perf: don't increment a column number

Rather than incrementing a column number as we go along, record the position when a new line starts and comput a column number as needed.
lddubeau · Sep 12, 2019 · 490fc24 · 490fc24
1 parent 4d66bbb
commit 490fc24
Show file tree

Hide file tree

Showing 2 changed files with 10 additions and 22 deletions.
diff --git a/lib/saxes.js b/lib/saxes.js
@@ -405,7 +405,7 @@ class SaxesParser {
     this.line = 1;
 
     /** The column the parser is currently looking at. */
-    this.column = 0;
+    this.positionAtNewLine = 0;
 
     this.fileName = this.opt.fileName;
     this.onready();
@@ -416,6 +416,10 @@ class SaxesParser {
     return this.chunkPosition + this.i;
   }
 
+  get column() {
+    return this.position - this.positionAtNewLine;
+  }
+
   /* eslint-disable class-methods-use-this */
   /**
    * Event handler for text data. The default implementation is a no-op.
@@ -610,8 +614,7 @@ class SaxesParser {
 
     this.i++;
     if (code < 0xD800) {
-      if (code >= SPACE) {
-        this.column++;
+      if (code >= SPACE || code === TAB) {
         return code;
       }
 
@@ -630,21 +633,16 @@ class SaxesParser {
         /* yes, fall through */
       case NL:
         this.line++;
-        this.column = 0;
+        this.positionAtNewLine = this.position;
         return NL;
-      case TAB:
-        this.column++;
-        return TAB;
       default:
-        this.column++;
         // If we get here, then code < SPACE and it is not NL CR or TAB.
         this.fail("disallowed character.");
         return code;
       }
     }
 
     if (code > 0xDBFF) {
-      this.column++;
       // This is a specialized version of isChar10 that takes into account
       // that in this context code > 0xDBFF and code <= 0xFFFF. So it does not
       // test cases that don't need testing.
@@ -662,7 +660,6 @@ class SaxesParser {
 
     const final = 0x10000 + ((code - 0xD800) * 0x400) +
           (chunk.charCodeAt(i + 1) - 0xDC00);
-    this.column += 2;
     this.i++;
 
     // This is a specialized version of isChar10 that takes into account that in
@@ -693,8 +690,7 @@ class SaxesParser {
 
     this.i++;
     if (code < 0xD800) {
-      if ((code > 0x1F && code < 0x7F) || code > 0x9F) {
-        this.column++;
+      if ((code > 0x1F && code < 0x7F) || code > 0x9F || code === TAB) {
         return code;
       }
 
@@ -715,20 +711,15 @@ class SaxesParser {
       case NEL: // 0x85
       case LS: // Ox2028
         this.line++;
-        this.column = 0;
+        this.positionAtNewLine = this.position;
         return NL;
-      case TAB: // 9
-        this.column++;
-        return TAB;
       default:
-        this.column++;
         this.fail("disallowed character.");
         return code;
       }
     }
 
     if (code > 0xDBFF) {
-      this.column++;
       // This is a specialized version of isCharAndNotRestricted that takes into
       // account that in this context code > 0xDBFF and code <= 0xFFFF. So it
       // does not test cases that don't need testing.
@@ -746,7 +737,6 @@ class SaxesParser {
 
     const final = 0x10000 + ((code - 0xD800) * 0x400) +
           (chunk.charCodeAt(i + 1) - 0xDC00);
-    this.column += 2;
     this.i++;
 
     // This is a specialized version of isCharAndNotRestricted that takes into
@@ -924,11 +914,9 @@ class SaxesParser {
     // If the initial character is 0xFEFF, ignore it.
     if (c === 0xFEFF) {
       this.i++;
-      this.column++;
     }
     else if (isS(c)) {
       this.i++;
-      this.column++;
       // An XML declaration cannot appear after initial spaces.
       this.xmlDeclPossible = false;
     }

diff --git a/test/trailing-non-whitespace.js b/test/trailing-non-whitespace.js
@@ -19,7 +19,7 @@ require(".").test({
       attributes: {},
       isSelfClosing: false,
     }],
-    ["error", "1:36: text data outside of root node."],
+    ["error", "1:37: text data outside of root node."],
     ["text", " to monkey land"],
     ["end", undefined],
     ["ready", undefined],