Merge pull request #2966 from idodeclare/feature/reuse_uri_code

Share implementation of trimUri()

Author: tarzanek

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

@@ -18,7 +18,7 @@
  */
 
 /*
- * Copyright (c) 2017-2018, Chris Fraire <[email protected]>.
+ * Copyright (c) 2017-2019, Chris Fraire <[email protected]>.
  */
 
 package org.opengrok.indexer.analysis;
@@ -27,6 +27,7 @@
 import java.util.Set;
 import java.util.regex.Pattern;
 import org.opengrok.indexer.util.StringUtils;
+import org.opengrok.indexer.util.UriUtils;
 
 /**
  * Represents an abstract base class for subclasses of
@@ -245,36 +246,15 @@ protected void onUriMatched(String uri, int start) {
      * may have been captured as valid URI characters but in a particular
      * context should mark the start of a pushback
      */
-    protected void onUriMatched(String uri, int start,
-        Pattern collateralCapture) {
-
-        int n = 0;
-        int subn;
-        do {
-            // An ending-pushback could be present before a collateral capture,
-            // so detect both in a loop (on a shrinking `url') until no more
-            // shrinking should occur.
-
-            subn = StringUtils.countURIEndingPushback(uri);
-            int ccn = StringUtils.countPushback(uri, collateralCapture);
-            if (ccn > subn) {
-                subn = ccn;
-            }
-
-            // Push back if positive, but not if equal to the current length.
-            if (subn > 0 && subn < uri.length()) {
-                uri = uri.substring(0, uri.length() - subn);
-                n += subn;
-            } else {
-                subn = 0;
-            }
-        } while (subn != 0);
-        if (n > 0) {
-            yypushback(n);
+    protected void onUriMatched(String uri, int start, Pattern collateralCapture) {
+        UriUtils.TrimUriResult result = UriUtils.trimUri(uri, true, collateralCapture);
+        if (result.getPushBackCount() > 0) {
+            yypushback(result.getPushBackCount());
         }
 
         NonSymbolMatchedListener l = nonSymbolListener;
         if (l != null) {
+            uri = result.getUri();
             LinkageMatchedEvent evt = new LinkageMatchedEvent(this, uri,
                 LinkageType.URI, start, start + uri.length());
             l.linkageMatched(evt);

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

@@ -20,7 +20,7 @@
 /*
  * Copyright (c) 2009, 2018, Oracle and/or its affiliates. All rights reserved.
  * Portions Copyright 2011 Jens Elkner.
- * Portions Copyright (c) 2017-2018, Chris Fraire <[email protected]>.
+ * Portions Copyright (c) 2017-2019, Chris Fraire <[email protected]>.
  */
 
 package org.opengrok.indexer.analysis;
@@ -41,6 +41,7 @@
 import org.opengrok.indexer.configuration.Project;
 import org.opengrok.indexer.configuration.RuntimeEnvironment;
 import org.opengrok.indexer.util.StringUtils;
+import org.opengrok.indexer.util.UriUtils;
 import org.opengrok.indexer.web.HtmlConsts;
 import org.opengrok.indexer.web.Util;
 
@@ -80,37 +81,14 @@ public class JFlexXrefUtils {
      * @throws IOException if an error occurs while appending
      */
     public static void appendLink(Writer out, JFlexLexer lexer, String url,
-        boolean doEndingPushback, Pattern collateralCapture)
-            throws IOException {
-
-        int n = 0;
-        int subn;
-        do {
-            // An ending-pushback could be present before a collateral capture,
-            // so detect both in a loop (on a shrinking `url') until no more
-            // shrinking should occur.
-
-            subn = 0;
-            if (doEndingPushback) {
-                subn = StringUtils.countURIEndingPushback(url);
-            }
-            int ccn = StringUtils.countPushback(url, collateralCapture);
-            if (ccn > subn) {
-                subn = ccn;
-            }
+            boolean doEndingPushback, Pattern collateralCapture) throws IOException {
 
-            // Push back if positive, but not if equal to the current length.
-            if (subn > 0 && subn < url.length()) {
-                url = url.substring(0, url.length() - subn);
-                n += subn;
-            } else {
-                subn = 0;
-            }
-        } while (subn != 0);
-        if (n > 0) {
-            lexer.yypushback(n);
+        UriUtils.TrimUriResult result = UriUtils.trimUri(url, doEndingPushback, collateralCapture);
+        if (result.getPushBackCount() > 0) {
+            lexer.yypushback(result.getPushBackCount());
         }
 
+        url = result.getUri();
         out.write("<a href=\"");
         Util.htmlize(url, out);
         out.write("\">");

opengrok-indexer/src/main/java/org/opengrok/indexer/util/UriUtils.java

@@ -0,0 +1,104 @@
+/*
+ * 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) 2017-2019, Chris Fraire <[email protected]>.
+ */
+
+package org.opengrok.indexer.util;
+
+import java.util.regex.Pattern;
+
+/**
+ * Represents a container for utility methods concerned with URIs.
+ */
+public class UriUtils {
+
+    /**
+     * Represents the immutable return value of
+     * {@link #trimUri(String, boolean, Pattern)}.
+     */
+    public static final class TrimUriResult {
+        private final String uri;
+        private final int pushBackCount;
+
+        TrimUriResult(String uri, int pushBackCount) {
+            this.uri = uri;
+            this.pushBackCount = pushBackCount;
+        }
+
+        public String getUri() {
+            return uri;
+        }
+
+        public int getPushBackCount() {
+            return pushBackCount;
+        }
+    }
+
+    /**
+     * Trims a URI, specifying whether to enlist the
+     * {@link StringUtils#countURIEndingPushback(String)} algorithm or the
+     * {@link StringUtils#countPushback(String, Pattern)} or both.
+     * <p>
+     * If the pushback count is equal to the length of {@code url}, then the
+     * pushback is set to zero -- in order to avoid a never-ending lexical loop.
+     *
+     * @param uri the URI string
+     * @param shouldCheckEnding a value indicating whether to call
+     * {@link StringUtils#countURIEndingPushback(String)}
+     * @param collateralCapture optional pattern to call with
+     * {@link StringUtils#countPushback(String, Pattern)}
+     * @return a defined instance
+     */
+    public static TrimUriResult trimUri(String uri, boolean shouldCheckEnding,
+            Pattern collateralCapture) {
+
+        int n = 0;
+        while (true) {
+            /*
+             * An ending-pushback could be present before a collateral capture,
+             * so detect both in a loop (on a shrinking `url') until no more
+             * shrinking should occur.
+             */
+
+            int subN = 0;
+            if (shouldCheckEnding) {
+                subN = StringUtils.countURIEndingPushback(uri);
+            }
+            int ccn = StringUtils.countPushback(uri, collateralCapture);
+            if (ccn > subN) {
+                subN = ccn;
+            }
+
+            // Increment if positive, but not if equal to the current length.
+            if (subN > 0 && subN < uri.length()) {
+                uri = uri.substring(0, uri.length() - subN);
+                n += subN;
+            } else {
+                break;
+            }
+        }
+        return new TrimUriResult(uri, n);
+    }
+
+    /** Private to enforce static. */
+    private UriUtils() {
+    }
+}