A user account is required in order to edit this wiki, but we've had to disable public user registrations due to spam.

To request an account, ask an autoconfirmed user on Chat (such as one of these permanent autoconfirmed members).

HTML vs. XHTML: Difference between revisions

From WHATWG Wiki
Jump to navigation Jump to search
(→‎Element-specific parsing: move underspecified item to end)
Line 291: Line 291:
* The <code>plaintext</code> element has a special parsing requirement in HTML. (It is, however, forbidden.) '''Do not use plaintext.'''
* The <code>plaintext</code> element has a special parsing requirement in HTML. (It is, however, forbidden.) '''Do not use plaintext.'''
* In HTML, a line feed that immediately follows a <code>pre</code>, <code>listing</code> or <code>textarea</code> start tag is ignored. '''Add line breaks before the element using HTML or CSS.'''
* In HTML, a line feed that immediately follows a <code>pre</code>, <code>listing</code> or <code>textarea</code> start tag is ignored. '''Add line breaks before the element using HTML or CSS.'''
* The following are void elements in HTML (see void elements in table): In head (<code>base</code>, <code>link</code>, <code>meta</code>), in body (<code>area</code>,<code>br</code>, <code>col</code>, <code>embed</code>, <code>hr</code>, <code>img</code>, <code>input</code>, <code>param</code>, and now also <code>link</code> and <code>meta</code>). '''Always use the self-closing version of these tags to avoid double-inclusion in HTML.'''
* <em>Many other special handling of edge cases and error conditions, not all of which are listed here, occur in HTML.</em> (such as?)
* <em>Many other special handling of edge cases and error conditions, not all of which are listed here, occur in HTML.</em> (such as?)
* The following are void elements in HTML (see void elements in table): In head (<code>base</code>, <code>link</code>, <code>meta</code>), in body (<code>area</code>,<code>br</code>, <code>col</code>, <code>embed</code>, <code>hr</code>, <code>img</code>, <code>input</code>, <code>param</code>, and now also <code>link</code> and <code>meta</code>). '''Always use the self-closing version of these tags to avoid double-inclusion in HTML.'''


==== HTML Elements with Optional Tags ====
==== HTML Elements with Optional Tags ====

Revision as of 08:11, 21 August 2011

Differences Between HTML and XHTML

This page is currently being revised. Some information is incomplete or missing.

Please note that the information in here is based upon the current spec for (X)HTML5. Some of the issues technically do not apply to previous versions of HTML.

Although HTML and XHTML appear to have similarities in their syntax, they are significantly different in many ways.

Note: As the current WHATWG document is a draft, this section will need to track to a moving target.

The document at http://dev.w3.org/html5/html-xhtml-author-guide/html-xhtml-authoring-guide.html provides a similar analysis.

Overlap Language

There is a community who find it valuable to be able to serve HTML5 documents which are also valid XML documents. They may, for example, use XML tools to generate the document, and they and others may process the document using XML tools. These documents are served as text/html.

This language is sometimes called "polyglot". It is the overlap language of documents which are both HTML5 documents and XML documents. Guidelines are listed below for how one can construct such a polyglot document which will work in either environment. Besides following the well-formedness rules of XML, there are some other restrictions to which one must adhere (for the sake of text/html documents).

This wiki web page is an example of such a document. You can parse it with an XML parser or an HTML parser.

MIME Types

Feature HTML Requirement XHTML Requirement Notes
Mime Type Must use text/html. Must use an XML MIME type, such as application/xml or application/xhtml+xml. It is the MIME type that determines what type of document you are using. Any document, including a document authored with the intention of being XHTML, served as text/html is technically an HTML document.

Note that XHTML 1.0 previously defined that documents adhering to the compatibility guidelines were allowed to be served as text/html, but HTML 5 now defines that such documents are HTML, not XHTML.

Syntax and Parsing

XHTML uses XML parsing requirements. HTML uses its own which are defined much more closely to the way browsers actually handle HTML today. The following table describes the differences between how each is parsed.

The column on "Guidance for XHTML-HTML compatibility" lists ways in which a document can be crafted to work in either XHTML or HTML. The item will be bolded if it is a requirement for XHTML-compliant code to be changed, since XHTML will otherwise usually work as HTML, at least if its full features are constrained.

Feature HTML Requirement XHTML Requirement Notes Guidance for XHTML-HTML compatibility
Parsing Modes Three parsing modes are defined: no quirks mode, quirks mode and limited quirks mode. The mode is only ever changed from the default by the HTML parser, based on the presence, absence, or value of the DOCTYPE string, respectively. XML parsing rules are used. There is only one mode. The parsing modes in HTML also have an effect upon script and stylesheet processing. XHTML is considered to be in no quirks mode for these purposes. Use an explicit <!DOCTYPE html> (case insensitively) or legacy-compat version <!DOCTYPE html SYSTEM "about:legacy-compat"> for the sake of HTML and thus trigger no quirks parsing.
Error Handling HTML does not have a well-formedness constraint, no errors are fatal. Graceful error handling and recovery procedures are thoroughly defined. Well-formedness errors are fatal Ensure there are no well-formedness errors.
Character Encoding (including XML Declaration, meta) The XML declaration is forbidden (treated as a bogus comment, but such style of comments are deprecated), but the meta element with a charset attribute may be used instead.

If the encoding is unspecified in HTML, it should be determined through implementation specific heuristics or fallback to a default value (Note: this section of the spec is not yet finished).

The XML declaration may be used to specify the character encoding, while meta is only allowed as case-insensitive "UTF-8" (and is ignored if included).

The default character encoding for XHTML is, according to XML rules, UTF-8 or UTF-16.

If you need to include XML 1.1-only markup, if you do not wish to convert the encoding of the document to UTF-8 or UTF-16 (since use of other encodings also requires a declaration), or if you wish to define an external SYSTEM DTD in the DOCTYPE but use standalone=yes (redundant?), you must use an XML Declaration for XHTML, but this may not be allowable in the future in HTML. For future compatibility, it would be best to avoid XML 1.1-only markup, convert to UTF-8 or UTF-16 (probably UTF-8 which could allow use of a meta tag), and avoid use of a SYSTEM DTD (rendering the standalone=yes unnecessary), respectively. Do not use a meta tag, unless it is UTF-8 (and included in the first 512 bytes of the document), in which case it is probably a good idea to include it for the sake of HTML (as <meta charset="UTF-8">) in case you cannot specify such in a content header.
Namespaced elements Elements and attributes for known vocabularies (HTML, SVG and MathML) are implicitly assigned to appropriate namespaces, according to the rules specified in the parsing algorithm. Elements in the HTML, SVG, or MathML namespaces may have an xmlns attribute explicitly specified, if, and only if, it has the exact value "http://www.w3.org/1999/xhtml" (see namespace declaration). The attribute has absolutely no effect. It is basically a talisman. It is allowed merely to make migration to and from XHTML mildly easier. When parsed by an HTML parser, the xmlns attribute itself ends up in no namespace. Foreign elements are also not treated as being in another namespace and will have no effect except for displaying by default as inline elements (and be aware that self-closing elements cannot be used as such since unrecognized elements will be treated as though they are non-void; thus one cannot, for example, type <caesura /> in HTML or it will be treated as though there is no immediate closing tag). Namespaced prefixes are not allowed on HTML elements; a prefixed xmlns attribute cannot be used even if it is defined in the XHTML namespace. The XHTML namespace must be declared for HTML elements according to the rules defined by the Namespaces in XML specification. Namespaces must be explicitly declared. The xmlns attribute ends up in the "http://www.w3.org/2000/xmlns" namespace. Foreign elements can be used independently of HTML elements, as long as they are assigned to their own namespace. Declare HTML namespaces (or other namespaces) explicitly and do not prefix XHTML elements. Do not depend on the behavior of foreign namespaced elements in an HTML setting; if you need to include these, you will probably wish to set this foreign markup via CSS to display:none. You should explicitly close (not self-close) all empty elements defined in a non-XHTML namespace, since otherwise when used in HTML, HTML will treat them as though they have not been closed.
Namespaced attributes on HTML elements Attributes of the form xmlns:prefix may not be used on HTML elements. The xmlns:prefix attributes end up in the "http://www.w3.org/2000/xmlns" namespace. Do not use namespaced attributes on HTML elements. Do not depend on the behavior of foreign attributes in an HTML setting.
Namespace attributes on foreign elements

Elements in the SVG namespace may have an xmlns attribute specified, if, and only if, it has the exact value "http://www.w3.org/2000/svg". The attribute is optional because the namespace is implied during parsing.

Elements in the MathML namespace may have an xmlns attribute specified, if, and only if, it has the exact value "http://www.w3.org/1998/Math/MathML". The attribute is optional because the namespace is implied during parsing.

Foreign elements may also have an xmlns:xlink attribute specified, if, and only if, it has the exact value "http://www.w3.org/1999/xlink". This attribute is optional, even if XLink attributes are used, because the namespaces for XLink attributes is implied during parsing.

When parsed by an HTML parser, the xmlns and xmlns:xlink attributes end up in the "http://www.w3.org/2000/xmlns" namespace.

The SVG and MathML namespaces must be declared for SVG and MathML elements, respectively, according to the rules defined by Namespaces in XML. The xmlns and xmlns:prefix attributes end up in the "http://www.w3.org/2000/xmlns" namespace.
XLink attributes Foreign elements may use the attributes xlink:actuate, xlink:arcrole, xlink:href, xlink:role, xlink:show, xlink:title and xlink:type. These attributes are placed in the "http://www.w3.org/1999/xlink". The prefix used must be "xlink". XLink attributes may be specified on foreign elements using any prefix, subject to the conformance rules defined by Namespaces in XML. The XLink namespace must be declared according to the conformance rules defined by Namespaces in XML if XLink attributes are used within the document. Do not use XLink attributes on HTML elements and do not depend on them on foreign elements as will not work as such in HTML. If being used, ensure they have the appropriate XLink namespace defined.
XML attributes

Foreign elements may use the attributes xml:lang, xml:id, xml:base and xml:space. These attributes are placed in the "http://www.w3.org/XML/1998/namespace". The prefix used must be "xml".

HTML elements may use the xml:lang attribute. The attribute in no namespace with no prefix and with the literal localname "xml:lang" has no effect on language processing (as does "lang". HTML elements must not use the xml:base, xml:space, or xml:id attributes.

Any element, including HTML elements, may use the attributes xml:lang, xml:id, xml:base and xml:space. These attributes are placed in the "http://www.w3.org/XML/1998/namespace". The prefix used must be "xml". Though they can be used on foreign elements, do not use xml:base, xml:id, or xml:space on HTML elements; use both xml:lang and lang attributes whenever one is to be needed on HTML elements.
Attributes Names are not case sensitive. Attribute minimization is allowed (i.e. omitting the equals sign and the value). Names are case sensitive (and lower case). Attribute minimization is not allowed. Use lower case attribute names. Do not minimize attributes. Non-namespaced attributes not belonging to HTML will be included in the DOM tree and accessible to script and stylesheets, but it is discouraged to use these due to the potential for future naming conflicts; data- attributes can be used instead, or if in an XML-only environment, namespaced attributes.
Attribute values White space characters are not normalized. Unquoted attribute values are allowed. Fixed or default attribute values ...? White space characters are normalized to single spaces (unless attribute is of CDATA type?). Unquoted attribute values are not allowed. Default attribute values could conceivably be defined with a DTD. Create whitespace in attribute values which is already normalized (converted to single spaces). Always quote attribute values. Do not rely on defining default or fixed attribute values (or elements with exclusively element content) in a DTD (unless it matches HTML behavior).
Space characters The space characters are defined as:
  • U+0009 CHARACTER TABULATION
  • U+000A LINE FEED
  • U+000C FORM FEED
  • U+000D CARRIAGE RETURN
  • U+0020 SPACE
The space characters are defined as:
  • U+0009 CHARACTER TABULATION
  • U+000A LINE FEED
  • U+000D CARRIAGE RETURN
  • U+0020 SPACE
The difference is the inclusion of Form Feed. Form feed characters are discouraged in XML 1.1. Do not use the form feed character.
The DOCTYPE

A DOCTYPE is a mostly useless, but required, header. The DOCTYPE is used during parsing to determing the parsing mode. The keywords "DOCTYPE", "PUBLIC" and "SYSTEM", and the name "html" are treated case insensitively. The system identifier "about:legacy-compat" (and the public and system identifiers for previous versions of HTML) are case sensitive.

Conforming HTML documents are required to use <!DOCTYPE html> (case insensitively) or the legacy-compat version <!DOCTYPE html SYSTEM "about:legacy-compat">.

When using the obsolete but conforming DOCTYPEs based on the HTML 4.0 and 4.01 Strict DTDs, the system identifier is optional. The obsolete but conforming DOCTYPEs based on XHTML 1.0 Strict and XHTML 1.1 may also be specified.

Use of an internal subset is forbidden. The system identifier is never de-referenced by HTML implementations.

The DOCTYPE is optional. XML rules for case sensitivity apply (everything is case sensitive).

Either of the DOCTYPEs defined in HTML5 may be used, or any other custom DOCTYPE. If the public identifier is specified, the system identifier must also be specified. The obsolete status of the obsolete permitted DOCTYPEs defined for HTML does not apply to XHTML. Any DOCTYPE may be used, subject to the conformance rules defined by XML.

Use of an internal subset is permitted according to the requirements of XML. Some validating XML processors may dereference the system identifier, if used, but most browsers use non-validating processors.

Use the empty DOCTYPE with no SYSTEM or PUBLIC identifiers and no use of internet subset.
Element names Element names are case insensitive. Element names are case sensitive and lower-case. Only use lower-case element names (as with attributes).
Void vs. Non-void Elements Void elements only have a start tag; end tags must not be specified for void elements, and it is impossible for them to contain any content. A trailing slash may optionally be inserted at the end of the element's tag, immediately before the closing greater-than sign. For non-void elements (e.g., <script>), the trailing slash is a parsing error (ignored and thus treated as unclosed). Void elements may use either the empty-element tag syntax (EmptyElemTag) or use a start tag immediately followed by an end tag, with no content in between. While it is possible for the element to contain content, this is non-conforming. For void elements (e.g., <br />), do not include content or use a closing tag; only use a self-closing element with closing slash at the end (with a space preceding it for the sake of older browsers). For non-void elements, i.e., where content can exist (e.g., <script>), always use an explicit closing tag (not a self-closing tag) even if there is no content.
Unexpected end tags Unexpected end tags (in HTML, an unexpected </br> or </p> can cause the start tag to be implied before it). Unexpected end tags are well-formedness errors. Do not add end tags unless there is an explicit and properly nested open tag before it.
End tag with attributes ? An end tag with attributes is not allowed. Do not use end tags with attributes.
Raw text elements
RCDATA elements
Foreign elements
Normal elements
Optional tags

For some elements, the start and/or end tags are optional and are implied by certain specified conditions. For example, the end tag for the p element is implied by a subsequent p element.

Omitting the end tag for other elements is a parse error and various error recovery procedures are applied appropriately.

End tags must be explicitly included for all elements, except empty elements using the EmptyElemTag syntax. Always use end tags (or self-closing tags for void elements).
Comment syntax Comments must start with the four character sequence "<!--" and must be ended by the three character sequence "-->" (bogus comments such as those beginning with "<?" are deprecated). The content of comments must not start with a single U+003E GREATER-THAN SIGN ('>') character, nor start with a U+002D HYPHEN-MINUS (-) character followed by a U+003E GREATER-THAN SIGN ('>') character, nor contain two consecutive U+002D HYPHEN-MINUS (-) characters, nor end with a U+002D HYPHEN-MINUS (-) character. Violating these constraints is a parse error and various error recovery procedures are applied appropriately. The content of comments must not contain two consecutive U+002D HYPHEN-MINUS (-) characters, nor end with a hyphen. Violating this is a well-formedness error. Only use comments of the "<!--...-->" variety. Do not use two consecutive U+002D HYPHEN-MINUS (-) characters in comment content or end with such a hyphen (especially for the sake of XML). Do not begin comments with a single U+003E GREATER-THAN SIGN ('>') character, nor with a U+002D HYPHEN-MINUS (-) character followed by a U+003E GREATER-THAN SIGN ('>') character.
Processing Instructions HTML does not allow processing instructions and deprecates the bogus comments which appear in their form, whether in the form <?foo ...> (without a closing '?') or <?foo ...?>. XHTML allows the use of XML processing instructions which are only closed by "?>". Avoid ">" inside processing instructions (as these will close the "instruction" (comment) prematurely) (or one must strip out processing instructions entirely). Processing instructions might need to be avoided entirely in case HTML may in future disallow them completely.
CDATA sections <![CDATA[...]]> is a a bogus comment. The sequence of characters "]]>" in content when it does not mark the end of a CDATA section is just regular character data. <![CDATA[...]]> is a CDATA section. The sequence of characters "]]>" in content when it does not mark the end of a CDATA section is a well-formedness error. Ensure sequence "]]>" in content is escaped (not necessary to escape in attribute values). Do not use CDATA sections.
Unescaped Special Characters

Unescaped ampersands (U+0026 AMPERSAND - &, instead of &amp;) are permitted within the content of normal elements, RCDATA elements, foreign elements and attribute values where they are not considered to be ambiguous ampersands, and within Raw text elements.

Unescaped less than signs (U+003C LESS-THAN SIGN - <, instead of &lt;) are permitted in Raw text elements, RCDATA elements and attribute values, excluding the unquoted attribute value syntax.

Unescaped ampersands and less-than signs may not appear within CharData or AttValue (basically, the normal text content of elements and attribute values.) Violation of this constraint is a well-formedness error. Always escape ampersands and less-than signs in text content and attribute values. See CDATA for need to escape sequence "]]>" in text content.
Character References
Entity References In HTML, all entity references are predefined and do not require a DTD. There is no formal DTD for XHTML5, but one could provide an exteranl DTD (if not an internal subset?) for use with one's entity-checking (or validating) parser, but be aware that browsers do not universally use external entity-checking (or validating) parsers and may not read the external DTD. (Some still have bugs in that they mistakenly create a well-formedness error out of such missing entities instead of showing them as missing, making them clickable, or using a entity-checking or validating parser.) Do not use entity references in XHTML (except for the 5 predefined entities: &amp;, &lt;, &gt;, &quot; and &apos;); use the equivalent Unicode or numeric character reference sequence instead.
Character data The valid set of unicode characters in XML 1.0 is limited beyond that in HTML (we need to specify this here).

Element-specific parsing

  • In HTML, the script and style elements are parsed as CDATA elements. (Note: the definition of CDATA differs from that in XML). In XML, they're parsed as normal elements (which means that things that look like comments are treated as real comments, and things that look like start tags actually are start tags).

The following code with escaping can ensure script and style elements will work in both XHTML and HTML, including older browsers.

In both cases, XML ignores the first comment and then uses the CDATA section to avoid the need for escaping special characters < and & within the rest of the contents (with subsequent JavaScript comments added within to ensure the HTML-oriented code is ignored by JavaScript).

In HTML, older browsers might display the content without the content being within a comment, so comments are used to hide this from them (while modern HTML browsers will run code inside the comments). The subsequent JavaScript comment is added to negate the text added for the sake of XHTML.

The <style> requires the /**/ comments since CSS does not support the single line ones.

   <script type="text/javascript"><!--//--><![CDATA[//><!--
       ...
   //--><!]]></script>


   <style type="text/css"><!--/*--><![CDATA[/*><!--*/
       ...
   /*]]>*/--></style>

If not concerned about much older browsers (from which one is hiding the HTML) one can use the simpler:

   <script>//<![CDATA[
   
   //]]></script>
   <style>/*<![CDATA[*/
   
   /*]]>*/</style>

Also note that the sequence "]]>" is not allowed within a CDATA section, so it cannot be used in true XHTML-embedded JavaScript without escaping.

  • In HTML, the title and textarea elements are parsed as RCDATA elements. (Note: The definition of RCDATA differs from that in SGML and there is no RCDATA in XML). Use &amp; and &lt; escape forms (and "]]&gt;" if the sequence "]]>" is required) within these elements even though HTML does not require them (CDATA sections apparently cannot be added here in a polyglot-supportive fashion).
  • In HTML, if scripting is enabled, the noscript element is parsed as an CDATA element. If scripting is disabled, it's parsed as a normal element. In XHTML, the element is always parsed as a normal element, and can't really be used to stop content from being present when script is disabled. Add content to the page which should be shown when JavaScript is disabled and use JavaScript to hide these elements when the page has loaded (DOMContentLoaded can be used for modern browsers).
  • In HTML, the iframe, noembed and noframes elements are parsed as CDATA elements. In XHTML, they are parsed as normal elements, and therefore do not stop content from being used. Do not add content for these elements or hide them by JavaScript.
  • In HTML, tags for certain elements, which appear out of context, are ignored. This includes caption, col, colgroup, frame, frameset, head, option, optgroup, tbody, td, tfoot, th, thead, tr. Do not use these elements out of context.
  • In XHTML, table elements may contain child tr elements. In the HTML serialisation, due to backwards compatibility constraints, this is not possible (though it may be done through DOM manipulation). One may use an explicit tbody to avoid potential confusion.
  • The plaintext element has a special parsing requirement in HTML. (It is, however, forbidden.) Do not use plaintext.
  • In HTML, a line feed that immediately follows a pre, listing or textarea start tag is ignored. Add line breaks before the element using HTML or CSS.
  • The following are void elements in HTML (see void elements in table): In head (base, link, meta), in body (area,br, col, embed, hr, img, input, param, and now also link and meta). Always use the self-closing version of these tags to avoid double-inclusion in HTML.
  • Many other special handling of edge cases and error conditions, not all of which are listed here, occur in HTML. (such as?)

HTML Elements with Optional Tags

Element Start Tag End Tag
html optional optional
head optional optional
body optional optional
li required optional
dt required optional
dd required optional
p required optional
colgroup optional optional
thead required optional
tbody optional optional
tfoot required optional
tr required optional
th required optional
td required optional
rt required optional
rp required optional
optgroup required optional
option required optional

Scripts

  • document.write() and document.writeln() cannot be used in XHTML, they can in HTML.
  • In XHTML, the use of the innerHTML property requires that the string be a well-formed fragment of XML.
  • DOM APIs are case sensitive in XHTML and some are case insensitive in HTML. (This does not apply to elements which are not in the HTML namespace)
    • Element.tagName and Node.nodeName return the value in uppercase in HTML but lower-case in XHTML (Node.localName is consistent now, as of HTML5).
    • Document.createElement() is case insensitive (the canonical form is lowercase).
    • Element.setAttributeNode() will change the attribute name to lowercase.
    • Element.setAttribute() is case insensitive (the canonical form is lowercase).
    • Document.getElementsByTagName() and Element.getElementsByTagName() are case insensitive.
    • Document.renameNode(). If the new namespace is the HTML namespace, then the new qualified name will be lowercased before the rename takes place.
  • In HTML, Document.createElement() will create an element in the HTML namespace. In XML (including XHTML), the namespace is defined by both DOM2 and DOM3 to be null.
    • In XHTML, browsers lack interoperability in this area. In Firefox and Safari, the namespace is dependent upon the MIME type. In Opera, it's dependent upon the root element.
  • XPath expressions targeted at pre-HTML5 browsers need to use the XHTML namespace for XHTML and null for HTML. (HTML5 browsers would use the XHTML namespace even in HTML.)

Stylesheets

  • Selectors, as used in CSS, match case sensitively in XHTML, but case insensitively in HTML.
  • CSS requires special handling of the body element in HTML for painting backgrounds on the canvas, which do not apply to XHTML.
  • For polyglot documents, use lower-case element selectors and style the html and body elements appropriately (?).

Differences Between HTML4 and HTML5

See HTML5 differences from HTML4.

Differences Between DOM Level 2.0, 3.0 and the HTML 5 DOM APIs

This section might belong on a separate page.

  • TODO (need to talk about the changes to the DOM API that HTML5 is making, compared with DOM2 and DOM3)

Translations