered Code, include this CDDL HEADER in each
+ * file and include the License file at LICENSE.txt.
+ * If applicable, add the following below this CDDL HEADER, with the
+ * fields enclosed by brackets "[]" replaced with your own identifying
+ * information: Portions Copyright [yyyy] [name of copyright owner]
+ *
+ * CDDL HEADER END
+ */
+
+/*
+ * Copyright (c) 2023, Oracle and/or its affiliates. All rights reserved.
+ */
+package org.opengrok.indexer.index;
+
+import java.nio.file.Path;
+import java.util.Map;
+
+/**
+ * Exception thrown when index contains duplicate live documents.
+ */
+public class IndexDocumentException extends IndexCheckException {
+    private static final long serialVersionUID = 5693446916108385595L;
+
+    private final Map<String, Integer> fileMap;
+
+    public IndexDocumentException(String s, Path path) {
+        super(s, path);
+        this.fileMap = null;
+    }
+
+    public IndexDocumentException(String s, Path path, Map<String, Integer> fileMap) {
+        super(s, path);
+        this.fileMap = fileMap;
+    }
+
+    @Override
+    public String toString() {
+        return getMessage() + ": " + (fileMap == null ? "" : fileMap);
+    }
+}

opengrok-indexer/src/main/java/org/opengrok/indexer/index/IndexDocumentException.java

@@ -0,0 +1,48 @@
+/*
+ * CDDL HEADER START
+ *
+ * The contents of this file are subject to the terms of the
+ * Common Development and Distribution License (the "License").
+ * You may not use this file except in compliance with the License.
+ *
+ * See LICENSE.txt included in this distribution for the specific
+ * language governing permissions and limitations under the License.
+ *
+ * When distributing Covered Code, include this CDDL HEADER in each
+ * file and include the License file at LICENSE.txt.
+ * If applicable, add the following below this CDDL HEADER, with the
+ * fields enclosed by brackets "[]" replaced with your own identifying
+ * information: Portions Copyright [yyyy] [name of copyright owner]
+ *
+ * CDDL HEADER END
+ */
+
+/*
+ * Copyright (c) 2023, Oracle and/or its affiliates. All rights reserved.
+ */
+package org.opengrok.indexer.index;
+
+import java.nio.file.Path;
+
+/**
+ * Exception thrown when index version does not match Lucene version.
+ */
+public class IndexVersionException extends IndexCheckException {
+
+    private static final long serialVersionUID = 5693446916108385595L;
+
+    private final int luceneIndexVersion;
+    private final int indexVersion;
+
+    public IndexVersionException(String s, Path path, int luceneIndexVersion, int indexVersion) {
+        super(s, path);
+        this.indexVersion = indexVersion;
+        this.luceneIndexVersion = luceneIndexVersion;
+    }
+
+    @Override
+    public String toString() {
+        return getMessage() + ": " + String.format("Lucene version = %d", luceneIndexVersion) + ", " +
+                String.format("index version = %d", indexVersion);
+    }
+}

opengrok-indexer/src/main/java/org/opengrok/indexer/index/IndexVersionException.java

@@ -265,30 +265,27 @@ public static void main(String[] argv) {
             env.setConfiguration(cfg, subFilePaths, CommandTimeoutType.INDEXER);
 
             // Let repository types to add items to ignoredNames.
-            // This changes env so is called after the setConfiguration()
-            // call above.
+            // This changes env so is called after the setConfiguration() call above.
             RepositoryFactory.initializeIgnoredNames(env);
 
-            // Check index(es). Exit with distinct return code upon failure.
+            // Check index(es) and exit. Use distinct return code upon failure.
             if (indexCheckMode.ordinal() > IndexCheck.IndexCheckMode.NO_CHECK.ordinal()) {
                 if (cfg.getDataRoot() == null || cfg.getDataRoot().isEmpty()) {
                     System.err.println("Empty data root in configuration");
                     System.exit(1);
                 }
 
-                try {
-                    if (!IndexCheck.isOkay(cfg, indexCheckMode, subFileArgs)) {
-                        System.err.printf("Index check failed%n");
-                        System.err.print("You might want to remove " +
-                                (!subFilePaths.isEmpty() ? "data for projects " + String.join(",", subFilePaths) :
-                                        "all data") + " under the data root and reindex\n");
-                        System.exit(1);
-                    }
+                try (IndexCheck indexCheck = new IndexCheck(cfg, subFileArgs)) {
+                    indexCheck.check(indexCheckMode);
                 } catch (IOException e) {
                     // Use separate return code for cases where the index could not be read.
                     // This avoids problems with wiping out the index based on the check.
                     LOGGER.log(Level.WARNING, String.format("Could not perform index check for '%s'", subFileArgs), e);
                     System.exit(2);
+                } catch (IndexCheckException e) {
+                    System.err.printf("Index check failed%n");
+                    System.err.print("You might want to remove " + e.getFailedPaths());
+                    System.exit(1);
                 }
 
                 System.exit(0);