CODEWITHSUNDEEP
htmlxhtml
Jack B.
Jack B.

Reputation: 957

Is HTML5 valid XML?

I am confused. A co-worker turned me on to the possibility that tags ending in />, such as <br /> can still be used in HTML5. I thought that only <br>-style could be used. All of the "talk" across the Internet is about using the latter.

Could someone please explain this to me? This seems very confusing and poorly documented.

And this brings up another question: Is HTML 5 considered to be well-formed XML?

Upvotes: 53

Views: 35859

Answers (5)

Ciro Santilli OurBigBook.com
Ciro Santilli OurBigBook.com

Reputation: 382762

No. Counter-examples:

These are valid HTML5 but invalid XHTML5:

  1. Some closing tags can be omitted:

    <p>First
<p>Second

    See: P-end-tag (</p>) is not needed in HTML

  2. script escape magic:

    <script><a></script>

    See: What is CDATA in HTML?

  3. Attributes without values (boolean attributes):

    <input type="text" disabled />

    See: What is the correct value for the disabled attribute?

  4. Attributes without quotes, e.g.:

    <div data-a=b></div>

    See: In XHTML 1.0 Strict do attribute values need to be surrounded with quotes?

  5. Implicit open elements and multiple top level tags.

    Some HTML elements are created implicitly. E.g. html. This allows the HTML to have "multiple top level elements":

    <!doctype html><title>a</title><p>a</p>

    See: Is it necessary to write HEAD, BODY and HTML tags?

Valid XHTML that is invalid HTML:

  1. Void elements with explicit closing tag:

    <hr></hr>

    See: HTML 5: Is it <br>, <br/>, or <br />?

  2. CDATA constructs with invalid tags inside

  3. ENTITY and other exclamation mark constructs, which allow for billion laughs: How does the billion laughs XML DoS attack work?

Valid HTML and XHTML but with different meanings:

  1. HTML has hundreds of named character references (e.g. &pound;, &copy;), XML has only 5 (quot, amp, apos, lt, gt).

Upvotes: 56

unor
unor

Reputation: 96577

You can markup your page as valid HTML5 and XHTML5: http://www.w3.org/TR/html-polyglot/

Polyglot markup that meets a well defined set of constraints is interpreted as compatible, regardless of whether they are processed as HTML or as XHTML, per the HTML5 specification.

The basic document could look like:

<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">
  <head>
    <title></title>
  </head>
  <body>
  </body>
</html>

Of course you'd have to follow some additional rules (like not to use the noscript element, for example), outlined in the linked working draft.

Upvotes: 11

ChrisLively
ChrisLively

Reputation: 88054

Under no condition should you expect any html document (regardless of version) to be "well-formed xml"

html != xml.

It is a different spec with different suggestions (I'm purposely avoiding the word "rules" here) on how it should be interpreted.

The HTML 5 spec has enough "do it this way, but it's okay if you don't" wiggle statements that it's a wonder that any of the browsers show the same thing at all.

Upvotes: 6

Kevin Ji
Kevin Ji

Reputation: 10489

HTML5 can be written with or without self-closing slashes; it is meant to be backwards-compatible with both HTML 4.01 and XHTML 1.0 code, so that it is easy to convert code into valid HTML5. There is an XML serialization called XHTML5, but for backwards-compatibility purposes with IE browsers, it is not recommended to be used. So technically, HTML5 is not considered to be well-formed XML.

Upvotes: 3

Matthew Flaschen
Matthew Flaschen

Reputation: 284796

There is an XML serialization of it, called XHTML5. Basically, you're free to use either HTML5 (HTML serialization) or XHTML5 (XML serialization). The draft spec says HTML5 "is the format suggested for most authors," mainly for the same reasons people recommend text/html for XHTML 1.1.

Upvotes: 12

Related Questions