Fix index computation after search() in <meta> extraction

openandclose · domenic · commit a417f6bfda15 · 2020-02-23T17:26:28.000-05:00
Closes #6. Closes #7.
diff --git a/lib/html-encoding-sniffer.js b/lib/html-encoding-sniffer.js
@@ -239,12 +239,12 @@ function extractCharacterEncodingFromMeta(string) {
   let position = 0;
 
   while (true) {
-    let subPosition = string.substring(position).search(/charset/i);
+    const indexOfCharset = string.substring(position).search(/charset/i);
 
-    if (subPosition === -1) {
+    if (indexOfCharset === -1) {
       return null;
     }
-    subPosition += "charset".length;
+    let subPosition = position + indexOfCharset + "charset".length;
 
     while (isSpaceCharacter(string[subPosition].charCodeAt(0))) {
       ++subPosition;
@@ -280,10 +280,11 @@ function extractCharacterEncodingFromMeta(string) {
     return null;
   }
 
-  let end = string.substring(position + 1).search(/\x09|\x0A|\x0C|\x0D|\x20|;/);
-  if (end === -1) {
-    end = string.length;
-  }
+  const indexOfASCIIWhitespaceOrSemicolon = string.substring(position + 1).search(/\x09|\x0A|\x0C|\x0D|\x20|;/);
+  const end = indexOfASCIIWhitespaceOrSemicolon === -1 ?
+    string.length :
+    position + indexOfASCIIWhitespaceOrSemicolon + 1;
+
   return whatwgEncoding.labelToName(string.substring(position, end));
 }
 
diff --git a/test/no-bom-charset-http-equiv-second-charset.html b/test/no-bom-charset-http-equiv-second-charset.html
@@ -0,0 +1,7 @@
+<!DOCTYPE html>
+<html>
+    <head>
+        <meta http-equiv=Content-Type content="text/html;charsetcharset=iso-8859-2">
+    </head>
+    <body></body>
+</html>
diff --git a/test/no-bom-charset-http-equiv-trailing-space.html b/test/no-bom-charset-http-equiv-trailing-space.html
@@ -0,0 +1,7 @@
+<!DOCTYPE html>
+<html>
+    <head>
+        <meta http-equiv=Content-Type content="text/html;charset=iso-8859-2 ">
+    </head>
+    <body></body>
+</html>
diff --git a/test/tests.js b/test/tests.js
@@ -233,6 +233,62 @@ describe("A file with no BOM and a <meta charset> preceeded by a short comment <
   });
 });
 
+describe("A file with no BOM and a <meta http-equiv> ending with a trailing space", () => {
+  const buffer = read("no-bom-charset-http-equiv-trailing-space.html");
+
+  it("should sniff as the charset value, given no options", () => {
+    const sniffedEncoding = htmlEncodingSniffer(buffer);
+
+    assert.strictEqual(sniffedEncoding, "ISO-8859-2");
+  });
+
+  it("should sniff as the transport layer encoding, given that", () => {
+    const sniffedEncoding = htmlEncodingSniffer(buffer, {
+      transportLayerEncodingLabel: "windows-1251",
+      defaultEncoding: "ISO-8859-16"
+    });
+
+    assert.strictEqual(sniffedEncoding, "windows-1251");
+  });
+
+
+  it("should sniff as the charset value, given only a default encoding", () => {
+    const sniffedEncoding = htmlEncodingSniffer(buffer, {
+      defaultEncoding: "ISO-8859-16"
+    });
+
+    assert.strictEqual(sniffedEncoding, "ISO-8859-2");
+  });
+});
+
+describe("A file with no BOM and a <meta http-equiv> with 'charsetcharset'", () => {
+  const buffer = read("no-bom-charset-http-equiv-second-charset.html");
+
+  it("should sniff as the charset value, given no options", () => {
+    const sniffedEncoding = htmlEncodingSniffer(buffer);
+
+    assert.strictEqual(sniffedEncoding, "ISO-8859-2");
+  });
+
+  it("should sniff as the transport layer encoding, given that", () => {
+    const sniffedEncoding = htmlEncodingSniffer(buffer, {
+      transportLayerEncodingLabel: "windows-1251",
+      defaultEncoding: "ISO-8859-16"
+    });
+
+    assert.strictEqual(sniffedEncoding, "windows-1251");
+  });
+
+
+  it("should sniff as the charset value, given only a default encoding", () => {
+    const sniffedEncoding = htmlEncodingSniffer(buffer, {
+      defaultEncoding: "ISO-8859-16"
+    });
+
+    assert.strictEqual(sniffedEncoding, "ISO-8859-2");
+  });
+});
+
 for (const utf16Encoding of ["utf-16be", "utf-16", "utf-16le"]) {
   describe(`A file with a BOM and a <meta charset> of ${utf16Encoding}`, () => {
     const buffer = read(`no-bom-charset-${utf16Encoding}.html`);
