From b3860a3d5795ac4ac33ce14ba76d87a06e6e4013 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Cl=C3=A9ment=20Fournier?= <clement.fournier76@gmail.com>
Date: Tue, 6 Aug 2019 09:42:48 +0200
Subject: [PATCH] Compute image of whitespace tokens lazily

Nobody cares and they make up a significant part of
token sequences. Alternatively we could intern them
which would most likely yield near 100% cache hit,
because of indentation patterns.
---
 .../pmd/lang/ast/impl/JavaccToken.java        |  2 +-
 .../pmd/lang/java/ast/ASTCompilationUnit.java |  5 ----
 .../pmd/lang/java/ast/JavaNode.java           |  1 +
 .../pmd/lang/java/ast/JavaTokenUtils.java     | 24 +++++++++++++++++++
 4 files changed, 26 insertions(+), 6 deletions(-)

diff --git a/pmd-core/src/main/java/net/sourceforge/pmd/lang/ast/impl/JavaccToken.java b/pmd-core/src/main/java/net/sourceforge/pmd/lang/ast/impl/JavaccToken.java
index 1b2197c3dc..ac407b6eee 100644
--- a/pmd-core/src/main/java/net/sourceforge/pmd/lang/ast/impl/JavaccToken.java
+++ b/pmd-core/src/main/java/net/sourceforge/pmd/lang/ast/impl/JavaccToken.java
@@ -67,7 +67,7 @@ public class JavaccToken implements GenericToken, java.io.Serializable {
     private final CharSequence image;
     private final int startInclusive;
     private final int endExclusive;
-    private final TokenDocument document;
+    protected final TokenDocument document;
 
     /** {@link #undefined()} */
     private JavaccToken() {
diff --git a/pmd-java/src/main/java/net/sourceforge/pmd/lang/java/ast/ASTCompilationUnit.java b/pmd-java/src/main/java/net/sourceforge/pmd/lang/java/ast/ASTCompilationUnit.java
index 078e35ab3b..b12f7c9e62 100644
--- a/pmd-java/src/main/java/net/sourceforge/pmd/lang/java/ast/ASTCompilationUnit.java
+++ b/pmd-java/src/main/java/net/sourceforge/pmd/lang/java/ast/ASTCompilationUnit.java
@@ -104,11 +104,6 @@ public final class ASTCompilationUnit extends AbstractJavaTypeNode implements Ro
         return classTypeResolver;
     }
 
-    @Override
-    public ASTCompilationUnit getRoot() {
-        return this;
-    }
-
     @InternalApi
     @Deprecated
     public void setClassTypeResolver(ClassTypeResolver classTypeResolver) {
diff --git a/pmd-java/src/main/java/net/sourceforge/pmd/lang/java/ast/JavaNode.java b/pmd-java/src/main/java/net/sourceforge/pmd/lang/java/ast/JavaNode.java
index 4e488027f7..cfc8bbad63 100644
--- a/pmd-java/src/main/java/net/sourceforge/pmd/lang/java/ast/JavaNode.java
+++ b/pmd-java/src/main/java/net/sourceforge/pmd/lang/java/ast/JavaNode.java
@@ -85,6 +85,7 @@ public interface JavaNode extends ScopedNode, TextAvailableNode {
     @Override
     JavaNode jjtGetParent();
 
+
     GenericToken jjtGetFirstToken();
 
 
diff --git a/pmd-java/src/main/java/net/sourceforge/pmd/lang/java/ast/JavaTokenUtils.java b/pmd-java/src/main/java/net/sourceforge/pmd/lang/java/ast/JavaTokenUtils.java
index 1e67b3093a..6ac09c6d35 100644
--- a/pmd-java/src/main/java/net/sourceforge/pmd/lang/java/ast/JavaTokenUtils.java
+++ b/pmd-java/src/main/java/net/sourceforge/pmd/lang/java/ast/JavaTokenUtils.java
@@ -32,6 +32,18 @@ final class JavaTokenUtils {
                 jcs.getEndOffset(),
                 jcs.getTokenDocument()
             );
+        case JavaParserConstants.WHITESPACE:
+            // unlikely that anybody cares about that, and since
+            // they're still 30% of all tokens best make this assumption
+
+            // btw 40% of all tokens have a compile-time string constant
+            // as image (jjstrLiteralImages) so they're shared.
+            return new LazyImageToken(
+                kind,
+                jcs.getStartOffset(),
+                jcs.getEndOffset(),
+                jcs.getTokenDocument()
+            );
         default:
             return new JavaccToken(
                 kind,
@@ -47,6 +59,18 @@ final class JavaTokenUtils {
         return token instanceof GTToken ? ((GTToken) token).realKind : token.kind;
     }
 
+    private static final class LazyImageToken extends JavaccToken {
+
+        public LazyImageToken(int kind, int startInclusive, int endExclusive, TokenDocument document) {
+            super(kind, null, startInclusive, endExclusive, document);
+        }
+
+        @Override
+        public String getImage() {
+            return document.getFullText().substring(getStartInDocument(), getEndInDocument());
+        }
+    }
+
     private static final class GTToken extends JavaccToken {
 
         final int realKind;