querystring: improve unescapeBuffer() performance

Before this, v8 would deopt when an out of bounds `inIndex` would get
passed to charCodeAt(). charCodeAt() returns NaN in such cases, so we
directly emulate that behavior as well.

Also, calls to charCodeAt() for constant strings have been replaced
by the raw character codes and parser state is now stored as an
integer instead of a string. Both of these provide a slight
performance increase.

PR-URL: #5012
Reviewed-By: James M Snell <[email protected]>
Reviewed-By: Roman Reiss <[email protected]>
Reviewed-By: Matteo Collina <[email protected]>

Author: mscdex

lib/querystring.js

@@ -6,63 +6,59 @@ const QueryString = exports;
 const Buffer = require('buffer').Buffer;
 
 
-function charCode(c) {
-  return c.charCodeAt(0);
-}
-
-
 // a safe fast alternative to decodeURIComponent
 QueryString.unescapeBuffer = function(s, decodeSpaces) {
   var out = new Buffer(s.length);
-  var state = 'CHAR'; // states: CHAR, HEX0, HEX1
+  var state = 0;
   var n, m, hexchar;
 
   for (var inIndex = 0, outIndex = 0; inIndex <= s.length; inIndex++) {
-    var c = s.charCodeAt(inIndex);
+    var c = inIndex < s.length ? s.charCodeAt(inIndex) : NaN;
     switch (state) {
-      case 'CHAR':
+      case 0: // Any character
         switch (c) {
-          case charCode('%'):
+          case 37: // '%'
             n = 0;
             m = 0;
-            state = 'HEX0';
+            state = 1;
             break;
-          case charCode('+'):
-            if (decodeSpaces) c = charCode(' ');
+          case 43: // '+'
+            if (decodeSpaces)
+              c = 32; // ' '
             // falls through
           default:
             out[outIndex++] = c;
             break;
         }
         break;
 
-      case 'HEX0':
-        state = 'HEX1';
+      case 1: // First hex digit
         hexchar = c;
-        if (charCode('0') <= c && c <= charCode('9')) {
-          n = c - charCode('0');
-        } else if (charCode('a') <= c && c <= charCode('f')) {
-          n = c - charCode('a') + 10;
-        } else if (charCode('A') <= c && c <= charCode('F')) {
-          n = c - charCode('A') + 10;
+        if (c >= 48/*0*/ && c <= 57/*9*/) {
+          n = c - 48/*0*/;
+        } else if (c >= 65/*A*/ && c <= 70/*F*/) {
+          n = c - 65/*A*/ + 10;
+        } else if (c >= 97/*a*/ && c <= 102/*f*/) {
+          n = c - 97/*a*/ + 10;
         } else {
-          out[outIndex++] = charCode('%');
+          out[outIndex++] = 37/*%*/;
           out[outIndex++] = c;
-          state = 'CHAR';
+          state = 0;
           break;
         }
+        state = 2;
         break;
 
-      case 'HEX1':
-        state = 'CHAR';
-        if (charCode('0') <= c && c <= charCode('9')) {
-          m = c - charCode('0');
-        } else if (charCode('a') <= c && c <= charCode('f')) {
-          m = c - charCode('a') + 10;
-        } else if (charCode('A') <= c && c <= charCode('F')) {
-          m = c - charCode('A') + 10;
+      case 2: // Second hex digit
+        state = 0;
+        if (c >= 48/*0*/ && c <= 57/*9*/) {
+          m = c -  48/*0*/;
+        } else if (c >= 65/*A*/ && c <= 70/*F*/) {
+          m = c - 65/*A*/ + 10;
+        } else if (c >= 97/*a*/ && c <= 102/*f*/) {
+          m = c - 97/*a*/ + 10;
         } else {
-          out[outIndex++] = charCode('%');
+          out[outIndex++] = 37/*%*/;
           out[outIndex++] = hexchar;
           out[outIndex++] = c;
           break;