This document is also available in this non-normative format: diff to previous version
Copyright © 2009-2013 W3C® (MIT, ERCIM, Keio, Beihang), All Rights Reserved. W3C liability, trademark and document use rules apply.
This specification defines rules and guidelines for adapting the RDFa Core 1.1 and RDFa Lite 1.1 specifications for use in HTML5 and XHTML5. The rules defined in this specification not only apply to HTML5 documents in non-XML and XML mode, but also to HTML4 and XHTML documents interpreted through the HTML5 parsing rules.
This section describes the status of this document at the time of its publication. Other documents may supersede this document. A list of current W3C publications and the latest revision of this technical report can be found in the W3C technical reports index at http://www.w3.org/TR/.
This specification had been jointly developed by the RDFa Working Group and the HTML Working Group. The document was previously published via the HTML Working Group, but has since been transitioned to the newly rechartered RDFa Working Group.
Changes in this version of the specification include:
link
and meta
elements validity in the body of a document to require the use of @property
. (non-substantive)@datetime
processing to be easier to understand for implementers. (editorial)@datetime
and rdf:HTML
features are non-normative features from non-REC documents and that once those features are published in REC documents that a Proposed Edited Recommendation will be published for HTML+RDFa 1.1 making the features normative. (non-substantive)This specification skipped the Candidate Recommendation phase due to having four fully interoperable implementations available before entering the Candidate Recommendation phase. The final implementation report considered by the Director has been made available to the public.
This specification is an extension to the HTML5 language. All normative content in the HTML5 specification, unless specifically overridden by this specification, is intended to be the basis for this specification.
There are two features in this
specification, @datetime
processing and rdf:HTML
literals, that are currently defined as non-normative features. The intent
is that these features will eventually become normative features once the
specification that describes the @datetime
attribute [HTML5]
and the specification that defines rdf:HTML
[RDF-CONCEPTS]
become W3C Recommendations.
Implementers should implement these features now;
a 2nd (or later) edition of this specification will clarify the long-term
stability of those features. Based on the discussion between the RDFa
Working Group, the HTML Working Group, and the RDF Working Group,
it is not expected that implementation changes will be necessary as
HTML5 and RDF 1.1 advance to Recommendation.
There has been a single formal objection filed on this specification, arguing that the use of prefixes is too complicated for a Web technology. The RDFa WG made changes based on the commenters feedback. The commenter did not respond to the changes made. Three requests for feedback were made over the course of several months with no response from the commenter. The RDFa WG does not believe the changes made would be enough to address the commenters request, which asked that that prefixes should be removed entirely or hard-coded.
A sample test harness is available for software developers. This set of tests is not intended to be exhaustive. A community-maintained website contains more information on further reading, developer tools, and software libraries that can be used to extract RDFa data from web documents.
This document was published by the RDFa Working Group as a Proposed Recommendation. This document is intended to become a W3C Recommendation. The W3C Membership and other interested parties are invited to review the document and send comments to public-rdfa-wg@w3.org (subscribe, archives) through 23 July 2013. Advisory Committee Representatives should consult their WBS questionnaires. Note that substantive technical comments were expected during the Last Call review period that ended 28 February 2013.
Publication as a Proposed Recommendation does not imply endorsement by the W3C Membership. This is a draft document and may be updated, replaced or obsoleted by other documents at any time. It is inappropriate to cite this document as other than work in progress. The Proposed Recommendation review period ends on 23 July 2013.
This document was produced by a group operating under the 5 February 2004 W3C Patent Policy. W3C maintains a public list of any patent disclosures made in connection with the deliverables of the group; that page also includes instructions for disclosing a patent. An individual who has actual knowledge of a patent which the individual believes contains Essential Claim(s) must disclose the information in accordance with section 6 of the W3C Patent Policy.
This section is non-normative.
Today's web is built predominantly for human readers. Even as machine-readable data begins to permeate the web, it is typically distributed in a separate file, with a separate format, and very limited correspondence between the human and machine versions. As a result, web browsers can provide only minimal assistance to humans in parsing and processing web pages: browsers only see presentation information. RDFa is intended to solve the problem of marking up machine-readable data in HTML documents. RDFa provides a set of HTML attributes to augment visual data with machine-readable hints. Using RDFa, authors may turn their existing human-visible text and links into machine-readable data without repeating content.
As well as sections marked as non-normative, all authoring guidelines, diagrams, examples, and notes in this specification are non-normative. Everything else in this specification is normative.
The key words MUST, MUST NOT, REQUIRED, SHOULD, SHOULD NOT, RECOMMENDED, MAY, and OPTIONAL in this specification are to be interpreted as described in [RFC2119].
There are two types of document conformance criteria for HTML documents containing RDFa semantics; HTML+RDFa and HTML+RDFa Lite.
The following conformance criteria apply to any HTML document including RDFa markup:
An example of a conforming HTML+RDFa document, with the RDFa portions highlighted in green:
<!DOCTYPE html> <html lang="en"> <head> <title>Example Document</title> </head> <body vocab="http://schema.org/"> <p typeof="Blog"> Welcome to my <a property="url" href="http://example.org/">blog</a>. </p> </body> </html>
[] a <http://schema.org/Blog>; <http://schema.org/url> <http://example.org/> .
Non-XML mode HTML+RDFa 1.1 documents SHOULD be labeled with the Internet
Media Type text/html
as defined in
section 12.1
of the HTML5 specification [HTML5].
XML mode XHTML5+RDFa 1.1 documents SHOULD be labeled with the Internet Media
Type application/xhtml+xml
as defined in
section 12.3
of the HTML5 specification [HTML5], MUST NOT use a DOCTYPE
declaration for XHTML+RDFa 1.0 or XHTML+RDFa 1.1, and SHOULD NOT use the
@version
attribute.
The RDFa processor conformance criteria are listed below, all of which are mandatory:
A user agent is considered to be a type of RDFa processor when the user agent stores or processes RDFa attributes and their values. The reason there are separate RDFa Processor Conformance and a User Agent Conformance sections is because one can be a valid HTML5 RDFa processor but not a valid HTML5 user agent (for example, by only providing a very small subset of rendering functionality).
The user agent conformance criteria are listed below, all of which are mandatory:
The RDFa Core 1.1 [RDFA-CORE] specification is the base document on which this specification builds. RDFa Core 1.1 specifies the attributes and syntax, in Section 5: Attributes and Syntax, and processing model, in Section 7: Processing Model, for extracting RDF from a web document. This section specifies changes to the attributes and processing model defined in RDFa Core 1.1 in order to support extracting RDF from HTML documents.
The requirements and rules, as specified in RDFa Core and further extended in this document, apply to all HTML5 documents. An RDFa processor operating on both HTML and XHTML documents, specifically on their resulting DOMs or infosets, MUST apply these processing rules for HTML4, HTML5 and XHTML5 serializations, DOMs and/or infosets.
Documents conforming to the rules in this specification are processed according to [RDFA-CORE] with the following extensions:
http://www.w3.org/2011/rdfa-context/html-rdfa-1.1
, which must
be applied after the initial context for [RDFA-CORE]
(http://www.w3.org/2011/rdfa-context/rdfa-1.1
).base
element. For XHTML5+RDFa 1.1 documents,
base can also be set using the @xml:base
attribute.@lang
or @xml:lang
attributes. When the @lang
attribute and the
@xml:lang
attribute are specified on the same element, the
@xml:lang
attribute takes precedence. When both
@lang
and @xml:lang
are specified on the same
element, they MUST have the same value.application/xhtml+xml
media type, a conforming
RDFa processor MUST look at the value in the DOCTYPE declaration of the
document. If a DOCTYPE declaration exists, then the
processing rules are:
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML+RDFa 1.0//EN" "http://www.w3.org/MarkUp/DTD/xhtml-rdfa-1.dtd">
, or<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML+RDFa 1.1//EN" "http://www.w3.org/MarkUp/DTD/xhtml-rdfa-2.dtd">
, orapplication/xhtml+xml
, that don't contain
a DOCTYPE declaration, and don't specify a @version
attribute MUST be interpreted
as XHTML5+RDFa 1.1 documents.@property
attribute and the @rel
and/or
@rev
attribute exists on the same element, the non-CURIE and
non-URI @rel
and @rev
values are ignored. If, after
this, the value of @rel
and/or @rev
becomes empty,
then the processor MUST act as if the respective attribute is not present.
@about
, @href
, @resource
, or
@src
), then first check to see if the element is the
head
or body
element. If it is, then set
new subject
to
parent object.
@datetime
attribute
MUST be utilized when generating
the current property value, unless @content
is also
present on the same element. Otherwise, if @datetime
is
present, the current property value must be generated as
follows. The literal value is the value contained in the
@datetime
attribute. If @datatype
is
present, it is to be used as defined in the RDFa Core [RDFA-CORE]
processing rules. Otherwise, if the value of
@datetime
lexically matches a valid
xsd:date
, xsd:time
, xsd:dateTime
,
xsd:duration
, xsd:gYear
, or
xsd:gYearMonth
a typed literal must be generated, with its
datatype set to the matching xsd datatype. Otherwise, a plain literal
MUST be generated, taking into account the
current language.
Implementers should note that the correct order of match testing should be:
xsd:duration
, xsd:dateTime
,
xsd:date
, xsd:time
,
xsd:gYearMonth
, and xsd:gYear
.
This feature is currently non-normative,
see the note on when it will become
normative.time
, and the element does not have @datetime
or @content
attributes, the processor MUST act as if there
is a @datetime
attribute containing exactly the elements
text value. This feature is currently non-normative,
see the note on when it will become
normative.@datatype
attribute is present and evaluates to
http://www.w3.org/1999/02/22-rdf-syntax-ns#HTML
,
the value of the HTML Literal is a string
created by serializing all child nodes to text. This applies to all nodes
that are descendants of the current
element, not including the element itself. The HTML Literal is
given a
datatype of http://www.w3.org/1999/02/22-rdf-syntax-ns#HTML
as defined in
Section 5.2: The rdf:HTML Datatype
of [RDF-CONCEPTS]. This feature is currently non-normative,
see the note on when it will become
normative.
The @version
attribute is not supported in HTML5 and is
non-conforming. However, if an HTML+RDFa document contains the
@version
attribute on the html
element, a conforming
RDFa processor MUST examine the value of this attribute. If the value matches
that of a defined version of RDFa, then the processing rules for that version
MUST be used. If the value does not match a defined version, or there is no
@version
attribute, then the processing rules for the most recent
version of RDFa 1.1 MUST be used.
RDFa's tree-based processing rules, outlined in Section 7.5: Sequence of the RDFa Core 1.1 specification [RDFA-CORE], allow an input document to be automatically corrected, cleaned-up, re-arranged, or modified in any way that is approved by the host language prior to processing. Element nesting issues in HTML documents SHOULD be corrected before the input document is translated into the DOM, a valid tree-based model, on which the RDFa processing rules will operate.
Any mechanism that generates a data structure equivalent to the HTML5 or XHTML5 DOM, such as the html5lib library, MAY be used as the mechanism to construct the tree-based model provided as input to the RDFa processing rules.
According to RDFa Core 1.1 the current language MAY be specified by the host language. In order to conform to this specification, RDFa processors MUST use the mechanism described in The lang and xml:lang attributes section of the [HTML5] specification to determine the language of a node.
If the final encapsulating MIME type for an HTML fragment is not decided
on while editing, it is RECOMMENDED that the author
specify both @lang
and @xml:lang
where the value in
both attributes is exactly the same.
The HTML5 specification takes the
Content-Language
HTTP header into account when determining the
language of an element. Some RDFa processor implementations, like those
written in JavaScript, may not have
access to this header and will be non-conforming in the edge case where
the language is only specified in the Content-Language
HTTP
header. In these instances, RDFa document authors are urged to
set the language in the document via the @lang
attribute on the html
element in order to ensure
that the document is interpreted correctly across all RDFa processors.
When generating literals of type XMLLiteral, the processor MUST ensure that the output XMLLiteral is a namespace well-formed XML fragment. A namespace well-formed XML fragment has the following properties:
@xmlns
and @xmlns:
that are stored in the
RDFa processor's current
evaluation context
in the
IRI mappings
MUST be preserved in the generated XMLLiteral. The PREFIX value for
@xmlns:PREFIX
MUST be entirely transformed into lower-case characters
when preserving the value in the XMLLiteral. All active namespaces declared
via @xmlns
, @xmlns:
, and @prefix
MUST be placed in each top-level element in the generated XMLLiteral,
taking care to not overwrite pre-existing namespace values.An RDFa processor that transforms the XML fragment MUST use the Coercing an HTML DOM into an infoset algorithm, as specified in the HTML5 specification, followed by the algorithm defined in the Serializing XHTML Fragments section of the HTML5 specification. If an error or exception occurs at any point during the transformation, the triple containing the XMLLiteral MUST NOT be generated.
Transformation to a namespace well-formed XML fragment is required because an application that consumes XMLLiteral data expects that data to be a namespace well-formed XML fragment.
The transformation requirement does not apply to plain text input data that are
text-only, such as literals that contain a @datatype
attribute
with an empty value (""
), or input data that contain only
text nodes.
An example transformation demonstrating the preservation of namespace values is provided below. The → symbol is used to denote that the line is a continuation of the previous line and is included purely for the purposes of readability:
<p xmlns:ex="http://example.org/vocab#" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"> Two rectangles (the example markup for them are stored in a triple): <svg xmlns="http://www.w3.org/2000/svg" property="ex:markup" datatype="rdf:XMLLiteral"> →<rect width="300" height="100" style="fill:rgb(0,0,255);stroke-width:1; stroke:rgb(0,0,0)"/> →<rect width="50" height="50" style="fill:rgb(255,0,0);stroke-width:2;stroke:rgb(0,0,0)"/></svg> </p>
The markup above SHOULD produce the following triple, which preserves the
xmlns declaration in the markup by injecting the @xmlns
attribute
in the rect
elements:
<> <http://example.org/vocab#markup> """<rect xmlns="http://www.w3.org/2000/svg" width="300" →height="100" style="fill:rgb(0,0,255);stroke-width:1; stroke:rgb(0,0,0)"/> →<rect xmlns="http://www.w3.org/2000/svg" width="50" →height="50" style="fill:rgb(255,0,0);stroke-width:2; →stroke:rgb(0,0,0)"/>"""^^<http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral> .
Since the ex
and rdf
namespaces are not used in either rect
element, they are not
preserved in the XMLLiteral.
Similarly, compound document elements that reside in different namespaces must have their namespace declarations preserved:
<p xmlns:ex="http://example.org/vocab#"
xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
xmlns:fb="http://www.facebook.com/2008/fbml">
This is how you markup a user in FBML:
<span property="ex:markup" datatype="rdf:XMLLiteral">
→<span><fb:user uid="12345">The User</fb:user></span>
→</span>
</p>
The markup above SHOULD produce the following triple, which preserves the
fb
namespace in the corresponding triple:
<>
<http://example.org/vocab#markup>
"""<span xmlns:fb="http://www.facebook.com/2008/fbml">
→<fb:user uid="12345"></fb:user>
→</span>"""^^<http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral> .
There are times when authors will find that they have many resources on a page that share a common set of properties. For example, several music events may have different performance times, but use the same location, band, and ticket prices. In this particular case, it is beneficial to have a short-hand notation to instruct the RDFa processor to include the location, band, and ticket price information without having to repeat all of the markup that expresses the data.
HTML+RDFa defines a property copying mechanism which allows
properties associated with a resource to be copied to another resource.
This mechanism can be activated by using the rdfa:copy
predicate.
The feature is illustrated in the following two examples:
<div vocab="http://schema.org/"> <p typeof="MusicEvent"> <link property="image" href="Muse1.jpg"/> <link property="image" href="Muse2.jpg"/> <link property="image" href="Muse3.jpg"/> <span property="name">Muse</span> at the United Center. <time property="startDate" datetime="20130403">March 3rd 2013</time>, <a property="location" href="#united">United Center, Chicago, Illinois</a> ... </p> <p typeof="MusicEvent"> <link property="image" href="Muse1.jpg"/> <link property="image" href="Muse2.jpg"/> <link property="image" href="Muse3.jpg"/> <span property="name">Muse</span> at the Target Center. <time property="startDate" datetime="20130703">March 7th 2013</time>, <a property="location" href="#target">Target Center, Minneapolis, Minnesota</a> ... </p> </div>
In this case, two music events are defined with image, name, startDate, and location properties. The image and name values are identical for the two events and are unnecessarily duplicated in the markup. Using RDFa's property copying feature, a pattern can be declared that expresses the common properties. This pattern can then be copied into other resources within the document:
<div vocab="http://schema.org/"> <div resource="#muse" typeof="rdfa:Pattern"> <link property="image" href="Muse1.jpg"/> <link property="image" href="Muse2.jpg"/> <link property="image" href="Muse3.jpg"/> <span property="name">Muse</span> </div> <p typeof="MusicEvent"> <link property="rdfa:copy" href="#muse"/> Muse at the United Center. <time property="startDate" datetime="20130403">March 3rd 2013</time>, <a property="location" href="#united">United Center, Chicago, Illinois</a> ... </p> <p typeof="MusicEvent"> <link property="rdfa:copy" href="#muse"/> Muse at the Target Center. <time property="startDate" datetime="20130703">March 7th 2013</time>, <a property="location" href="#target">Target Center, Minneapolis, Minnesota</a> ... </p> </div>
In this case, the common properties for all of the events are expressed in
the first div
. The common properties are copied into each
event resource via the rdfa:copy
predicate. The output for the
previous two examples is the same:
@prefix schema: <http://schema.org/> . [] a schema:MusicEvent; schema:image <Muse1.jpg>, <Muse2.jpg>, <Muse3.jpg>; schema:name "Muse"; schema:startDate "March 3rd 2013"; schema:location <#united> . [] a schema:MusicEvent; schema:image <Muse1.jpg>, <Muse2.jpg>, <Muse3.jpg>; schema:name "Muse"; schema:startDate "March 7th 2013"; schema:location <#target> .
The copy process is iterative, so that resources may copy other resources that copy other resources. For example:
<div vocab="http://schema.org/"> <div typeof="Person"> <link property="rdfa:copy" href="#lennon"/> <link property="rdfa:copy" href="#band"/> </div> <p resource="#lennon" typeof="rdfa:Pattern"> Name: <span property="name">John Lennon</span> <p> <div resource="#band" typeof="rdfa:Pattern"> <div property="band" typeof="MusicGroup"> <link property="rdfa:copy" href="#beatles"/> </div> </div> <div resource="#beatles" typeof="rdfa:Pattern"> <p>Band: <span property="name">The Beatles</span></p> <p>Size: <span property="size">4</span> players</p> </div> </div>
In the example above, the properties from #lennon
and
#band
are copied into the first resource. Then the
properties from #beatles
are copied into
#band
. Subsequently, those properties are again copied into
the first resource yielding the following output:
@prefix schema: <http://schema.org/> . [ a schema:Person; schema:name "John Lennon" ; schema:band [ a schema:MusicGroup; schema:name "The Beatles"; schema:size "4" ] ] .
Similar to Vocabulary Expansion as defined in [RDFA-CORE], RDFa Property Copying operates on the output graph after document processing is complete.
Once the output graph is generated following the processing steps defined in Section 7.5: Sequence of the RDFa Core 1.1 specification [RDFA-CORE], and the Extensions to the HTML5 Syntax defined in this specification, processors MUST update the output graph using the following rules:
rdfa:copy
statement
in the
output graph,
and for each new rdfa:copy
statement added as a result of
property copying until no new triples are added to the
output graph:
Rule name | If output graph contains | then add |
---|---|---|
pattern-copy |
?subject rdfa:copy ?target ?target rdf:type rdfa:Pattern ?target ?predicate ?object | ?subject ?predicate ?object |
rdfa:copy
statements and rdfa:Pattern
resources from the
output graph:
Rule name | If output graph contains | then remove |
---|---|---|
pattern-clean |
?subject rdfa:copy ?target ?target rdf:type rdfa:Pattern ?target ?predicate ?object |
?subject rdfa:copy ?target ?subject rdf:type rdfa:Pattern ?target ?predicate ?object |
Implementers should be aware that a simplistic implementation of the pattern-copy rule may lead to an infinite loop when handling circular dependencies. A processor should cease the pattern-copy rule when no unique triples are generated.
Alternate rules may be used to update the output graph as long as the end result is the same.
There are a few attributes that are added as extensions to the HTML5 syntax in order to fully support RDFa:
@vocab
,
@typeof
, @property
, @resource
, and
@prefix
. All other attributes that RDFa may process, such as
@href
and @src
, are only allowed when consistent
with the content model for that element,
as defined in the HTML5 specification.@vocab
,
@typeof
, @property
, @resource
,
@prefix
, @content
, @about
,
@rel
, @rev
, @datatype
, and
@inlist
. All other attributes that RDFa may process, such as
@href
and @src
, are only allowed when consistent
with the content model for that element,
as defined in the HTML5 specification.@property
RDFa attribute is present on the
link
or meta
elements, they MUST be viewed as
conforming if used in the body
of the document.
More specifically,
when link
or meta
elements contain the
RDFa @property
attribute and are used in the
body
of an HTML5 document, they MUST be considered
flow content.@property
attribute is present on the link
element, the @rel
attribute is not required.@resource
attribute is present on the link
element, the @href
attribute is not required.@property
attribute is present on the meta
element, neither the @name
, @http-equiv
, nor @charset
attributes are required
and the @content
attribute MUST be specified.RDFa Core 1.1 deprecates the usage of @xmlns:
in RDFa 1.1
documents. Web page authors SHOULD NOT use @xmlns:
to express
prefix mappings in RDFa 1.1 documents. Web page authors SHOULD use
the @prefix
attribute to specify prefix mappings.
However, there are times when XHTML+RDFa 1.0 documents are served by web
servers using the text/html
MIME type. In these instances, the
HTML5 specification asserts that the document is processed according to the
non-XML mode HTML5 processing rules. In these particular cases, it is
important that the prefixes declared via @xmlns:
are preserved
for the RDFa processors to ensure backwards-compatibility with RDFa 1.0
documents. The following sections elaborate upon the backwards compatibility
requirements for RDFa processor implementations.
@xmlns:
-Prefixed AttributesThe RDFa Core 1.1 [RDFA-CORE] specification deprecates the
use of the @xmlns:
mechanism to declare CURIE prefix mappings in
favor of the @prefix
attribute. However, there are instances
where its use is unavoidable. For example, publishing legacy documents as HTML5 or
supporting older XHTML+RDFa 1.0 documents that rely on the @xmlns:
attribute.
CURIE prefix mappings specified using attributes prepended with
@xmlns:
MUST be processed using the algorithm defined in
section 4.4.1:
Extracting URI Mappings from Infosets
for infoset-based processors, or section 4.5.1:
Extracting URI Mappings from DOMs
for DOM Level 2-based processors. For CURIE prefix mappings using the
@prefix
attribute,
Section 7.5: Sequence, step 3
MUST be used to process namespace values.
Since CURIE prefix mappings have been specified using
@xmlns:
, and since HTML attribute names are case-insensitive,
CURIE prefix names declared using the @xmlns:
attribute-name
pattern xmlns:<PREFIX>="<URI>"
SHOULD be specified
using only lower-case characters. For example, the text
"@xmlns:
" and the text in "<PREFIX>"
SHOULD
be lower-case only. This is to ensure that prefix mappings are interpreted
in the same way between HTML (case-insensitive attribute names) and XHTML
(case-sensitive attribute names) document types.
@xmlns:
-Prefixed AttributesSince RDFa 1.0 documents may contain attributes starting with
@xmlns:
to specify CURIE prefixes, any attribute starting with
a case-insensitive match on the text string "@xmlns:
" MUST be
preserved in the DOM or other tree-like model that is passed to the RDFa
Processor.
For documents conforming to this specification, attributes with
names that have a case insensitive prefix matching "@xmlns:
"
MUST be considered conforming. Conformance checkers SHOULD
accept attribute names that have a case insensitive prefix matching
"@xmlns:
" as conforming. Conformance checkers SHOULD generate
warnings noting that the use of @xmlns:
is deprecated.
Conformance checkers MAY report the use of xmlns: as an error.
All attributes starting with a case insensitive prefix matching
"@xmlns:
" MUST conform to the production rules outlined in
Namespaces in XML [XML-NAMES11],
Section 3: Declaring Namespaces.
Documents that contain @xmlns:
attributes that do not conform to
Namespaces in XML MUST NOT be accepted as conforming.
RDFa 1.0 documents may contain the @xmlns:
pattern to
declare prefix mappings, it is important that namespace information that
is declared in non-XML mode HTML5 documents are mapped to an infoset
correctly. In order to ensure this mapping is performed correctly, the
"Coercing an HTML DOM into an infoset" rules defined in [HTML5]
must be extended to include the following rule:
If the XML API is namespace-aware, the tool must ensure that
([namespace
name], [local name],
[normalized
value]) namespace tuples are created when converting the non-XML mode
DOM into an infoset. Given a standard @xmlns:
definition,
xmlns:foo="http://example.org/bar#"
, the [namespace name]
is http://www.w3.org/2000/xmlns/
,
the [local name] is foo
, and the
[normalized value] is http://example.org/bar#
, thus the
namespace tuple would be (http://www.w3.org/2000/xmlns/
,
foo
, http://example.org/bar#
).
For example, given the following input text:
<div xmlns:com="https://w3id.org/commerce#">
The div
element above, when coerced from an HTML DOM into
an infoset, should contain an attribute in the [namespace
attributes] list with a [namespace name] set to
"http://www.w3.org/2000/xmlns/
", a [local name] set to
com
, and a [normalized value] of
"https://w3id.org/commerce#
".
While the intent of the RDFa processing instructions is to provide a set of rules that are as language and toolchain independent as possible, for the sake of clarity, detailed methods of extracting RDFa content from processors operating on an XML Information Set are provided below.
Extracting URI Mappings declared via @xmlns:
while operating from within an infoset-based RDFa processor can be achieved
using the following algorithm:
While processing an element as described in [RDFA-CORE], Section 7.5: Sequence, Step #2:
@xmlns:
, create an [IRI mapping] by
storing the [local name] part with the @xmlns:
characters
removed as the value to be mapped, and the [normalized
value] as the value to map.
This step is unnecessary if the infoset coercion rules preserve namespaces specified in non-XML mode.
For example, assume that the following markup is processed by an infoset-based RDFa processor:
<div xmlns:ps="https://w3id.org/payswarm#" ...
After the markup is processed, there should exist a [URI mapping] in
the [local list of URI mappings] that contains a mapping from
ps
to https://w3id.org/payswarm#
.
There are a number of non-prefixed attributes that are associated with RDFa Processing in HTML5. If an XML Information Set based RDFa processor is used to process these attributes, the following algorithm should be used to detect and extract the values of the attributes.
While processing Infoset Attribute Information Items in Element Information Items as described in [RDFA-CORE], Section 7.5: Sequence, Step #4 through Step #9:
Most DOM-aware RDFa processors are capable of accessing DOM Level 1
[DOM-LEVEL-1]
methods to process attributes on elements. To discover all
@xmlns:
-specified CURIE prefix mappings, the
Node.attributes
NamedNodeMap can be iterated over. Each
Attr.name that
starts with the text string @xmlns:
specifies a CURIE prefix
mapping. The value to be mapped is the string after the @xmlns:
substring in the Attr.name variable and the value to be mapped is
the value of the Attr.value variable.
The intent of the RDFa processing instructions are to provide a set of rules that are as language and toolchain independent as possible. If a developer chooses to not use the DOM1 environment mechanism outlined in the previous paragraph, they may use the following DOM2 [DOM-LEVEL-2-CORE] environment mechanism.
Extracting URI Mappings declared via @xmlns:
while operating
from within a DOM Level 2 based RDFa processor can be achieved using the
following algorithm:
While processing each DOM2 [Element] as described in [RDFA-CORE], Section 7.5: Sequence, Step #2:
@xmlns
, create an [IRI mapping] by
storing the [local
name] as the value to be mapped, and the [Node.nodeValue]
as the value to map.@xmlns:
, create an [IRI mapping] by
storing the [local name] part with the @xmlns:
characters
removed as the value to be mapped, and the [Node.nodeValue]
as the value to map.
This step is unnecessary if the XML and non-XML mode DOMs are namespace consistent.
For example, assume that the following markup is processed by a DOM2-based RDFa processor:
<div xmlns:com="https://w3id.org/commerce#" ...
After the markup is processed, there should exist a [URI mapping] in
the [local list of URI mappings] that contains a mapping from
com
to https://w3id.org/commerce#
.
There are a number of non-prefixed attributes that are associated with RDFa processing in HTML5. If an DOM2-based RDFa processor is used to process these attributes, the following algorithm should be used to detect and extract the values of the attributes.
While processing an element as described in [RDFA-CORE], Section 5.5: Sequence, Step #3 through Step #9:
When extracting values from @href
and
@src
, web authors and developers should
note that certain values MAY be transformed if accessed via the DOM versus
a non-DOM processor. The rules for modification of URL values can be
found in the main HTML5 specification under
Section 2.6.2: Parsing URLs.
This section is non-normative.
In early 2004, Mark Birbeck published a document named "RDF in XHTML" via the XHTML2 Working Group wherein he laid the groundwork for what would eventually become RDFa (The Resource Description Framework in Attributes).
In 2006, the work was co-sponsored by the Semantic Web Deployment Working Group, which began to formalize a technology to express semantic data in XHTML. This technology was successfully developed and reached consensus at the W3C, later published as an official W3C Recommendation. While HTML provides a mechanism to express the structure of a document (title, paragraphs, links), RDFa provides a mechanism to express the meaning in a document (people, places, events).
The document, titled "RDF in XHTML: Syntax and Processing" [XHTML-RDFA], defined a set of attributes and rules for processing those attributes that resulted in the output of machine-readable semantic data. While the document applied to XHTML, the attributes and rules were always intended to operate across any tree-based structure containing attributes on tree nodes (such as HTML4, SVG and ODF).
While RDFa was initially specified for use in XHTML, adoption by a number of large organizations on the web spurred RDFa's use in non-XHTML languages. Its use in HTML4, before an official specification was developed for those languages, caused concern regarding document conformance.
Over the years, the members of the RDFa Community had discussed the possibility of applying the same attributes and processing rules outlined in the XHTML+RDFa specification to all HTML family documents. By design, the possibility of a unified semantic data expression mechanism between all HTML and XHTML family documents was squarely in the realm of possibility.
An RDFa Working Group was created in 2010 to address the issues concerning multiple language implementations of RDFa. The XHTML+RDFa document was split into a base specification, called RDFa Core 1.1 [RDFA-CORE], and thin specifications that layer above RDFa Core 1.1. The XHTML+RDFa 1.1 specification [XHTML-RDFA] is an example of such a thin specification. This document, also a thin specification, is targeted at HTML4, HTML5 and XHTML5.
This document describes the extensions to the RDFa Core 1.1 specification that permits the use of RDFa in all HTML family documents. By using the attributes and processing rules described in the RDFa Core 1.1 specification and heeding the minor changes in this document, authors can generate markup that produces the same semantic data output in HTML4, HTML5 and XHTML5.
This section is non-normative.
2009-10-15: First version of the RDFa for HTML4, HTML5 and XHTML5.
2010-03-04: Updated HTML5 coercion to infoset rules, preservation of namespaces in infoset and DOM2-based processors, clarifying how to extract RDFa attributes via infoset, how to extract RDFa attributes via DOM2.
2010-05-02: Inheritance of basic processing rules from RDFa 1.1 [RDFA-CORE], instead of XHTML+RDFa 1.0 [RDFA-SYNTAX], inclusion of the HTML Default Vocabulary Terms, inclusion of a HTML 4.01 + RDFa 1.1 DTD for validation purposes.
2010-06-24: Inheritance of basic processing rules from RDFa 1.1 [RDFA-CORE], instead of XHTML+RDFa 1.0 [RDFA-SYNTAX], inclusion of the HTML Default Vocabulary Terms, added HTML 4.01 + RDFa 1.1 DTD for validation purposes, added normative definition of @version attribute.
2010-10-19: Removal of @version attribute, migrated HTML Vocabulary Terms to RDFa Profile document, added statement to send comments to the HTML WG bug tracker.
2011-01-11: Removed decentralized extensibility issue markers, added DOM Level 1 prefix mapping extraction algorithm.
2011-04-05: Moved all xmlns: rules into a section titled Backwards Compatibility and brought spec in-line with latest RDFa Core 1.1 spec.
2011-05-12: Generated Last Call document, no substantive changes.
2011-12-30: Addition of normative dependency for RDFa Lite 1.1.
Addition of rules to allow meta
and
link
elements in flow and phrasing content as long as they
contain at least one RDFa-specific attribute. Added support for
@datetime
and value
processing.
2012-03-10: Clarification of where each RDFa attribute is allowed to be used. Feature at risk warning for HTML4+RDFa DTD-based validation.
2012-09-10: Publishing control of the HTML+RDFa document is handed over from the HTML WG to the newly re-chartered RDFa WG. DTD-based validation is removed from the specification.
2012-12-13: Addition of new HTML-specific processing rules for dealing with XHTML5 vs. HTML5 documents, xml:base, HTML Literals, property and rel/rev on the same element, and more types for the datetime attribute.
2012-12-27: Added Property Copying section and special processing for head and body.
2013-01-19: Removed @value processing, added @content overriding @datetime if present, cleanup and prep for Last Call publication in RDFa WG.
2013-06-06: Applied all Last Call comments and editorial fixes in preparation for Proposed Recommendation phase.
This section is non-normative.
At the time of publication, the members of the RDFa Working Group were:
Ivan Herman (staff contact), Shane McCarron, Gregg Kellogg, Niklas Lindström, Steven Pemberton, Manu Sporny (chair), Ted Thibodeau, and Stéphane Corlosquet.
A great deal of thanks to everyone that provided feedback on the specification (most of whom are listed below):
Adam Powell, Alex Milowski, Andy Seaborne, Arto Bendiken, Austin William, BAI Xi, Benjamin Adrian, Benjamin Nowack, Bjoern Hoehrmann, Christian Langanke, Christoph Lange, Cindy Lewis, Corey Mwamba, Crisfer Inmobiliaria, Dan Brickley, Daniel Friesen, Dave Beckett, David Wood, D. Grant, Dominik Tomaszuk, Dominique Hazael-Massieux, Doug Schepers, Dr. Olaf , Edward O'Connor, Faye Harris, Felix Sasaki, Gavin Carothers, Grant Robertson, Guus Schreiber, Harry Halpin, Michael Hausenblas, Henri Bergius, Henri Sivonen, Henry Story, Holger Knublauch, Ian Hickson, Irene Celino, Alexander Kroener, Knud Möller, Philip Jägenstedt, Reto Bachmann-Gmür, Ivan Mikhailov, James Leigh, Jeff Sonstein, Jeni Tennison, Jens Haupert, Jochen Rau, John Breslin, John Cowan, John O'Donovan, Jonathan Rees, Julian Reschke, KANZAKI Masahide, Kingsley Idehen, Knud Hinnerk, Landong Zuo, Leif Halvard Silli, Liam R., Lin Clark, Maciej Stachowiak, Mark Nottingham, Markus Gylling, Martin Hepp, Martin McEvoy, Matthias Tylkowski, Darin McBeath, Melvin Carvalho, Michael Chan, Michael Hausenblas, Michael Steidl, Michael™ Smith, Mischa Tuffield, Misha Wolf, Nathan Rixham, Nathan Yergler, Nicholas Stimpson, Noah Mendelsohn, Paul Cotton, Paul Sparrow, Pete Cordell, Peter Frederick, Peter Mika, Peter Occil, Phil Archer, Reece Dunn, Richard Cyganiak, Robert Leif, Robert Weir, Ramanathan V. Guha, Sami Korhonen, Sam Ruby, Sandro Hawke, Sebastian Germesin, Sebastian Heath, Shelley Powers, Simon Grant, Simon Reinhardt, Stefan Schumacher, Tab Atkins Jr., Thomas Adamich, Thomas Baker, Thomas Roessler, Thomas Steiner, Tim Berners-Lee, Toby Inkster, Tom Adamich, Tantek Çelik, Ville Skyttä, Wayne Smith, and Will Clark