From 04855d6eb519fa2cd4edf03f8f5a475d40ed047a Mon Sep 17 00:00:00 2001
From: Louis-Dominique Dubeau <ldd@lddubeau.com>
Date: Thu, 23 Aug 2018 19:02:13 -0400
Subject: [PATCH] perf: introduce a specialized version of captureWhile

captureName performs faster for those cases where we just want to capture the
characters in an XML name.
---
 lib/saxes.js | 66 +++++++++++++++++++++++++---------------------------
 1 file changed, 32 insertions(+), 34 deletions(-)

diff --git a/lib/saxes.js b/lib/saxes.js
index 071cd63f..85aa8468 100644
--- a/lib/saxes.js
+++ b/lib/saxes.js
@@ -571,12 +571,7 @@ class SaxesParser {
    */
 
   /**
-   * Capture characters into a buffer while a condition is true. A sequence of
-   * ``write`` calls may require the capture of text into a buffer as multiple
-   * "fragments". For instance, given ``write("<x>Multiple")`` and
-   * ``write("parts</x>")``, the text which is part of the ``x`` element will be
-   * recorded in two steps: one recording ``"Multiple"`` and one recording
-   * ``"parts"``. These are two fragments.
+   * Capture characters into a buffer while a condition is true.
    *
    * @private
    *
@@ -607,6 +602,35 @@ class SaxesParser {
     return undefined;
   }
 
+  /**
+   * Capture characters that satisfy ``isNameChar`` into a buffer.
+   *
+   * @private
+   *
+   * @param {ChunkState} chunkState The current chunk state.
+   *
+   * @param {string} buffer The name of the buffer to save into.
+   *
+   * @return {string|undefined} The character that made the test fail, or
+   * ``undefined`` if we hit the end of the chunk.
+   */
+  captureName(chunkState, buffer) {
+    const { limit, chunk, i: start } = chunkState;
+    while (chunkState.i < limit) {
+      const c = this.getCode(chunkState);
+      if (!isNameChar(c)) {
+        // This is faster than adding codepoints one by one.
+        this[buffer] += chunk.substring(start,
+                                        chunkState.i - (c <= 0xFFFF ? 1 : 2));
+        return c;
+      }
+    }
+
+    // This is faster than adding codepoints one by one.
+    this[buffer] += chunk.substring(start);
+    return undefined;
+  }
+
   /**
    * Skip characters while a condition is true.
    *
@@ -1223,20 +1247,7 @@ class SaxesParser {
   sOpenTag(chunkState) {
     // We don't need to check with isNameStartChar here because the first
     // character of tagName is fed elsewhere, and the check is done there.
-    const c = this.captureWhile(
-      chunkState,
-      (cx) => {
-        if (cx !== GREATER && !isS(cx) && cx !== FORWARD_SLASH) {
-          if (!isNameChar(cx)) {
-            this.fail("disallowed character in tag name.");
-          }
-
-          return true;
-        }
-
-        return false;
-      },
-      "tagName");
+    const c = this.captureName(chunkState, "tagName");
     if (!c) {
       return;
     }
@@ -1305,20 +1316,7 @@ class SaxesParser {
   sAttribName(chunkState) {
     // We don't need to check with isNameStartChar here because the first
     // character of attribute is fed elsewhere, and the check is done there.
-    const c = this.captureWhile(
-      chunkState,
-      (cx) => {
-        if (cx !== EQUAL && !isS(cx) && cx !== GREATER) {
-          if (!isNameChar(cx)) {
-            this.fail("disallowed characer in attribute name.");
-          }
-
-          return true;
-        }
-
-        return false;
-      },
-      "attribName");
+    const c = this.captureName(chunkState, "attribName");
     if (c === EQUAL) {
       this.state = S_ATTRIB_VALUE;
     }