From 63b06ebde8b016f811b906f50d45af53516ab287 Mon Sep 17 00:00:00 2001
From: Johannes Schindelin <johannes.schindelin@gmx.de>
Date: Sat, 9 Nov 2013 14:13:19 -0600
Subject: [PATCH] Regex: optimize matching characters

Instead of having an opcode 'CHAR', let's have the opcodes that fall
within the range of a char *be* the opcode 'match this character'.

While at it, break the ranges of the different types of opcodes apart
into ranges so that related operations are clustered.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
---
 test/regex/Pattern.java       | 10 +++++-----
 test/regex/PikeVM.java        | 13 +++++++------
 test/regex/PikeVMOpcodes.java | 13 +++++++------
 3 files changed, 19 insertions(+), 17 deletions(-)

diff --git a/test/regex/Pattern.java b/test/regex/Pattern.java
index 0089b45564..d10fe8fa3b 100644
--- a/test/regex/Pattern.java
+++ b/test/regex/Pattern.java
@@ -52,13 +52,13 @@ public abstract class Pattern implements PikeVMOpcodes {
     if ("a(bb)?a".equals(regex)) {
       int[] program = new int[] {
         SAVE_OFFSET, 0,
-        CHAR, 'a',
-        SPLIT, 14,
+        'a',
+        SPLIT, 11,
         SAVE_OFFSET, 2,
-        CHAR, 'b',
-        CHAR, 'b',
+        'b',
+        'b',
         SAVE_OFFSET, 3,
-        /* 14 */ CHAR, 'a',
+        /* 11 */ 'a',
         SAVE_OFFSET, 1
       };
       return new RegexPattern(regex, flags, new PikeVM(program, 1));
diff --git a/test/regex/PikeVM.java b/test/regex/PikeVM.java
index c4b14e07b7..9219280c68 100644
--- a/test/regex/PikeVM.java
+++ b/test/regex/PikeVM.java
@@ -256,12 +256,6 @@ class PikeVM implements PikeVMOpcodes {
 
         int opcode = program[pc];
         switch (opcode) {
-        /* Possible optimization: make all opcodes <= 0xffff implicit chars */
-        case CHAR:
-          if (c == (char)program[pc + 1]) {
-            current.queueNext(pc, pc + 2, next);
-          }
-          break;
         case DOT:
           if (c != '\0' && c != '\r' && c != '\n') {
             current.queueNext(pc, pc + 1, next);
@@ -270,6 +264,7 @@ class PikeVM implements PikeVMOpcodes {
         case DOTALL:
           current.queueNext(pc, pc + 1, next);
           break;
+        /* immediate opcodes, i.e. thread continues within the same step */
         case SAVE_OFFSET:
           int index = program[pc + 1];
           current.saveOffset(pc, index, i);
@@ -283,6 +278,12 @@ class PikeVM implements PikeVMOpcodes {
           current.queueImmediately(pc, program[pc + 1], false);
           break;
         default:
+          if (program[pc] >= 0 && program[pc] <= 0xffff) {
+            if (c == (char)program[pc]) {
+              current.queueNext(pc, pc + 1, next);
+            }
+            break;
+          }
           throw new RuntimeException("Invalid opcode: " + opcode
             + " at pc " + pc);
         }
diff --git a/test/regex/PikeVMOpcodes.java b/test/regex/PikeVMOpcodes.java
index 4518130700..c12ad99427 100644
--- a/test/regex/PikeVMOpcodes.java
+++ b/test/regex/PikeVMOpcodes.java
@@ -19,10 +19,11 @@ package regex;
  * @author Johannes Schindelin
  */
 interface PikeVMOpcodes {
-  final static int CHAR = 1;
-  final static int DOT = 2;
-  final static int DOTALL = 3;
-  final static int SAVE_OFFSET = 4;
-  final static int SPLIT = 5;
-  final static int JMP = 6;
+  final static int DOT = -1;
+  final static int DOTALL = -2;
+
+  final static int SAVE_OFFSET = -40;
+
+  final static int SPLIT = -50;
+  final static int JMP = -51;
 }