Draft version of OCaml indexer.

opengrok-indexer/src/main/java/org/opengrok/indexer/analysis/AnalyzerGuru.java

@@ -89,6 +89,7 @@
 import org.opengrok.indexer.analysis.kotlin.KotlinAnalyzerFactory;
 import org.opengrok.indexer.analysis.lisp.LispAnalyzerFactory;
 import org.opengrok.indexer.analysis.lua.LuaAnalyzerFactory;
+import org.opengrok.indexer.analysis.ocaml.OCamlAnalyzerFactory;
 import org.opengrok.indexer.analysis.pascal.PascalAnalyzerFactory;
 import org.opengrok.indexer.analysis.perl.PerlAnalyzerFactory;
 import org.opengrok.indexer.analysis.php.PhpAnalyzerFactory;
@@ -298,6 +299,7 @@ public class AnalyzerGuru {
                 new HaskellAnalyzerFactory(),
                 new GolangAnalyzerFactory(),
                 new LuaAnalyzerFactory(),
+                new OCamlAnalyzerFactory(),
                 new PascalAnalyzerFactory(),
                 new AdaAnalyzerFactory(),
                 new RubyAnalyzerFactory(),

opengrok-indexer/src/main/java/org/opengrok/indexer/analysis/ocaml/Consts.java

@@ -0,0 +1,101 @@
+/*
+ * 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) 2015, 2018, Oracle and/or its affiliates. All rights reserved.
+ * Portions Copyright (c) 2017, 2020, Chris Fraire <cfraire@me.com>.
+ */
+package org.opengrok.indexer.analysis.ocaml;
+
+import java.util.HashSet;
+import java.util.Set;
+
+/**
+ * Represents a container for a set of OCaml keywords.
+ */
+public class Consts {
+
+    static final Set<String> kwd = new HashSet<>();
+
+    /* From parsing/lexer.mll of OCaml 5.3.0. */
+    static {
+        kwd.add("and");
+        kwd.add("as");
+        kwd.add("assert");
+        kwd.add("begin");
+        kwd.add("class");
+        kwd.add("constraint");
+        kwd.add("do");
+        kwd.add("done");
+        kwd.add("downto");
+        kwd.add("effect");
+        kwd.add("else");
+        kwd.add("end");
+        kwd.add("exception");
+        kwd.add("external");
+        kwd.add("false");
+        kwd.add("for");
+        kwd.add("fun");
+        kwd.add("function");
+        kwd.add("functor");
+        kwd.add("if");
+        kwd.add("in");
+        kwd.add("include");
+        kwd.add("inherit");
+        kwd.add("initializer");
+        kwd.add("lazy");
+        kwd.add("let");
+        kwd.add("match");
+        kwd.add("method");
+        kwd.add("module");
+        kwd.add("mutable");
+        kwd.add("new");
+        kwd.add("nonrec");
+        kwd.add("object");
+        kwd.add("of");
+        kwd.add("open");
+        kwd.add("or");
+        kwd.add("parser");
+        kwd.add("private");
+        kwd.add("rec");
+        kwd.add("sig");
+        kwd.add("struct");
+        kwd.add("then");
+        kwd.add("to");
+        kwd.add("true");
+        kwd.add("try");
+        kwd.add("type");
+        kwd.add("val");
+        kwd.add("virtual");
+        kwd.add("when");
+        kwd.add("while");
+        kwd.add("with");
+        kwd.add("lor");
+        kwd.add("lxor");
+        kwd.add("mod");
+        kwd.add("land");
+        kwd.add("lsl");
+        kwd.add("lsr");
+        kwd.add("asr");
+    }
+
+    /** Private to enforce static. */
+    private Consts() {
+    }
+}

opengrok-indexer/src/main/java/org/opengrok/indexer/analysis/ocaml/OCamlAnalyzer.java

@@ -0,0 +1,76 @@
+/*
+ * 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) 2015, 2018, Oracle and/or its affiliates. All rights reserved.
+ * Portions Copyright (c) 2017, 2020, Chris Fraire <cfraire@me.com>.
+ */
+package org.opengrok.indexer.analysis.ocaml;
+
+import org.opengrok.indexer.analysis.AbstractAnalyzer;
+import org.opengrok.indexer.analysis.FileAnalyzerFactory;
+import org.opengrok.indexer.analysis.JFlexTokenizer;
+import org.opengrok.indexer.analysis.JFlexXref;
+import org.opengrok.indexer.analysis.plain.AbstractSourceCodeAnalyzer;
+
+import java.io.Reader;
+
+/**
+ * Represents an analyzer for the OCaml language.
+ */
+@SuppressWarnings("java:S110")
+public class OCamlAnalyzer extends AbstractSourceCodeAnalyzer {
+
+    /**
+     * Creates a new instance of {@link OCamlAnalyzer}.
+     * @param factory instance
+     */
+    protected OCamlAnalyzer(FileAnalyzerFactory factory) {
+        super(factory, () -> new JFlexTokenizer(new OCamlSymbolTokenizer(
+                AbstractAnalyzer.DUMMY_READER)));
+    }
+
+    /**
+     * @return {@code "ocaml"}
+     */
+    @Override
+    public String getCtagsLang() {
+        return "ocaml";
+    }
+
+    /**
+     * Gets a version number to be used to tag processed documents so that
+     * re-analysis can be re-done later if a stored version number is different
+     * from the current implementation.
+     * @return 20250403_00
+     */
+    @Override
+    protected int getSpecializedVersionNo() {
+        return 20250403_00; // Edit comment above too!
+    }
+
+    /**
+     * Creates a wrapped {@link OCamlXref} instance.
+     * @return a defined instance
+     */
+    @Override
+    protected JFlexXref newXref(Reader reader) {
+        return new JFlexXref(new OCamlXref(reader));
+    }
+}

opengrok-indexer/src/main/java/org/opengrok/indexer/analysis/ocaml/OCamlAnalyzerFactory.java

@@ -0,0 +1,56 @@
+/*
+ * 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) 2015, 2018, Oracle and/or its affiliates. All rights reserved.
+ * Portions Copyright (c) 2017, 2020, Chris Fraire <cfraire@me.com>.
+ */
+package org.opengrok.indexer.analysis.ocaml;
+
+import org.opengrok.indexer.analysis.AbstractAnalyzer.Genre;
+import org.opengrok.indexer.analysis.FileAnalyzer;
+import org.opengrok.indexer.analysis.FileAnalyzerFactory;
+
+/**
+ * Represents a factory to create {@link OCamlAnalyzer} instances.
+ */
+public class OCamlAnalyzerFactory extends FileAnalyzerFactory {
+
+    private static final String NAME = "OCaml";
+
+    private static final String[] SUFFIXES = {"ML", "MLI"};
+
+    /**
+     * Initializes a factory instance to associate a file extensions ".ml",
+     * ".mli" with {@link OCamlAnalyzer}.
+     */
+    public OCamlAnalyzerFactory() {
+        super(null, null, SUFFIXES, null, null, "text/plain", Genre.PLAIN,
+                NAME, true);
+    }
+
+    /**
+     * Creates a new {@link OCamlAnalyzer} instance.
+     * @return a defined instance
+     */
+    @Override
+    protected FileAnalyzer newAnalyzer() {
+        return new OCamlAnalyzer(this);
+    }
+}

opengrok-indexer/src/main/java/org/opengrok/indexer/analysis/ocaml/OCamlLexer.java

@@ -0,0 +1,59 @@
+/*
+ * 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) 2015, 2018, Oracle and/or its affiliates. All rights reserved.
+ * Portions Copyright (c) 2017, 2020, Chris Fraire <cfraire@me.com>.
+ */
+package org.opengrok.indexer.analysis.ocaml;
+
+import org.opengrok.indexer.analysis.JFlexJointLexer;
+import org.opengrok.indexer.analysis.JFlexSymbolMatcher;
+import org.opengrok.indexer.analysis.Resettable;
+
+/**
+ * Represents an abstract base class for OCaml lexers.
+ */
+@SuppressWarnings("Duplicates")
+abstract class OCamlLexer extends JFlexSymbolMatcher
+        implements JFlexJointLexer, Resettable {
+
+    /**
+     * Calls {@link #phLOC()} if the yystate is not COMMENT or SCOMMENT.
+     */
+    public void chkLOC() {
+        if (yystate() != COMMENT() && yystate() != SCOMMENT()) {
+            phLOC();
+        }
+    }
+
+    /**
+     * Subclasses must override to get the constant value created by JFlex to
+     * represent COMMENT.
+     */
+    @SuppressWarnings("java:S100")
+    abstract int COMMENT();
+
+    /**
+     * Subclasses must override to get the constant value created by JFlex to
+     * represent SCOMMENT.
+     */
+    @SuppressWarnings("java:S100")
+    abstract int SCOMMENT();
+}