[Pkg-javascript-commits] [node-acorn-jsx] 423/484: Speed-up reading words, strings and templates.

[Bastien Roucariès]

This is an automated email from the git hooks/post-receive script.

rouca pushed a commit to branch master
in repository node-acorn-jsx.

commit cdd444eff12a4e6eb0744606ce12d6b6043133b6
Author: Ingvar Stepanyan <me at rreverser.com>
Date:   Sat Jan 24 02:24:55 2015 +0200

    Speed-up reading words, strings and templates.
    
    Now identifiers and strings even with escaped chars are read
    in optimized way by reading entire chunks delimited by escape
    chars (and not bailing to deopt mode on first one).
---
 acorn.js | 61 ++++++++++++++++++++++++++++++++-----------------------------
 1 file changed, 32 insertions(+), 29 deletions(-)

diff --git a/acorn.js b/acorn.js
index 01f7037..91284e0 100644
--- a/acorn.js
+++ b/acorn.js
@@ -1119,36 +1119,33 @@
   }
 
   function readString(quote) {
-    ++tokPos;
-    var out = "";
+    var out = "", chunkStart = ++tokPos;
     for (;;) {
       if (tokPos >= inputLen) raise(tokStart, "Unterminated string constant");
       var ch = input.charCodeAt(tokPos);
-      if (ch === quote) {
-        ++tokPos;
-        return finishToken(_string, out);
-      }
+      if (ch === quote) break;
       if (ch === 92) { // '\'
+        out += input.slice(chunkStart, tokPos);
         out += readEscapedChar();
+        chunkStart = tokPos;
       } else {
+        if (isNewLine(ch)) raise(tokStart, "Unterminated string constant");
         ++tokPos;
-        if (isNewLine(ch)) {
-          raise(tokStart, "Unterminated string constant");
-        }
-        out += String.fromCharCode(ch); // '\'
       }
     }
+    out += input.slice(chunkStart, tokPos++);
+    return finishToken(_string, out);
   }
 
   // Reads template string tokens.
 
   function readTmplToken() {
-    var out = "", start = tokPos;
+    var out = "", chunkStart = tokPos;
     for (;;) {
       if (tokPos >= inputLen) raise(tokStart, "Unterminated template");
       var ch = input.charCodeAt(tokPos);
       if (ch === 96 || ch === 36 && input.charCodeAt(tokPos + 1) === 123) { // '`', '${'
-        if (tokPos === start && tokType === _template) {
+        if (tokPos === tokStart && tokType === _template) {
           if (ch === 36) {
             tokPos += 2;
             return finishToken(_dollarBraceL);
@@ -1157,23 +1154,29 @@
             return finishToken(_backQuote);
           }
         }
+        out += input.slice(chunkStart, tokPos);
         return finishToken(_template, out);
       }
       if (ch === 92) { // '\'
+        out += input.slice(chunkStart, tokPos);
         out += readEscapedChar();
-      } else {
+        chunkStart = tokPos;
+      } else if (isNewLine(ch)) {
+        out += input.slice(chunkStart, tokPos);
         ++tokPos;
-        if (isNewLine(ch)) {
-          if (ch === 13 && input.charCodeAt(tokPos) === 10) {
-            ++tokPos;
-            ch = 10;
-          }
-          if (options.locations) {
-            ++tokCurLine;
-            tokLineStart = tokPos;
-          }
+        if (ch === 13 && input.charCodeAt(tokPos) === 10) {
+          ++tokPos;
+          out += "\n";
+        } else {
+          out += String.fromCharCode(ch);
+        }
+        if (options.locations) {
+          ++tokCurLine;
+          tokLineStart = tokPos;
         }
-        out += String.fromCharCode(ch);
+        chunkStart = tokPos;
+      } else {
+        ++tokPos;
       }
     }
   }
@@ -1228,20 +1231,19 @@
   // Read an identifier, and return it as a string. Sets `containsEsc`
   // to whether the word contained a '\u' escape.
   //
-  // Only builds up the word character-by-character when it actually
-  // containeds an escape, as a micro-optimization.
+  // Incrementally adds only escaped chars, adding other chunks as-is
+  // as a micro-optimization.
 
   function readWord1() {
     containsEsc = false;
-    var word, first = true, start = tokPos;
+    var word = "", first = true, chunkStart = tokPos;
     for (;;) {
       var ch = input.charCodeAt(tokPos);
       if (isIdentifierChar(ch)) {
-        if (containsEsc) word += input.charAt(tokPos);
         ++tokPos;
       } else if (ch === 92) { // "\"
-        if (!containsEsc) word = input.slice(start, tokPos);
         containsEsc = true;
+        word += input.slice(chunkStart, tokPos);
         if (input.charCodeAt(++tokPos) != 117) // "u"
           raise(tokPos, "Expecting Unicode escape sequence \\uXXXX");
         ++tokPos;
@@ -1251,12 +1253,13 @@
         if (!(first ? isIdentifierStart(esc) : isIdentifierChar(esc)))
           raise(tokPos - 4, "Invalid Unicode escape");
         word += escStr;
+        chunkStart = tokPos;
       } else {
         break;
       }
       first = false;
     }
-    return containsEsc ? word : input.slice(start, tokPos);
+    return word + input.slice(chunkStart, tokPos);
   }
 
   // Read an identifier or keyword token. Will check for reserved

-- 
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/pkg-javascript/node-acorn-jsx.git