Please refer to the errata for this document, which may include normative corrections.
See also translations.
Copyright © 2018 W3C® (MIT, ERCIM, Keio, Beihang). W3C liability, trademark and document use rules apply.
This document specifies Timed Text Markup Language (TTML), Version 1, also known as TTML1, in terms of a vocabulary and semantics thereof.
The Timed Text Markup Language is a content type that represents timed text media for the purpose of interchange among authoring systems. Timed text is textual information that is intrinsically or extrinsically associated with timing information.
It is intended to be used for the purpose of transcoding or exchanging timed text information among legacy distribution content formats presently in use for subtitling and captioning functions.
In addition to being used for interchange among legacy distribution content
formats, TTML Content may be used directly as a distribution format,
for example, providing a standard content format to reference from a <track>
element in an HTML5 document, or a <text>
or
<textstream>
media element in a [SMIL 2.1] document.
This edition of the Timed Text Markup Language (TTML) 1.0 [TTML10] clarifies ambiguities and corrects errors identified in the previous edition, which this edition supersedes. No feature is added or removed in this edition.
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 https://www.w3.org/TR/.
This document was published by the Timed Text Working Group as a Recommendation.
GitHub Issues are preferred for discussion of this specification. Alternatively, you can send comments to our mailing list. Please send them to public-tt@w3.org
(subscribe,
archives) with [ttml1]
at the start of your email's subject.
Please see the Working Group's implementation report.
Substantive changes applied since the previous version of this specification are listed at substantive-changes-summary.txt.
For convenience, a diff between this and the previous edition is offered at the W3C HTML Diff service.
This document has been reviewed by W3C Members, by software developers, and by other W3C groups and interested parties, and is endorsed by the Director as a W3C Recommendation. It is a stable document and may be used as reference material or cited from another document. W3C's role in making the Recommendation is to draw attention to the specification and to promote its widespread deployment. This enhances the functionality and interoperability of the Web.
This document was produced by a group operating under the 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 document is governed by the 1 February 2018 W3C Process Document.
Unless specified otherwise, this section and its sub-sections are non-normative.
The Timed Text Markup Language (TTML) Version 1, also referred to as TTML1, provides a standardized representation of a particular subset of textual information with which stylistic, layout, and timing semantics are associated by an author or an authoring system for the purpose of interchange and processing.
TTML is expressly designed to meet only a limited set of requirements established by [TTAF1-REQ], and summarized in I Requirements. In particular, only those requirements which service the need of performing interchange with existing, legacy distribution systems are satisfied.
In addition to being used for interchange among legacy distribution content
formats, TTML Content may be used directly as a distribution format, providing, for
example, a standard content format to reference from a <track>
element in an HTML5 document, or a <text>
or
<textstream>
media element in a [SMIL 2.1] document. Certain properties of TTML support
streamability of content, as described in L Streaming TTML Content.
Note:
While TTML is not expressly designed for direct (embedded) integration into an HTML or a SMIL document instance, such integration is not precluded.
Note:
In some contexts of use, it may be appropriate to employ animated content to depict sign language representations of the same content as expressed by a Timed Text Document Instance. This use case is not explicitly addressed by TTML mechanisms, but may be addressed by some external multimedia integration technology, such as SMIL.
Note:
In the first edition of this specification, the version number 1.0
was
used to refer to this version of TTML. In this second edition, this version number is simplified
to 1
as a result of a determination to call the next (future) version 2
instead of 1.1
.
Note:
In previous drafts of this specification, TTML was referred to as DFXP (Distribution
Format Exchange Profile). This latter term is retained for historical reasons in
certain contexts, such as profile names and designators, and the short name
ttaf1-dfxp
used in older URLs to refer to this specification. Starting in
this edition of the first version of TTML, the short name ttml1
will be
used, and subsequent versions of TTML are expected to use short names of
ttml2
, etc., with integer version numbers.
Use of TTML is intended to function in a wider context of Timed Text Authoring and Distribution mechanisms that are based upon the system model depicted in Figure 1 – System Model, wherein the Timed Text Markup Language serves as a bidirectional interchange format among a heterogeneous collection of authoring systems, and as a unidirectional interchange format to a heterogeneous collection of distribution formats after undergoing transcoding or compilation to the target distribution formats as required, and where one particular distribution format is TTML (labeled as "DFXP" in the figure).
A TTML Document Instance consists of a tt
document element that contains
a header and a body, where the header specifies document level metadata, styling
definitions and layout definitions, and the body specifies text content intermixed
with references to style and layout information and inline styling and timing information.
<tt xml:lang="" xmlns="http://www.w3.org/ns/ttml"> <head> <metadata/> <styling/> <layout/> </head> <body/> </tt> |
Document level metadata may specify a document title, description, and copyright information. In addition, arbitrary metadata drawn from other namespaces may be specified.
<metadata xmlns:ttm="http://www.w3.org/ns/ttml#metadata"> <ttm:title>Timed Text TTML Example</ttm:title> <ttm:copyright>The Authors (c) 2006</ttm:copyright> </metadata> |
Styling information may be specified in the form of style specification definitions that are referenced by layout and content information, specified inline with content information, or both.
In Example Fragment – TTML Styling, four style sets of specifications are defined, with one set serving as a collection of default styles.
<styling xmlns:tts="http://www.w3.org/ns/ttml#styling"> <!-- s1 specifies default color, font, and text alignment --> <style xml:id="s1" tts:color="white" tts:fontFamily="proportionalSansSerif" tts:fontSize="22px" tts:textAlign="center" /> <!-- alternative using yellow text but otherwise the same as style s1 --> <style xml:id="s2" style="s1" tts:color="yellow"/> <!-- a style based on s1 but justified to the right --> <style xml:id="s1Right" style="s1" tts:textAlign="end" /> <!-- a style based on s2 but justified to the left --> <style xml:id="s2Left" style="s2" tts:textAlign="start" /> </styling> |
Layout information defines one or more regions into which content
is intended to be presented. A region definition may reference one or
more sets of style specifications in order to permit content flowed in the
region to inherit from these styles.
In Example Fragment – TTML Layout, the region definition makes
reference to style specification s1
augmented by specific inline
styles which, together, allow content flowed into the region to inherit
from the region's styles (in the case
that a style is not already explicitly specified on content or inherited
via the content hierarchy.)
<layout xmlns:tts="http://www.w3.org/ns/ttml#styling"> <region xml:id="subtitleArea" style="s1" tts:extent="560px 62px" tts:padding="5px 3px" tts:backgroundColor="black" tts:displayAlign="after" /> </layout> |
The content of a Document Instance is expressed in its
body, which is organized in terms of block and inline text elements.
The hierarchical organization of content elements serves a primary role in
determining both spatial and temporal relationships. For example, in
Example Fragment – TTML Body, each paragraph (p
element)
is flowed into its target region in the specified lexical order; furthermore,
the active time interval of each paragraph is timed in accordance to its
parent or sibling according to the applicable time containment semantics —
in this case, the division parent is interpreted (by default) as a parallel time
container.
<body region="subtitleArea"> <div> <p xml:id="subtitle1" begin="0.76s" end="3.45s"> It seems a paradox, does it not, </p> <p xml:id="subtitle2" begin="5.0s" end="10.0s"> that the image formed on<br/> the Retina should be inverted? </p> <p xml:id="subtitle3" begin="10.0s" end="16.0s" style="s2"> It is puzzling, why is it<br/> we do not see things upside-down? </p> <p xml:id="subtitle4" begin="17.2s" end="23.0s"> You have never heard the Theory,<br/> then, that the Brain also is inverted? </p> <p xml:id="subtitle5" begin="23.0s" end="27.0s" style="s2"> No indeed! What a beautiful fact! </p> <p xml:id="subtitle6a" begin="28.0s" end="34.6s" style="s2Left"> But how is it proved? </p> <p xml:id="subtitle6b" begin="28.0s" end="34.6s" style="s1Right"> Thus: what we call </p> <p xml:id="subtitle7" begin="34.6s" end="45.0s" style="s1Right"> the vertex of the Brain<br/> is really its base </p> <p xml:id="subtitle8" begin="45.0s" end="52.0s" style="s1Right"> and what we call its base<br/> is really its vertex, </p> <p xml:id="subtitle9a" begin="53.5s" end="58.7s"> it is simply a question of nomenclature. </p> <p xml:id="subtitle9b" begin="53.5s" end="58.7s" style="s2"> How truly delightful! </p> </div> </body> |
The first subtitle Subtitle 1 – Time Interval [0.76, 3.45) is
presented during the time interval 0.76 to 3.45 seconds.
This
subtitle inherits its font family, font size,
foreground color, and text alignment from the region into which it is
presented. Since no region is explicitly specified on the paragraph,
the nearest ancestor that specifies a region determines the
targeted region. Note also that content is presented
at the bottom (after edge) of the containing region due to the tts:displayAlign="after"
being specified on the region definition.
Note:
The notation "[X,Y]" denotes a closed interval from X to Y, including X and Y; "[X,Y)" denotes a right half-open interval from X to Y, including X but not including Y; "(X,Y]" denotes a left half-open interval from X to Y, not including X but including Y; "(X,Y)" denotes an open interval from X to Y, not including X or Y.
Note:
In this example, the p
element is used as a presentational
element rather than as a semantic element, i.e., as a linguistic paragraph.
It is up to an author to determine which TTML elements are used to convey
the intended meaning. For instance, this example could be written to use
timing on span
elements in order to preserve the integrity of
semantic paragraphs.
The second subtitle continues with the default style, except that it contains
two lines of text with an intervening author-specified line break. Note the effects of
the use of tts:textAlign="center"
to specify the paragraph's alignment
in the inline progression direction.
The third subtitle continues, using a variant style which overrides the default style's foreground color with a different color.
The fourth subtitle reverts to the default style.
The fifth subtitle continues, again using a variant style which overrides the default style's foreground color with a different color.
During the next active time interval, two distinct subtitles are simultaneously active, with the paragraphs expressing each subtitle using different styles that override color and paragraph text alignment of the default style. Note that the flow order is determined by the lexical order of elements as they appear in the content hierarchy.
The next subtitle is specified in a similar manner using a style override to give the paragraph right (end) justification in the inline progression direction.
The eighth subtitle uses the same style override as the previous subtitle in order to maintain the right (end) justification of the paragraph.
During the final (ninth) active time interval, two distinct subtitles are again simultaneously active, but with a different style applied to the second paragraph to override the default color. Note that the flow order is determined by the lexical order of elements as they appear in the content hierarchy.
The examples shown above demonstrate the primary types of information that may be authored using TTML: metadata, styling, layout, timing, and content. In typical cases, styling and layout information are separately specified in a Document Instance. Content information is expressed in a hierarchical fashion that embodies the organization of both spatial (flow) and timing information. Content makes direct or indirect references to styling and layout information and may specify inline overrides to styling.
DFXP | Distribution Format Exchange Profile |
TT | Timed Text |
TTML | Timed Text Markup Language |
TTAF | Timed Text Authoring Format |
TTWG | Timed Text Working Group |
An instance of an abstract data set as represented by a Reduced XML Infoset.
A set of constraints that defines a class of XML Information Sets [XML InfoSet].
Each specified or defaulted attribute of an XML document corresponds with an attribute information item as defined by [XML InfoSet], §2.3.
Each data character appearing in an XML document corresponds with a character information item as defined by [XML InfoSet], §2.6.
The size (extent) of a cell computed by dividing the width of the Root Container
Region by the column count, i.e., the number of cells in the horizontal axis, and by dividing
the height of the Root Container Region by the row count, i.e., the number of cells in the vertical
axis, where the column and row counts are determined by the ttp:cellResolution
parameter attribute.
A processing system capable of importing (receiving) Timed Text Markup Language content for the purpose of transforming, presenting, or otherwise processing the content.
A logical region into which rendered content is placed when modeling or performing presentation processing.
A concrete realization of a Timed Text Markup Language document, where the concrete form is specific to the context of reference. For example, a sequence of bytes that represents an XML serialization of a Timed Text document, an internal, parsed representation of such a Timed Text document, etc.
The implied context or environment external to a Content Processor in which document interchange occurs, and in which out-of-band protocols or specifications may define certain behavioral defaults, such as an implied profile.
The implied context or environment internal to a Content Processor in which document processing occurs, and in which out-of-band protocols or specifications may define certain behavioral defaults, such as the establishment or creation of a Synthetic Document Syncbase.
Each element appearing in an XML document corresponds with an element information item as defined by [XML InfoSet], §2.2.
A profile of content that serves a set of needs for content interchange.
A syntactic or semantic expression or capability that is defined and labeled (using a extension designation) in another (public or private) specification.
A syntactic or semantic expression or capability that is defined and labeled (using a feature designation) in this specification (or a future revision of this specification).
A Content Processor which purpose is to layout, format, and render, i.e., to present, Timed Text Markup Language content by applying the presentation semantics defined in this specification.
See Content Processor.
A document that defines a specific collection of features and extensions for which support is required or optional in a recipient content processor.
A logical construct that models authorial intention regarding desired or potential presentation processing, and which is represented as a rectangular area of a presentation surface, i.e., a region area, into which content is composed and rendered during presentation processing.
An XML Information Set [XML InfoSet] that satisfies the constraints specify by A Reduced XML Infoset.
A (possibly null) media object associated with or otherwise related to a Document Instance. For example, an aggregate audio/video media object for which a Document Instance provides caption or subtitle information, and with which that Document Instance is associated.
When a non-null Related Media Object exists, the region of this media object, expressed in the coordinate system that applies to the Document Instance that is associated with the related media object.
A logical region that establishes a coordinate system into which Document Instance content regions are placed and optionally clipped.
The temporal extent (interval) defined by the temporal beginning and ending of a Document Instance in relationship with some external application or presentation context.
A time code whose format and semantics are established by [SMPTE 12M], which may be embedded into or otherwise associated with media content, such as a broadcast audio/video stream.
A document level syncbase [SMIL 2.1], § 10.7.1, synthesized or otherwise established by the Document Processing Context in accordance with the Related Media Object or other processing criteria.
A Synthetic Document Syncbase constructed from SMPTE Time Code values embedded in or associated with the Related Media Object or otherwise determined by the Document Processing Context.
A syntactic or semantic feature, e.g., an element or the presentation of an element, is Temporally Active when the current time of selected time base intersects with the active time interval of the feature.
A Region that is Temporally Active.
Textual information that is intrinsically or extrinsically associated with timing information.
A content type that represents timed text media for the purpose of interchange among authoring systems.
A content authoring system capable of importing and exporting Timed Text Markup Language content.
A Content Processor which purpose is to transform or otherwise rewrite Timed Text Markup Language content to either Timed Text Markup Language or to another (arbitrary) content format. An example of the first is a processor that removes or rewrites TTML features so as to conform to a profile of TTML. An example of the latter is a processor that translates TTML into a completely different timed text format. Because this specification does not otherwise define a target profile or format for transformation processing, no further transformation semantics are defined by this specification.
An Abstract Document Instance which has been assessed for validity and found to be valid as defined by 4 Document Types.
Within normative prose in this specification, the words may, should, and must are defined as follows:
Conforming documents and/or TTML processors are permitted to, but need not behave as described.
Conforming documents and/or TTML processors are strongly recommended to, but need not behave as described.
Conforming documents and/or TTML processors are required to behave as described; otherwise, they are in error.
If normative specification language takes an imperative form, then it is to be treated as if the term must applies. Furthermore, if normative language takes a declarative form, and this language is governed by must, then it is also to be treated as if the term must applies.
Note:
For example, the phrases "treat X as an error" and "consider X as an error" are to be read as mandatory requirements in the context of use. Similarly, if the specification prose is "X must apply", "X applies", or "X is mandatory", and "X" is further defined as "X is Y and Z", then, by transitive closure, this last declarative phrase is to be read as "Y is mandatory" and "Z is mandatory" in the context of use.
All normative syntactic definitions of XML representations and other related terms are depicted with a light yellow-orange background color and labeled as "XML Representation" or "Syntax Representation", such as in the following:
<example
count = integer
size = (large|medium|small) : medium>
Content: (all | any*)
</example>
|
In an XML representation, bold-face attribute names (e.g. count above) indicate a required attribute
information item, and the rest are optional. Where an attribute
information item has an enumerated type definition, the values are
shown separated by vertical bars, as for size
above; if
there is a default value, it is shown following a colon. Where an
attribute information item has a built-in simple type definition
defined in [XML Schema Part 2], a hyperlink to its definition
therein is given.
In an XML representation, the expression {any attribute not in default or any TT namespace} applies only to namespace qualified attributes; unqualified attributes are not permitted unless explicitly defined in this specification.
The allowed content of the information item is shown as a grammar
fragment, using the Kleene operators ?
, *
and +
. Each element name therein is a hyperlink to its
own illustration.
The term linear white-space (LWSP) is to be interpreted as a non-empty sequence of
SPACE (U+0020), TAB (U+0009), CARRIAGE RETURN (U+000D), or LINE FEED (U+000A),
which corresponds to production [3] S
as defined by [XML 1.0].
All content of this specification that is not explicitly marked as non-normative is considered to be normative. If a section or appendix header contains the expression "Non-Normative", then the entirety of the section or appendix is considered non-normative.
All paragraphs marked as a Note are considered non-normative.
Example code fragments are depicted with a light blue-green background color and labeled as "Example Fragment", such as in the following:
<tt xml:lang="" xmlns="http://www.w3.org/ns/ttml"> <head> <metadata/> <styling/> <layout/> </head> <body/> </tt> |
This section specifies the general conformance requirements for TTML Content and processors.
A TTML Document Instance conforms to this specification if the following criteria are satisfied:
When transporting a Document Instance in a Document Interchange Context in which a
Media Type [Media Types] identifies the content type of the interchanged Document Instance, then the specified media type is
application/ttml+xml
in conformance with [XML Media Types] § 7, with which an optional
profile
parameter may appear, the value of which
conforms to
a profile designator as
defined by
5.2 Profiles.
The Document Instance is or can be represented as a Reduced XML Infoset as defined by A Reduced XML Infoset.
The Reduced XML Infoset that corresponds to the Document Instance is or can be associated with one of the Abstract Document Types defined by 4 Document Types.
The Reduced XML Infoset that corresponds to the Document Instance is a Valid Abstract Document Instance of the associated Abstract Document Type.
The Reduced XML Infoset satisfies all additional mandatory syntactic and semantic constraints defined by this specification. In addition, this Infoset should satisfy the web content accessibility guidelines specified by [WCAG].
A TTML Content Processor conforms to this specification if the following generic processor criteria are satisfied:
The processor provides at least one mechanism for notionally instantiating a Reduced XML Infoset representation of a conformant Document Instance.
If a process does or can perform validation of a candidate Document Instance, then it provides at least one mechanism to implicitly or explicitly associate the Reduced XML Infoset representation of a conformant Document Instance with one of the Abstract Document Types defined by 4 Document Types.
The processor does not a priori reject or abort the processing of a conformant Document Instance unless the processor does not support some required (mandatory) feature specified or implied by a TTML profile declared to apply to the Document Instance.
The processor supports all mandatory processing semantics defined by this specification.
Note:
The phrase mandatory semantics refers to all explicit use of the conformance key phrases must and must not as well as any declarative statement that can be reasonably inferred from such key phrases. For example, these mandatory semantics include support for all features marked as mandatory in D.2 Feature Support.
If the processor supports some optional processing semantics defined by this specification, then it does so in a manner consistent with the defined semantics.
Note:
The phrase optional semantics refers to all explicit use of the conformance key phrases should, should not, may, and may not, as well as any declarative statement that can be reasonably inferred from such key phrases. For example, these optional semantics include support for all features marked as optional in D.2 Feature Support.
A TTML Content Processor is a conformant TTML Transformation Processor if the following criteria are satisfied:
The processor satisfies all requirements specified by 3.2.1 Generic Processor Conformance.
The processor supports the DFXP Transformation profile as specified by F.1 DFXP Transformation Profile.
A TTML Content Processor is a conformant TTML Presentation Processor if the following criteria are satisfied:
The processor satisfies all requirements specified by 3.2.1 Generic Processor Conformance.
The processor supports the DFXP Presentation profile as specified by F.2 DFXP Presentation Profile.
Any claim of compliance with respect to the conformance of a TTML Document Instance or Content Processor must make reference to an implementation compliance statement (ICS).
An implementation compliance statement must identify all mandatory and optional features of this specification that are satisfied by the document instance or the content processor implementation. In particular, the statement must identify the utilized or supported TTML vocabulary profile(s) as defined by 5.2 Profiles, and, if a subset or superset profile is used or supported, then what features are excluded or included in the subset or superset profile.
A Document Instance for which a compliance claim is made
must specify
either (1)
a ttp:profile
attribute on its root
tt
element
as defined by 6.2.8 ttp:profile
or (2)
a ttp:profile
element as a child of the
head
element as defined by
6.1.1 ttp:profile.
This section defines the following TTML Abstract Document Types:
Each Abstract Document Type consists of the following constraints:
a non-empty collection of element types, where each element type consists of a name, a (possibly empty) collection of attributes, and a content specification
a non-empty collection of element types that may appear as the document element
An Abstract Document Instance may be assessed in terms of validity, and is considered to be a Valid Abstract Document Instance if it satisfies the following condition: if after
pruning all element information items whose names are not members of the collection of element types defined by the associated Abstract Document Type, then
pruning character information item children from any remaining element in case that all character children of the element denote XML whitespace characters and the element's type is defined as empty in the associated Abstract Document Type, and then
pruning all attribute information items having expanded names such that the namespace URI of the expanded names are not listed in Table 1 – Namespaces, or, if listed in Table 1 – Namespaces, are not members of the collection of attributes defined by the associated Abstract Document Type for use with the owning element information item,
then the document element is one of the document element types permitted by the associated Abstract Document Type, the descendants of the document element satisfy their respective element type's content specifications, all required attributes are present, and the declared value of each attribute satisfies the type declared by the associated Abstract Document Type.
Note:
While a conformant processor may not a priori reject a conformant Document Instance, a given Document Instance may be constrained by the author or authoring tool to satisfy a more restrictive definition of validity.
Note:
As illustrated in the following example, an Abstract Document Instance can be a Valid Abstract Document Instance even if it includes elements and attributes whose namespace names are listed in Table 1 – Namespaces but whose local names
are not part of the vocabulary defined by this version of the specification. Specifically, the element foo
and the attribute tts:foo
are pruned by above steps (1) and (3), respectively, because they are not members of the associated Abstract Document Type, even though their
namespace names are listed in Table 1 – Namespaces.
<tt xmlns="http://www.w3.org/ns/ttml" xmlns:tts="http://www.w3.org/ns/ttml#styling" xml:lang="en"> <body> <foo>Foo</foo> <div> <p tts:foo="bar"> Bar </p> </div> </body> </tt> |
TTML Content is an Abstract Document Type of a profile of the Timed Text Markup Language intended to be used for interchange among distribution systems. This document type is defined in terms of the element and attribute vocabulary specified in 5 Vocabulary.
This specification references two types of schemas that may be used to validate a superset/subset of conformant TTML Content Document Instances:
The (root) document element of a TTML Content Document Instance must be a
tt
element, as defined by 7.1.1 tt.
Note:
The schemas referenced by this specification do not validate all syntactic constraints
defined by this specification, and, as such, represent a superset of conformant
TTML Content. In particular, performing validation with one of the above
referenced schemas may result in a false positive indication of validity.
For example, both the RNC and XSD schemas specify
that a tts:fontFamily
attribute must
satisfy the xs:string
XSD data type; however, this data type is a superset
of the values permitted to be used with the tts:fontFamily
attribute.
In addition, the RNC schema may produce a false negative
indication of validity when using the xml:id
attribute with an element in
a foreign namespace, thus representing a subset of conformant TTML Content. This is
due to a specific limitation in expressing wildcard patterns involving xsd:ID
typed attributes in Relax NG schemas. Note that this specification defines the
formal validity of a Document Instance to be based on an Abstract Document Instance from
which all foreign namespace elements and attributes have been removed. Therefore, the
exceptional reporting of this false negative does not impact the formal assessment
of Document Instance validity.
Note:
The schemas referenced by this specification are intended for use after the pruning steps (1)-(3) specified by 4 Document Types have been applied.
This section defines the namespaces, profiles, and vocabulary (as an element and attribute catalog) of the Timed Text Markup Language (TTML) as follows:
The Timed Text Markup Language (TTML) employs a number of XML Namespaces [XML Namespaces 1.0] for elements and certain global attributes. The following table specifies this set of namespaces and indicates the default prefix used within this specification and the normative URI that denotes each namespace.
Note:
In a specific Document Instance, it is not required that the default prefixes shown below are used. Any prefix or namespace binding that satisfies the constraints of XML Namespaces [XML Namespaces 1.0] may be used that is associated with the specified namespace URI.
Name | Prefix | Value |
---|---|---|
TT | tt: | http://www.w3.org/ns/ttml |
TT Parameter | ttp: | http://www.w3.org/ns/ttml#parameter |
TT Style | tts: | http://www.w3.org/ns/ttml#styling |
TT Metadata | ttm: | http://www.w3.org/ns/ttml#metadata |
TT Profile | none | http://www.w3.org/ns/ttml/profile/ |
TT Feature | none | http://www.w3.org/ns/ttml/feature/ |
TT Extension | none | http://www.w3.org/ns/ttml/extension/ |
Note:
If a reference to an element type is used in this specification and the name of the element type is not namespace qualified, then the TT Namespace applies.
For certain namespaces defined above, the default prefix is specified as none if no XML vocabulary is defined in the namespace by this specification (nor expected to be defined in a future version of this specification). In such cases, the use of the namespace URI is for purposes other than defining XML vocabulary, e.g., for designating profiles, features, extensions and for dereferencing standard profile definitions.
All TTML Namespaces are mutable [NSState]; all undefined names in these namespaces are reserved for future standardization by the W3C.
The Timed Text Markup Language (TTML) employs a number of standard, predefined profiles of its vocabulary and associated semantics. The following table specifies this set of profiles, indicating a normative name and designator for each predefined profile, and where each of these profiles is formally elaborated in F Profiles or in another TTWG specification.
Name | Designator |
---|---|
DFXP Transformation | http://www.w3.org/ns/ttml/profile/dfxp-transformation |
DFXP Presentation | http://www.w3.org/ns/ttml/profile/dfxp-presentation |
DFXP Full | http://www.w3.org/ns/ttml/profile/dfxp-full |
SDP US | http://www.w3.org/ns/ttml/profile/sdp-us |
A profile designator must adhere to the
xsd:anyURI
data type defined by [XML Schema Part 2],
§3.2.17. If the profile designator is expressed as a relative URI,
then it must be absolutized by using the TT Profile Namespace value as
the base URI.
Note:
For example, if a profile designator is expressed as
dfxp-presentation
, then the absolutized profile designator
would be
http://www.w3.org/ns/ttml/profile/dfxp-presentation
.
All profile designators which have the TT Profile Namespace as a prefix but are otherwise not listed in Table 2 – Profiles are reserved for future standardization, and must not appear in a conformant Document Instance. Nothwithstanding this constraint, a profile designator is not restricted to the set of designators enumerated in Table 2 – Profiles, but may be any URI that feasibly dereferences a TTML Profile Definition Document provided it does not use the TT Profile Namespace as a prefix.
The profile of TTML that must be supported by a
TTML Content Processor in order to process a Document Instance is determined either (1) by specifying a
ttp:profile
attribute on the root tt
element, as
defined by 6.2.8 ttp:profile, or (2) by
including one or more ttp:profile
elements in the head
element, in
accordance with 6.1.1 ttp:profile.
If a ttp:profile
element appears as a
descendant of the tt
element, then the ttp:profile
attribute should not be specified on the tt
element. If both
a ttp:profile
element and a ttp:profile
attribute are present (in a given Document Instance), then the
ttp:profile
attribute must be ignored for the purpose of
determining the declared profile requirements.
If more than one ttp:profile
element appears in
a Document Instance, then all specified profiles apply
simultaneously. In such a case, if some feature or some extension is
specified by one profile to be used
(mandatory and enabled) and by
another profile to be required
(mandatory) or optional
(voluntary),
then that feature or extension must be considered to be used
(mandatory and enabled); if some feature or some extension is
specified by one profile to be merely required
(mandatory) and by
another profile to be optional
(voluntary), then that feature or
extension must be considered to be required
(mandatory).
If neither
ttp:profile
attribute nor
ttp:profile
element is present in a Document Instance, and if the Document
Interchange Context does not make an implicit or explicit
reference to a pre-defined profile or does not specify a Profile Definition
Document or another equivalent set of feature designations,
then the DFXP Transformation profile applies.
Note:
It is not a requirement on a conformant Document Instance that a profile be internally
defined by use of a ttp:profile
element or internally referenced by a
ttp:profile
attribute. More specifically, it is permitted that the Document
Interchange Context determines the applicable profile through private
agreement, out-of-band protocol, or common use (between sender and receiver) of a profile defined
by an external specification.
Note:
It is intended that the ttp:profile
attribute be used when
the author wishes to reference one of the standard, predefined profiles of
TTML Content, and does not wish to modify (by supersetting or subsetting)
that profile. This attribute may also be used by an author to indicate the use
of a non-standard profile, in which case the specified profile designator expresses
a URI that denotes an externally defined Profile Definition Document.
However, it is not required that a conformant TTML Content Processor be able
to dereference such an externally specified profile definition.
In contrast, it is intended that the ttp:profile
element be
used when the author wishes to make use of a modified predefined profile or
wishes to include in the Document Instance a non-standard profile definition
not based upon one of the predefined profiles.
A predefined profile is supersetted by specifying some feature or extension to
be required
(mandatory) that was either not specified in the underlying, baseline
profile or was specified as optional
(voluntary) in the baseline profile.
A predefined profile is subsetted by specifying some feature or extension to
be optional
(voluntary) that was specified as required
(mandatory)
in the underlying, baseline profile.
When a baseline profile is modified by subsetting, the resulting, derived profile is referred to as a subtractive profile; when modified by supersetting, the result is referred to as an additive profile. It is also possible to define a derived profile that is simultaneously subtractive and additive.
If a Document Instance makes use of a feature defined by D.1 Feature Designations and if the intended use of the document requires the recognition and processing of that feature, then the document must include a required feature or a used feature specification in one of its declared or referenced profiles. If a Document Instance makes use of an extension designatable by E.1 Extension Designations and if the intended use of the document requires the recognition and processing of that extension, then the document must include a required extension or a used extension specification in one of its declared or referenced profiles.
Note:
A required feature or used feature
specification is expressed directly (or indirectly by referring to a
profile) by means of a ttp:feature
element where the value of its value
attribute is
required
or use
,
respectively. A required extension or used
extension specification is expressed directly (or indirectly by
referring to a profile) by means of a ttp:extension
element where the value of its value
attribute is
required
or use
,
respectively.
An example of an author defined additive, derived profile of the DFXP Presentation profile is shown below in Example Fragment – DFXP Additive Profile.
<tt xml:lang="" xmlns="http://www.w3.org/ns/ttml"> <head> <profile use="dfxp-presentation" xmlns="http://www.w3.org/ns/ttml#parameter"> <features xml:base="http://www.w3.org/ns/ttml/feature/"> <feature value="required">#fontStyle-italic</feature> </features> </profile> </head> <body/> </tt> |
Note:
In the above example, the baseline profile is declared to be
the DFXP Presentation profile, which is then additively modified by making the
#fontStyle-italic
feature required (rather than
optional as it is defined
in F.2 DFXP Presentation Profile). Note also the resetting of the
default XMLNS binding on the profile
element to the TT Parameter Namespace.
The vocabulary of the Timed Text Markup Language (TTML) is defined in the following major catalogs (divisions of vocabulary):
The core catalog defines the baseline, core vocabulary of TTML, and, in particular, the vocabulary of TTML Content. The extension catalog serves as a placeholder for extensions to the core vocabulary defined by TTML.
The core vocabulary catalog is intended to satisfy the needs of TTML while providing a baseline vocabulary for future profiles. This vocabulary is divided into distinct categories, specified in detail in the following sections:
The core element vocabulary specified for use with a Document Instance is enumerated in Table 3 – Element Vocabulary.
Module | Elements |
---|---|
Animation | set |
Content | body, div, p, span, br |
Document | tt |
Head | head |
Layout | layout, region |
Metadata | metadata |
Metadata Items | ttm:actor, ttm:agent, ttm:copyright, ttm:desc, ttm:name, ttm:title |
Parameter Items | ttp:profile, ttp:features, ttp:feature, ttp:extensions, ttp:extension |
Styling | styling, style |
Element vocabulary groups that are used in defining content models for TTML element types are enumerated in Table 4 – Element Vocabulary Groups.
Group | Elements |
---|---|
Animation.class | set |
Block.class | div | p |
Inline.class |
span |
br |
#PCDATA
|
Metadata.class | metadata | ttm:agent | ttm:copyright | ttm:desc | ttm:title |
Parameters.class | ttp:profile |
The attribute vocabulary specified for use with the core vocabulary catalog is enumerated in Table 5 – Attribute Vocabulary.
Module | Attributes |
---|---|
Core Attributes | xml:id, xml:lang, xml:space |
Layout | region |
Metadata Attributes | ttm:agent, ttm:role |
Parameter Attributes | ttp:cellResolution, ttp:clockMode, ttp:dropMode, ttp:frameRate, ttp:frameRateMultiplier, ttp:markerMode, ttp:pixelAspectRatio, ttp:profile, ttp:subFrameRate, ttp:tickRate, ttp:timeBase |
Styling | style |
Styling Attributes | tts:backgroundColor, tts:color, tts:direction, tts:display, tts:displayAlign, tts:extent, tts:fontFamily, tts:fontSize, tts:fontStyle, tts:fontWeight, tts:lineHeight, tts:opacity, tts:origin, tts:overflow, tts:padding, tts:showBackground, tts:textAlign, tts:textDecoration, tts:textOutline, tts:unicodeBidi, tts:visibility, tts:wrapOption, tts:writingMode, tts:zIndex |
Timing Attributes | begin, dur, end, timeContainer |
Note:
Only those attributes defined as either (1) global, i.e., namespace qualified, or (2) shared element-specific, i.e., not namespace qualified but shared across multiple element types, are listed in Table 5 – Attribute Vocabulary above.
Note:
All vocabulary defined by TTML consistently makes use of the so-called lowerCamelCase naming convention. In some cases, this results in the change of a name when the name was based upon another specification that used a different naming convention.
The extension vocabulary catalog is intended for use by future profiles of TTML, and is not further defined by this version of this specification.
In addition to standardized extension vocabulary, a conforming Document Instance may contain arbitrary namespace qualified elements that reside in any namespace other than those namespaces defined for use with this specification. Furthermore, a conforming Document Instance may contain arbitrary namespace qualified attributes on TTML defined vocabulary where such attributes reside in any namespace other than those defined for use with this specification.
This section specifies the parameters matter of the core vocabulary catalog, where parameters are to be understood as information that is either (1) essential or (2) of significant importance for the purpose of interpreting the semantics of other types of information expressed by core vocabulary items or for establishing a Document Processing Context by means of which TTML Content can be related to an external environment.
The following elements, all defined in the TT Parameter Namespace, specify parametric information that applies to a Document Instance or Content Processor:
The ttp:profile
element is used to specify a collection of
used (mandatory and enabled), required (mandatory),
and optional (voluntary) features and extensions that must or
may be supported by a Content Processor in order to process a
Document Instance that makes (or may make) use of such features and
extensions.
Note:
The difference between a feature and an extension is where it is defined and how it is labeled: if defined in this specification (or a future revision thereof) and labeled with a feature designation in D Features, then it is considered to be a feature; if defined in another specification and labeled there with an extension designation, then it is considered to be an extension. In general, features are expected to be defined by the W3C standards process, while extensions are expected to be defined by third parties.
This specification defines two distinct uses of the ttp:profile
element:
as a child of the head
element within a TTML Document Instance;
as the root element of a TTML Profile Definition Document instance;
When a ttp:profile
element appears within a TTML Document Instance,
its purpose is to express authorial intentions about which
features and extensions must or may be supported by a recipient
content processor. In addition, the element indirectly expresses
information about the set of features or extensions that are (or may
expected to be) used by the Document Instance.
When a ttp:profile
element is used by a TTML Profile Definition Document instance,
it serves to publish a machine
readable specification of a specific TTML profile, of which
this specification defines three such Profile Definition Documents in
F Profiles.
The ttp:profile
element accepts as its children zero or more
elements in the Metadata.class
element group, followed by
zero or more ttp:features
elements, followed by
zero or more ttp:extensions
elements.
<ttp:profile use = string xml:id = ID {any attribute not in default or any TT namespace}> Content: Metadata.class*, ttp:features*, ttp:extensions* </ttp:profile> |
If specified, the use
attribute must adhere to the
xsd:anyURI
data type defined by [XML Schema Part 2],
§3.2.17, and, furthermore, must denote a profile designator in
accordance with 5.2 Profiles. In this case,
the profile designator must refer to (1) a standard, predefined
Profile Definition Document as defined by F Profiles,
or (2) a feasibly dereferenceable resource representing a valid Profile Definition
Document instance. In either case, the referenced profile serves as the
baseline profile of the specifying ttp:profile
element.
If the use
attribute is not specified, then the baseline
profile of the ttp:profile
element must be considered to be
the empty (null) profile, i.e., a profile definition containing no
feature or extension specifications.
The collection of features and extensions of a profile are determined according to the following ordered rules:
initialize the features and extensions of the profile to the empty set;
if a use
attribute is present, then augment the profile
with the set of features and extensions specified by the referenced
baseline profile;
for each ttp:feature
and ttp:extension
element
descendant of the ttp:profile
element, using a post-order
traversal, merge the specified feature or extension with the features
and extensions of the profile, where merging a feature or extension
entails replacing an existing feature or extension specification, if
it already exists, or adding a new feature or extension specification,
if it does not yet exist in the profile;
A conformant TTML processor is not required to be able to dereference a Profile Definition Document that is not one of the standard, predefined profiles defined by F Profiles. Furthermore, a conformant TTML processor may make use of a built-in, static form of each standard, predefined profile so as not to require dereferencing a network resource.
If a TTML processor is unable to dereference a non-standard Profile Definition Document, then it must not further process the document without the presence of an explicit override from an end-user or some implementation specific parameter traceable to an end-user or to a user or system configuration setting. If a TTML processor aborts processing of a Document Instance due to the inability to reference a non-standard Profile Definition Document, then some end-user notification should be given unless the end-user or system has disabled such a notification, or if the processor does not permit or entail the intervention of an end-user.
The ttp:profile
element is illustrated by the following example.
<ttp:profile use="dfxp-presentation"> <ttp:features xml:base="http://www.w3.org/ns/ttml/feature/"> <ttp:feature>#text-outline</ttp:feature> </ttp:features> </ttp:profile> |
Note:
In the above example, the DFXP presentation profile is used as the
baseline profile. This baseline profile is then supersetted (thus
creating an additive derived profile) by requiring support for
#text-outline
feature.
The ttp:features
element is a container element used to group
infomation about feature support requirements.
The ttp:features
element accepts as its children zero or more
elements in the Metadata.class
element group, followed by
zero or more ttp:feature
elements.
<ttp:features xml:base = string : TT Feature Namespace xml:id = ID {any attribute not in default or any TT namespace}> Content: Metadata.class*, ttp:feature* </ttp:features> |
If specified, the xml:base
attribute must (1) adhere to the
xsd:anyURI
data type defined by [XML Schema Part 2],
§3.2.17, (2) express an absolute URI that
adheres to [XML Base] and,
(3) express a feature namespace as defined by
D.1 Feature Designations. If not specified, the
xml:base
attribute's default
value applies, which is the TT Feature Namespace.
The xml:base
attribute is
used to permit the abbreviation of feature designation URIs expressed
by child ttp:feature
elements.
The ttp:feature
element is used to specify
infomation about support requirements for a particular feature.
The children of the ttp:feature
element must express a non-empty
sequence of character information items that adheres to the
xsd:anyURI
data type defined by [XML Schema Part 2],
§3.2.17.
<ttp:feature value = (optional|required|use) : required xml:id = ID {any attribute not in default or any TT namespace}> Content: #PCDATA </ttp:feature> |
If the URI expressed by the content of the ttp:feature
element is a relative URI, then, when combined with the
feature namespace value expressed by the
xml:base
attribute of the
nearest ancestor ttp:features
element, it must express an
absolute URI. In either case (original absolute URI or resulting
absolutized URI), the URI expressed by the ttp:feature
element must further adhere to the syntax of a feature designation as
defined by D.1 Feature Designations, and, furthermore,
the specific designation that appears in this URI, i.e., the portion of the feature designation
that starts with the fragment identifier separator '#', must
be defined by this specification or some published version thereof (that
has achieved REC status).
If the URI expressed by the content of the ttp:feature
element
is a relative URI, then an xml:base
attribute should be
specified on the nearest ancestor ttp:features
element.
The value
attribute specifies whether a conforming TTML
processor must or may implement the designated feature in order to
process the document. If the value of the value
attribute
is optional
, then the processor need
not implement or otherwise support the feature in order to process the
document; if the value is required
, then the processor
must implement or otherwise support the feature, irrespective of
whether the feature is enabled or disabled, in order to process
the document; if the value is
use
, then the processor must both (1) implement or
otherwise support the feature and (2) have enabled (activated) use of the
feature.
Note:
The default value of the value
attribute is
required
, as indicated in the above element information
item definition. Therefore, if a value
attribute is not
specified on a ttp:feature
element, it is equivalent to
specifying that support for the feature is required.
If the value of the value
attribute is
required
or use
and the TTML processor implementation does
not support the feature, or if the value
attribute is
use
and the TTML processor implementation supports but has disabled
that feature, then it must not further process the document
without the presence of an explicit override from an end-user or some
implementation specific parameter traceable to an end-user or to a
user or system configuration setting. If a TTML processor aborts
processing of a Document Instance due to the specification of a
required, but unsupported feature by this element, then some end-user
notification should be given unless the end-user or system has
disabled such a notification, or if the processor does not permit or
entail the intervention of an end-user.
If the value of the value
attribute is
optional
, and if the TTML processor implementation does
not support the feature, then it may further process the document even
in the case that some use of the feature is present in the document.
In the case of actual use of a feature designated as optional, the
default semantics associated with that feature apply; that is, the
processor may behave as if the feature were not actually used or
referenced by the document. Notwithstanding the above, the syntactic
presence or reference to an optional feature by a document must not be
considered to be a violation of document validity or a barrier to
further processing if the syntactic expression is well-formed and
valid.
If some defined (i.e., standardized) or otherwise well known feature is not specified by
a ttp:feature
element in a given profile, then it must be interpreted as if the feature were specified
with the value
attribute equal to optional
.
Note:
In particular, if some feature is not present in a profile definition, then it is not to be interpreted as meaning the use of that feature (in a Document Instance) is disallowed or otherwise prohibited.
The ttp:feature
element is illustrated by the following example.
<ttp:profile use="http://www.w3.org/ns/ttml/profile/dfxp-presentation"> <ttp:features xml:base="http://www.w3.org/ns/ttml/feature/"> <ttp:feature value="required">#fontStyle-italic</ttp:feature> <ttp:feature value="required">#textDecoration-under</ttp:feature> </ttp:features> </ttp:profile> |
Note:
In the above example, the DFXP presentation profile is used as the
baseline profile. This baseline profile is then modified by two
ttp:feature
elements in order to
superset the baseline profile (since neither
#fontStyle-italic
nor #textDecoration-under
are required by the DFXP presentation profile).
The effect of this example is to express authorial intentions that italic font style and text underlining must be supported.
The ttp:extensions
element is a container element used to group
infomation about extension support requirements.
The ttp:extensions
element accepts as its children zero or more
elements in the Metadata.class
element group, followed by
zero or more ttp:extension
elements.
<ttp:extensions xml:base = string : TT Extension Namespace xml:id = ID {any attribute not in default or any TT namespace}> Content: Metadata.class*, ttp:extension* </ttp:extensions> |
If specified, the xml:base
attribute must (1) adhere to the
xsd:anyURI
data type defined by [XML Schema Part 2],
§3.2.17, (2) express an absolute URI that
adheres to [XML Base] and,
(3) express an extension namespace as defined by
E.1 Extension Designations. If not specified, the
xml:base
attribute's default
value applies, which is the TT Extension Namespace.
The xml:base
attribute is
used to permit the abbreviation of feature designation URIs expressed
by child ttp:extension
elements.
The ttp:extension
element is used to specify
infomation about support requirements for a particular extension.
The children of the ttp:extension
element must express a non-empty
sequence of character information items that adheres to the
xsd:anyURI
data type defined by [XML Schema Part 2],
§3.2.17.
<ttp:extension value = (optional|required|use) : required xml:id = ID {any attribute not in default or any TT namespace}> Content: #PCDATA </ttp:extension> |
If the URI expressed by the content of the ttp:extension
element is a relative URI, then, when combined with the
extension namespace value expressed by the
xml:base
attribute of the
nearest ancestor ttp:extensions
element, it must express an
absolute URI. In either case (original absolute URI or resulting
absolutized URI), the URI expressed by the ttp:extension
element must further adhere to the syntax of an extension designation
as defined by E.1 Extension Designations.
If the URI expressed by the content of the
ttp:extension
element is a relative URI, then an
xml:base
attribute should be specified on the nearest
ancestor ttp:extensions
element.
The value
attribute specifies whether a conforming TTML
processor must or may implement the designated extension in order to
process the document. If the value of the value
attribute
is optional
, then the processor need
not implement or otherwise support the extension in order to process the
document; if the value is required
, then the processor
must implement or otherwise support the extension in order to process
the document; if the value is
use
, then the processor must both (1) implement or
otherwise support the extension and (2) enable (activate) use of the
extension.
Note:
The default value of the value
attribute is
required
, as indicated in the above element information
item definition. Therefore, if a value
attribute is not
specified on a ttp:extension
element, it is equivalent to
specifying that support for the extension is required.
If the value of the value
attribute is
required
or use
and the TTML processor implementation does
not support the extension, or if the value
attribute is
use
and the TTML processor implementation supports but has disabled
that extension, then it must not further process the document without the
presence of an explicit override from an end-user or some
implementation specific parameter traceable to an end-user or to a
user or system configuration setting. If a TTML processor aborts
processing of a Document Instance due to the specification of a
required, but unsupported extension by this element, then some
end-user notification should be given unless the end-user or system
has disabled such a notification, or if the processor does not permit
or entail the intervention of an end-user.
If the value of the value
attribute is
optional
, and if the TTML processor implementation does
not support the extension, then it may further process the document even
in the case that some use of the extension is present in the document.
In the case of actual use of an extension designated as optional, the
default semantics associated with that extension apply; that is, the
processor may behave as if the extension were not actually used or
referenced by the document. Notwithstanding the above, the syntactic
presence or reference to an optional extension by a document must not be
considered to be a violation of document validity or a barrier to
further processing if the syntactic expression is well-formed and
valid.
If some well known extension is not specified by
a ttp:extension
element in a given profile, then it must be interpreted as if the extension were specified
with the value
attribute equal to optional
.
Note:
In particular, if some extension is not present in a profile definition, then it is not to be interpreted as meaning the use of that extension (in a Document Instance) is disallowed or otherwise prohibited.
The ttp:extension
element is illustrated by the following example.
<ttp:profile use="http://www.w3.org/ns/ttml/profile/dfxp-transformation">
<ttp:extensions xml:base="http://example.org/ttml/extension/">
<ttp:extension value="use">#prefilter-by-language</ttp:extension>
</ttp:extensions>
</ttp:profile>
|
Note:
In the above example, the DFXP transformation profile is used as the baseline profile. This baseline profile is then supersetted by specifying that support and use is required for a private extension defined in a third party namespace.
The effect of this example is to express authorial intentions that a recipient processor must support the DFXP transformation profile and must also support and enable an extension defined by a third party.
The following attributes are defined in the TT Parameter Namespace.
Unless explicitly stated otherwise, linear white-space (LWSP) must appear between adjacent non-terminal components of a TT Parameter value unless some other delimiter is permitted and used.
The ttp:cellResolution
attribute may be used by an author
to express the number of horizontal and vertical cells into which the Root Container Region
area is divided for the purpose of expressing presentation semantics in terms of
a uniform grid.
If specified, the value of this attribute must adhere to the following syntax:
ttp:cellResolution : columns rows // columns != 0; rows != 0 columns | rows : <digit>+ |
If not specified, the number of columns and rows must be considered to be 32 and 15, respectively. If specified, then columns or rows must not be zero (0).
Note:
The choice of values 32 and 15 are based on this being the maximum number of columns and rows defined by [CEA-608-E].
A ttp:cellResolution
attribute is considered to be significant only
when specified on the tt
element.
Note:
The use of a uniform grid is employed only for the purpose of measuring lengths and expressing coordinates. In particular, it is not assumed that the presentation of text or the alignment of individual glyph areas is coordinated with this grid. Such alignment is possible, but requires the use of a monospaced font and a font size whose EM square exactly matches the cell size.
Except where indicated otherwise, when a <length> expressed in cells denotes a dimension parallel to the inline or block progression dimension, the cell's dimension in the inline or block progression dimension applies, respectively.
Note:
For example, if padding (on all four edges) is specified as 0.1c, the cell resolution is 20 by 10, and the extent of the Root Container Region is 640 by 480, then, assuming top to bottom, left to right writing mode, the start and end padding will be (640 / 20) * 0.1 pixels and the before and after padding will be (480 / 10) * 0.1 pixels.
The ttp:clockMode
attribute is used to specify the
interpretation of time expressions as real-time time coordinates when
operating with time base of clock
as defined by 6.2.11 ttp:timeBase.
Note:
See 10.3 Time Value Expressions for the specification of time expression syntax and semantics.
If specified, the value of this attribute must adhere to the following syntax:
ttp:clockMode : "local" | "gps" | "utc" |
If the time base, defined by 6.2.11 ttp:timeBase, is designated as
clock
, then this parameter applies as follows: if the
parameter's value is local
, then time expressions are
interpreted as local wall-clock time coordinates;
if utc
, then time expressions are
interpreted as UTC time coordinates
[UTC];
if gps
, then time expressions are
interpreted as GPS time coordinates
[GPS].
Note:
The primary difference between GPS time and UTC time is that GPS time is not adjusted for leap seconds, while UTC time is adjusted as follows: UTC = TAI (Temp Atomique International) + leap seconds accumulated since 1972. TAI is maintained by the Bureau International des Poids et Mesures (BIPM) in Sevres, France. The GPS system time is steered to a Master Clock (MC) at the US Naval Observatory which is kept within a close but unspecified tolerance of TAI.
If not specified, the value of this parameter must be considered
to be utc
.
A ttp:clockMode
attribute is considered to be significant only
when specified on the tt
element.
The ttp:dropMode
attribute is used to specify
constraints on the interpretation and use of frame counts
that correspond with [SMPTE 12M] time coordinates when operating
with time base of smpte
as defined by
6.2.11 ttp:timeBase.
If specified, the value of this attribute must adhere to the following syntax:
ttp:dropMode : "dropNTSC" | "dropPAL" | "nonDrop" |
If the time base, defined by 6.2.11 ttp:timeBase, is designated as
smpte
, then this parameter applies as follows: if the
parameter's value is nonDrop
, then, within any given
second of a time expression, frames count from 0 to
N−1, where N is the value specified by
the ttp:frameRate
parameter, but while ignoring any value
specified by the ttp:frameRateMultiplier
parameter.
Note:
When operating in nonDrop
mode, a second of a time expression may or may not be equal to a
second of real time during normal (1x speed) forward playback. If the
ttp:frameRateMultiplier
parameter is specified and is not
equal to 1:1, then a second of a time expression will either be
shorter or longer than a second of elapsed play in real
time.
Note:
See 10.3 Time Value Expressions regarding the absence of defined semantics for both the
offset-time form and the fraction component of time expressions when the smpte
time base applies.
If this parameter's value is dropNTSC
, then, within any
given second of a time expression except the second 00
,
frames count from 0 to N−1, where N is
the value specified by the ttp:frameRate
parameter, but
while ignoring any value specified by the
ttp:frameRateMultiplier
parameter. If the second of a time
expression is 00
and the minute of the time expression is
not 00
, 10
, 20
, 30
,
40
, or 50
, then frame codes 00
and 01
are dropped during that second; otherwise, these
frame codes are not dropped.
Note:
For example, when operating in dropNTSC
mode with ttp:frameRate
of 30
, a discontinuity
in frame count occurs between consecutive frames as shown in the
following sequence of time expressions: 01:08:59:28
,
01:08:59:29
, 01:09:00:02
,
01:09:00:03
.
If this parameter's value is dropPAL
, then, within any
given second of a time expression except the second 00
,
frames count from 0 to N−1, where N is
the value specified by the ttp:frameRate
parameter, but
while ignoring any value specified by the
ttp:frameRateMultiplier
parameter. If the second of a time
expression is 00
and the minute of the time expression is
even but not 00
, 20
, or 40
,
then frame codes 00
through 03
are dropped
during that second; otherwise, these frame codes are not dropped.
Note:
For example, when operating in dropPAL
mode with ttp:frameRate
of 30
, a discontinuity
in frame count occurs between consecutive frames as shown in the
following sequence of time expressions: 01:09:59:28
,
01:09:59:29
, 01:10:00:04
,
01:10:00:05
.
Note:
The dropPAL
mode is also known as the
M/PAL or PAL (M) drop-frame
code, which uses PAL modulation with the NTSC frame rate of ~29.97
frames/second. The M/PAL system is used primarily in Brazil.
If not specified, then nonDrop
must be assumed to apply.
A ttp:dropMode
attribute is considered to be significant only
when specified on the tt
element.
The ttp:frameRate
attribute is used to specify the frame rate of a related
media object or the intrinsic frame rate of a Document Instance in case it is intended
to function as an independent media object.
If specified, the value of this attribute must adhere to the following syntax:
ttp:frameRate : <digit>+ // value > 0 |
The frame rate that applies to a Document Instance is used to interpret time expressions that are expressed in frames as defined by 10.3.1 <timeExpression>.
If the media
time base applies and the effective frame rate is
integral, then a frame is interpreted as a division of a second of media
time, such that if the frame rate is specified as F, then
a second of media time is divided into F intervals of
equal duration, where each interval is labeled as frame f,
with f ∈ [0…F−1].
Note:
See N.2 Media Time Base for further details
on the interpretation of time expressions for the media
time base.
If not specified, the frame rate must be considered to be equal to some application defined frame rate, or if no application defined frame rate applies, then thirty (30) frames per second. If specified, then the frame rate must be greater than zero (0).
A ttp:frameRate
attribute is considered to be significant only
when specified on the tt
element.
The ttp:frameRateMultiplier
attribute is used to
specify a multiplier to be applied to the frame rate specified by a
ttp:frameRate
attribute in order to compute the effective
frame rate.
If specified, the value of this attribute must adhere to the following syntax:
ttp:frameRateMultiplier : numerator denominator // numerator != 0; denominator != 0 numerator | denominator : <digit>+ |
A frame rate multiplier is used when the desired frame rate cannot be expressed as an integral number of frames per second.
If not specified, the frame rate multiplier must be considered to be equal to some application defined frame rate multiplier, or if no application defined frame rate multiplier applies, then one (1:1). Both numerator and denominator must be non-zero.
A ttp:frameRateMultiplier
attribute is considered to be significant only
when specified on the tt
element.
Note:
The frame rate multiplier used for synchronizing with NTSC [SMPTE 170M] formatted video objects at 30 frames per second is nominally 1000:1001. The nominal frame rate of NTSC video is defined as the chrominance sub-carrier frequency of 3,579,545.45…Hz (= 5.0MHz × 63/88) times the ratio 2/455 divided by the number of horizontal lines per frame of 525, which yields a frame rate of 29.970029970029… (= 30 × 1000/1001) frames per second. Other frame rate multipliers apply to different regions of usage and video format standards.
Note:
Except in the case of PAL/M, the frame rate multiplier used for synchronizing with PAL formatted video objects at 25 frames per second is nominally 1:1.
The ttp:markerMode
attribute is used to specify
constraints on the interpretation and use of time expressions
that correspond with [SMPTE 12M] time coordinates when operating
with time base of smpte
as defined by
6.2.11 ttp:timeBase.
If specified, the value of this attribute must adhere to the following syntax:
ttp:markerMode : "continuous" | "discontinuous" |
If the time base, defined by 6.2.11 ttp:timeBase, is designated as
smpte
, then this parameter applies as follows: if the
parameter's value is continuous
, then [SMPTE 12M] time coordinates
may be assumed to be linear and either monotonically increasing or
decreasing; however, if discontinuous
, then any assumption
must not be made regarding linearity or monotonicity of time coordinates.
If not specified, the value of this parameter must be considered
to be discontinuous
.
Note:
The default value for this parameter was originally specified
(in TTML 1.0 First Edition)
as continuous
; however, further evaluation of the state of the industry indicates this
choice was incorrect, and that the most common default is discontinuous
.
Note:
Due to lack of industry consensus on the utility and interpretation of the
continuous
marker mode, authors are advised to avoid its use. Furthermore, the
ttp:markerMode
is being considered for deprecation in the next revision of this
specification.
A ttp:markerMode
attribute is considered to be significant only
when specified on the tt
element.
If a value of continuous
applies, then time
expressions may be converted to real time
coordinates by taking into account the computed frame rate
and drop mode as expressed by the ttp:dropMode
parameter.
In this case, the Content Processor must create and maintain a
Synthetic SMPTE Document Syncbase within which these time
expressions are interpreted as further described in
10.4 Time Intervals.
Note:
When operating with smpte
time base and
continuous
marker mode, there is an implied time coordinate
space, the Synthetic SMPTE Document Syncbase,
defined by the monotonically increasing (or decreasing)
[SMPTE 12M] time coordinates, while taking into account
the computed frame rate and drop mode. All time expressions are interpreted
in relationship to this time coordinate space based upon
SMPTE Time Code synchronization events (markers),
where the Document Processing Context emits these events with implied constraints
regarding time coordinate monoticity and resynchronization in
the presence of dropped frames.
Use of continuous
marker mode with the smpte
time base is different from using the media
time base
because (1) the semantics of the ttp:dropMode
parameter
apply to the former, but not the latter, and (2) [SMPTE 12M]
time coordinates may be applied monotonically to media which has been
subjected to dilation in time, constriction in time, or reversal in
time.
If a value of discontinuous
applies, then time
expressions must not be converted to real time
coordinates, arithmetical operators (addition, multiplication) are not
defined on time expressions, and, consequently, any (well-formed)
expression of a duration must be considered to be invalid.
Note:
When operating with smpte
time base and
discontinuous
marker mode, there is no effective time coordinate
space; rather, all time expressions are interpreted as labeled
synchronization events (markers), where the Document Processing Context
emits these events, which, when they correspond with time
expressions that denote the same label, cause a temporal interval to
begin or end accordingly.
An additional side-effect of operating in discontinuous
mode is that time expressions of children have no necessary
relationship with time expressions of their temporal container; that
is, temporal containers and children of these containers are
temporally activated and inactivated independently based on the
occurrence of a labeled synchronization (marker) event.
Note:
The notion of marker discontinuity as captured by this parameter
is logically independent from the method used to count frames as
expressed by the ttp:dropMode
parameter. In particular,
even if the ttp:dropMode
parameter is specified as
dropNTSC
or dropPAL
, the marker mode may be
specified as continuous
, even in the presence of frame count
discontinuities induced by the frame counting method, unless there
were some other non-linearity or discontinuity in marker labeling, for
example, two consecutive frames labeled as 10:00:00:00
and 10:00:01:00
.
The ttp:pixelAspectRatio
attribute may be used by a content author
to express the aspect ratio of non-square pixels in the production of content that
makes use of pixel coordinates.
If specified, the value of this attribute must adhere to the following syntax:
ttp:pixelAspectRatio : width height // width != 0; height != 0 width | height : <digit>+ |
If not specified, then square pixels (i.e., aspect ratio 1:1) must be assumed to apply. If specified, then both width and height must be non-zero.
A ttp:pixelAspectRatio
attribute is considered to be significant only
when specified on the tt
element.
Note:
This parameter may be used by a content transcoder or translator in order to convert pixel measurements between different pixel aspect ratios while still maintaining authorial layout intentions.
The ttp:profile
attribute may be used by a content author
to express the profile of the Timed Text Markup Language (TTML) used
in a Document Instance.
If specified, the value of this attribute must adhere to the
xsd:anyURI
data type defined by [XML Schema Part 2],
§3.2.17, and, further, must specify a
profile designator in accordance with
5.2 Profiles.
A ttp:profile
attribute is considered to be significant only when specified on the
tt
element.
The ttp:subFrameRate
attribute is used to specify the sub-frame rate of a related
media object or the intrinsic sub-frame rate of a Document Instance in case it is intended
to function as an independent media object.
If specified, the value of this attribute must adhere to the following syntax:
ttp:subFrameRate : <digit>+ // value > 0 |
The sub-frame rate that applies to a Document Instance is used to interpret time expressions that are expressed in sub-frames as defined by 10.3.1 <timeExpression>.
If the media
time base applies and the effective frame rate is
integral, a sub-frame is interpreted as a division of a frame of media
time, such that if the sub-frame rate is specified as S, then
a frame of media time is divided into S intervals of
equal duration, where each interval is labeled as sub-frame s,
with s ∈ [0…S−1].
Note:
See N.2 Media Time Base for further details
on the interpretation of time expressions for the media
time base.
If not specified, the sub-frame rate must be considered to be equal to some application defined sub-frame rate, or if no application defined sub-frame rate applies, then one (1). If specified, then the sub-frame rate must be greater than zero (0).
A ttp:subFrameRate
attribute is considered to be significant only
when specified on the tt
element.
Note:
The sub-frame is sometimes referred to as a field in the context of synchronization with an interlaced video media object.
The ttp:tickRate
attribute is used to specify the tick rate of a related
media object or the intrinsic tick rate of content of a Document Instance in case it is intended
to function as an independent media object.
If specified, the value of this attribute must adhere to the following syntax:
ttp:tickRate : <digit>+ // value > 0 |
The tick rate that applies to a Document Instance is used to
interpret time expressions that are expressed in ticks by
using the t
metric as defined by 10.3.1 <timeExpression>.
If the media
time base applies,
a tick is interpreted as an integral division of a second of media
time, such that if the tick rate is specified as T, then
a second of media time is divided into T intervals of
equal duration, where each interval is labeled as tick t,
with t ∈ [0…T−1].
Note:
See N.2 Media Time Base for further details
on the interpretation of time expressions for the media
time base.
If not specified, then if a frame rate is specified, the tick rate must be considered to be the effective frame rate multiplied by the sub-frame rate (i.e., ticks are interpreted as sub-frames); or, if no frame rate is specified, the tick rate must be considered to be one (1) tick per second of media time. If specified, then the tick rate must not be zero (0).
Note:
There is no predefined relationship between ticks and frames or sub-frames. Ticks are an arbitrary division of seconds that permit use of fixed point arithmetic rather than fractional (and potentially inexact) expressions of seconds.
A ttp:tickRate
attribute is considered to be significant only
when specified on the tt
element.
The ttp:timeBase
attribute is used to specify the
temporal coordinate system by means of which time expressions are
interpreted in a Document Instance.
If specified, the value of this attribute must adhere to the following syntax:
ttp:timeBase : "media" | "smpte" | "clock" |
If the time base is designated as media
, then a time
expression denotes a coordinate in some media object's time line,
where the media object may be an external media object with which the
content of a Document Instance is to be synchronized, or it may
be the content of a Document Instance itself in a case where
the timed text content is intended to establish an independent time
line.
Note:
When using a media time base, if that time base is paused or scaled positively or negatively, i.e., the media play rate is not unity, then it is expected that the presentation of associated Timed Text content will be similarly paused, accelerated, or decelerated, respectively. The means for controlling an external media time base is outside the scope of this specification.
If the time base is designated as smpte
,
then a time expression denotes a [SMPTE 12M] time
coordinate with which the content of a Document Instance is to
be synchronized. In this case, the value of the ttp:markerMode
and ttp:dropMode
parameters apply, as defined by 6.2.6 ttp:markerMode and 6.2.3 ttp:dropMode, respectively.
Note:
When the time base is designated as smpte
,
every time expression denotes a media marker value akin to that
defined by [SMIL 2.1], § 10.4.1, except instead
of using an opaque marker name, a structured [SMPTE 12M]
time coordinate serves as the marker name.
If the time base is designated as clock
, then the time
expression denotes a coordinate in some real-world time line as
established by some real-time clock, such as the local wall-clock time
or UTC (Coordinated Universal Time) or GPS (Global Positioning System)
time lines.
If not specified, the default time base must be considered to be
media
.
A ttp:timeBase
attribute is considered to be significant only
when specified on the tt
element.
When operating with either media
or smpte
time bases, a diachronic presentation of a Document Instance may
be subject to transformations of the controlling time line, such as
temporal reversal, dilation (expansion), or constriction
(compression); however, when operating with the clock
time base, no transformations are permitted, and diachronic
presentation proceeds on a linear, monotonically increasing time line
based on the passage of real time.
Note:
Due to there being only one time base parameter that applies to a given Document Instance, the interpretation of time expressions is uniform throughout the Document Instance.
Note:
See N Time Expression Semantics for further details on the interpretation of time expressions according to the designated time base.
This section specifies the content matter of the core vocabulary catalog.
The following elements specify the structure and principal content aspects of a Document Instance:
The tt
element serves as the root document element of a Document Instance.
The tt
element accepts as its children zero or one head
element followed by
zero or one body
element.
The Root Temporal Extent, i.e., the time interval over which a Document Instance is active, has an implicit duration that is equal to the
implicit duration of the body
element of the document, if the body
element is present, or zero, if the body
element is absent.
If the tts:extent
attribute is specified on the tt
element, then it must adhere to 8.2.7 tts:extent, in which case it
specifies the spatial extent of the Root Container Region in
which content regions are located and presented. If no tts:extent
attribute is specified, then the spatial extent of the Root Container Region is
considered to be determined by the Document Processing Context.
The origin of the Root Container Region is determined by the Document Processing Context.
Note:
In the absence of other requirements, and if a Related Media Object exists, then it is recommended that the Document Processing Context determine that:
if no tts:extent
is specified on the root tt
element, the extent of the Root Container Region be
established as equal to the extent of the Related Media Object Region; and
the origin of the Root Container Region be established so that this region is centered in the Related Media Object Region.
Note:
If an author desires to signal the (storage or image) aspect ratio of the Root Container Region
without specifying its resolution, then this may be accomplished by using metadata specified
in an external namespace, such as m708:aspectRatio
as defined in [SMPTE 2052-11], §5.4.4.
This would permit, for example, the interchange of information that reflects the
the semantics of [CEA-708-E] , §4.5 “Caption Service Metadata”, “ASPECT RATIO”.
An xml:lang
attribute must be specified on the tt
element.
If the attribute value is empty, it signifies that there is no default language that applies
to the text contained within the Document Instance.
If no xml:space
attribute is specified upon the tt
element, then it must
be considered as if the attribute had been specified with a value of default
.
The head
element is a container element used to group
header matter, including metadata,
profile, styling, and layout information.
The head
element accepts as its children zero or more
elements in the Metadata.class
element group,
followed by zero or more elements in the
Parameters.class
element group,
followed by zero or one styling
element,
followed by zero or one layout
element.
Any metadata specified by children in the Metadata.class
element group applies semantically to the Document Instance as a
whole, and not just the head
element.
Any parameters specified by children in the Parameters.class
element group applies semantically to the Document Instance as a
whole, and not just the head
element.
A styling
child element is used to specify style constructs
that are referenced from other style constructs, by layout constructs, and by
Content elements.
A layout
child element is used to specify layout constructs that
are referenced by Content elements.
<head xml:id = ID xml:lang = string xml:space = (default|preserve) {any attribute not in default or any TT namespace}> Content: Metadata.class*, Parameters.class*, styling?, layout? </head> |
To the extent that time semantics apply to the content of the
head
element, the implied time interval of this element is
defined to be coterminous with the Root Temporal Extent.
The body
element functions as a logical container and a temporal
structuring element for a sequence of textual content units represented as
logical divisions.
The body
element accepts as its children zero or more
elements in the Metadata.class
element group,
followed by zero or more
elements in the Animation.class
element group,
followed by
zero or more div
elements.
Any metadata specified by children in the Metadata.class
element group applies semantically to the body
element and its descendants as a whole.
Any animation elements specified by children in the Animation.class
element group apply semantically to the body
element.
<body begin = <timeExpression> dur = <timeExpression> end = <timeExpression> region = IDREF style = IDREFS timeContainer = (par|seq) xml:id = ID xml:lang = string xml:space = (default|preserve) {any attribute in TT Metadata namespace} {any attribute in TT Style namespace} {any attribute not in default or any TT namespace}> Content: Metadata.class*, Animation.class*, div* </body> |
An author may specify a temporal interval for a body
element using
the begin
, dur
, and end
attributes. If the begin point
of this interval remains unspecified, then the begin point is interpreted as the beginning
point of the Root Temporal Extent. Similarly, if the end point of this interval remains unspecified, then the
end point is interpreted as the ending point of the Root Temporal Extent.
Note:
A Document Instance referenced from a SMIL presentation is expected to follow the same timing rules as apply to other SMIL media objects.
If relative begin or end times are specified on the body
element, then these times are resolved by reference to the beginning and
ending time of the Root Temporal Extent.
If the Root Temporal Extent is shorter than the computed duration of the
body
element, then the active time interval of a body
element is
truncated to the active end point of the Root Temporal Extent.
An author may associate a set of style properties with a
body
element by means of either the style
attribute or inline style attributes or a combination thereof.
Note:
Style properties that are associated with a body
element
in a Document Instance are available for style inheritance by
descendant Content elements such as div
, p
,
span
and br
.
If no timeContainer
attribute is specified on a
body
element, then it must be interpreted as having
parallel
time containment semantics.
The div
element functions as a logical container and a temporal
structuring element for a sequence of textual content units represented as
logical sub-divisions or paragraphs.
Note:
When rendered on a continuous (non-paged) visual presentation medium,
a div
element is expected to generate
one or more block
areas
that contain zero or more child block areas
generated by the div
element's
descendant
p
elements.
If some
block area generated by a div
element does
not contain any child areas, then it is not expected to be presented.
The div
element accepts as its children zero or more
elements in the Metadata.class
element group,
followed by zero or more
elements in the Animation.class
element group,
followed by
zero or more
div
or
p
elements.
Any metadata specified by children in the Metadata.class
element group applies semantically to the div
element and its descendants as a whole.
Any animation elements specified by children in the Animation.class
element group apply semantically to the div
element.
<div begin = <timeExpression> dur = <timeExpression> end = <timeExpression> region = IDREF style = IDREFS timeContainer = (par|seq) xml:id = ID xml:lang = string xml:space = (default|preserve) {any attribute in TT Metadata namespace} {any attribute in TT Style namespace} {any attribute not in default or any TT namespace}> Content: Metadata.class*, Animation.class*, Block.class* </div> |
An author may associate a set of style properties with a div
element by means of either
the style
attribute or inline style attributes or a combination thereof.
Note:
Style properties that are associated
with
a div
element in a Document Instance
are available for style inheritance by descendant Content elements such as
div
,
p
,
span
, and
br
.
If no timeContainer
attribute is specified on
a div
element, then it must be interpreted as having
parallel time containment semantics.
A p
element represents a logical paragraph, serving as
a transition between block level and inline level formatting semantics.
The p
element accepts as its children zero or more
elements in the Metadata.class
element group,
followed by zero or more
elements in the Animation.class
element group,
followed by
zero or more intermixed
span
elements,
br
elements,
or text nodes.
Any metadata specified by children in the Metadata.class
element group applies semantically to the p
element and its descendants as a whole.
Any animation elements specified by children in the Animation.class
element group apply semantically to the p
element.
<p begin = <timeExpression> dur = <timeExpression> end = <timeExpression> region = IDREF style = IDREFS timeContainer = (par|seq) xml:id = ID xml:lang = string xml:space = (default|preserve) {any attribute in TT Metadata namespace} {any attribute in TT Style namespace} {any attribute not in default or any TT namespace}> Content: Metadata.class*, Animation.class*, Inline.class* </p> |
An author may associate a set of style properties with a
p
element by means of either the style
attribute or inline style attributes or a combination thereof.
Note:
Style properties that are associated with a p
element
in a Document Instance are available for style inheritance by
descendant Content elements such as span
and
br
.
If no timeContainer
attribute is specified on
a p
element, then it must be interpreted as having
parallel time containment semantics.
If a sequence of children of a p
element
consists solely of character information items, then that sequence must
be considered to be an anonymous span for the purpose of
applying style properties that apply to span
elements.
Note:
The presentation semantics of TTML effectively
implies that a p
element constitutes a line break. In particular,
it is associated with a block-stacking constraint both before the first
generated line area and after the last generated line area. See
9.3.4 Synchronic Flow Processing for further details.
The span
element functions as a logical container and a temporal
structuring element for a sequence of textual content units having inline
level formatting semantics.
When presented on a visual medium, a span
element is intended to
generate a sequence of inline areas, each containing one or more glyph areas.
The span
element accepts as its children zero or more
elements in the Metadata.class
element group,
followed by zero or more
elements in the Animation.class
element group,
followed by
zero or more intermixed
span
elements,
br
elements,
or text nodes.
Any metadata specified by children in the Metadata.class
element group applies semantically to the span
element and its descendants as a whole.
Any animation elements specified by children in the Animation.class
element group apply semantically to the span
element.
<span begin = <timeExpression> dur = <timeExpression> end = <timeExpression> region = IDREF style = IDREFS timeContainer = (par|seq) xml:id = ID xml:lang = string xml:space = (default|preserve) {any attribute in TT Metadata namespace} {any attribute in TT Style namespace} {any attribute not in default or any TT namespace}> Content: Metadata.class*, Animation.class*, Inline.class* </span> |
An author may associate a set of style properties with a
span
element by means of either the style
attribute or inline style attributes or a combination thereof.
Note:
Style properties that are associated with a span
element
in a Document Instance are available for style inheritance by
descendant Content elements such as span
and
br
.
If no timeContainer
attribute is specified on
a span
element, then it must be interpreted as having
parallel time containment semantics.
The br
element denotes an explicit line break.
Any metadata specified by children in the Metadata.class
element group applies semantically to the br
element and its descendants as a whole.
Any animation elements specified by children in the Animation.class
element group apply semantically to the br
element.
Note:
While permitted, the use of Animation.class
children is not recommended since no style property
applies to the br
element. Instead, animating a br
element can be achieved by wrapping it in an animated span
element.
<br style = IDREFS xml:id = ID xml:lang = string xml:space = (default|preserve) {any attribute in TT Metadata namespace} {any attribute in TT Style namespace} {any attribute not in default or any TT namespace}> Content: Metadata.class*, Animation.class* </br> |
When presented on a visual medium, the presence of a br
element must be
interpreted as a forced line break.
Note:
The visual presentation of a br
element is
intended to produce the same effect as the control character
CR
(U+000D)
followed by the control code LF
(U+000A) when presented on a teletype device.
Therefore, two br
elements in sequence will produce a
different effect than a single br
element.
This section defines the following common attributes used with many or all element types in the core vocabulary catalog:
The xml:id
attribute is used as defined by
[XML ID].
The xml:id
attribute may be used with any element in the
core vocabulary catalog.
The xml:lang
attribute is used as defined by
[XML 1.0], §2.12, Language Identification.
The xml:lang
attribute must be specified on the tt
element and may be specified by an instance of any other element type
in the core vocabulary catalog except parameter vocabulary.
The xml:space
attribute is used as defined by
[XML 1.0], §2.10, White Space Handling.
The xml:space
attribute may be used with any element in
the core vocabulary catalog except parameter vocabulary.
The semantics of the value default
are fixed to mean that
when performing presentation processing of a Document Instance as
described by 9.3.4 Synchronic Flow Processing,
processing must occur as
if the following properties were specified on the affected
elements of an equivalent intermediate XSL-FO document:
suppress-at-line-break="auto"
linefeed-treatment="treat-as-space"
white-space-collapse="true"
white-space-treatment="ignore-if-surrounding-linefeed"
Similarly, the semantics of the value preserve
are fixed to mean that
when performing presentation processing, processing must occur as
if the following properties were specified on the affected
elements of an equivalent intermediate XSL-FO document:
suppress-at-line-break="retain"
linefeed-treatment="preserve"
white-space-collapse="false"
white-space-treatment="preserve"
When performing other types of processing intended to eventually
result in a visual presentation by means other than those described
in this specification, the semantics of space collapsing and
preservation as described above should be respected. For other types
of processing, the treatment of the xml:space
attribute
is processor dependent, but should respect the semantics described
above if possible.
Note:
The semantics of the above four cited XSL-FO properties are defined by by [XSL 1.1], § 7.17.3, 7.16.7, 7.16.12, and 7.16.8, respectively.
Note:
No presentation semantics are specified for the TAB (U+0009) character. Furthermore, the TAB (U+0009) character
can generate a glyph area. As a result, the use of the TAB (U+0009) character
in #PCDATA
content within p
and span
elements is not recommended.
This section specifies the styling matter of the core vocabulary catalog, where styling is to be understood as a separable layer of information that applies to content and that denotes authorial intentions about the presentation of that content.
Styling attributes are included in TTML to enable authorial intent of presentation to be included within a self-contained document. This section describes the semantics of style presentation in terms of a standard processing model. TTML Processors are not required to present Document Instances in any particular way; but an implementation of this model by a TTML Presentation Processor that provides externally observable results that are consistent with this model is likely to lead to a user experience that closely resembles the experience intended by the documents' authors.
Note:
The semantics of TTML style presentation are described in terms of the layout and formatting model defined in [XSL 1.1]. The effects of the attributes in this section are intended to be compatible with this model; however, presentation agents may use any technology to satisfy the authorial intent of the document. For example, a [CSS2] processor may be used to implement features it has in common with this model.
No normative use of an
<?xml-stylesheet ... ?>
processing instruction is defined
by this specification.
The following elements specify the structure and principal styling aspects of a Document Instance:
The styling
element is a container element used to group
styling matter, including metadata that applies to styling matter.
The styling
element accepts as its children zero or more
elements in the Metadata.class
element group, followed by
zero or more style
elements.
<styling xml:id = ID xml:lang = string xml:space = (default|preserve) {any attribute not in default or any TT namespace}> Content: Metadata.class*, style* </styling> |
To the extent that time semantics apply to the content of the
styling
element, the implied time interval of this element is
defined to be coterminous with the Root Temporal Extent.
The style
element is used to define
a set of style specifications expressed as a
specified style set in accordance with
8.4.4.2 Specified Style Set Processing.
If a style
element appears as a descendant of a
region
element, then the style
element must be
ignored for the purpose of computing referential styles as defined by
8.4.1.2 Referential Styling and 8.4.1.3 Chained Referential Styling.
Note:
That is to say, when referential styling is used by an element to
refer to a style
element, then the referenced style
element must appear as a descendant of the styling
element, and
not in any other context.
This section defines the 8.2.1 style attribute used with both style definition elements as well as Content elements.
In addition, this section specifies the following attributes in the TT Style Namespace for use with style definition elements, certain layout elements, and Content elements that support inline style specifications:
Unless explicitly stated otherwise, linear white-space (LWSP) must appear between adjacent non-terminal components of a value of a TT Style or TT Style Extension Property value unless some other delimiter is permitted and used.
Note:
This specification makes use of lowerCamelCased local names for style attributes that are based upon like-named properties defined by [XSL 1.1]. This convention is likewise extended to token values of such properties.
Note:
A style property may be expressed as a specified attribute on any Content element type independently of whether the property applies to that element type. This capability permits the expression of an inheritable style property on ancestor elements to which the property does not apply.
Note:
Due to the general syntax of this specification (and the schemas it references) with respect to how style attributes are specified, particularly for the purpose of supporting inheritance, it is possible for an author to inadvertently specify a non-inheritable style attribute on an element that applies neither to that element or any of its descendants while still remaining conformant from a content validity perspective. Content authors may wish to make use of TTML content verification tools that detect and warn about such usage.
The style
attribute is used by referential style association
to reference one or more style
elements each of which define a style (property) set.
The style
attribute may be specified by an instance of the following
element types:
If specified, the value of a style
attribute must
adhere to the IDREFS
data type defined by [XML Schema Part 2],
§ 3.3.10, and, furthermore, each IDREF must reference a style
element
which has a styling
element as an ancestor.
If the same IDREF, ID1, appears more than one time in the value of a style
attribute, then there should be an intervening IDREF, ID2, where
ID2 is not equal to ID1.
Note:
This constraint is intended to discourage the use of redundant referential styling while still allowing the same style to be referenced multiple times in order to potentially override prior referenced styles, e.g., when an intervening, distinct style is referenced in the IDREFS list.
Note:
See the specific element type definitions that permit use of the
style
attribute, as well as 8.4.1.2 Referential Styling and 8.4.1.3 Chained Referential Styling, for
further information on its semantics.
The tts:backgroundColor
attribute is used to specify a style property that
defines the background color of a region or an area generated by content flowed into a region.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: | <color> |
Initial: | transparent |
Applies to: |
body ,
div ,
p ,
region ,
span
|
Inherited: | no |
Percentages: | N/A |
Animatable: | discrete |
For the purpose of determining applicability of this style property,
each character child of a p
element is considered to be enclosed in an anonymous
span.
The tts:backgroundColor
style is illustrated by the following example.
<region xml:id="r1"> <style tts:extent="306px 114px"/> <style tts:backgroundColor="red"/> <style tts:color="white"/> <style tts:displayAlign="after"/> <style tts:padding="3px 40px"/> </region> ... <p region="r1" tts:backgroundColor="purple" tts:textAlign="center"> Twinkle, twinkle, little bat!<br/> How <span tts:backgroundColor="green">I wonder</span> where you're at! </p> |
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.8.2.
The tts:color
attribute is used to specify a style property that
defines the foreground color of marks associated with an area generated by content flowed into a region.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: | <color> |
Initial: | see prose |
Applies to: |
span
|
Inherited: | yes |
Percentages: | N/A |
Animatable: | discrete |
For the purpose of determining applicability of this style property,
each character child of a p
element is considered to be enclosed in an anonymous
span.
The initial value of the tts:color
property
is considered to be implementation dependent. In the absence of
end-user preference information, a conformant presentation processor
should use an initial value that is highly contrastive to the
background color of the Root Container Region.
The tts:color
style is illustrated by the following example.
<region xml:id="r1"> <style tts:backgroundColor="black"/> <style tts:color="white"/> <style tts:displayAlign="after"/> <style tts:textAlign="center"/> </region> ... <p region="r1"> In spring, when woods are <span tts:color="green">getting green</span>,<br/> I'll try and tell you what I mean. </p> |
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.18.1.
The tts:direction
attribute is used to specify a style property that, depending on the context of use,
determines (1) the bidirectional paragraph level, or (2) the directionality of a bidirectional embedding or override.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
ltr |
rtl
|
Initial: |
ltr
|
Applies to: |
p ,
span
|
Inherited: | yes |
Percentages: | N/A |
Animatable: | discrete |
For the purpose of determining applicability of this style property,
each character child of a p
element is considered to be enclosed in an anonymous
span.
When applied to a p
element, the computed value of this property explicitly establishes the
paragraph level as specified by [UAX9], § 4.3, Higher Level Protocol HL1.
When applied to a span
element (or anonymous span), the computed value of this property, in combination with the computed value
of the tts:unicodeBidi
style property, determines the directionality of a bidirectional embedding or override as specified by
[UAX9], § 4.3, Higher Level Protocol HL3.
If a computed value of the property associated with this attribute is not supported,
then a presentation processor must use the value ltr
.
The tts:direction
style is illustrated by the following example.
<region xml:id="r1">
<style tts:extent="265px 84px"/>
<style tts:padding="5px"/>
<style tts:backgroundColor="black"/>
<style tts:color="white"/>
<style tts:displayAlign="after"/>
<style tts:textAlign="center"/>
</region>
...
<p region="r1">
Little birds are playing<br/>
Bagpipes on the shore,<br/>
<span tts:unicodeBidi="bidiOverride" tts:direction="rtl">where the tourists snore.</span>
</p>
|
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.29.1.
The tts:display
attribute is used to specify a style property that
defines whether an element is a candidate for layout and composition
in a region.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
auto |
none
|
Initial: |
auto
|
Applies to: |
body ,
div ,
p ,
region ,
span
|
Inherited: | no |
Percentages: | N/A |
Animatable: | discrete |
For the purpose of determining applicability of this style property,
each character child of a p
element is considered to be enclosed in an anonymous
span.
If the value of this attribute is auto
,
then the affected element is a candidate for region layout and
presentation; however, if the value is none
, then the
affected element and its descendants must be considered ineligible for region
layout and presentation.
The tts:display
style is illustrated by the following example.
<region xml:id="r1"> <style tts:extent="369px 119px" tts:backgroundColor="black" tts:color="white" tts:displayAlign="before" tts:textAlign="start"/> </region> ... <div region="r1"> <p dur="5s"> [[[ <span tts:display="none"> <set begin="1s" dur="1s" tts:display="auto"/> Beautiful soup, </span> <span tts:display="none"> <set begin="2s" dur="1s" tts:display="auto"/> so rich and green, </span> <span tts:display="none"> <set begin="3s" dur="1s" tts:display="auto"/> waiting in a hot tureen! </span> ]]] </p> </div> |
Note:
The semantics of the style property represented by this attribute are based upon that defined by [CSS2], § 9.2.4.
The tts:displayAlign
attribute is used to specify a style property that
defines the alignment of block areas in the block progression direction.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
before |
center |
after
|
Initial: |
before
|
Applies to: |
region
|
Inherited: | no |
Percentages: | N/A |
Animatable: | discrete |
If a computed value of the property associated with this attribute is not supported,
then a presentation processor must use the value before
.
The tts:displayAlign
style is illustrated by the following example.
<region xml:id="r1"> <style tts:extent="128px 66px" tts:origin="0px 0px" tts:backgroundColor="black" tts:color="white"/> <style tts:displayAlign="before"/> <style tts:textAlign="start"/> </region> <region xml:id="r2"> <style tts:extent="192px 66px" tts:origin="128px 66px"/> tts:backgroundColor="green" tts:color="white"/> <style tts:displayAlign="after"/> <style tts:textAlign="start"/> </region> <region xml:id="r3"> <style tts:extent="128px 66px"/> style tts:origin="0px 132px" tts:backgroundColor="black" tts:color="white"/> <style tts:displayAlign="before"/> <style tts:textAlign="start"/> </region> <region xml:id="r4"> <style tts:extent="192px 66px" tts:origin="128px 198px"/> tts:backgroundColor="green" tts:color="white"/> <style tts:displayAlign="after"/> <style tts:textAlign="start"/> </region> ... <div> <p region="r1">I sent a message to the fish:</p> <p region="r2">I told them<br/> "This is what I wish."</p> <p region="r3">The little fishes of the sea,</p> <p region="r4">They sent an<br/> answer back to me.</p> </div> |
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.14.4.
The tts:extent
attribute is used to specify the
width and height of a region area
(which may be the Root Container Region).
Note:
If padding is applied to a region, then the region's extent includes that padding, i.e., padding is applied as an inset, and, therefore, is interior to the region's extent.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
auto |
<length> <length>
|
Initial: | auto |
Applies to: |
tt ,
region
|
Inherited: | no |
Percentages: | relative to width and height of Root Container Region |
Animatable: | discrete |
If the value of this attribute consists of two <length> specifications, then they must be interpreted as width and height, where the first specification is the width, and the second specification is the height.
The <length> value(s) used to express extent must be non-negative.
If the value of this attribute is auto
, then the computed value
of the style property must be considered to be the same as the extent of the Root Container Region.
The extent of the Root Container Region is determined either by a
tts:extent
specified on the tt
element, if
present, or as described by 7.1.1 tt if not present.
If tts:extent
is specified on the tt
element, then
the width and height must be expressed in terms of two <length> specifications,
and these specifications must be expressed as non-percentage, definite lengths using pixel units.
If a computed value of the property associated with this attribute is not supported, then a presentation processor must use the closest supported value.
Note:
In this context, the phrase closest supported value means the value for which the Euclidean distance between the computed extent and the supported extent is minimized. If there are multiple closest supported values equally distant from the computed value, then the value most distant from [0,0], i.e., of greatest extent, is used.
This rule for resolving closest supported value for the tts:extent
attribute makes use of the
nearest larger rather than nearest smaller supported distance. The rationale for this difference
in treatment is that use of a larger extent ensures that the affected content will be contained in
the region area without causing region overflow, while use of a smaller extent makes region
overflow more likely.
The tts:extent
style is illustrated by the following example.
<region xml:id="r1">
<style tts:extent="330px 122px"/>
<style tts:backgroundColor="black"/>
<style tts:color="white"/>
<style tts:displayAlign="after"/>
<style tts:textAlign="center"/>
</region>
...
<p region="r1">
'Tis the voice of the Lobster:<br/>
I heard him declare,<br/>
"You have baked me too brown,<br/>
I must sugar my hair."
</p>
|
The tts:fontFamily
attribute is used to specify a style property that
defines the font family from which glyphs are selected for glyph areas generated
by content flowed into a region.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
(<familyName> | <genericFamilyName>)
(","
(<familyName> | <genericFamilyName>))*
|
Initial: | default |
Applies to: |
span
|
Inherited: | yes |
Percentages: | N/A |
Animatable: | discrete |
Note:
While permitted, the use of linear whitespace (LWSP) around the comma delimiters in the value of this attribute is not recommended. The rationale for this recommendation is that some existing processors may not be tolerant of this usage.
Note:
The initial value, default
, is a generic font family name, and is
further described in 8.3.6 <genericFamilyName> below.
For the purpose of determining applicability of this style property,
each character child of a p
element is considered to be enclosed in an anonymous
span.
If a computed value of the property associated with this attribute is not supported, then a
presentation processor must attempt to map the computed font family
to a supported font family that has similar typographic
characteristics, or, in the absence of such a mapping, it must
use the value default
.
The tts:fontFamily
style is illustrated by the following example.
<region xml:id="r1"> <style tts:extent="474px 146px"/> <style tts:backgroundColor="black"/> <style tts:color="white"/> <style tts:displayAlign="center"/> <style tts:textAlign="start"/> <style tts:fontFamily="proportionalSansSerif"/> </region> ... <div region="r1"> <p> "The time has come," the Walrus said,<br/> "to talk of many things: </p> <p tts:textAlign="end" tts:fontFamily="monospaceSerif"> Of shoes, and ships, and sealing wax,<br/> Of cabbages and kings, </p> <p> And why the sea is boiling hot,<br/> and whether pigs have wings." </p> </div> |
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.9.2.
The tts:fontSize
attribute is used to specify a style property
that defines the font size for glyphs that are selected for glyph areas
generated by content flowed into a region.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: | <length> <length>? |
Initial: | 1c |
Applies to: |
span
|
Inherited: | yes |
Percentages: | if not region element, then relative to the parent element's computed font size; otherwise, relative to the Computed Cell Size. |
Animatable: | discrete |
The computed value of the property associated with this attribute consists of a 2-tuple the entries of which denote the computed values of the width and height components of the font size respectively.
If a single <length> value is specified, then this length applies equally to horizontal and vertical size of a glyph's EM square; if two <length> values are specified, then the first expresses the horizontal size and the second expresses vertical size.
Note:
A glyph's EM square is conventionally defined as the EM square of the font that contains the glyph. That is, glyphs do not have an EM square that is distinct from their font's EM square.
Note:
Anamorphic font scaling, i.e., fonts scaled to independent (and distinct) horizontal and vertical sizes, is expected to be used in a number of contexts, such as when an author desires to synthesize a narrow or wide font face, when an author desires to employ font sizes based on non-square cell units, etc.
If horizontal and vertical sizes are expressed independently, then the units of the <length> values must be the same.
Relative <length> values
that appear in this attribute, i.e., values expressed in percentage (%
), cell (c
), or EM (em
) units,
are resolved in relation to a referenced 2-dimensional size value consisting of two components, a width component and
a height component.
When a single relative <length> value is specified, then this <length> is resolved in terms of the height component of the referenced value; when two relative <length> values are specified, the first <length> is resolved in terms of the width component of the referenced value and the second <length> is resolved in terms of the height component of the referenced value.
If the relative unit is cell (c
), then the referenced value is the
Computed Cell Size.
Note:
For example, consider a paragraph (p
) element P
.
If the Computed Cell Size is (24px,36px), and if
tts:fontSize="1c"
is specified on P
, then the computed value of tts:fontSize
resolves to (36px,36px), while if tts:fontSize="1c 1c"
is specified on P
,
then the computed value resolves to (24px,36px).
If the relative unit is EM (em
), then the referenced value is determined as if percentage units
were used, where 1em
equals 100%
.
Note:
For example, consider a span element S
, a child of a paragraph (p
) element P
.
If the computed value of tts:fontSize
on P
is (18px,24px), and if
tts:fontSize="1em"
is specified on S
, then this is equivalent to specifying 100%
,
which resolves to (24px,24px). However, if tts:fontSize="1em 1em"
is specified on S
,
then this is equivalent to specifying 100% 100%
which resolves to (18px,24px).
The <length> value(s) used to express font size must be non-negative.
For the purpose of determining applicability of this style property,
each character child of a p
element is considered to be enclosed in an anonymous
span.
If a computed value of the property associated with this attribute is not supported, then a presentation processor must use the closest supported value.
Note:
In this context, the phrase closest supported value means the value for which the Euclidean distance between the computed font size and the supported font size is minimized. If there are multiple closest supported values equally distant from the computed value, then the value most distant from 0 (single length specification) or [0,0] (two length specifications) is used, i.e., the largest font size, is used.
The tts:fontSize
style is illustrated by the following example.
<region xml:id="r1"> <style tts:extent="299px 97px"/> <style tts:backgroundColor="black"/> <style tts:color="white"/> <style tts:displayAlign="center"/> <style tts:textAlign="center"/> <style tts:fontFamily="proportionalSansSerif"/> <style tts:fontSize="18px"/> </region> ... <p region="r1"> Then fill up the glasses<br/> with treacle and ink,<br/> Or anything else<br/> that is <span tts:fontSize="24px">pleasant</span> to drink. </p> |
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.9.4. The addition of a second length component to permit specifying font width and height independently is an extension introduced by TTML.
The tts:fontStyle
attribute is used to specify a style property that
defines the font style to apply to glyphs that are selected for glyph areas generated
by content flowed into a region, where the mapping from font
style value to specific font face or style parameterization is not determined
by this specification.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
normal |
italic |
oblique
|
Initial: | normal |
Applies to: |
span
|
Inherited: | yes |
Percentages: | N/A |
Animatable: | discrete |
For the purpose of determining applicability of this style property,
each character child of a p
element is considered to be enclosed in an anonymous
span.
Use of the value oblique
denotes a shear transformation (at an unspecified
angle) in the inline progression dimension.
If a computed value of the property associated with this attribute is not supported,
then a presentation processor must use the value normal
.
The tts:fontStyle
style is illustrated by the following example.
<region xml:id="r1">
<style tts:extent="331px 84px"/>
<style tts:backgroundColor="black"/>
<style tts:color="white"/>
<style tts:displayAlign="after"/>
<style tts:textAlign="center"/>
<style tts:fontFamily="proportionalSansSerif"/>
</region>
...
<p region="r1">
In autumn, when the leaves are brown,<br/>
Take pen and ink, and <span tts:fontStyle="italic">write it down.</span>
</p>
|
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.9.7.
The tts:fontWeight
attribute is used to specify a style property
that defines the font weight to apply to glyphs that are selected for glyph
areas generated by content flowed into a region, where the mapping from font
weight value to specific font face or weight parameterization is not determined
by this specification.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
normal |
bold
|
Initial: | normal |
Applies to: |
span
|
Inherited: | yes |
Percentages: | N/A |
Animatable: | discrete |
For the purpose of determining applicability of this style property,
each character child of a p
element is considered to be enclosed in an anonymous
span.
If a computed value of the property associated with this attribute is not supported,
then a presentation processor must use the value normal
.
The tts:fontWeight
style is illustrated by the following example.
<region xml:id="r1"> <style tts:extent="376px 95px"/> <style tts:backgroundColor="black"/> <style tts:color="white"/> <style tts:displayAlign="center"/> <style tts:textAlign="center"/> <style tts:fontFamily="proportionalSansSerif"/> </region> ... <p region="r1"> They told me you had been to her,<br/> <span tts:fontWeight="bold">and mentioned me to him:</span><br/> She gave me a good character<br/> <span tts:fontWeight="bold">but said I could not swim.</span> </p> |
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.9.9.
The tts:lineHeight
attribute is used to specify a style property that
defines the inter-baseline separation between line areas generated by content flowed into a region.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
normal |
<length>
|
Initial: | normal |
Applies to: |
p
|
Inherited: | yes. See also special inheritance semantics. |
Percentages: | relative to this element's font size |
Animatable: | discrete |
If the value of this attribute is normal
, then the
used value of this style property should be determined as follows:
Let P be the p
element to which this style property applies.
Let FF be the computed value of the tts:fontFamily
style property of P.
Let FS be the computed value of the tts:fontSize
style property of P.
Let F0 be the first font obtained when sequentially mapping each font family in FF to a set of available fonts,
where this set of available fonts is constrained as needed to satisfy the computed values of the
tts:fontStyle
and
tts:fontWeight
style properties of P.
If F0 is associated with font metrics that specify altitude A, descent D, and line gap G, then set LH to the sum of scaled(A), scaled(D), and scaled(G), where scaled(X) denotes font metric X scaled according to font size FS.
Otherwise, LH is considered to be implementation dependent; however, in the absence of implementation specific requirements, LH is recommended to be set to 125% of FS.
Set the used value of this style property to LH.
Note:
If a content author wishes to avoid the possibility of different interpretations of normal
, for example, due to differences in
the set of available fonts, then it is recommended that a
<length> value expression be used to explicitly specify line height value.
It is the intention of this specification that the above algorithm be compatible with [XSL 1.1] and [CSS2].
If specified as a <length>, then the length must be non-negative.
If a computed value of the property associated with this attribute is not supported, then a presentation processor must use the closest supported value.
Note:
In this context, the phrase closest supported value means the value for which the Euclidean distance between the computed line height and the supported line height is minimized. If there are multiple closest supported values equally distant from the computed value, then the value most distant from 0, i.e., the largest line height, is used.
The tts:lineHeight
style is illustrated by the following example.
<region xml:id="r1">
<style tts:extent="255px 190px"/>
<style tts:backgroundColor="black"/>
<style tts:color="white"/>
<style tts:displayAlign="center"/>
<style tts:textAlign="start"/>
<style tts:fontFamily="proportionalSansSerif"/>
<style tts:fontSize="16px"/>
<style tts:lineHeight="32px"/>
</region>
...
<p region="r1">
He thought he saw an elephant,<br/>
That practised on a fife:<br/>
He looked again, and found it was<br/>
A letter from his wife.<br/>
"At length I realise," he said,<br/>
"The bitterness of Life.
"</p>
|
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.16.4. Furthermore, it is the intention of this specification that the allocation rectangle of a line be consistent with the per-inline-height-rectangle as defined by [XSL 1.1], § 4.5, i.e., that a CSS-style line box stacking strategy be used.
When the value of the attribute is normal
and when applying inheritance semantics, the value normal
, not the computed value, is inherited. However, when this computed value is to be used to perform paragraph layout, it is further processed to obtain a used value as described above.
The tts:opacity
attribute is used to specify a style property that
defines the opacity (or conversely, the transparency)
of marks associated with a region.
When presented onto a visual medium, the opacity of the region is applied uniformly and on a linear scale to all marks produced by content targeted to the region.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: | <alpha> |
Initial: |
1.0
|
Applies to: |
region
|
Inherited: | no |
Percentages: | N/A |
Animatable: | discrete |
The tts:opacity
style is illustrated by the following example.
<region xml:id="r1" dur="5s"> <set begin="0s" dur="1s" tts:opacity="1.00"/> <set begin="1s" dur="1s" tts:opacity="0.75"/> <set begin="2s" dur="1s" tts:opacity="0.50"/> <set begin="3s" dur="1s" tts:opacity="0.25"/> <set begin="4s" dur="1s" tts:opacity="0.00"/> <style tts:extent="304px 77px"/> <style tts:backgroundColor="black"/> <style tts:color="white"/> <style tts:displayAlign="after"/> <style tts:textAlign="center"/> </region> ... <p region="r1"> The sun was shining on the sea </p> |
Note:
The semantics of the style property represented by this attribute are based upon that defined by [CSS3 Color], § 3.2.
The tts:origin
attribute is used to specify the
x and y coordinates of the origin of a
region area with respect to the origin of the Root Container Region.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
auto |
<length> <length>
|
Initial: | auto |
Applies to: |
region
|
Inherited: | no |
Percentages: | relative to width and height of Root Container Region |
Animatable: | discrete |
If the value of this attribute consists of two <length> specifications, then they must be interpreted as x and y coordinates, where the first specification is the x coordinate, and the second specification is the y coordinate.
If the value of this attribute is auto
, then the computed value
of the style property must be considered to be the same as the origin of the Root Container Region.
If a computed value of the property associated with this attribute is not supported, then a presentation processor must use the closest supported value.
Note:
In this context, the phrase closest supported value means the value for which the Euclidean distance between the computed origin and the supported origin is minimized. If there are multiple closest supported values equally distant from the computed value, then the value least distant from [0,0], i.e., closest to the coordinate space origin, is used.
The tts:origin
style is illustrated by the following example.
<region xml:id="r1">
<style tts:origin="40px 40px"/>
<style tts:extent="308px 92px"/>
<style tts:backgroundColor="black"/>
<style tts:color="white"/>
<style tts:displayAlign="center"/>
<style tts:textAlign="center"/>
</region>
...
<p region="r1">
"To dine!" she shrieked in dragon-wrath.<br/>
"To swallow wines all foam and froth!<br/>
To simper at a table-cloth!"
</p>
|
The tts:overflow
attribute is used to specify a style property that
defines whether a region area is clipped or not if the descendant areas of the region overflow
its extent.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
visible |
hidden
|
Initial: | hidden |
Applies to: |
region
|
Inherited: | no |
Percentages: | N/A |
Animatable: | discrete |
If the value of this attribute is visible
, then content should
not be clipped outside of the affected region, and region composition and layout
must be performed as if the region's width and height were
unconstrained,
but with
a well-defined origin. If the value is hidden
, then content should
be clipped outside of the affected region.
Note:
This attribute has no impact on presentation processing when no
overflow condition applies. An overflow condition applies when the
bounding box of some descendant line area extends outside of the
containing region's nominal content area extent in either or both 1) the
inline and 2) the block progression dimensions, where the nominal
content area extent in both dimensions is determined by the computed
values of the tts:extent
and tts:padding
style properties of the
containing region. Overflow in the inline progression dimension can
occur only if tts:wrapOption
is noWrap
. Furthermore, when an overflow
condition applies, it is not intended that the effective extent of the
region be modified for the purpose of presentation processing. For
example, the area painted with the region's background color is not
extended in either dimension to enclose the overflowed content.
Note that, in particular, the normative text in the previous paragraph "region composition and layout must be performed as if the region's width and height were unconstrained" refers to the process of determining the effective extent and origin of descendant line areas produced in either (or both) of the two overflow contexts described here, and is not intended to imply that the region extent is altered for the purpose of determining the region's padding area insets or the extent of its background color. More specifically, the normative language above is not intended to override or contradict the semantics of [XSL 1.1], § 7.21.2, or of [CSS2], § 11.1.1, on which the former is based.
Note:
Unless a manual line break element br
is used by the content author,
a paragraph of a given region will generate no more than one line area in that
region if the value of the tts:overflow
style that applies to the
region is visible
and if the applicable tts:wrapOption
style is noWrap
.
If a computed value of the property associated with this attribute is not supported,
then a presentation processor must use the value hidden
.
The tts:overflow
style is illustrated by the following example.
<?xml version="1.0"?> <tt xmlns="http://www.w3.org/ns/ttml" xmlns:tts="http://www.w3.org/ns/ttml#styling" xml:lang="en" tts:extent="320px 240px"> <head> <styling> <style xml:id="s1" tts:backgroundColor="black" tts:padding="6px"/> <style xml:id="s2" tts:color="red" tts:backgroundColor="white" tts:wrapOption="noWrap" tts:fontFamily="proportionalSansSerif" tts:fontSize="10px"/> </styling> <layout> <region xml:id="r1" style="s1" tts:extent="100px 40px" tts:origin="20px 20px" tts:overflow="visible"/> <region xml:id="r2" style="s1" tts:extent="100px 40px" tts:origin="20px 180px" tts:overflow="hidden"/> </layout> </head> <body> <div> <p region="r1" style="s2"> "But wait a bit," the Oysters cried,<br/> "Before we have our chat; </p> <p region="r2" style="s2"> For some of us are out of breath,<br/> And all of us are fat!" </p> </div> </body> </tt> |
Note:
In the above example, the tts:noWrap
is set to
noWrap
to prevent automatic line wrapping (breaking);
if this were not specified, then overflow would occur in the block
progression direction as opposed to the inline progression direction.
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.21.2.
The tts:padding
attribute is used to specify padding (or inset)
space on all sides of a region area.
Note:
Padding is applied as an inset to a region area, which is to say, the content rectangle of a region area is reduced by the presence of padding applied to the region.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: | <length> | <length> <length> | <length> <length> <length> | <length> <length> <length> <length> |
Initial: | 0px |
Applies to: |
region
|
Inherited: | no |
Percentages: | relative to width and height of region |
Animatable: | discrete |
If the value of this attribute consists of one <length> specification, then that length applies to all edges of the affected areas. If the value consists of two <length> specifications, then the first applies to the before and after edges, and the second applies to the start and end edges. If three <length> specifications are provided, then the first applies to the before edge, the second applies to the start and end edges, and the third applies to the after edge. If four <length> specifications are provided, then they apply to before, end, after, and start edges, respectively.
Note:
Percentage values are relative to the dimension of the region to which they apply. For example, if the before and after edges correspond
to the top and bottom edges of the region, as is the case for Latin text where tts:writingMode
is equal to "lrtb"
,
then percentage values that apply to either of the two edges are relative to the height of the region. Conversely, if the before and after
edges correspond to the right and left edges of the region, as is the case for Japanese text where tts:writingMode
is equal to
"tbrl"
, then percentage values that apply to either of the two edges are relative to the width of the region.
The <length> value(s) used to express padding must be non-negative.
If a computed value of the property associated with this attribute is not supported, then a presentation processor must use the closest supported value.
Note:
In this context, the phrase closest supported value means the value for which the one-dimensional Euclidean distance between the computed padding and the supported padding is minimized on a per-edge basis. If there are multiple closest supported values equally distant from the computed value for a given edge, then the value least distant from 0, i.e., the least padding, is used.
The tts:padding
style is illustrated by the following example.
<region xml:id="r1">
<style tts:extent="446px 104px"/>
<style tts:backgroundColor="black"/>
<style tts:color="white"/>
<style tts:displayAlign="after"/>
<style tts:textAlign="center"/>
<style tts:padding="10px 40px"/>
</region>
...
<p region="r1" tts:backgroundColor="red">
Just the place for a Snark! I have said it twice:<br/>
That alone should encourage the crew.<br/>
Just the place for a Snark! I have said it thrice:<br/>
What I tell you three times is true.
</p>
|
When rendering an area to which padding applies, the background color that applies to the area is rendered into the padded portion of the area.
Note:
The above example depicts how padding is applied as an inset to a region area. In particular,
10px
of padding is applied to the before (top) and after (bottom) edges, and
40px
of padding is applied at the start (left) and end (right) edges. Subtracting
these from the extent of the region area results in the region's content rectangle having 366px
width
and 84px
height. The black background color of the region appears in the region's padding rectangle
while the red background color of the paragraph appears in the region's content rectangle.
The tts:showBackground
attribute is used to specify
constraints on when the background color of a region is intended to be
presented.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
always |
whenActive
|
Initial: | always |
Applies to: |
region
|
Inherited: | no |
Percentages: | N/A |
Animatable: | discrete |
If the value of this attribute is always
, then the
background color of a region is always rendered when performing presentation
processing on a visual medium; if the value is whenActive
,
then the background color of a region is rendered only when some
content is flowed into the region.
A region satisfies the whenActive
case if (1) it is a Temporally Active Region and (2) content is selected into the region, where that content is also Temporally Active.
If a computed value of the property associated with this attribute is not supported,
then a presentation processor must use the value always
.
The tts:showBackground
style is illustrated by the following example.
<region xml:id="r1"> <style tts:origin="0px 0px"/> <style tts:extent="265px 100px"/> <style tts:backgroundColor="black"/> <style tts:showBackground="always"/> <style tts:color="white"/> <style tts:displayAlign="before"/> <style tts:textAlign="start"/> </region> <region xml:id="r2"> <style tts:origin="205px 60px"/> <style tts:extent="290px 100px"/> <style tts:backgroundColor="red"/> <style tts:color="white"/> <style tts:displayAlign="before"/> <style tts:textAlign="end"/> <style tts:showBackground="whenActive"/> </region> |
Note:
The semantics of the style property represented by this attribute are based upon that defined by [SMIL 2.1], § 5.3.3.
The tts:textAlign
attribute is used to specify a style property that
defines how inline areas are aligned within a containing block area in the inline progression
direction.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
left |
center |
right |
start |
end
|
Initial: | start |
Applies to: |
p
|
Inherited: | yes |
Percentages: | N/A |
Animatable: | discrete |
If a computed value of the property associated with this attribute is not supported,
then a presentation processor must use the value start
.
The tts:textAlign
style is illustrated by the following example.
<region xml:id="r1"> <style tts:extent="355px 43px"/> <style tts:origin="0px 0px"/> <style tts:backgroundColor="black"/> <style tts:color="white"/> <style tts:textAlign="start"/> </region> <region xml:id="r2"> <style tts:extent="355px 43px"/> <style tts:origin="0px 47px"/> <style tts:backgroundColor="black"/> <style tts:color="white"/> <style tts:textAlign="end"/> </region> ... <p region="r1"> Beware the Jabberwock, my son!<br/> The jaws that bite, the claws that catch! </p> <p region="r2"> Beware the Jubjub bird, and shun<br/> The frumious Bandersnatch! </p> |
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.16.9.
The tts:textDecoration
attribute is used to specify a style property that
defines a text decoration effect to apply to glyph areas or other inline
areas that are generated by content flowed into a region.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
none |
[ [
underline |
noUnderline
] || [
lineThrough |
noLineThrough
] || [
overline |
noOverline
] ]
|
Initial: | none |
Applies to: | span |
Inherited: | see special semantics |
Percentages: | N/A |
Animatable: | discrete |
For the purpose of determining applicability of this
style property, each character child of a p
element is
considered to be enclosed in an anonymous span.
If a computed value of the property associated with this attribute is not supported,
then a presentation processor must use the value none
.
Note:
The syntax used above in defining the value of this property is based on the
value component syntax defined in [CSS2], § 1.4.2.1. In essence,
one or more of the values separated by ||
may appear in the property
value in any order, such as "noUnderline overline lineThrough"
.
The tts:textDecoration
style is illustrated by the following example.
<region xml:id="r1"> <style tts:extent="385px 82px"/> <style tts:origin="0px 0px"/> <style tts:backgroundColor="black"/> <style tts:color="white"/> <style tts:padding="5px 2px"/> <style tts:textDecoration="underline"/> </region> ... <p region="r1"> The sea was wet<span tts:textDecoration="noUnderline"> as </span>wet <span tts:textDecoration="noUnderline"> could be,<br/> The sand was dry as dry.<br/> <span tts:textDecoration="lineThrough">There weren't any</span> You <span tts:textDecoration="lineThrough">couldn't</span> could not see a cloud<br/> Because no cloud was in the sky. </span> </p> |
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.17.4.
The computed value of the property associated with this attribute consists of a 3-tuple, each of which denotes, respectively, whether an underline, line through, or overline decoration is to be applied to the affected text. Furthermore, when applying inheritance semantics, each of these sub-values is considered to be inherited separately from the others.
Note:
For example, if the computed value of tts:textDecoration
that applies to a div
(division) element
is the tuple (noUnderline, lineThrough, overline)
and a
p
(paragraph) element child of that div
specifies tts:textDecoration="noLineThrough"
, then
the aggregate computed value that applies to the p
element is (noUnderline, noLineThrough, overline)
, which value
is then inherited by children of the p
element.
The tts:textOutline
attribute is used to specify a style property that
defines a text outline effect to apply to glyphs that are selected for glyph areas generated
by content flowed into a region.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
none |
<color>?
<length>
<length>?
|
Initial: | none |
Applies to: |
span
|
Inherited: | yes |
Percentages: | relative to this element's font size |
Animatable: | discrete |
For the purpose of determining applicability of this style property,
each character child of a p
element is considered to be enclosed in an anonymous
span.
The value of this attribute consists of an optional <color> term
followed by one or two <length> terms. If a color term is
present, then it denotes the outline color; if no color term is
present, the computed value of the tts:color
applies. The first
length term denotes the outline thickness and the second length
term, if present, indicates the blur radius.
When a relative <length> value is specified for outline thickness or blur radius, then it is resolved in terms of the height component of the referenced font size or Computed Cell Size.
The <length> value(s) used to express thickness and blur radius must be non-negative.
If no blur radius is specified, i.e., only one <length> term is present,
then the computed value of 0px
applies.
Note:
When a <length> expressed in
cells is used in a tts:textOutline
value,
the cell's height applies.
For example, if text outline thickness is specified as 0.1c, the cell resolution
is 20 by 10, and the extent of the Root Container Region is 640 by 480, then the
outline thickness will be a nominal 480 / 10 * 0.1 pixels, i.e., 4.8px,
without taking into account rasterization effects.
If a computed value of the property associated with this attribute is not supported,
then a presentation processor must use the value none
.
The tts:textOutline
style is illustrated by the following
example.
<region xml:id="r1">
<style tts:backgroundColor="transparent"/>
<style tts:color="yellow"/>
<style tts:textOutline="black 2px 0px"/>
<style tts:fontFamily="proportionalSansSerif"/>
<style tts:fontSize="24px"/>
</region>
...
<p>
How doth the little crocodile<br/>
Improve its shining tail,<br/>
And pour the waters of the Nile<br/>
On every golden scale!<br/>
How cheerfully he seems to grin,<br/>
How neatly spreads his claws,<br/>
And welcomes little fishes in,<br/>
With gently smiling jaws!
</p>
|
The tts:unicodeBidi
attribute is used to specify a style property that
defines a directional embedding or override according to
the Unicode bidirectional algorithm.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
normal |
embed |
bidiOverride
|
Initial: |
normal
|
Applies to: |
p ,
span
|
Inherited: | no |
Percentages: | N/A |
Animatable: | discrete |
For the purpose of determining applicability of this style property,
each character child of a p
element is considered to be enclosed in an anonymous
span.
If a computed value of the property associated with this attribute is not supported,
then a presentation processor must use the value normal
.
The tts:unicodeBidi
style is illustrated by the following example.
<region xml:id="r1">
<style tts:extent="265px 84px"/>
<style tts:padding="5px"/>
<style tts:backgroundColor="black"/>
<style tts:color="white"/>
<style tts:displayAlign="after"/>
<style tts:textAlign="center"/>
</region>
...
<p region="r1">
Little birds are playing<br/>
Bagpipes on the shore,<br/>
<span tts:unicodeBidi="bidiOverride" tts:direction="rtl">where the tourists snore.</span>
</p>
|
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.29.6.
The tts:visibility
attribute is used to specify a style property that
defines whether generated areas are visible or not when rendered on a visual
presentation medium.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
visible |
hidden
|
Initial: |
visible
|
Applies to: |
body ,
div ,
p ,
region ,
span
|
Inherited: | yes |
Percentages: | N/A |
Animatable: | discrete |
For the purpose of determining applicability of this style property,
each character child of a p
element is considered to be enclosed in an anonymous
span.
The tts:visibility
style has no affect on content layout or
composition, but merely determines whether composed content is visible or not.
If the computed value of the property associated with this attribute is visible
, then areas generated
by this element are rendered visible when presented on a visual medium. If the computed value is hidden
, then they are not
rendered visible, i.e., they are hidden; notwithstanding the foregoing, a descendant element of a hidden element is rendered
visible if the computed value of this property is visible
on the descendant.
If a computed value of the property associated with this attribute is not supported,
then a presentation processor must use the value visible
.
The tts:visibility
style is illustrated by the following example.
<region xml:id="r1"> <style tts:extent="398px 121px"/> <style tts:backgroundColor="black"/> <style tts:color="white"/> <style/> </region> ... <p region="r1" dur="4s"> <span tts:visibility="hidden"> <set begin="1s" tts:visibility="visible"/> Curiouser </span> <span tts:visibility="hidden"> <set begin="2s" tts:visibility="visible"/> and </span> <span tts:visibility="hidden"> <set begin="3s" tts:visibility="visible"/> curiouser! </span> </p> |
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.30.17.
The tts:wrapOption
attribute is used to specify a style property that
defines whether or not automatic line wrapping (breaking) applies within the context of
the affected element.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
wrap |
noWrap
|
Initial: | wrap |
Applies to: |
span
|
Inherited: | yes |
Percentages: | N/A |
Animatable: | discrete |
For the purpose of determining applicability of this style property,
each character child of a p
element is considered to be enclosed in an anonymous
span.
If a computed value of the property associated with this attribute is not supported,
then a presentation processor must use the value wrap
.
The tts:wrapOption
style is illustrated by the following example.
<region xml:id="r1">
<style tts:extent="192px 117px"/>
<style tts:backgroundColor="black"/>
<style tts:color="white"/>
<style tts:displayAlign="after"/>
<style tts:overflow="hidden"/>
<style tts:wrapOption="noWrap"/>
</region>
...
<p>
I'll tell thee everything I can:<br/>
There's little to relate.<br/>
I saw an aged aged man,<br/>
A-sitting on a gate.
</p>
|
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.16.13.
The tts:writingMode
attribute is used to specify a style property that
defines the block and inline progression directions to be used for the purpose of
stacking block and inline areas within a region area.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
lrtb |
rltb |
tbrl |
tblr |
lr |
rl |
tb
|
Initial: | lrtb |
Applies to: |
region
|
Inherited: | no |
Percentages: | N/A |
Animatable: | discrete |
If a computed value of the property associated with this attribute is not supported,
then a presentation processor must use the value lrtb
.
The tts:writingMode
style is illustrated by the following example.
<region xml:id="r1"> <style tts:extent="50px 570px"/> <style tts:origin="0px 0px"/> <style tts:padding="10px 3px"/> <style tts:backgroundColor="black"/> <style tts:color="white"/> <style tts:writingMode="tbrl"/> </region> <region xml:id="r2"> <style tts:extent="310px 50px"/> <style tts:origin="70px 120px"/> <style tts:padding="10px 3px"/> <style tts:backgroundColor="black"/> <style tts:color="white"/> <style tts:writingMode="rltb"/> </region> ... <p region="r1"> I sometimes dig for buttered rolls,<br/> Or set limed twigs for crabs: </p> <p region="r2" tts:direction="rtl" tts:unicodeBidi="bidiOverride"> I sometimes search the grassy knolls for the wheels of Hansom-cabs. </p> |
Note:
In the second paragraph in the above example that targets region
r2
, the tts:unicodeBidi
and
tts:direction
properties are set to
bidiOverride
and rtl
, respectively, in order
to override the normally left-to-right directionality of characters in
the Latin script.
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.29.7.
The tts:zIndex
attribute is used to specify a style property that
defines the front-to-back ordering of region areas in the case that they overlap.
This attribute may be specified by any element type that permits use of attributes in the TT Style Namespace; however, this attribute applies as a style property only to those element types indicated in the following table.
Values: |
auto |
<integer>
|
Initial: | auto |
Applies to: |
region
|
Inherited: | no |
Percentages: | N/A |
Animatable: | discrete |
If two areas are associated with the same Z-index value, then, if those areas overlap in space, the area(s) generated by lexically subsequent elements must be rendered over area(s) generated by lexically prior elements, where lexical order is defined as the postorder traversal of a Document Instance.
The tt
element is considered to establish the root of the stacking context, as defined by
[XSL 1.1], § 7.30.18.
If a computed value of the property associated with this attribute is not supported, then a presentation processor must use the closest supported value.
Note:
In this context, the phrase closest supported value means the value for which the Euclidean distance between the computed z-index and the supported z-index is minimized. If there are multiple closest supported values equally distant from the computed value, then the value least distant from 0, i.e., closest to the base stack level of the current stacking context, is used.
The tts:zIndex
style is illustrated by the following example.
<region xml:id="r1"> <style tts:origin="0px 0px"/> <style tts:extent="400px 100px"/> <style tts:padding="5px"/> <style tts:backgroundColor="black"/> <style tts:color="white"/> <style tts:zIndex="0"/> </region> <region xml:id="r2"> <style tts:origin="100px 60px"/> <style tts:extent="400px 100px"/> <style tts:padding="5px"/> <style tts:backgroundColor="red"/> <style tts:color="white"/> <style tts:textAlign="end"/> <style tts:zIndex="1"/> </region> <region xml:id="r3"> <style tts:origin="0px 120px"/> <style tts:extent="400px 100px"/> <style tts:padding="5px"/> <style tts:backgroundColor="black"/> <style tts:color="white"/> <style tts:zIndex="2"/> </region> <region xml:id="r4"> <style tts:origin="100px 180px"/> <style tts:extent="400px 100px"/> <style tts:padding="5px"/> <style tts:backgroundColor="red"/> <style tts:color="white"/> <style tts:textAlign="end"/> <style tts:zIndex="3"/> </region> ... <p region="r1"> I passed by his garden, and marked, with one eye,<br/> How the Owl and the Panther were sharing a pie. </p> <p region="r2"> The Panther took pie-crust, and gravy, and meat,<br/> While the Owl had the dish as its share of the treat. </p> <p region="r3"> When the pie was all finished, the Owl, as a boon,<br/> Was kindly permitted to pocket the spoon: </p> <p region="r4"> While the Panther received knife and fork<br/> with a growl,<br/> And concluded the banquet by... </p> |
Note:
The semantics of the style property represented by this attribute are based upon that defined by [XSL 1.1], § 7.30.18.
Style property values include the use of the following expressions:
In the syntax representations defined in this section, no linear whitespace (LWSP) is implied or permitted between tokens unless explicitly specified.
An <alpha> expression is used to express an opacity value, where 0 means fully transparent and 1 means fully opaque.
<alpha> : float |
In the above syntax representation, the syntactic element
float
must adhere to the lexical
representation defined by [XML Schema Part 2] §
3.2.4.1. If the value represented is less than 0.0,
then it must be interpreted as equal to 0.0; similarly, if the value
represented is greater than 1.0, then it must be interpreted as
1.0. The value NaN must be interpreted as 0.0.
A specified value for <alpha> should not be NaN
,
less than 0, or greater than 1.
If a presentation processor does not support a specific, valid opacity value, then it must interpret it as being equal to the closest supported value.
Note:
In this context, the phrase closest supported value means the value for which the Euclidean distance between the computed opacity and the supported opacity is minimized. If there are multiple closest supported values equally distant from the computed value, then the value most distant from 0, i.e., the greatest opacity, is used.
A <color> expression is used to specify a named color, exact RGB color triple, or exact RGBA color tuple, where the alpha component, if expressed, is maximum (255) at 100% opacity and minimum (0) at 0% opacity, and where the applicable color space is defined by [SRGB].
<color> : "#" rrggbb | "#" rrggbbaa | "rgb(" r-value "," g-value "," b-value ")" | "rgba(" r-value "," g-value "," b-value "," a-value ")" | <namedColor> rrggbb : <hexDigit>{6} rrggbbaa : <hexDigit>{8} r-value | g-value | b-value | a-value : component-value component-value : non-negative-integer // valid range: [0,255] non-negative-integer : <digit>+ |
Note:
While permitted, the use of linear whitespace (LWSP) around the comma delimiters in
<color>
expressions is not recommended. The rationale for this recommendation
is that some existing processors may not be tolerant of this usage.
When expressing RGB component values, these values are considered to not be premultiplied by alpha.
For the purpose of performing presentation processing such that non-opaque or non-transparent alpha or opacity values apply, then the semantics of compositing functions are defined with respect to the use of the [SRGB] color space for both inputs and outputs of the composition function.
Note:
The use of [SRGB] for the stated semantics of composition is not meant to prevent an actual processor from using some other color space either for internal or external purposes. For example, a presentation processor may ultimately convert the SRGB values used here to the YUV color space for rendition on a television device.
If a presentation processor does not support a specific, valid color or alpha value, then it must interpret it as being equal to the closest supported value.
Note:
In this context, the phrase closest supported value means the value for which the Euclidean distance between
the computed color or alpha and the supported color or alpha in the RGB color space is minimized. If there are multiple closest supported values equally distant from
the computed value, then the value least distant from opaque black rgba(0,0,0,255)
, i.e., the closest to opaque black, is used.
A <digit> is used to express integers and other types of numbers or tokens.
<digit> : "0" | "1" | "2" | "3" | "4" | "5" | "6" | "7" | "8" | "9" |
This section is non-normative.
Note:
The information from this section has been removed due to lack of any normative use within this specification. This section is retained in its empty form in order to prevent section renumbering.
A <familyName> expression specifies a font family name.
<familyName> : unquoted-string | quoted-string unquoted-string : identifier ( lwsp identifier )* lwsp : ( ' ' | '\t' | '\n' | '\r' )+ identifier : [-]? identifier-start identifier-following* identifier-start : [_a-zA-Z] | non-ascii-or-c1 | escape identifier-following : [_a-zA-Z0-9-] | non-ascii-or-c1 | escape non-ascii-or-c1 : [^\0-\237] escape : '\\' char quoted-string : double-quoted-string | single-quoted-string double-quoted-string : '"' ( [^"\\] | escape )+ '"' single-quoted-string : "'" ( [^'\\] | escape )+ "'" |
In addition to adhering to the syntax rules specified above, the following semantic rules apply:
the semantic value of a <familyName> expression is the semantic value of its unquoted-string
or quoted-string
non-terminal, according to whichever applies;
the semantic value of an unquoted-string
non-terminal is a pair <
quoted, content>
,
where quoted is a boolean false
, and
where content is the result of appending the value of each identifier
non-terminal, in lexical order, where the value of each identifier is preceded by a single SPACE (U+0020) character if it is not the first
identifier;
the semantic value of a quoted-string
non-terminal is a pair <
quoted, content>
,
where quoted is a boolean true
, and
where content is the unquoted content of the quoted string, i.e., the sequence of characters between the delimiting quotes.
the semantic value of an escape
non-terminal is the value of the escaped char
;
a <familyName> that takes the form of an unquoted-string
that contains an identifier
that
starts with two -
HYPHEN-MINUS (U+002D) characters must be considered to be invalid;
a <familyName> that takes the form of an unquoted-string
that contains a single identifier
that
matches (by case sensitive comparison) a <genericFamilyName> must be interpreted as that <genericFamilyName>;
a <familyName> that takes the form of a quoted-string
whose content (unquoted value)
matches (by case sensitive comparison) a <genericFamilyName> must not be interpreted as that <genericFamilyName>, but as
the actual name of a non-generic font family.
Note:
The syntactic element char
is to be interpreted according
to the Char
production defined by [XML 1.0] §2.2.
Note:
The {unicode}
escape mechanism defined by [CSS2] §4.1.1
is not supported by this syntax; rather, authors are expected to either (1) directly encode the character
using the document encoding or (2) use an XML character reference according to
[XML 1.0] §4.1. When a syntactically significant character needs to be used
without its normal syntactic interpretation, it may be be escaped using the backslash
(reverse solidus) escape
non-terminal specified above.
When using the backslash (reverse solidus) escape
non-terminal, the above syntax does
not place any restriction on what character may be escaped, e.g., \\[\n\r\f0-9a-f]
are
permitted. If one of these latter escapes appears in a <familyName> expression, then it will
need to be converted to a {unicode}
escape if it is to be used with a standard XSL-FO
or CSS parser. In particular, a backslash followed by a newline is ignored by CSS, while it is not
ignored by the above syntax. Such an unignored escaped newline would need to be represented using
an equivalent {unicode}
escape, such as \a
, to order to express in CSS.
A <genericFamilyName> expression specifies a font family using a general token that indicates a class of font families.
The resolution of a generic family name to a concrete font instance is considered to be implementation dependent, both in the case of content authoring and content interpretation.
<genericFamilyName> : "default" | "monospace" | "sansSerif" | "serif" | "monospaceSansSerif" | "monospaceSerif" | "proportionalSansSerif" | "proportionalSerif" |
The mapping between a generic (font) family name and an actual font is not determined by this specification; however, the distinction of monospace versus proportional and serif versus sans-serif should be maintained if possible when performing presentation.
If a generic (font) family name of monospace
is specified,
then it may be interpreted as equivalent to either monospaceSansSerif
or monospaceSerif
. The generic family names sansSerif
and serif
are to be interpreted as equivalent to proportionalSansSerif
and proportionalSerif
, respectively.
If the generic family name default
is specified (or implied by
an initial value), then its typographic characteristics are considered to be implementation
dependent; however, it is recommended that this default font family be mapped to
a monospaced, sans-serif font.
A <hexDigit> is used to express integers and other types of numbers or tokens that employ base 16 arithmetic.
For the purpose of parsing, a distinction must not be made between lower and upper case.
<hexDigit> : <digit> | "a" | "b" | "c" | "d" | "e" | "f" | "A" | "B" | "C" | "D" | "E" | "F" |
An <integer> expression is used to express an arbitrary, signed integral value.
<integer> : ( "+" | "-" )? <digit>+ |
A <length> expression is used to express either a coordinate component of point in a cartesian space or a distance between two points in a cartesian space.
<length> : scalar | percentage scalar : number units percentage : number "%" sign : "+" | "-" number : sign? non-negative-number non-negative-number : non-negative-integer | non-negative-real non-negative-integer : <digit>+ non-negative-real : <digit>* "." <digit>+ units : "px" | "em" | "c" // abbreviation of "cell" |
It is an error to omit the units component of a scalar length value.
The semantics of
the unit of measure px
(pixel)
are as defined by [XSL 1.1], § 5.9.13.
When specified relative to a font whose size is expressed as a single length
measure or as two length measures of equal length, the unit of measure em
is considered to be
identical to that defined by [XSL 1.1], § 5.9.13;
however, when specified relative to a font whose size is expressed as
two length measures of non-equal lengths, then one em
is
equal to the inline progression dimension of the anamorphically
scaled font when used to specify lengths in the inline progression
direction and equal to the block progression dimension of the scaled
font when used to specify lengths in the block progression direction.
The semantics of the unit of measure c
(cell) are
defined by the parameter 6.2.1 ttp:cellResolution.
A <namedColor> is used to express an RGBA color with a convenient name, and where the applicable color space is defined by [SRGB].
For the purpose of parsing, a distinction must not be made between lower and upper case.
<namedColor> : "transparent" // #00000000 | "black" // #000000ff | "silver" // #c0c0c0ff | "gray" // #808080ff | "white" // #ffffffff | "maroon" // #800000ff | "red" // #ff0000ff | "purple" // #800080ff | "fuchsia" // #ff00ffff | "magenta" // #ff00ffff (= fuchsia) | "green" // #008000ff | "lime" // #00ff00ff | "olive" // #808000ff | "yellow" // #ffff00ff | "navy" // #000080ff | "blue" // #0000ffff | "teal" // #008080ff | "aqua" // #00ffffff | "cyan" // #00ffffff (= aqua) |
Note:
Except for transparent
, the set of named colors specified above constitutes a proper
subset of the set of named colors specified by [SVG 1.1], § 4.2.
This section is non-normative.
Note:
The information from this section has been incorporated into 8.3.5 <familyName>. This section is retained in its empty form in order to prevent section renumbering.
This section is non-normative.
Note:
The information from this section has been incorporated into 8.3.5 <familyName>. This section is retained in its empty form in order to prevent section renumbering.
This section defines the semantics of style resolution in terms of a standard processing model as follows:
Any implementation of this model is permitted provided that the externally observable results are consistent with the results produced by this model.
Note:
The semantics of style resolution employed here are based upon [XSL 1.1], § 5.
Style association is a sub-process of 8.4.4 Style Resolution Processing used to determine the specified style set of each content and layout element.
Style matter may be associated with content and layout matter in a number of ways:
In addition to the above, style matter may be associated with layout matter using:
Style properties may be expressed in an inline manner by direct specification of an attribute from the TT Style Namespace on the affected element. When expressed in this manner, the association of style information is referred to as inline styling.
Style properties associated by inline styling are afforded a higher priority than all other forms of style association.
<p tts:color="white">White 1 <span tts:color="yellow">Yellow</span> White 2</p> |
Note:
In the above example, the two text fragments "White 1 "
and
" White 2"
, which are interpreted as anonymous spans, are not
associated with a color style property; rather, they inherit their color style
from their parent p
element as described in 8.4.2.1 Content Style Inheritance below.
Style properties may be expressed in an out-of-line manner and referenced by the
affected element using the style
attribute. When expressed in this manner,
the association of style information is referred to as referential styling.
If a style
attribute specifies multiple references, then those references are
evaluated in the specified order, and that order applies to resolution of the value of a
style property in the case that it is specified along multiple reference paths.
The use of referential styling is restricted to making
reference to style
element descendants of a styling
element. It is considered an error to reference a style
element that is a descendant of a layout
element.
Note:
The use of referential styling encourages the reuse of style specifications while sacrificing locality of reference.
Note:
A single Content element may be associated with style properties by a hybrid mixture of inline and referential styling, in which case inline styling is given priority as described above by 8.4.1.1 Inline Styling.
<style xml:id="s1" tts:color="white"/> <style xml:id="s2" tts:color="yellow"/> ... <p style="s1">White 1 <span style="s2">Yellow</span> White 2</p> |
Note:
In the above example, the two text fragments "White 1 "
and
" White 2"
, which are interpreted as anonymous spans, are not
associated with a color style property; rather, they inherit their color style
from their parent p
element as described in 8.4.2.1 Content Style Inheritance below.
Style properties may be expressed in an out-of-line manner and may themselves reference other out-of-line style properties, thus creating a chain of references starting at the affected element. When expressed in this manner, the association of style information is referred to as chained referential styling.
A loop in a sequence of chained style references must be considered an error.
The use of referential styling is restricted to making
reference to style
element descendants of a styling
element. It is considered an error to reference a style
element that is a descendant of a layout
element.
Note:
The use of chained referential styling encourages the grouping of style specifications into general and specific sets, which further aids in style specification reuse.
Note:
A single Content element may be associated with style properties by a hybrid mixture of inline, referential styling, and chained referential styling, in which case inline styling is given priority as described above by 8.4.1.1 Inline Styling.
<style xml:id="s1" tts:color="white" tts:fontFamily="monospaceSerif"/> <style xml:id="s2" style="s1" tts:color="yellow"/> ... <p style="s1">White Monospace</p> <p style="s2">Yellow Monospace</p> |
Note:
In the above example, the text of the second paragraph is yellow, since tts:color='yellow'
effectively overwrites (is merged over) the tts:color='white'
that style s2
obtains by a
reference to style s1
.
Style properties may be expressed in a nested manner by direct specification of one or
more style
element children of the affected element.
When expressed in this manner, the association of style information
is referred to as nested styling.
Style properties associated by nested styling are afforded a lower priority than inline styling but with higher priority than referential styling.
<region xml:id="r1"> <style tts:extent="128px 66px"/> <style tts:origin="0px 0px"/> <style tts:displayAlign="center"/> </region> |
Note:
In this version of this specification, nested styling applies only
to the region
element.
Style inheritance is a sub-process of 8.4.4 Style Resolution Processing used to determine the specified style set of each content and layout element.
Styles are further propagated to content matter using:
For the purpose of determining inherited styles, the element hierarchy of an intermediate synchronic document form of a Document Instance must be used, where such intermediate forms are defined by 9.3.3 Intermediate Synchronic Document Construction.
Note:
The intermediate synchronic document form is utilized rather than the original form in order to facilitate region inheritance processing.
Style properties are inherited from ancestor Content elements within an intermediate synchronic document if a style property is not associated with a Content element (or an anonymous span) and the style property is designated as inheritable.
Note:
The tt
element is not a Content element;
consequently, the body
element is the outermost element from which content style inheritance occurs.
If a style property is determined to require inheritance, then the inherited value must be the value of the same named style property in the computed style set of the element's immediate ancestor element within the applicable intermediate synchronic document.
<p tts:fontFamily="monospaceSansSerif">
<span tts:color="yellow">Yellow Monospace</span>
</p>
|
Note:
In the above example, the span
element that encloses the character items
Yellow Monospace
is not associated with a tts:fontFamily
style
property and this property is inheritable; therefore, the value of the
tts:fontFamily
style
is inherited from the computed style
set of the ancestor p
element,
and is added to the
specified
style
set of the span
element.
Style properties are inherited from a region element in the following case:
if a style property P is not associated with a Content element or an anonymous span E and the style property is designated as inheritable, and
if that style property P is in the computed style set of region R, and
if that element E is flowed into (presented within) region R.
<region xml:id="r1"> <style tts:color="yellow"/> <style tts:fontFamily="monospaceSerif"/> </region> ... <p region="r1">Yellow Monospace</p> |
Note:
In the above example, the anonymous span that encloses the character
items Yellow Monospace
effectively inherits the tts:color
and tts:fontFamily
styles specified on the region
element into which the p
element is flowed (presented).
During style resolution, layout, and presentation processing, three categories of style property values are distinguished as follows:
Values of style properties that are associated with or inherited by an element or anonymous span are referred to as specified values. The set of all specified style properties of a given element is referred to as the specified style set of that element.
When style properties are specified using relative value expressions, such as a named color, a relative unit (e.g., cell), or a percentage, then they need to be further resolved into absolute units, such as an RGB triple, pixels, etc.
During the style resolution process, all specified style values are reinterpreted (or recalculated) in absolute terms, and then recorded as computed values. The set of all computed style properties of a given element is referred to as the computed style set of that element.
When a style value is inherited, either explicitly or implicitly, it is the computed value of the style that is inherited from an ancestor element. This is required since the resolution of certain relative units, such as percentage, require evaluating the expression in the immediate (local) context of reference, and not in a distant (remote) context of reference where the related (resolving) expression is not available.
Subsequent to style inheritance, a computed value may require further resolution at layout or presentation time when such resolution is required by the semantics of a specific property or the information required to perform the resolution is not available until layout or presentation processing occurs. Such a value is referred to as a used value, which need not be the same value as the final actual value described below.
Note:
The determination of a used value is always preceded by the inheritance of a computed value; that is to say, used value computation is nominally performed after the 8.4.4 Style Resolution Processing completes.
Note:
See also [CSS2], §6.1.3..
During the actual presentation process, other transformations occur that map some value expressions to concrete, physical values. For example, the colors of computed style values are further subjected to closest color approximation and gamma correction during the display process. In addition, length value expressions that use pixels in computed style values are considered to express logical rather than physical (device) pixels. Consequently, these logical pixels are subject to being further transformed or mapped to physical (device) pixels during presentation.
The final values that result from the logical to device mapping process are referred to as actual values. The set of all actual style properties of a given element is referred to as the actual style set of that element.
Note:
More than one set of actual values may be produced during the process of presentation. For example, a TTML presentation processor device may output an RGBA component video signal which is then further transformed by an NTSC or PAL television to produce a final image. In this case, both color and dimensions may further be modified prior to presentation.
Note:
In general, a TTML presentation processor will not have access to actual style set values; as a consequence, no further use or reference to actual values is made below when formally describing the style resolution process.
The process of style resolution is defined herein as the procedure (and results thereof) for resolving (determining) the computed values of all style properties that apply to content and layout elements:
The process described here forms an integral sub-process of 9.3 Region Layout and Presentation.
For the purpose of interpreting the style resolution processing model specified below, the following conceptual definitions apply:
a style property, P, is considered to consist of a
tuple [name, value]
, where the name of the
property is a tuple [namespace value,
unqualified name]
and the value of the property is a
tuple [category, type, value expression]
[ ["http://www.w3.org/ns/ttml#styling", "color"], ["specified", color, "red"] ] |
a style (property) set consists of an unordered collection of style properties, where no two style properties within the set have an identical name, where by "identical name" is meant equality of namespace value of name tuple and unqualified name of name tuple;
in a specified style (property) set, the category of each style property is "specified"; a specified style (property) set of an element E is referred to as SSS(E);
{ [ ["http://www.w3.org/ns/ttml#styling", "backgroundColor"], ["specified", color, 0x00FF00 ] ], [ ["http://www.w3.org/ns/ttml#styling", "color"], ["specified", color, "red" ] ], [ ["http://www.w3.org/ns/ttml#styling", "fontSize"], ["specified", length, "1c" ] ], [ ["http://www.w3.org/ns/ttml#styling", "lineHeight"], ["specified", length, "117%" ] ] } |
in a computed style (property) set, the category of each style property is either "specified" or "computed"; a computed style (property) set of an element E is referred to as CSS(E);
{ [ ["http://www.w3.org/ns/ttml#styling", "backgroundColor"], ["specified", color, 0x00FF00 ] ], [ ["http://www.w3.org/ns/ttml#styling", "color"], ["computed", color, 0xFF0000 ] ], [ ["http://www.w3.org/ns/ttml#styling", "fontSize"], ["computed", length, "24px" ] ], [ ["http://www.w3.org/ns/ttml#styling", "lineHeight"], ["computed", length, "28px" ] ] } |
a style property Pnew is merged into a style (property) set, SS, as follows: if a style property Pold is already present in SS where the name of Pnew is identical to the name of Pold, then replace Pold in SS with Pnew; otherwise, add Pnew to SS;
a style (property) set SSnew is merged into an existing style (property) set SSold as follows: for each style property Pnew in SSnew, merge Pnew into SSold;
The specified style set SSS of an element or anonymous span E, SSS(E), is determined according to the following ordered rules:
[initialization] initialize the specified style set SSS of E to the empty set;
[referential and chained referential
styling] for each style
element SREF
referenced by a style
attribute specified on
E, and in the order specified in the style
attribute, then, if SREF is a descendant
of a styling
element,
merge the specified style set of
SREF, SSS(SREF), into
the specified style set of E,
SSS(E);
[nested styling] for each
nested style
element child SNEST of
E, and in the specified order of child elements, merge
the specified style set of SNEST,
SSS(SNEST), into the specified style set of
E, SSS(E);
[inline styling] for each style property P expressed as a specified styling attribute of E, merge P into the specified style set of E, SSS(E);
[animation styling] for each
style property P expressed as a specified styling
attribute of an immediate animation (set
) element child of element E,
merge P into the specified style
set of E, SSS(E);
[implicit inheritance] if the
element type of E is not the styling element type
style
, then for each style property
P in the set of style properties defined
above in 8.2 Styling Attribute Vocabulary, perform the
following ordered sub-steps:
if P is present in the specified style set of E, SSS(E), then continue to the next style property;
if P is inheritable and E is a region
element, or if P is not inheritable and applies to E, then set P′ to the initial value of
property P if such an initial value is defined, where the initial value of a property is
determined according to the specific property definition found above
in 8.2 Styling Attribute Vocabulary;
if P is inheritable and the element type of E is a Content element type or anonymous span, then set P′ to the result of looking up the value of P in the computed style set of the immediate ancestor element of E, i.e., CSS(PARENT(E));
if the value of P′ is not undefined, then merge P′ into the specified style set of E, SSS(E).
The computed style set CSS of an element or anonymous span E, CSS(E), is determined according to the following ordered rules:
[resolve specified styles] determine (obtain) the specified style set SSS of E, namely, SSS(E), in accordance with 8.4.4.2 Specified Style Set Processing;
[initialization] initialize CSS(E) to a (deep) copy of SSS(E);
[filter] if
E is a style
element, then return
CSS(E) as the resulting computed style set without
further resolution; otherwise, continue with the next rule;
[relative value resolution] for each style property P in CSS(E), where the value type of P is relative, perform the following ordered sub-steps:
replace the relative value of P with an equivalent, non-relative (computed) value;
set the category of P to "computed";
Note:
As a result of the filtering rule above, the computed style set of
a style
element includes only specified values, in which case
relative value expressions remain relative; consequently, the
resolution of relative value expressions (that may be assigned by
means of referential style association) always takes place in the
context of a layout or Content element which has a presentation
context, and not in the non-presentation, declaration context
of a referentiable style
element.
Note:
See individual style properties for additional inheritance semantics, if any.
The top-level style resolution process is defined as follows: using a preorder traversal of each element and anonymous span, E, of an intermediate synchronic document, DOCinter, perform the following ordered sub-steps:
[filter] if the element type
of E is not the styling element type style
, is
not the layout element type region
, and is not one of the
Content element types body
, div
, p
,
span
, br
, or anonymous span, then continue to the
next element in the preorder traversal;
[resolve computed styles] determine (obtain) the computed style set CSS of E, namely, CSS(E), in accordance with 8.4.4.3 Computed Style Set Processing.
This section specifies the layout matter of the core vocabulary catalog, where layout is to be understood as a separable layer of information that applies to content and that denotes authorial intentions about the presentation of that content.
Note:
The two layers of layout and style matter are considered to be independently separable. Layout matter specifies one or more spaces or areas into which content is intended to be presented, while style matter specifies the manner in which presentation occurs within the layout.
In certain cases, a content author may choose to embed (inline) style matter directly into layout or content matter. In such cases, an alternative exists – use of referential styling – in which the style matter is not embedded (inlined).
The following elements specify the structure and principal layout aspects of a Document Instance:
The layout
element is a container element used to group
layout matter, including metadata that applies to layout matter.
The layout
element accepts as its children zero or more
elements in the Metadata.class
element group, followed by
zero or more region
elements.
<layout xml:id = ID xml:lang = string xml:space = (default|preserve) {any attribute not in default or any TT namespace}> Content: Metadata.class*, region* </layout> |
To the extent that time semantics apply to the content of the
layout
element, the implied time interval of this element is
defined to be coterminous with the Root Temporal Extent.
The region
element is used to define a rectangular space or area into which content is
to be flowed for the purpose of presentation.
Note:
The rectangular area of a region is explicitly not constrained to be contained within the Root Container Region. In particular, the origin components of a region may be negative, and the extent (width and height) components of a region may be greater than the width and height of the Root Container Region. Whether a presentation processor clips such a region to the Root Container Region is implementation dependent, and not prescribed by this specification.
In addition, and in accordance with
8.4.2.2 Region Style Inheritance,
the region
element may be used to specify inheritable style properties to be
inherited by content that is flowed into it.
The region
element accepts as its children zero or more
elements in the Metadata.class
element group,
followed by zero or more
elements in the Animation.class
element group,
followed by
zero or more style
elements.
Any metadata specified by children in the Metadata.class
element group applies semantically to the region
element and its descendants as a whole.
Any animation elements specified by children in the Animation.class
element group apply semantically to the region
element.
Any style
child element must be considered a local style
definition that applies only to the containing region
element, i.e., does not apply for resolving referential styling (but does apply for region
style inheritance).
<region begin = <timeExpression> dur = <timeExpression> end = <timeExpression> style = IDREFS timeContainer = (par|seq) ttm:role = string xml:id = ID xml:lang = string xml:space = (default|preserve) {any attribute in TT Style namespace} {any attribute not in default or any TT namespace}> Content: Metadata.class*, Animation.class*, style* </region> |
If begin
and (or) end
attributes are specified on a
region
element, then they specify the beginning and (or) ending points
of a time interval during which the region is eligible for activation and with
respect to which animation child elements of the region are timed.
If specified, these begin and end points are relative to the time
interval of the nearest ancestor element associated with a time
interval, irregardless of whether that interval is explicit or implied.
The nearest ancestor element of a region
element that is associated
with a time interval is the layout
element.
If a dur
attribute is specified on the region
element, then it
specifies the simple duration of the region.
For the purpose of determining the semantics of presentation processing, a region that is temporally inactive must not produce any visible marks when presented on a visual medium.
Note:
A region
element may be associated with a time interval for two
purposes: (1) in order to temporally bound the presentation of the region and
its content, and (2) to provide a temporal context in which animations of region
styles may be effected.
For example, an author may wish to specify a region that is otherwise empty, but may have a visible background color to be presented starting at some time and continuing over the region's duration. The simple duration of the region serves additionally to scope the presentation effects of content that is targeted to the region. An author may also wish to move a region within the Root Container Region or change a region's background color by means of animation effects. In both of these cases, it is necessary to posit an active time interval for a region.
If no timeContainer
attribute is specified on
a region
element, then it must be interpreted as having
parallel time containment semantics.
If a ttm:role
attribute is
specified on a region
element, then it must adhere to the
value syntax defined by Syntax Representation – ttm:role, and where the roles identified
by this attribute express the semantic roles of the region
independently from the semantic roles of any content targeted to
(associated with) the region.
This section defines the 9.2.1 region attribute used with Content elements.
The region
attribute is used to reference a region
element which defines a space or area into which a Content element is intended to be flowed.
If specified, the value of a region
attribute must
adhere to the IDREF
data type defined by [XML Schema Part 2],
§ 3.3.9, and, furthermore, this IDREF must reference a region
element
which has a layout
element as an ancestor.
The region
attribute may be specified by an instance of the following
element types:
Note:
See 9.3 Region Layout and Presentation below for further information on content flow in a region.
This section defines the semantics of region layout and presentation in terms of a standard processing model as follows:
Any implementation is permitted provided that the externally observable results are consistent with the results produced by this model.
If a Document Instance does not specify a
region
element,
then a default region is implied with the following characteristics:
the identity of the default region is considered to be anonymous;
the extent of the default region is the same as the extent of the Root Container Region;
the temporal interval of the default region is the same as the interval defined by the Root Temporal Extent;
Furthermore, if no region
element is
specified, then the region
attribute must
not be specified on any Content element in the Document Instance.
If a default region is implied for a given Document Instance, then
the body
element
is implicitly targeted to (associated with) the default region.
When implying a default region, the Document Instance is to be
treated as if a region
element and its
parent layout
element were specified in a head
element, and a matching region
attribute were
specified on the body
element
as shown in the following example:
<tt xml:lang="" xmlns="http://www.w3.org/ns/ttml"> <head> <layout> <region xml:id="anonymous"/> </layout> </head> <body region="anonymous"/> </tt> |
Note:
In the above example, a default region
element and region
attribute are implied. In addition, a layout
container element is implied for
the implied region
element.
For the purposes of performing presentation processing, anonymous spans are created according to the following [construct anonymous spans] procedure:
for each significant text node in a Content element, synthesize an anonymous span to enclose the text node, substituting the new anonymous span for the original text node child in its sibling and parent hierarchy;
for each contiguous sequence of anonymous spans, replace the sequence with a single anonymous span which contains a sequence of text nodes representing the individual text node children of the original sequence of anonymous spans;
for each span element whose child is a single anonymous span, replace the anonymous span with its sequence of child text nodes;
For the purposes of performing presentation processing, the active time duration of a Document Instance is divided into a sequence of time coordinates where at each time coordinate, some element becomes temporally active or inactive, then, at each such time coordinate, a Document Instance is mapped from its original, source form, DOCsource , to an intermediate synchronic document form, DOCinter , according to the [construct intermediate document] procedure:
for each temporally active region R, replicate the
sub-tree
of DOCsource
headed by the body
element;
evaluating this sub-tree in a postorder traversal, prune elements if any of the following conditions is true:
if the pruned sub-tree is non-empty, then reparent it to the R element;
finally, after completing the above steps, prune the original body
element from the intermediate document, then prune all region
,
begin
, end
, and dur
attributes, which are no
longer semantically relevant;
Note:
In this section, the term prune, when used in reference to an element, means that the element is to be removed from its parent's children, which, in turn, implies that the descendants of the pruned element will no longer be descendants of the element's parent. When prune is used in reference to an attribute, it means that attribute is to be removed from its associated (owning) element node.
A Content element is associated with a region according to the following ordered rules, where the first rule satisfied is used and remaining rules are skipped:
if the element specifies a region
attribute,
then the element is associated with the region referenced by that
attribute;
if some ancestor of that element specifies a region
attribute,
then the element is associated with the region referenced by the most
immediate ancestor that specifies this attribute;
if the element contains a descendant element that
specifies a region
attribute,
then the element is associated with the region referenced by that
attribute;
if a default region was implied (due to the absence of any
region
element),
then the element is associated with the default region;
the element is not associated with any region.
The result of performing the processing described above will be a sequence of N intermediate synchronic Document Instances, DOCinter0 … DOCinterN−1.
Note:
Where an implementation is able to detect significant similarity between two adjacent synchronic Document Instances, DOCinterN DOCinterN−1, then the implementation can apply processing to make the transition between presenting the two instances as smooth as possible, e.g., as described by [CEA-608-E], § C.3, and [CC-DECODER-REQ].
Subsequent to performing a temporal (synchronic) slice and subsequent remapping of regionally selected content hierarchy, the resulting intermediate synchronic document is subjected to a flow transformation step that produces a rooted flow object tree represented as an XSL FO document instance as defined by [XSL 1.1], and semantically extended by TTML specific style properties that have no XSL FO counterpart.
Note:
In this section, the use of XSL FO is intended to be conceptual only, employed solely for the purpose of defining the normative presentation semantics of TTML. An actual implementation of this algorithm is not required to create or process XSL-FO representations. In particular, it is possible to implement these semantics using alternative presentation models, such as Cascading Style Sheets (CSS).
Each intermediate synchronic document produced by 9.3.3 Intermediate Synchronic Document Construction is mapped to an XSL FO document instance, F, as follows:
perform the [construct anonymous spans] procedure;
resolve styles according to 8.4.4.4 Style Resolution Process;
map the tt
element to an fo:root
element, populated
initially with an fo:layout-master-set
element that contains a valid
fo:simple-page-master
that, in turn, contains an
fo:region-body
child, where the extent of the Root Container Region expressed on the
tt
element is mapped to page-width
and
page-height
attributes on the fo:simple-page-master
element;
map the layout
element to an fo:page-sequence
element and a
child fo:flow
element that reference the page master and page region
defined by the simple page master produced above;
map each non-empty region
element to an fo:block-container
element with an absolute-position
attribute with value
absolute
, with top
, left
, bottom
, and right
attributes that express a rectangle equivalent to the region's origin,
extent (including padding), and with a line-stacking-strategy
attribute with value line-height
;
Note:
The region's extent corresponds with the allocation rectangle of the block area generated by the fo:block-container
.
for each body
, div
, and p
element that is not
associated with a tts:display
style property with the value
none
, map the element to a distinct fo:block
element,
populating the style properties of fo:block
by using the computed
style set associated with each original TTML Content element;
for the resulting fo:block
formatting object
produced in the previous step that corresponds to the body
element, perform the following ordered sub-steps:
if the display-align
style property of this fo:block
has the value center
or after
,
then synthesize and insert as the first child of this fo:block
an empty fo:block
with the following attributes:
space-after.optimum
, space-after.maximum
, and
space-after.conditionality
, where the value of the former
two attributes is the height or width of the containing
fo:block-container
element, whichever of these is designated
as the block progression dimension, and where the value of the last is
retain
;
if the display-align
style property of this fo:block
has the value center
or before
,
then synthesize and insert as the last child of this fo:block
an empty fo:block
with the following attributes:
space-after.optimum
, space-after.maximum
, and
space-after.conditionality
, where the value of the former
two attributes is the height or width of the containing
fo:block-container
element, whichever of these is designated
as the block progression dimension, and where the value of the last is
retain
;
for each span
element that is not
associated with a tts:display
style property with the value
none
and for each anonymous
span that is a child of a p
or span
element,
map the element or
sequence of character items to a distinct fo:inline
element,
populating the style properties of fo:inline
by using the computed
style set associated with each original TTML Content element
or anonymous span;
for each br
element that is not
associated with a tts:display
style property with the value
none
, map the element to
a distinct fo:character
element having the following
properties:
character="
"
suppress-at-line-break="retain"
for each TTML style property attribute in some computed style set that has no counterpart in [XSL 1.1], map that attribute directly through to the relevant formatting object produced by the input TTML Content element to which the style property applies;
optionally, synthesize a unique id
attribute on each resulting
formatting object element that relates that element to the input element that
resulted in that formatting object element;
For each resulting document instance F, if processing requires presentation on a visual medium, then apply formatting and rendering semantics consistent with that prescribed by [XSL 1.1].
Note:
In an XSL FO area tree produced by formatting F using
an [XSL 1.1] formatting processor,
the page-viewport-area
,
which is generated by fo:page-sequence
element by
reference to the sole generated fo:simple-page-master
element, would correspond to the Root Container Region defined
above in 2 Definitions.
Note:
When mapping a region
element to fo:block-container
, it may
be necessary to use a negative offset as a value for one or more of the
top
, left
, bottom
, and right
XSL-FO properties in case the region extends outside of its containing block.
Note:
Due to the possible presence of TTML style properties or style property values in a given Document Instance for which there is no [XSL 1.1] counterpart, Implementors should recognize that it is the layout model of [XSL 1.1] that is being referenced by this specification, not the requirement to use a compliant [XSL 1.1] formatting processor, since such would not necessarily be sufficient to satisfy the full presentation semantics defined by this specification, and would contain a large number of features not needed to implement the presentation semantics of TTML.
Note:
The purpose of inserting additional, collapsible space in the block progression
dimension of the fo:block
that corresponds with the body
element
is to ensure that the before and after edges of this fo:block
are coincident
with the before and after edges of the fo:block-container
that corresponds
to the containing region
, while simultaneously taking into account the needs
to satisfy alignment in the block progression dimension. For example, this assures that
the background color associated with the body
element, if not transparent
,
will fill the containing region wholly.
An example of the processing steps described above is elaborated below, starting with Example – Sample Source Document.
<tt tts:extent="640px 480px" xml:lang="en" xmlns="http://www.w3.org/ns/ttml" xmlns:tts="http://www.w3.org/ns/ttml#styling"> <head> <layout> <region xml:id="r1"> <style tts:origin="10px 100px"/> <style tts:extent="620px 96px"/> <style tts:fontSize="40px"/> <style tts:fontWeight="bold"/> <style tts:backgroundColor="black"/> <style tts:color="red"/> <style tts:textAlign="center"/> <style tts:displayAlign="center"/> </region> <region xml:id="r2"> <style tts:origin="10px 300px"/> <style tts:extent="620px 96px"/> <style tts:fontSize="40px"/> <style tts:fontWeight="bold"/> <style tts:backgroundColor="black"/> <style tts:color="yellow"/> <style tts:textAlign="center"/> <style tts:displayAlign="center"/> </region> </layout> </head> <body xml:id="b1"> <div xml:id="d1" begin="0s" dur="2s"> <p xml:id="p1" region="r1">Text 1</p> <p xml:id="p2" region="r2">Text 2</p> </div> <div xml:id="d2" begin="1s" dur="2s"> <p xml:id="p3" region="r2">Text 3</p> <p xml:id="p4" region="r1">Text 4</p> </div> </body> </tt> |
In the above document, the content hierarchy consists of two divisions, each containing two paragraphs. This content is targeted (associated with) one of two non-overlapping regions that are styled identically except for their position and their foreground colors, the latter of which is inherited by and applies to the (and, in this case, anonymous) spans reparented into the regions.
The following, first intermediate document shows the synchronic state for time
interval [0,1), during which time only division d1
is temporally active,
and where paragraphs p1
and p2
(and their ancestors)
are associated with regions r1
and r2
, respectively.
Note:
The intermediate documents shown below are not valid Document Instances, but rather, are representations of possible internal processing states used for didactic purposes.
<tt tts:extent="640px 480px" xml:lang="en" xmlns="http://www.w3.org/ns/ttml" xmlns:tts="http://www.w3.org/ns/ttml#styling"> <head> <layout> <region xml:id="r1"> <style tts:origin="10px 100px"/> <style tts:extent="620px 96px"/> <style tts:fontSize="40px"/> <style tts:fontWeight="bold"/> <style tts:backgroundColor="black"/> <style tts:color="red"/> <style tts:textAlign="center"/> <style tts:displayAlign="center"/> <body xml:id="b1-1"> <div xml:id="d1-1"> <p xml:id="p1">Text 1</p> </div> </body> </region> <region xml:id="r2"> <style tts:origin="10px 300px"/> <style tts:extent="620px 96px"/> <style tts:fontSize="40px"/> <style tts:fontWeight="bold"/> <style tts:backgroundColor="black"/> <style tts:color="yellow"/> <style tts:textAlign="center"/> <style tts:displayAlign="center"/> <body xml:id="b1-2"> <div xml:id="d1-2"> <p xml:id="p2">Text 2</p> </div> </body> </region> </layout> </head> </tt> |
An XSL FO document instance that would yield rendering consistent with TTML, and which may be produced by performing flow processing upon the first intermediate document is illustrated below.
<fo:root xmlns:fo="http://www.w3.org/1999/XSL/Format"> <fo:layout-master-set> <fo:simple-page-master master-name="m1" page-width="640px" page-height="480px"> <fo:region-body/> </fo:simple-page-master> </fo:layout-master-set> <fo:page-sequence master-reference="m1"> <fo:flow flow-name="xsl-region-body"> <!-- region (r1) --> <fo:block-container id="r1" absolute-position="absolute" left="10px" top="100px" width="620px" height="96px" background-color="black" display-align="center"> <!-- body (b1) --> <fo:block id="b1-1"> <!-- body's space (before) filler --> <fo:block space-after.optimum="96px" space-after.maximum="96px" space-after.conditionality="retain"/> <!-- div (d1) --> <fo:block id="d1-1"> <!-- p (p1) --> <fo:block id="p1" text-align="center"> <fo:inline font-size="40px" font-weight="bold" color="red">Text 1</fo:inline> </fo:block> </fo:block> <!-- body's space (after) filler --> <fo:block space-after.optimum="96px" space-after.maximum="96px" space-after.conditionality="retain"/> </fo:block> </fo:block-container> <!-- region (r2) --> <fo:block-container id="r2" absolute-position="absolute" left="10px" top="300px" width="620px" height="96px" background-color="black" display-align="center"> <!-- body (b1) --> <fo:block id="b1-2"> <!-- body's space (before) filler --> <fo:block space-after.optimum="96px" space-after.maximum="96px" space-after.conditionality="retain"/> <!-- div (d1) --> <fo:block id="d1-2"> <!-- p (p2) --> <fo:block id="p2" text-align="center"> <fo:inline font-size="40px" font-weight="bold" color="yellow">Text 2</fo:inline> </fo:block> </fo:block> <!-- body's space (after) filler --> <fo:block space-after.optimum="96px" space-after.maximum="96px" space-after.conditionality="retain"/> </fo:block> </fo:block-container> </fo:flow> </fo:page-sequence> </fo:root> |
The following, second intermediate document shows the synchronic state for
time interval [1,2), during which time both divisions d1
and
d2
are temporally active, and where paragraphs p1
and
p4
(and their ancestors) are associated with region r1
and paragraphs p2
and p3
(and their ancestors) are
associated with region r2
.
<tt tts:extent="640px 480px" xml:lang="en" xmlns="http://www.w3.org/ns/ttml" xmlns:tts="http://www.w3.org/ns/ttml#styling"> <head> <layout> <region xml:id="r1"> <style tts:origin="10px 100px"/> <style tts:extent="620px 96px"/> <style tts:fontSize="40px"/> <style tts:fontWeight="bold"/> <style tts:backgroundColor="black"/> <style tts:color="red"/> <style tts:textAlign="center"/> <style tts:displayAlign="center"/> <body xml:id="b1-1"> <div xml:id="d1-1"> <p xml:id="p1">Text 1</p> </div> <div xml:id="d2-1"> <p xml:id="p4">Text 4</p> </div> </body> </region> <region xml:id="r2"> <style tts:origin="10px 300px"/> <style tts:extent="620px 96px"/> <style tts:fontSize="40px"/> <style tts:fontWeight="bold"/> <style tts:backgroundColor="black"/> <style tts:color="yellow"/> <style tts:textAlign="center"/> <style tts:displayAlign="center"/> <body xml:id="b1-2"> <div xml:id="d1-2"> <p xml:id="p2">Text 2</p> </div> <div xml:id="d2-2"> <p xml:id="p3">Text 3</p> </div> </body> </region> </layout> </head> </tt> |
The following, third intermediate document shows the synchronic state for time
interval [2,3), during which time only division d2
is temporally active,
and where paragraphs p4
and p3
(and their ancestors)
are associated with regions r1
and r2
, respectively.
<tt tts:extent="640px 480px" xml:lang="en" xmlns="http://www.w3.org/ns/ttml" xmlns:tts="http://www.w3.org/ns/ttml#styling"> <head> <layout> <region xml:id="r1"> <style tts:origin="10px 100px"/> <style tts:extent="620px 96px"/> <style tts:fontSize="40px"/> <style tts:fontWeight="bold"/> <style tts:backgroundColor="black"/> <style tts:color="red"/> <style tts:textAlign="center"/> <style tts:displayAlign="center"/> <body xml:id="b1-1"> <div xml:id="d2-1"> <p xml:id="p4">Text 4</p> </div> </body> </region> <region xml:id="r2"> <style tts:origin="10px 300px"/> <style tts:extent="620px 96px"/> <style tts:fontSize="40px"/> <style tts:fontWeight="bold"/> <style tts:backgroundColor="black"/> <style tts:color="yellow"/> <style tts:textAlign="center"/> <style tts:displayAlign="center"/> <body xml:id="b1-2"> <div xml:id="d2-2"> <p xml:id="p3">Text 3</p> </div> </body> </region> </layout> </head> </tt> |
If a profile that applies to a Document Instance
requires use of the #lineBreak-uax14 feature
(i.e., the value attribute for the feature is specified as use
), then
the recommendations defined by Line
Breaking Algorithm [UAX14] apply when performing
line layout on the content of the Document Instance.
This section specifies the timing matter of the core vocabulary catalog, where timing is to be understood as a separable layer of information that applies to content and that denotes authorial intentions about the temporal presentation of that content.
No timing related element vocabulary is defined for use in the core vocabulary catalog.
This section defines the following basic timing attributes for use with timed elements:
In addition, this section defines the 10.2.4 timeContainer attribute for use with timed elements that serve simultaneously as timing containers.
The begin
attribute is used to specify the begin
point of a temporal interval associated with a timed element. If
specified, the value of a begin
attribute must adhere to
a <timeExpression> specification as defined by 10.3.1 <timeExpression>.
The begin point of a temporal interval is included in the interval; i.e., the interval is left-wise closed.
The semantics of the begin
attribute are those defined
by [SMIL 2.1], § 10.4.1,
while taking into account any overriding semantics defined by this specification.
Note:
See [SMIL 2.1], § 10.4.1, Begin value semantics when no begin
attribute is specified.
The end
attribute is used to specify the ending point
of a temporal interval associated with a timed element. If
specified, the value of an end
attribute must adhere to a
<timeExpression> specification as defined by 10.3.1 <timeExpression>.
The ending point of a temporal interval is not included in the interval; i.e., the interval is right-wise open.
The presentation effects of a non-empty active temporal interval include the frame immediately prior to the frame (or tick) equal to or immediately following the time specified by the ending point, but do not extend into this latter frame (or tick).
Note:
For example, if an active interval is [10s,10.33333s), and the frame rate is 30 frames per second, then the presentation effects of the interval are limited to frames 300 through 309 only (assuming that 0s corresponds with frame 0). The same holds if the active interval is specified as [300f,310f).
The semantics of the end
attribute are those defined
by [SMIL 2.1], § 10.4.1,
while taking into account any overriding semantics defined by this specification.
The dur
attribute is used to specify the duration of a
temporal interval associated with a timed element. If specified,
the value of a dur
attribute must adhere to a
<timeExpression> specification as defined by 10.3.1 <timeExpression>.
Note:
When the clock-time
form of a <timeExpression> specification
is used with a dur
attribute, it is intended to be interpreted as
a difference between two implied clock time expressions.
When a Document Instance specifies the use of the
smpte
time base and discontinuous
marker
mode, a (well-formed) dur
attribute must not be specified on any
element.
The semantics of the dur
attribute are those defined
by [SMIL 2.1], § 10.4.1,
while taking into account any overriding semantics defined by this specification. In a deliberate divergence from [SMIL 2.1], § 10.4.1, the value of the dur
attribute is permitted to be zero (0).
Note:
In the context of the subset of [SMIL 2.1] semantics supported by
this specification, the active duration of an element that specifies both
end
and dur
attributes is equal to the lesser of the value of
the dur
attribute and the difference between the value of the
end
attribute and the element's begin time.
The timeContainer
attribute is used to specify a local temporal
context by means of which timed child elements are temporally situated.
If specified, the value of a timeContainer
attribute must be one
of the following:
par
seq
If the time container semantics of an element instance is par
,
then the temporal intervals of child elements are considered to apply in
parallel, i.e., simultaneously in time. Furthermore, the specification of
the time interval of each child element is considered to be relative to the temporal
interval of the container element instance. For the purpose of determining the
[SMIL 2.1] endsync
semantics of a par
time container, a default value of all
applies.
Note:
The use of a default value of all
for the endsync
behavior is distinct from [SMIL 2.1] which uses a default value
of last
.
If the time container semantics of an element instance is seq
,
then the temporal intervals of child elements are considered to apply in
sequence, i.e., sequentially in time. Furthermore, the specification of
the time interval of each child element is considered to be relative to the temporal
interval of its sibling elements, unless it is the first child element, in which case
it is considered to be relative to the temporal interval of the container
element instance.
Each time container is considered to constitute an independent time base, i.e., time coordinate system.
If a timeContainer
attribute is not specified on an element that
has time container semantics, then par
time container semantics must apply.
Time container semantics applies only to the following element types:
The semantics of parallel and sequential time containment are those defined by [SMIL 2.1], § 10.4.2, while taking into account any overriding semantics defined by this specification.
Timing attribute values include the use of the following expressions:
A <timeExpression> is used to specify a coordinate within
some time base, where the applicable time base is determined by the
ttp:timeBase
parameter, and where the semantics defined
by N Time Expression Semantics apply.
Note:
See 6.2.4 ttp:frameRate, 6.2.9 ttp:subFrameRate, 6.2.10 ttp:tickRate, and 6.2.11 ttp:timeBase for further information on explicit specification of frame rate, sub-frame rate, tick rate, and time base.
<timeExpression> : clock-time | offset-time clock-time : hours ":" minutes ":" seconds ( fraction | ":" frames ( "." sub-frames )? )? offset-time : time-count fraction? metric hours : <digit> <digit> | <digit> <digit> <digit>+ minutes | seconds : <digit> <digit> frames : <digit> <digit> | <digit> <digit> <digit>+ sub-frames : <digit>+ fraction : "." <digit>+ time-count : <digit>+ metric : "h" // hours | "m" // minutes | "s" // seconds | "ms" // milliseconds | "f" // frames | "t" // ticks |
If a <timeExpression> is expressed in terms of a
clock-time, then leading zeroes are used when expressing hours,
minutes, seconds, and frames less than 10. Minutes are constrained to
[0…59], while seconds (including any fractional part) are constrained to the closed
interval [0,60], where the value 60 applies only to leap seconds. Except when ttp:timeBase
is clock
and ttp:clockMode
is local
or utc
, use of the special value of 60 seconds
to denote a leap second is undefined and should not be used.
If a <timeExpression> is expressed in terms of a
clock-time and a frames term is specified,
then the value of this term must be constrained to the interval
[0…F-1], where F is the frame rate
determined by the ttp:frameRate
parameter as defined
by 6.2.4 ttp:frameRate.
It is considered an error if a frames term or
f
(frames) metric is specified when the clock
time base applies.
If a <timeExpression> is expressed in terms of a
clock-time and a sub-frames term is
specified, then the value of this term must be constrained to the
interval [0…S-1], where S is the
sub-frame rate determined by the ttp:subFrameRate
parameter as defined by 6.2.9 ttp:subFrameRate.
It is considered an error if a sub-frames term is specified
when the clock
time base applies.
If the governing time base is smpte
, then (1) the offset-time
form is not defined and should not be used, and (2) the fraction
seconds component in a clock-time form is not defined and should
not be used.
The semantics of time containment, durations, and intervals defined by [SMIL 2.1] apply to the interpretation of like-named timed elements and timing vocabulary defined by this specification, given the following constraints:
The implicit duration of an anonymous span, a br
element, a set
element, or a span
element whose children
are exclusively text nodes, is defined as follows: if
the parent time container is a parallel time container, then
the implicit duration is equivalent to the indefinite
duration value as defined by [SMIL 2.1]; if the
parent time container is a sequential time container, then the implicit duration
is equivalent to zero.
The implicit duration of a body
element, div
element, p
element,
or span
element whose children are not exclusively text nodes, is determined in
accordance to (1) whether the element is a parallel or sequential time container, (2) the default
endsync
semantics defined above by 10.2.4 timeContainer,
and (3) the semantics of [SMIL 2.1] as applied to these time
containers.
The implicit duration of the region
element is defined to
be equivalent to the indefinite
duration value as defined by [SMIL 2.1].
If the governing time base is clock
, then time
expressions are considered to be equivalent to wall-clock
based timing in [SMIL 2.1], where the specific semantics of
N.1 Clock Time Base apply.
If the governing time base is media
, then time
expressions are considered to be equivalent to offset based timing
in [SMIL 2.1], where the specific semantics of
N.2 Media Time Base apply.
If the governing time base is smpte
, then time
expressions are considered to be equivalent to either offset
based timing or event based timing in [SMIL 2.1], where the specific semantics of
N.3 SMPTE Time Base apply.
For the purposes of performing presentation processing, the [construct anonymous spans] procedure is applied before resolving begin and end times of content elements.
The following example illustrates the timing semantics of anonymous spans that are children of sequential and parallel time containers.
The words Hello and Allo are both contained in anonymous spans that are children of a sequential
time container (the p
element). As such, neither Hello nor Allo will be displayed
since the implicit duration of the anonymous spans is 0. In contrast, the anonymous span that contains the word Guten
and the innermost span
that contains the word Tag are children of an element with parallel time container
semantics (the outermost span
element).
As a result, both Guten and Tag will be displayed since the implicit duration of their respective parent anonymous span
and span
element are indefinite, resulting in the implicit duration of
its span
, p
, div
and body
ancestors to also be indefinite.
... <body> <div> <p timeContainer="seq"> Hello <span>Guten <span>Tag</span></span> Allo </p> </div> </body> |
This section specifies the animation matter of the core vocabulary catalog, where animation is to be understood as a separable layer of information that combines timing and styling in order to denote authorial intention about (temporally) dynamic styling of content.
The following elements specify the structure and principal animation aspects of a Document Instance:
The set
element is used as a child element of a
Content element or a region
element
in order to express a discrete change of some style parameter value
that applies over some time interval.
The set
element accepts as its children zero or more
elements in the Metadata.class
element group.
<set begin = <timeExpression> dur = <timeExpression> end = <timeExpression> xml:id = ID xml:lang = string xml:space = (default|preserve) {a single attribute in TT Style namespace} {any attribute not in default or any TT namespace}> Content: Metadata.class* </set> |
Note:
The use of multiple set
element children may be used to effect fade and
position transitions.
An example of using the set
element to animate content
styling is illustrated below:
... <p dur="5s" tts:color="yellow"> <set begin="1s" dur="1s" tts:color="red"/> <set begin="2s" dur="1s" tts:color="green"/> <set begin="3s" dur="1s" tts:color="red"/> Text with Flashing Colors! </p> ... |
Note:
In the above example, the foreground color of the content "Text with Flashing Colors" is animated from yellow, to red, to green, to red, then back to yellow over a 5 second period.
An example of using the set
element to animate region
styling is illustrated below:
<tt xml:lang="" xmlns="http://www.w3.org/ns/ttml" xmlns:ttp="http://www.w3.org/ns/ttml#parameter" xmlns:tts="http://www.w3.org/ns/ttml#styling" ttp:cellResolution="40 16"> <head> <layout> <region xml:id="r1" timeContainer="seq"> <set dur="10s" tts:origin=" 8c 14c"/> <set dur="2s" tts:origin=" 2c 2c"/> <set dur="3s" tts:origin=" 8c 14c"/> <set dur="2s" tts:origin="14c 4c"/> <set dur="10s" tts:origin=" 8c 14c"/> <style tts:extent="24c 2c"/> </region> </layout> </head> <body region="r1">...</body> </tt> |
Note:
In the above example, the Root Container Region is divided into a
cell grid of 40 columns and 16 rows. A region, r1
, with
dimensions of 24 columns and 2 rows is then positioned within the
Root Container Region, with its position varying over time in order to
create an effect of moving the region, which may be desirable so as
to avoid obscuring characters in an underlying video with captions.
Note:
The semantics of the set
element
are based upon that defined by [SVG 1.1],
§ 6.2.13.
This section specifies the metadata matter of the core vocabulary catalog, where metadata is to be understood as a separable layer of information that applies to parameters, content, style, layout, timing, and even metadata itself, where the information represented by metadata takes one of two forms: (1) metadata defined by this specification for standardized use in a Document Instance, and (2) arbitrary metadata defined outside of the scope of this specification, whose use and semantics depend entirely upon an application's use of TTML Content.
The 12.1.1 metadata element serves as a generic container element for grouping metadata information.
In addition, the following elements, all defined in the TT Metadata Namespace, provide standard representations for metadata that is expected to be commonly used in a Document Instances:
The metadata
element functions as a generic container for metadata
information.
Metadata information may be expressed with a metadata
element by specifying
(1) one or more metadata attributes on the metadata
element,
(2) one or more metadata child elements in the metadata
element, or
(3) a combination of metadata attributes and metadata child elements. Both types
of metadata information are referred to in this document as metadata
items.
Note:
The meaning of a specific metadata item must be evaluated in the
context where it appears. The core vocabulary catalog permits an
arbitrary number of metadata
element children on any
Content element type. See specific element
vocabulary definitions for any constraints that apply to such
usage.
The use of document metadata is illustrated by the following example.
... <head> <metadata xmlns:ttm="http://www.w3.org/ns/ttml#metadata"> <ttm:title>Document Metadata Example</ttm:title> <ttm:desc>This document employs document metadata.</ttm:desc> </metadata> </head> ... |
The use of element metadata is illustrated by the following example.
... <div> <metadata xmlns:ttm="http://www.w3.org/ns/ttml#metadata"> <ttm:title>Chapter 6 – Sherlock Holmes Gives a Demonstration</ttm:title> <ttm:desc>Holmes shows Watson how the murderer entered the window.</ttm:desc> </metadata> </div> ... |
The use of metadata attribute items is illustrated by the following example.
... <div xmlns:ext="http://example.org/ttml#metadata"> <metadata ext:ednote="remove this division prior to publishing"/> </div> ... |
Note:
In the above example, a global attribute from a foreign (external)
namespace is used to express a metadata attribute that applies
semantically to the containing div
element. Note that the
attribute may also be expressed directly on the div
element;
however, in this case the author wishes to segregate certain metadata
attributes by expressing them indirectly on metadata
elements.
The use of foreign element metadata is illustrated by the following example.
... <metadata xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:dcterms="http://purl.org/dc/terms/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"> <dc:title>Foreign Element Metadata Example</dc:title> <dc:description>Express metadata using elements in foreign namespace.</dc:description> <dc:format xsi:type="dcterms:IMT">application/ttml+xml</dc:format> </metadata> ... |
Note:
In the above example, a number of elements defined by the Dublin Core metadata vocabulary are used to express document level metadata.
The ttm:title
element is used to express a human-readable title of
a specific element instance.
Note:
No specific use of the ttm:title
element is defined by this specification.
Examples of the ttm:title
element are shown above in
Example Fragment – Document Metadata and
Example Fragment – Element Metadata.
The ttm:desc
element is used to express a human-readable description of
a specific element instance.
Note:
No specific use of the ttm:desc
element is defined by this specification.
Examples of the ttm:desc
element are shown above in
Example Fragment – Document Metadata and
Example Fragment – Element Metadata.
The ttm:copyright
element is used to express a human-readable copyright that
applies to some scoping level.
A copyright statement that applies to a document as a whole should appear as a
child of the head
element.
Note:
No specific use of the ttm:copyright
element is defined by this specification.
The ttm:agent
element is used to define an agent for the purpose of associating
content information with an agent who is involved in the production or expression of that content.
The ttm:agent
element accepts as its children zero or more ttm:name
elements followed by
zero or one ttm:actor
element.
At least one ttm:name
element child should be specified that expresses a name for the
agent, whether it be the name of a person, character, group, or organization.
A type
attribute must be specified on each ttm:agent
element, and,
if specified, must have one of the following values:
person
character
group
organization
other
If the value of the type
attribute is character
, then the
ttm:agent
element instance should specify a ttm:actor
child that
specifies the agent that plays the role of the actor.
A ttm:agent
metadata item is considered to be significant only
when specified as a child of the head
element or as a child of a
metadata
element child of the head
element.
Note:
A ttm:agent
element instance is typically referenced using a ttm:agent
attribute on a Content element type.
Note:
If a character agent is played by multiple actors, then multiple character agents may be specified (and referenced) wherein different definitions of the character specify different actors.
The use of agent metadata is illustrated by the following example.
<tt xml:lang="en" xmlns="http://www.w3.org/ns/ttml" xmlns:ttm="http://www.w3.org/ns/ttml#metadata"> <head> <ttm:agent xml:id="connery" type="person"> <ttm:name type="family">Connery</ttm:name> <ttm:name type="given">Thomas Sean</ttm:name> <ttm:name type="alias">Sean</ttm:name> <ttm:name type="full">Sir Thomas Sean Connery</ttm:name> </ttm:agent> <ttm:agent xml:id="bond" type="character"> <ttm:name type="family">Bond</ttm:name> <ttm:name type="given">James</ttm:name> <ttm:name type="alias">007</ttm:name> <ttm:actor agent="connery"/> </ttm:agent> </head> <body> <div> ... <p ttm:agent="bond">I travel, a sort of licensed troubleshooter.</p> ... </div> </body> </tt> |
Note:
In the above example, two agents, a real (person) agent, Sean
Connery, and a fictitious (character) agent, James Bond, are defined,
where the latter is linked to the former by means of the a
ttm:actor
element. A reference is then made from content (the
p
element) to the character agent associated with
(responsible for producing) that content. Note that in this example
the ttm:agent
metadata items are specified as immediate
children of the document's head
element rather than being
placed in a container metadata
element.
The ttm:name
element is used to specify a name of a person, character,
group, or organization.
A type
attribute must be specified on each ttm:name
element, and,
if specified, must have one of the following values:
full
family
given
alias
other
The relationship between the type of a name and the syntactic expression of the name is not defined by this specification.
Two examples of the ttm:name
element are shown above in
Example Fragment – Agent Metadata.
The ttm:actor
element is used to link the definition of a (role-based) character
agent with another agent that portrays the character.
The agent
attribute of a ttm:actor
element must
reference a significant ttm:agent
element that denotes the agent
acting the part of a character.
An ttm:actor
element referencing its parent ttm:agent
element should be considered an error.
Note:
The semantics of an ttm:actor
element referencing its parent
ttm:agent
element are not defined. While discouraged, the relationship is syntactically permitted. It is therefore recommended that implementations be prepared to process this
syntax, e.g. by avoiding infinite loops
etc.
Note:
In the case of an actor playing themselves,
like Steve Coogan in The Trip playing Steve Coogan, the expected practice is to use two
ttm:agent
elements. The first element (corresponding to
Steve Coogan the person) has type
attribute equal to person
,
while the second element (corresponding to Steve Coogan the character) has (i) a type
attribute equal to character
and (ii) a child ttm:actor
element that
references the first ttm:agent
element.
An example of the ttm:actor
element is shown above in
Example Fragment – Agent Metadata.
This section specifies the following attributes in the TT Metadata Namespace
for use with the metadata
element and with certain
Content element types:
Note:
Only certain metadata item attributes may be used with Content elements. See the definitions of Content elements to determine permissible usage.
The ttm:agent
attribute takes an IDREFS
value, and is used with certain Content elements to designate the
agents that perform or are
involved in the performance of the content.
If specified, a ttm:agent
attribute must reference significant
ttm:agent
element instances.
The same IDREF, ID, should not appear more than once in the value of a ttm:agent
attribute.
Note:
This constraint is intended to discourage the use of redundant agent references.
An example of the ttm:agent
attribute is shown above in
Example Fragment – Agent Metadata.
The ttm:role
attribute may be used by a content author
to express
the roles, functions, or characteristics
of some Content element that is so labeled.
If specified, the value of this attribute must adhere to the following
syntax, where
the syntactic element S must adhere to production
[3] S
as defined by [XML 1.0] §
2.3:
ttm:role role ( S role )* role : "action" | "caption" | "description" | "dialog" | "expletive" | "kinesic" | "lyrics" | "music" | "narration" | "quality" | "sound" | "source" | "suppressed" | "reproduction" | "thought" | "title" | "transcription" | extension-role extension-role : "x-" token-char+ token-char : { XML NameChar } // XML 1.1 Production [4a] |
The same role
token, R, should not appear more than once in the value of a ttm:role
attribute.
Note:
This constraint is intended to discourage the use of redundant role tokens.
Note:
All values of ttm:role
that do not start with the prefix
x-
are reserved for future standardization.
Note:
If using a custom x-
prefixed form of ttm:role
, it
is recommended that an organization unique infix be used as well in order to
prevent collisions. For example, x-example-org-custom-role
. Furthermore,
a registry for role values is available at
http://www.w3.org/wiki/TTML/RoleRegistry
in order to promote interoperability and collision avoidance.
This appendix is normative.
For the purposes of this specification, a Reduced XML Infoset is an XML Information Set [XML InfoSet] that consists of only the following information items and information item properties:
This appendix is normative.
This appendix specifies the following schemas for use with Document Instances:
Relax NG, Compact Syntax (RNC) Schema
XML Schema Definition (XSD) Schema
In any case where a schema specified by this appendix differs from the normative definitions of document type, element type, or attribute type as defined by the body of this specification, then the body of this specification takes precedence.
A Relax NG Compact Syntax (RNC) [RELAX NG] based schema for TTML Content is available at ZIP Archive. This schema does not normatively define the validity of Timed Text Markup Language content as defined by this specification. In particular, the formal validity of TTML Content is defined by 3.1 Content Conformance.
A W3C XML Schema Definition (XSD) [XML Schema Part 1] based schema for TTML Content is available at ZIP Archive. This schema does not normatively define the validity of Timed Text Markup Language content as defined by this specification. In particular, the formal validity of TTML Content is defined by 3.1 Content Conformance.
This appendix is normative.
This appendix registers a new media type, "application/ttml+xml" in conformance with BCP 13 and W3CRegMedia. The information in this appendix is being submitted to the Internet Engineering Steering Group (IESG) for review, approval, and registration with the Internet Assigned Numbers Authority (IANA).
application
ttml+xml
None.
If specified, the charset
parameter must match the XML encoding declaration, or if absent, the actual encoding.
The document profile of a TTMLDocument Instance may be specified using
an optional profile
parameter, which, if specified,
the value of which must adhere to the syntax and semantics of
ttp:profile
parameter defined by
Section
6.2.8 ttp:profile
of the published specification.
binary
As with other XML types and as noted in [XML Media Types] (http://www.rfc-editor.org/rfc/rfc3023.txt), Section 10, repeated expansion of maliciously constructed XML entities can be used to consume large amounts of memory, which may cause XML processors in constrained environments to fail.
In addition, because of the extensibility features for TTML and of XML in general, it is possible that "application/ttml+xml" may describe content that has security implications beyond those described here. However, TTML does not provide for any sort of active or executable content, and if the processor follows only the normative semantics of the published specification, this content will be outside TTML namespaces and may be ignored. Only in the case where the processor recognizes and processes the additional content, or where further processing of that content is dispatched to other processors, would security issues potentially arise. And in that case, they would fall outside the domain of this registration document.
Although not prohibited, there are no expectations that XML signatures or encryption would normally be employed.
The published specification describes processing semantics that dictate behavior that must be followed when dealing with, among other things, unrecognized elements and attributes, both in TTML namespaces and in other namespaces.
Because TTML is extensible, conformant "application/ttml+xml" processors may expect (and enforce) that content received is well-formed XML, but it cannot be guaranteed that the content is valid to a particular DTD or Schema or that the processor will recognize all of the elements and attributes in the document.
This media type registration is extracted from Appendix C Media Type Registration of the Timed Text Markup Language (TTML) 1.0 specification: Timed Text Markup Language 1 (TTML1).
TTML is used in the television industry for the purpose of authoring, transcoding and exchanging timed text information and for delivering captions, subtitles, and other metadata for television material repurposed for the Web or, more generally, the Internet.
There is partial and full support of TTML in components used by several Web browsers plugins, and in a number of caption authoring tools.
.ttml
"TTML"
For documents labeled as application/ttml+xml, the fragment identifier notation is intended to be used with xml:id attributes, as described in section 7.2.1 of the Timed Text Markup Language 1 (TTML1) specification.
Timed Text Working Group (public-tt@w3.org)
COMMON
None
The published specification is a work product of the World Wide Web Consortium's Timed Text (TT) Working Group.
The W3C has change control over this specification.
This appendix is normative.
This appendix specifies (1) a set of feature designations, each of which labels one or more syntactic and/or semantic features defined by this specification, and (2) for each designated feature, whether the feature is mandatory or optional for a transformation or presentation processor.
Note:
A TTML processor is said to implement the transformation semantics or implement the presentation semantics of feature designation F if it satisfies the requirements of this appendix with respect to the definition of feature designation F as pertains to transformation or presentation processing, respectively.
A feature designation is expressed as a string that adheres to the following form:
feature-designation : feature-namespace designation feature-namespace : TT Feature Namespace // http://www.w3.org/ns/ttml/feature/ designation : "#" token-char+ token-char : { XML NameChar } // XML 1.1 Production [4a]
All values of feature-designation
not defined by this specification are reserved
for future standardization.
The following sub-sections define all feature designations, expressed as relative URIs (fragment identifiers) with respect to the TT Feature Namespace base URI.
A TTML transformation processor supports the #animation
feature if it recognizes and is capable of transforming the following
vocabulary defined by 11 Animation:
A TTML presentation processor supports the #animation
feature if it implements presentation semantic support for the
same vocabulary enumerated above.
A TTML transformation processor supports the
#backgroundColor
feature if it recognizes and is capable of
transforming the tts:backgroundColor
attribute.
A TTML presentation processor supports the
#backgroundColor
feature if it (1) implements presentation semantic support
for the tts:backgroundColor
attribute
and (2) is capable of displaying or generating an output display
signal that distinguishes between at least sixteen (16) values of color,
including all primary and secondary colors of the SRGB color space.
Note:
Support for this feature is intended to imply support for the following features: #backgroundColor-block
,
#backgroundColor-region
, and #backgroundColor-inline
.
A TTML transformation processor supports the
#backgroundColor-block
feature if it recognizes and is
capable of transforming all defined values of the tts:backgroundColor
attribute when applied to a content element that would generate a
block area during presentation processing.
A TTML presentation processor supports the
#backgroundColor-block
feature if it (1) implements presentation semantic
support for the tts:backgroundColor
attribute when applied to a content element that generates a
block area
and (2) is capable of displaying or generating an output display
signal that distinguishes between at least sixteen (16) values of color,
including all primary and secondary colors of the SRGB color space.
A TTML transformation processor supports the
#backgroundColor-inline
feature if it recognizes and is
capable of transforming all defined values of the tts:backgroundColor
attribute when applied to a content element that would generate an
inline area during presentation processing.
A TTML presentation processor supports the
#backgroundColor-inline
feature if it (1) implements presentation semantic
support for the tts:backgroundColor
attribute when applied to a content element that generates an
inline area
and (2) is capable of displaying or generating an output display
signal that distinguishes between at least sixteen (16) values of color,
including all primary and secondary colors of the SRGB color space.
A TTML transformation processor supports the
#backgroundColor-region
feature if it recognizes and is
capable of transforming all defined values of the tts:backgroundColor
attribute when applied to a region
element.
A TTML presentation processor supports the
#backgroundColor-region
feature if it (1) implements presentation semantic
support for the tts:backgroundColor
attribute when applied to a region
element
and (2) is capable of displaying or generating an output display
signal that distinguishes between at least sixteen (16) values of color,
including all primary and secondary colors of the SRGB color space
.
A TTML transformation processor supports the
#cellResolution
feature if it recognizes and is capable of
transforming the ttp:cellResolution
attribute.
A TTML presentation processor supports the
#cellResolution
feature if it implements presentation semantic support
for the ttp:cellResolution
attribute.
A TTML transformation processor supports the
#clockMode
feature if it recognizes and is capable of
transforming the ttp:clockMode
attribute.
A TTML presentation processor supports the
#clockMode
feature if it implements presentation semantic support
for the ttp:clockMode
attribute.
A TTML transformation processor supports the
#clockMode-gps
feature if it recognizes and is capable of
transforming the gps
value of the ttp:clockMode
attribute.
A TTML presentation processor supports the
#clockMode-gps
feature if it implements presentation semantic support
for the gps
value of the ttp:clockMode
attribute.
A TTML transformation processor supports the
#clockMode-local
feature if it recognizes and is capable of
transforming the local
value of the ttp:clockMode
attribute.
A TTML presentation processor supports the
#clockMode-local
feature if it implements presentation semantic support
for the local
value of the ttp:clockMode
attribute.
A TTML transformation processor supports the
#clockMode-utc
feature if it recognizes and is capable of
transforming the utc
value of the ttp:clockMode
attribute.
A TTML presentation processor supports the
#clockMode-utc
feature if it implements presentation semantic support
for the utc
value of the ttp:clockMode
attribute.
A TTML transformation processor supports the
#color
feature if it recognizes and is capable of
transforming the tts:color
attribute.
A TTML presentation processor supports the
#color
feature if it (1) implements presentation semantic support
for the tts:color
attribute
and (2) is capable of displaying or generating an output display
signal that distinguishes between at least sixteen (16) values of color,
including all primary and secondary colors of the SRGB color space.
A TTML transformation processor supports the #content
feature if it recognizes and is capable of transforming the following
vocabulary defined by 7 Content:
A TTML presentation processor supports the #content
feature if it implements presentation semantic support for the
same vocabulary enumerated above.
A TTML transformation processor supports the #core
feature if it recognizes and is capable of transforming the following
core attributes vocabulary defined by 7 Content:
A TTML presentation processor supports the #core
feature if it implements presentation semantic support for the
same vocabulary enumerated above.
A TTML transformation processor supports the
#direction
feature if it recognizes and is capable of
transforming all defined values of the tts:direction
attribute.
A TTML presentation processor supports the
#direction
feature if it implements presentation semantic support
for all defined values of the tts:direction
attribute.
A TTML transformation processor supports the
#display
feature if it recognizes and is capable of
transforming all defined values of the tts:display
attribute.
A TTML presentation processor supports the
#display
feature if it implements presentation semantic support
for all defined values of the tts:display
attribute.
A TTML transformation processor supports the
#display-block
feature if it recognizes and is
capable of transforming all defined values of the tts:display
attribute when applied to a content element that would generate a
block area during presentation processing.
A TTML presentation processor supports the
#display-block
feature if it implements presentation semantic
support for all defined values of the tts:display
attribute when applied to a content element that generates a
block area.
A TTML transformation processor supports the
#display-inline
feature if it recognizes and is
capable of transforming all defined values of the tts:display
attribute when applied to a content element that would generate an
inline area during presentation processing.
A TTML presentation processor supports the
#display-inline
feature if it implements presentation semantic
support for all defined values of the tts:display
attribute when applied to a content element that generates an
inline area.
A TTML transformation processor supports the
#display-region
feature if it recognizes and is
capable of transforming all defined values of the tts:display
attribute when applied to a region
element.
A TTML presentation processor supports the
#display-region
feature if it implements presentation semantic
support for all defined values of the tts:display
attribute when applied to a region
element.
A TTML transformation processor supports the
#displayAlign
feature if it recognizes and is capable of
transforming all defined values of the tts:displayAlign
attribute.
A TTML presentation processor supports the
#displayAlign
feature if it implements presentation semantic support
for all defined values of the tts:displayAlign
attribute.
A TTML transformation processor supports the
#dropMode
feature if it recognizes and is capable of
transforming the ttp:dropMode
attribute.
A TTML presentation processor supports the
#dropMode
feature if it implements presentation semantic support
for the ttp:dropMode
attribute.
A TTML transformation processor supports the
#dropMode-dropNTSC
feature if it recognizes and is capable of
transforming the dropNTSC
value of the ttp:dropMode
attribute.
A TTML presentation processor supports the
#dropMode-dropNTSC
feature if it implements presentation semantic support
for the dropNTSC
value of the ttp:dropMode
attribute.
A TTML transformation processor supports the
#dropMode-dropPAL
feature if it recognizes and is capable of
transforming the dropPAL
value of the ttp:dropMode
attribute.
A TTML presentation processor supports the
#dropMode-dropPAL
feature if it implements presentation semantic support
for the dropPAL
value of the ttp:dropMode
attribute.
A TTML transformation processor supports the
#dropMode-nonDrop
feature if it recognizes and is capable of
transforming the nonDrop
value of the ttp:dropMode
attribute.
A TTML presentation processor supports the
#dropMode-nonDrop
feature if it implements presentation semantic support
for the nonDrop
value of the ttp:dropMode
attribute.
A TTML transformation processor supports the
#extent
feature if it recognizes and is capable of
transforming the tts:extent
attribute.
A TTML presentation processor supports the
#extent
feature if it implements presentation semantic support for
the tts:extent
attribute.
A TTML transformation processor supports the
#extent-region
feature if it recognizes and is capable of
transforming the tts:extent
attribute when applied to a region
element.
A TTML presentation processor supports the
#extent-region
feature if it implements presentation semantic support for
the tts:extent
attribute when applied to a region
element.
A TTML transformation processor supports the
#extent-root
feature if it recognizes and is capable of
transforming the tts:extent
attribute when applied to the tt
element.
A TTML presentation processor supports the
#extent-root
feature if it implements presentation semantic support for
the tts:extent
attribute when applied to a tt
element.
A TTML transformation processor supports the
#fontFamily
feature if it recognizes and is capable of
transforming the tts:fontFamily
attribute.
A TTML presentation processor supports the
#fontFamily
feature if it implements presentation semantic support
for the tts:fontFamily
attribute.
A TTML transformation processor supports the
#fontFamily-generic
feature if it recognizes and is
capable of transforming <genericFamilyName>
values when used with the tts:fontFamily
attribute.
A TTML presentation processor supports the
#fontFamily-generic
feature if it implements presentation semantic
support for <genericFamilyName>
values when used with the tts:fontFamily
attribute.
A TTML transformation processor supports the
#fontFamily-non-generic
feature if it recognizes and is
capable of transforming <familyName>
values when used with the tts:fontFamily
attribute.
A TTML presentation processor supports the
#fontFamily-non-generic
feature if it implements presentation semantic
support for <familyName>
values when used with the tts:fontFamily
attribute.
A TTML transformation processor supports the
#fontSize
feature if it recognizes and is capable of
transforming the tts:fontSize
attribute.
A TTML presentation processor supports the
#fontSize
feature if it implements presentation semantic support
for the tts:fontSize
attribute.
A TTML transformation processor supports the
#fontSize-anamorphic
feature if it recognizes and is
capable of transforming values of the tts:fontSize
attribute that consist of two <length> specifications.
A TTML presentation processor supports the
#fontSize-anamorphic
feature if it implements presentation semantic
support for defined values of the tts:fontSize
attribute that consist of two <length> specifications.
A TTML transformation processor supports the
#fontSize-isomorphic
feature if it recognizes and is
capable of transforming values of the tts:fontSize
attribute that consist of a single <length> specification.
A TTML presentation processor supports the
#fontSize-isomorphic
feature if it implements presentation semantic
support for defined values of the tts:fontSize
attribute that consist of a single <length> specification.
A TTML transformation processor supports the
#fontStyle
feature if it recognizes and is capable of
transforming all defined values of the tts:fontStyle
attribute.
A TTML presentation processor supports the
#fontStyle
feature if it implements presentation semantic support
for all defined values of the tts:fontStyle
attribute.
A TTML transformation processor supports the
#fontStyle-italic
feature if it recognizes and is
capable of transforming the italic
value of the tts:fontStyle
attribute.
A TTML presentation processor supports the
#fontStyle-italic
feature if it implements presentation semantic
support for the italic
of the tts:fontStyle
attribute.
A TTML transformation processor supports the
#fontStyle-oblique
feature if it recognizes and is
capable of transforming the oblique
value of the tts:fontStyle
attribute.
A TTML presentation processor supports the
#fontStyle-oblique
feature if it implements presentation semantic
support for the oblique
of the tts:fontStyle
attribute.
A TTML transformation processor supports the
#fontWeight
feature if it recognizes and is capable of
transforming all defined values of the tts:fontWeight
attribute.
A TTML presentation processor supports the
#fontWeight
feature if it implements presentation semantic support
for all defined values of the tts:fontWeight
attribute.
A TTML transformation processor supports the
#fontWeight-bold
feature if it recognizes and is
capable of transforming bold
value of the tts:fontWeight
attribute.
A TTML presentation processor supports the
#fontWeight-bold
feature if it implements presentation semantic
support for the bold
of the tts:fontWeight
attribute.
A TTML transformation processor supports the
#frameRate
feature if it recognizes and is capable of
transforming the ttp:frameRate
attribute.
A TTML presentation processor supports the
#frameRate
feature if it implements presentation semantic support
for the ttp:frameRate
attribute.
A TTML transformation processor supports the
#frameRateMultiplier
feature if it recognizes and is capable of
transforming the ttp:frameRateMultiplier
attribute.
A TTML presentation processor supports the
#frameRateMultiplier
feature if it implements presentation semantic support
for the ttp:frameRateMultiplier
attribute.
A TTML transformation processor supports the #layout
feature if it (1) recognizes and is capable of transforming the following
vocabulary defined by 9 Layout:
and (2) supports the following attributes when applied to the
region
element:
A TTML presentation processor supports the #layout
feature if it implements presentation semantic support for the
same vocabulary and features enumerated above.
A TTML transformation processor supports the #length
feature if it recognizes and is capable of transforming all defined
values of the <length>
style value expression.
A TTML presentation processor supports the #length
feature if it implements presentation semantic support for all defined values of
the <length> style value
expression.
Note:
Support for #length
is intended to imply support for the following features:
#length-integer
,
#length-real
,
#length-positive
,
#length-negative
,
#length-cell
,
#length-em
,
#length-percentage
, and
#length-pixel
.
A TTML transformation processor supports the
#length-cell
feature if it recognizes and is capable of
transforming scalar values of the <length> style value expression
that use c
(cell) units.
A TTML presentation processor supports the
#length-cell
feature if it implements presentation semantic support
for scalar values of the <length> style value expression
that use c
(cell) units.
Note:
Support for #length-cell
does not, by itself, imply support for
#length-integer
,
#length-real
,
#length-positive
, or
#length-negative
features.
A TTML transformation processor supports the
#length-em
feature if it recognizes and is capable of
transforming scalar values of the <length> style value expression
that use em
(EM) units.
A TTML presentation processor supports the
#length-em
feature if it implements presentation semantic support
for scalar values of the <length> style value expression
that use em
(EM) units.
Note:
Support for #length-em
does not, by itself, imply support for
#length-integer
,
#length-real
,
#length-positive
, or
#length-negative
features.
A TTML transformation processor supports the
#length-integer
feature if it recognizes and is
capable of transforming integer values of the <length> style value
expression.
A TTML presentation processor supports the
#length-integer
feature if it implements presentation semantic
support for integer values of the <length> style value
expression.
Note:
Support for #length-integer
does not, by itself, imply support for
#length-positive
or
#length-negative
features.
A TTML transformation processor supports the
#length-negative
feature if it recognizes and is
capable of transforming negative values of the <length> style value
expression.
A TTML presentation processor supports the
#length-negative
feature if it implements presentation semantic
support for negative values of the <length> style value
expression.
Note:
Support for #length-negative
does not, by itself, imply support for
#length-integer
or
#length-real
features.
A TTML transformation processor supports the
#length-percentage
feature if it recognizes and is
capable of transforming percentage values of the <length> style value
expression.
A TTML presentation processor supports the
#length-percentage
feature if it implements presentation semantic
support for percentage values of the <length> style value
expression.
Note:
Support for #length-percentage
does not, by itself, imply support for
#length-integer
,
#length-real
,
#length-positive
, or
#length-negative
features.
A TTML transformation processor supports the
#length-pixel
feature if it recognizes and is capable of
transforming scalar values of the <length> style value expression
that use px
(pixel) units.
A TTML presentation processor supports the
#length-pixel
feature if it implements presentation semantic support
for scalar values of the <length> style value expression
that use px
(pixel) units.
Note:
Support for #length-pixel
does not, by itself, imply support for
#length-integer
,
#length-real
,
#length-positive
, or
#length-negative
features.
A TTML transformation processor supports the
#length-positive
feature if it recognizes and is
capable of transforming positive values of the <length> style value
expression.
A TTML presentation processor supports the
#length-positive
feature if it implements presentation semantic
support for positive values of the <length> style value
expression.
Note:
Support for #length-positive
is intended to imply support for zero valued
<length> style value
expressions.
Note:
Support for #length-positive
does not, by itself, imply support for
#length-integer
or
#length-real
features.
A TTML transformation processor supports the
#length-real
feature if it recognizes and is
capable of transforming real values of the <length> style value
expression.
A TTML presentation processor supports the
#length-real
feature if it implements presentation semantic
support for real values of the <length> style value
expression.
Note:
Support for #length-real
is intended to imply support for
integer valued <length> style value
expressions as well as real valued expressions.
Note:
Support for #length-real
does not, by itself, imply support for
#length-positive
or
#length-negative
features.
A TTML transformation processor supports the
#lineBreak-uax14
feature if it recognizes and is capable of
transforming requirements expressed by [UAX14] into its
target document space.
A TTML presentation processor supports the
#lineBreak-uax14
feature if it implements presentation semantic support
for [UAX14] as applies to line breaking.
A TTML transformation processor supports the
#lineHeight
feature if it recognizes and is capable of
transforming the tts:lineHeight
attribute.
A TTML presentation processor supports the
#lineHeight
feature if it implements presentation semantic support for
the tts:lineHeight
attribute.
A TTML transformation processor supports the
#markerMode
feature if it recognizes and is capable of
transforming the ttp:markerMode
attribute.
A TTML presentation processor supports the
#markerMode
feature if it implements presentation semantic support
for the ttp:markerMode
attribute.
A TTML transformation processor supports the
#markerMode-continuous
feature if it recognizes and is capable of
transforming the continuous
value of the ttp:markerMode
attribute.
A TTML presentation processor supports the
#markerMode-continuous
feature if it implements presentation semantic support
for the continuous
value of the ttp:markerMode
attribute.
A TTML transformation processor supports the
#markerMode-discontinuous
feature if it recognizes and is capable of
transforming the discontinuous
value of the ttp:markerMode
attribute.
A TTML presentation processor supports the
#markerMode-discontinuous
feature if it implements presentation semantic support
for the discontinuous
value of the ttp:markerMode
attribute.
A TTML transformation processor supports the #metadata
feature if it recognizes and is capable of transforming the following
vocabulary defined by 12 Metadata:
A TTML presentation processor supports the #metadata
feature if it recognizes and is capable of presenting the information
expressed by the same vocabulary enumerated above.
Note:
This specification does not define a standardized form for the presentation of metadata information. The presentation or ability to present metadata information is considered to be implementation dependent.
A TTML transformation processor supports the #nested-div
feature if it recognizes and is capable of transforming nested
div
elements.
A TTML presentation processor supports the #nested-div
feature if it implements presentation semantic support for nested
div
elements.
A TTML transformation processor supports the #nested-span
feature if it recognizes and is capable of transforming nested
span
elements.
A TTML presentation processor supports the #nested-span
feature if it implements presentation semantic support for nested
span
elements.
A TTML transformation processor supports the
#opacity
feature if it recognizes and is capable of
transforming the tts:opacity
attribute.
A TTML presentation processor supports the
#opacity
feature if it (1) implements presentation semantic support for
the tts:opacity
attribute and (2) is capable of displaying or generating an output display
signal that distinguishes between at least eight (8) values of opacity.
A TTML transformation processor supports the
#origin
feature if it recognizes and is capable of
transforming the tts:origin
attribute.
A TTML presentation processor supports the
#origin
feature if it implements presentation semantic support for
the tts:origin
attribute.
A TTML transformation processor supports the
#overflow
feature if it recognizes and is capable of
transforming all defined values of the tts:overflow
attribute.
A TTML presentation processor supports the
#overflow
feature if it implements presentation semantic support for
all defined values of the tts:overflow
attribute.
A TTML transformation processor supports the
#overflow-visible
feature if it recognizes and is capable of
transforming the visible
value of the tts:overflow
attribute.
A TTML presentation processor supports the
#overflow-visible
feature if it implements presentation semantic support for
the visible
value of the tts:overflow
attribute.
A TTML transformation processor supports the
#padding
feature if it recognizes and is capable of
transforming the tts:padding
attribute.
A TTML presentation processor supports the
#padding
feature if it implements presentation semantic support for
the tts:padding
attribute.
A TTML transformation processor supports the
#padding-1
feature if it recognizes and is capable of
transforming values of the tts:padding
attribute
that consist of one <length> specification.
A TTML presentation processor supports the #padding-1
feature if it implements presentation semantic support for values of the tts:padding
attribute
that consist of one <length> specification.
A TTML transformation processor supports the
#padding-2
feature if it recognizes and is capable of
transforming values of the tts:padding
attribute
that consist of two <length> specification.
A TTML presentation processor supports the #padding-2
feature if it implements presentation semantic support for values of the tts:padding
attribute
that consist of two <length> specification.
A TTML transformation processor supports the
#padding-3
feature if it recognizes and is capable of
transforming values of the tts:padding
attribute
that consist of three <length> specification.
A TTML presentation processor supports the #padding-3
feature if it implements presentation semantic support for values of the tts:padding
attribute
that consist of three <length> specification.
A TTML transformation processor supports the
#padding-4
feature if it recognizes and is capable of
transforming values of the tts:padding
attribute
that consist of four <length> specification.
A TTML presentation processor supports the #padding-4
feature if it implements presentation semantic support for values of the tts:padding
attribute
that consist of four <length> specification.
A TTML transformation processor supports the
#pixelAspectRatio
feature if it recognizes and is capable of
transforming the ttp:pixelAspectRatio
attribute.
A TTML presentation processor supports the
#pixelAspectRatio
feature if it implements presentation semantic support
for the ttp:pixelAspectRatio
attribute.
A TTML processor supports the
#presentation
feature if it
(1) satisfies the generic processor criteria defined by
3.2.1 Generic Processor Conformance,
(2) implements support for the region and line layout
semantics defined by 9.3 Region Layout and Presentation and
9.4 Line Layout, respectively, and
(3) implements presentation semantics for the following features:
In addition, a TTML processor that supports the
#presentation
feature should satisfy the user agent
accessibility guidelines specified by [UAAG].
A TTML transformation processor supports the #profile
feature if it recognizes and is capable of transforming the
ttp:profile
attribute on the tt
element and
transforming the following vocabulary defined by
6.1 Parameter Element Vocabulary:
A TTML presentation processor supports the
#profile
feature if it implements presentation semantic support
for the same vocabulary specified above.
A TTML transformation processor supports the
#showBackground
feature if it recognizes and is capable
of transforming all defined values of the tts:showBackground
attribute.
A TTML presentation processor supports the
#showBackground
feature if it implements presentation semantic support
for all defined values of the tts:showBackground
attribute.
A TTML transformation processor supports the #structure
feature if it recognizes and is capable of transforming the following
vocabulary defined by 7 Content:
A TTML presentation processor supports the #structure
feature if it implements presentation semantic support for the
same vocabulary enumerated above.
A TTML transformation processor supports the #styling
feature if it recognizes and is capable of transforming the following
vocabulary defined by 8 Styling:
A TTML presentation processor supports the #styling
feature if it implements presentation semantic support for the
same vocabulary enumerated above.
A TTML transformation processor supports the
#styling-chained
feature if it recognizes and is capable
of transforming chained style association as defined by 8.4.1.3 Chained Referential Styling.
A TTML presentation processor supports the
#styling-chained
feature if it implements presentation semantic
support for chained style association as defined by 8.4.1.3 Chained Referential Styling.
A TTML transformation processor supports the
#styling-inheritance
feature if it recognizes and is
capable of transforming content style inheritance as defined by
8.4.2.1 Content Style Inheritance.
A TTML presentation processor supports the
#styling-inheritance-content
feature if it implements
presentation semantic support for content style inheritance as defined by 8.4.2.1 Content Style Inheritance.
A TTML transformation processor supports the
#styling-inheritance
feature if it recognizes and is
capable of transforming region style inheritance as defined by
8.4.2.2 Region Style Inheritance.
A TTML presentation processor supports the
#styling-inheritance-region
feature if it implements
presentation semantic support for region style inheritance as defined by 8.4.2.2 Region Style Inheritance.
A TTML transformation processor supports the
#styling-inline
feature if it recognizes and is capable
of transforming inline style association as defined by 8.4.1.1 Inline Styling.
A TTML presentation processor supports the
#styling-inline
feature if it implements presentation semantic support
for inline style association as defined by 8.4.1.1 Inline Styling.
A TTML transformation processor supports the
#styling-nested
feature if it recognizes and is capable
of transforming nested style association as defined by 8.4.1.4 Nested Styling.
A TTML presentation processor supports the
#styling-nested
feature if it implements presentation semantic support
for nested style association as defined by 8.4.1.4 Nested Styling.
A TTML transformation processor supports the
#styling-referential
feature if it recognizes and is capable
of transforming referential style association as defined by 8.4.1.2 Referential Styling.
A TTML presentation processor supports the
#styling-referential
feature if it implements presentation semantic
support for referential style association as defined by 8.4.1.2 Referential Styling.
A TTML transformation processor supports the
#subFrameRate
feature if it recognizes and is capable of
transforming the ttp:subFrameRate
attribute.
A TTML presentation processor supports the
#subFrameRate
feature if it implements presentation semantic support
for the ttp:subFrameRate
attribute.
A TTML transformation processor supports the
#textAlign
feature if it recognizes and is capable
of transforming all defined values of the tts:textAlign
attribute.
A TTML presentation processor supports the
#textAlign
feature if it implements presentation semantic support
for all defined values of the tts:textAlign
attribute.
A TTML transformation processor supports the
#textAlign-absolute
feature if it recognizes and is
capable of transforming the left
, center
,
and right
values of the tts:textAlign
attribute.
A TTML presentation processor supports the
#textAlign-absolute
feature if it implements presentation semantic
support for the left
, center
, and
right
values of the tts:textAlign
attribute.
A TTML transformation processor supports the
#textAlign-relative
feature if it recognizes and is
capable of transforming the start
, center
,
and end
values of the tts:textAlign
attribute.
A TTML presentation processor supports the
#textAlign-relative
feature if it implements presentation semantic
support for the start
, center
, and
end
values of the tts:textAlign
attribute.
A TTML transformation processor supports the
#textDecoration
feature if it recognizes and is capable
of transforming all defined values of the tts:textDecoration
attribute.
A TTML presentation processor supports the
#textDecoration
feature if it implements presentation semantic support
for all defined values of the tts:textDecoration
attribute.
A TTML transformation processor supports the
#textDecoration-over
feature if it recognizes and is
capable of transforming the overline
and
noOverline
values of the tts:textDecoration
attribute.
A TTML presentation processor supports the
#textDecoration-over
feature if it implements
presentation semantic support for the overline
and
noOverline
values of the tts:textDecoration
attribute.
A TTML transformation processor supports the
#textDecoration-through
feature if it recognizes and is
capable of transforming the lineThrough
and
noLineThrough
values of the tts:textDecoration
attribute.
A TTML presentation processor supports the
#textDecoration-through
feature if it implements
presentation semantic support for the lineThrough
and
noLineThrough
values of the tts:textDecoration
attribute.
A TTML transformation processor supports the
#textDecoration-under
feature if it recognizes and is
capable of transforming the underline
and
noUnderline
values of the tts:textDecoration
attribute.
A TTML presentation processor supports the
#textDecoration-under
feature if it implements
presentation semantic support for the underline
and
noUnderline
values of the tts:textDecoration
attribute.
A TTML processor supports the #textOutline
feature if it
supports the following features:
A TTML transformation processor supports the
#textOutline-blurred
feature if it recognizes and is capable of
transforming values of the tts:textOutline
attribute that includes a blur radius specification.
A TTML presentation processor supports the
#textOutline-blurred
feature if it implements presentation semantic support
for values of the tts:textOutline
attribute that includes a blur radius specification.
A TTML transformation processor supports the
#textOutline-unblurred
feature if it recognizes and is capable of
transforming values of the tts:textOutline
attribute that does not include a blur radius specification.
A TTML presentation processor supports the
#textOutline-unblurred
feature if it implements presentation semantic support
for values of the tts:textOutline
attribute that does not include a blur radius specification.
A TTML transformation processor supports the
#tickRate
feature if it recognizes and is capable of
transforming the ttp:tickRate
attribute.
A TTML presentation processor supports the
#tickRate
feature if it implements presentation semantic support
for the ttp:tickRate
attribute.
A TTML transformation processor supports the
#timeBase-clock
feature if it recognizes and is capable
of transforming the clock
value of the ttp:timeBase
attribute and if it supports the #clockMode
feature.
A TTML presentation processor supports the
#timeBase-clock
feature if it implements presentation
semantic support for the clock
value of the ttp:timeBase
attribute and if it supports the #clockMode
feature.
A TTML transformation processor supports the
#timeBase-media
feature if it recognizes and is capable
of transforming the media
value of the ttp:timeBase
attribute.
A TTML presentation processor supports the
#timeBase-media
feature if it implements presentation
semantic support for the media
value of the ttp:timeBase
attribute.
A TTML transformation processor supports the
#timeBase-smpte
feature if it recognizes and is capable
of transforming the smpte
value of the ttp:timeBase
attribute and if it supports the #dropMode
feature.
A TTML presentation processor supports the
#timeBase-smpte
feature if it implements presentation
semantic support for the smpte
value of the ttp:timeBase
attribute and if it supports the #dropMode
feature.
A TTML transformation processor supports the
#timeContainer
feature if it recognizes and is capable of
transforming the timeContainer
attribute.
A TTML presentation processor supports the
#timeContainer
feature if it implements presentation semantic support
for the timeContainer
attribute.
A TTML transformation processor supports the
#time-clock
feature if it recognizes and is capable of
transforming all values of the <timeExpression> that
satisfy the following subset of time expression syntax:
<timeExpression> : hours ":" minutes ":" seconds ( fraction )?
A TTML presentation processor supports the
#time-clock
feature if it implements presentation
semantic support for the same syntax specified above.
A TTML transformation processor supports the
#time-clock-with-frames
feature if it supports the
#frameRate
,
#frameRateMultiplier
, and
#subFrameRate
features
and if it recognizes and is capable of
transforming all values of the <timeExpression> that
satisfy the following subset of time expression syntax:
<timeExpression> : hours ":" minutes ":" seconds ( fraction | ":" frames ( "." sub-frames )? )?
A TTML presentation processor supports the
#time-clock-with-frames
feature if it implements
presentation semantic support for the same features and syntax
specified above.
A TTML transformation processor supports the
#time-offset
feature if it recognizes and is capable of
transforming all values of the <timeExpression> that
satisfy the following subset of time expression syntax:
<timeExpression> : time-count fraction? ( "h" | "m" | "s" | "ms" )
A TTML presentation processor supports the
#time-offset
feature if it implements presentation
semantic support for the same syntax specified above.
A TTML transformation processor supports the
#time-offset-with-frames
feature if it supports the
#frameRate
,
#frameRateMultiplier
, and
#subFrameRate
features
and if it recognizes and is capable of
transforming all values of the <timeExpression> that
satisfy the following subset of time expression syntax:
<timeExpression> : time-count fraction? "f"
A TTML presentation processor supports the
#time-offset-with-frames
feature if it implements
presentation semantic support for the same features and syntax
specified above.
A TTML transformation processor supports the
#time-offset-with-ticks
feature if it supports the
#tickRate
feature
and if it recognizes and is capable of
transforming all values of the <timeExpression> that
satisfy the following subset of time expression syntax:
<timeExpression> : time-count fraction? "t"
A TTML presentation processor supports the
#time-offset-with-ticks
feature if it implements presentation
semantic support for the same features and syntax specified above.
A TTML transformation processor supports the #timing
feature if it recognizes and is capable of transforming the following
vocabulary defined by 10 Timing:
A TTML presentation processor supports the #timing
feature if it implements presentation semantic support for the
same vocabulary enumerated above.
A TTML processor supports the
#transformation
feature if it
(1) satisfies the generic processor criteria defined by
3.2.1 Generic Processor Conformance and
(2) implements the transformation semantics of the following features:
A TTML transformation processor supports the
#unicodeBidi
feature if it recognizes and is capable of
transforming all defined values of the tts:unicodeBidi
attribute.
A TTML presentation processor supports the
#unicodeBidi
feature if it implements presentation semantic support
for all defined values of the tts:unicodeBidi
attribute.
A TTML transformation processor supports the
#visibility
feature if it recognizes and is capable
of transforming all defined values of the tts:visibility
attribute.
A TTML presentation processor supports the
#visibility
feature if it implements presentation semantic support
for all defined values of the tts:visibility
attribute.
A TTML transformation processor supports the
#visibility-block
feature if it recognizes and is
capable of transforming all defined values of the tts:visibility
attribute when applied to a content element that would generate a
block area during presentation processing.
A TTML presentation processor supports the
#visibility-block
feature if it implements presentation semantic
support for all defined values of the tts:visibility
attribute when applied to a content element that generates a
block area.
A TTML transformation processor supports the
#visibility-inline
feature if it recognizes and is
capable of transforming all defined values of the tts:visibility
attribute when applied to a content element that would generate an
inline area during presentation processing.
A TTML presentation processor supports the
#visibility-inline
feature if it implements presentation semantic
support for all defined values of the tts:visibility
attribute when applied to a content element that generates an
inline area.
A TTML transformation processor supports the
#visibility-region
feature if it recognizes and is
capable of transforming all defined values of the tts:visibility
attribute when applied to a region
element.
A TTML presentation processor supports the
#visibility-region
feature if it implements presentation semantic
support for all defined values of the tts:visibility
attribute when applied to a region
element.
A TTML transformation processor supports the
#wrapOption
feature if it recognizes and is capable
of transforming all defined values of the tts:wrapOption
attribute.
A TTML presentation processor supports the
#wrapOption
feature if it implements presentation semantic support
for all defined values of the tts:wrapOption
attribute.
A TTML processor supports the #writingMode
feature if it
supports the following features:
A TTML transformation processor supports the
#writingMode-vertical
feature if it recognizes and is
capable of transforming the tbrl
, tblr
, and
tb
values of the tts:writingMode
attribute.
A TTML presentation processor supports the
#writingMode-vertical
feature if it implements presentation semantic
support for the tbrl
, tblr
, and
tb
values of the tts:writingMode
attribute.
A TTML processor supports the #writingMode-horizontal
feature if it
supports the following features:
A TTML transformation processor supports the
#writingMode-horizontal
feature if it recognizes and is
capable of transforming the lrtb
and lr
values of the tts:writingMode
attribute.
A TTML presentation processor supports the
#writingMode-horizontal-lr
feature if it implements presentation semantic
support for the lrtb
and lr
values of the
tts:writingMode
attribute.
A TTML transformation processor supports the
#writingMode-horizontal
feature if it recognizes and is
capable of transforming the rltb
and rl
values of the tts:writingMode
attribute.
A TTML presentation processor supports the
#writingMode-horizontal-rl
feature if it implements presentation semantic
support for the rltb
and rl
values of the
tts:writingMode
attribute.
A TTML transformation processor supports the
#zIndex
feature if it recognizes and is capable of
transforming the tts:zIndex
attribute.
A TTML presentation processor supports the
#zIndex
feature if it implements presentation semantic support for
the tts:zIndex
attribute.
The following table, Table D-1 – Feature Support, enumerates every defined feature designation (expressed without the TT Feature Namespace), and, for each designated feature, specifies whether the feature must be implemented, i.e., is mandatory (M), or may be implemented, i.e., is optional (O), for transformation and presentation processors.
For the sake of convenience, the following table, Table D-2 – Mandatory Features - Transformation, enumerates all mandatory features for a TTML Transformation Processor, providing additional comments to summarize the context of usage or the nature of the feature. The Profile Definition Document that defines the corresponding DFXP Transformation Profile is specified in F.1 DFXP Transformation Profile.
Feature | Comments |
#content |
body ,
div ,
p ,
span ,
br
|
#core |
@xml:id ,
@xml:lang ,
@xml:space
|
#profile | |
#structure |
tt ,
head
|
#time-offset | |
#timing |
@begin ,
@dur ,
@end
|
#transformation |
For the sake of convenience, the following table, Table D-3 – Mandatory Features - Presentation, enumerates all mandatory features for a TTML Presentation Processor, providing additional comments to summarize the context of usage or the nature of the feature. The Profile Definition Document that defines the corresponding DFXP Presentation Profile is specified in F.2 DFXP Presentation Profile.
Feature | Comments |
#content |
body ,
div ,
p ,
span ,
br
|
#core |
@xml:id ,
@xml:lang ,
@xml:space
|
#profile | |
#presentation | |
#structure |
tt ,
head
|
#time-offset | |
#timing |
@begin ,
@dur ,
@end
|
This appendix is normative.
This appendix specifies the syntactic form of extension designations, which are used to express authorial intent regarding the support for extension mechanisms in a TTML processor.
An extension designation is expressed as a string that adheres to the following form:
extension-designation : extension-namespace designation extension-namespace : TT Extension Namespace // http://www.w3.org/ns/ttml/extension/ | Other Extension Namespace // expressed as an absolute URI designation : "#" token-char+ token-char : { XML NameChar } // XML 1.1 Production [4a]
If the extension namespace of an extension designation is the TT
Extension Namespace, then all values of the following
designation
token are reserved for future
standardization.
If the extension namespace of an extension designation is not the
TT Extension Namespace, i.e., is an Other Extension Namespace,
then the extension namespace must be expressed as an absolute URI
capable of serving as a base URI used in combination with a
designation
token that takes the form of a fragment
identifier.
This appendix is normative.
This appendix specifies the following standard TTML profiles:
The SDP US profile is defined in TTML Simple Delivery Profile for Closed Captions (US).
Each TTML profile is defined in terms of a Profile Definition Document, which is expressed as an XML document wherein the root element adheres to 6.1.1 ttp:profile.
The DFXP Transformation Profile is intended to be used to express minimum compliance for transformation processing.
<?xml version="1.0" encoding="utf-8"?> <!-- this file defines the "dfxp-transformation" profile of ttml --> <profile xmlns="http://www.w3.org/ns/ttml#parameter"> <features xml:base="http://www.w3.org/ns/ttml/feature/"> <!-- required (mandatory) feature support --> <feature value="required">#content</feature> <feature value="required">#core</feature> <feature value="required">#profile</feature> <feature value="required">#structure</feature> <feature value="required">#time-offset</feature> <feature value="required">#timing</feature> <feature value="required">#transformation</feature> <!-- optional (voluntary) feature support --> <feature value="optional">#animation</feature> <feature value="optional">#backgroundColor-block</feature> <feature value="optional">#backgroundColor-inline</feature> <feature value="optional">#backgroundColor-region</feature> <feature value="optional">#backgroundColor</feature> <feature value="optional">#bidi</feature> <feature value="optional">#cellResolution</feature> <feature value="optional">#clockMode</feature> <feature value="optional">#clockMode-gps</feature> <feature value="optional">#clockMode-local</feature> <feature value="optional">#clockMode-utc</feature> <feature value="optional">#color</feature> <feature value="optional">#direction</feature> <feature value="optional">#display</feature> <feature value="optional">#display-block</feature> <feature value="optional">#display-inline</feature> <feature value="optional">#display-region</feature> <feature value="optional">#displayAlign</feature> <feature value="optional">#dropMode</feature> <feature value="optional">#dropMode-dropNTSC</feature> <feature value="optional">#dropMode-dropPAL</feature> <feature value="optional">#dropMode-nonDrop</feature> <feature value="optional">#extent</feature> <feature value="optional">#extent-region</feature> <feature value="optional">#extent-root</feature> <feature value="optional">#fontFamily</feature> <feature value="optional">#fontFamily-generic</feature> <feature value="optional">#fontFamily-non-generic</feature> <feature value="optional">#fontSize</feature> <feature value="optional">#fontSize-anamorphic</feature> <feature value="optional">#fontSize-isomorphic</feature> <feature value="optional">#fontStyle</feature> <feature value="optional">#fontStyle-italic</feature> <feature value="optional">#fontStyle-oblique</feature> <feature value="optional">#fontWeight</feature> <feature value="optional">#fontWeight-bold</feature> <feature value="optional">#frameRate</feature> <feature value="optional">#frameRateMultiplier</feature> <feature value="optional">#layout</feature> <feature value="optional">#length</feature> <feature value="optional">#length-cell</feature> <feature value="optional">#length-em</feature> <feature value="optional">#length-integer</feature> <feature value="optional">#length-negative</feature> <feature value="optional">#length-percentage</feature> <feature value="optional">#length-pixel</feature> <feature value="optional">#length-positive</feature> <feature value="optional">#length-real</feature> <feature value="optional">#lineBreak-uax14</feature> <feature value="optional">#lineHeight</feature> <feature value="optional">#markerMode</feature> <feature value="optional">#markerMode-continuous</feature> <feature value="optional">#markerMode-discontinuous</feature> <feature value="optional">#metadata</feature> <feature value="optional">#nested-div</feature> <feature value="optional">#nested-span</feature> <feature value="optional">#opacity</feature> <feature value="optional">#origin</feature> <feature value="optional">#overflow</feature> <feature value="optional">#overflow-visible</feature> <feature value="optional">#padding</feature> <feature value="optional">#padding-1</feature> <feature value="optional">#padding-2</feature> <feature value="optional">#padding-3</feature> <feature value="optional">#padding-4</feature> <feature value="optional">#pixelAspectRatio</feature> <feature value="optional">#presentation</feature> <feature value="optional">#showBackground</feature> <feature value="optional">#styling</feature> <feature value="optional">#styling-chained</feature> <feature value="optional">#styling-inheritance-content</feature> <feature value="optional">#styling-inheritance-region</feature> <feature value="optional">#styling-inline</feature> <feature value="optional">#styling-nested</feature> <feature value="optional">#styling-referential</feature> <feature value="optional">#subFrameRate</feature> <feature value="optional">#textAlign</feature> <feature value="optional">#textAlign-absolute</feature> <feature value="optional">#textAlign-relative</feature> <feature value="optional">#textDecoration</feature> <feature value="optional">#textDecoration-over</feature> <feature value="optional">#textDecoration-through</feature> <feature value="optional">#textDecoration-under</feature> <feature value="optional">#textOutline</feature> <feature value="optional">#textOutline-blurred</feature> <feature value="optional">#textOutline-unblurred</feature> <feature value="optional">#tickRate</feature> <feature value="optional">#time-clock-with-frames</feature> <feature value="optional">#time-clock</feature> <feature value="optional">#time-offset-with-frames</feature> <feature value="optional">#time-offset-with-ticks</feature> <feature value="optional">#timeBase-clock</feature> <feature value="optional">#timeBase-media</feature> <feature value="optional">#timeBase-smpte</feature> <feature value="optional">#timeContainer</feature> <feature value="optional">#unicodeBidi</feature> <feature value="optional">#visibility</feature> <feature value="optional">#visibility-block</feature> <feature value="optional">#visibility-inline</feature> <feature value="optional">#visibility-region</feature> <feature value="optional">#wrapOption</feature> <feature value="optional">#writingMode</feature> <feature value="optional">#writingMode-horizontal-lr</feature> <feature value="optional">#writingMode-horizontal-rl</feature> <feature value="optional">#writingMode-horizontal</feature> <feature value="optional">#writingMode-vertical</feature> <feature value="optional">#zIndex</feature> </features> <extensions xml:base="http://www.w3.org/ns/ttml/extension/"> <!-- required (mandatory) extension support --> <!-- optional (voluntary) extension support --> </extensions> </profile>
The DFXP Presentation Profile is intended to be used to express minimum compliance for presentation processing.
<?xml version="1.0" encoding="utf-8"?> <!-- this file defines the "dfxp-presentation" profile of ttml --> <profile xmlns="http://www.w3.org/ns/ttml#parameter"> <features xml:base="http://www.w3.org/ns/ttml/feature/"> <!-- required (mandatory) feature support --> <feature value="required">#content</feature> <feature value="required">#core</feature> <feature value="required">#presentation</feature> <feature value="required">#profile</feature> <feature value="required">#structure</feature> <feature value="required">#time-offset</feature> <feature value="required">#timing</feature> <!-- optional (voluntary) feature support --> <feature value="optional">#animation</feature> <feature value="optional">#backgroundColor</feature> <feature value="optional">#backgroundColor-block</feature> <feature value="optional">#backgroundColor-inline</feature> <feature value="optional">#backgroundColor-region</feature> <feature value="optional">#bidi</feature> <feature value="optional">#cellResolution</feature> <feature value="optional">#clockMode</feature> <feature value="optional">#clockMode-gps</feature> <feature value="optional">#clockMode-local</feature> <feature value="optional">#clockMode-utc</feature> <feature value="optional">#color</feature> <feature value="optional">#direction</feature> <feature value="optional">#display</feature> <feature value="optional">#display-block</feature> <feature value="optional">#display-inline</feature> <feature value="optional">#display-region</feature> <feature value="optional">#displayAlign</feature> <feature value="optional">#dropMode</feature> <feature value="optional">#dropMode-dropNTSC</feature> <feature value="optional">#dropMode-dropPAL</feature> <feature value="optional">#dropMode-nonDrop</feature> <feature value="optional">#extent</feature> <feature value="optional">#extent-region</feature> <feature value="optional">#extent-root</feature> <feature value="optional">#fontFamily</feature> <feature value="optional">#fontFamily-generic</feature> <feature value="optional">#fontFamily-non-generic</feature> <feature value="optional">#fontSize</feature> <feature value="optional">#fontSize-anamorphic</feature> <feature value="optional">#fontSize-isomorphic</feature> <feature value="optional">#fontStyle</feature> <feature value="optional">#fontStyle-italic</feature> <feature value="optional">#fontStyle-oblique</feature> <feature value="optional">#fontWeight</feature> <feature value="optional">#fontWeight-bold</feature> <feature value="optional">#frameRate</feature> <feature value="optional">#frameRateMultiplier</feature> <feature value="optional">#layout</feature> <feature value="optional">#length</feature> <feature value="optional">#length-cell</feature> <feature value="optional">#length-em</feature> <feature value="optional">#length-integer</feature> <feature value="optional">#length-negative</feature> <feature value="optional">#length-percentage</feature> <feature value="optional">#length-pixel</feature> <feature value="optional">#length-positive</feature> <feature value="optional">#length-real</feature> <feature value="optional">#lineBreak-uax14</feature> <feature value="optional">#lineHeight</feature> <feature value="optional">#markerMode</feature> <feature value="optional">#markerMode-continuous</feature> <feature value="optional">#markerMode-discontinuous</feature> <feature value="optional">#metadata</feature> <feature value="optional">#nested-div</feature> <feature value="optional">#nested-span</feature> <feature value="optional">#opacity</feature> <feature value="optional">#origin</feature> <feature value="optional">#overflow</feature> <feature value="optional">#overflow-visible</feature> <feature value="optional">#padding</feature> <feature value="optional">#padding-1</feature> <feature value="optional">#padding-2</feature> <feature value="optional">#padding-3</feature> <feature value="optional">#padding-4</feature> <feature value="optional">#pixelAspectRatio</feature> <feature value="optional">#showBackground</feature> <feature value="optional">#styling</feature> <feature value="optional">#styling-chained</feature> <feature value="optional">#styling-inheritance-content</feature> <feature value="optional">#styling-inheritance-region</feature> <feature value="optional">#styling-inline</feature> <feature value="optional">#styling-nested</feature> <feature value="optional">#styling-referential</feature> <feature value="optional">#subFrameRate</feature> <feature value="optional">#textAlign</feature> <feature value="optional">#textAlign-absolute</feature> <feature value="optional">#textAlign-relative</feature> <feature value="optional">#textDecoration</feature> <feature value="optional">#textDecoration-over</feature> <feature value="optional">#textDecoration-through</feature> <feature value="optional">#textDecoration-under</feature> <feature value="optional">#textOutline</feature> <feature value="optional">#textOutline-blurred</feature> <feature value="optional">#textOutline-unblurred</feature> <feature value="optional">#tickRate</feature> <feature value="optional">#time-clock</feature> <feature value="optional">#time-clock-with-frames</feature> <feature value="optional">#time-offset-with-frames</feature> <feature value="optional">#time-offset-with-ticks</feature> <feature value="optional">#timeBase-clock</feature> <feature value="optional">#timeBase-media</feature> <feature value="optional">#timeBase-smpte</feature> <feature value="optional">#timeContainer</feature> <feature value="optional">#transformation</feature> <feature value="optional">#unicodeBidi</feature> <feature value="optional">#visibility</feature> <feature value="optional">#visibility-block</feature> <feature value="optional">#visibility-inline</feature> <feature value="optional">#visibility-region</feature> <feature value="optional">#wrapOption</feature> <feature value="optional">#writingMode</feature> <feature value="optional">#writingMode-horizontal-lr</feature> <feature value="optional">#writingMode-horizontal-rl</feature> <feature value="optional">#writingMode-horizontal</feature> <feature value="optional">#writingMode-vertical</feature> <feature value="optional">#zIndex</feature> </features> <extensions xml:base="http://www.w3.org/ns/ttml/extension/"> <!-- required (mandatory) extension support --> <!-- optional (voluntary) extension support --> </extensions> </profile>
The DFXP Full Profile is intended to be used to express maximum compliance for both transformation and presentation processing.
<?xml version="1.0" encoding="utf-8"?> <!-- this file defines the "dfxp-full" profile of ttml --> <profile xmlns="http://www.w3.org/ns/ttml#parameter"> <features xml:base="http://www.w3.org/ns/ttml/feature/"> <!-- required (mandatory) feature support --> <feature value="required">#animation</feature> <feature value="required">#backgroundColor-block</feature> <feature value="required">#backgroundColor-inline</feature> <feature value="required">#backgroundColor-region</feature> <feature value="required">#backgroundColor</feature> <feature value="required">#bidi</feature> <feature value="required">#cellResolution</feature> <feature value="required">#clockMode-gps</feature> <feature value="required">#clockMode-local</feature> <feature value="required">#clockMode-utc</feature> <feature value="required">#clockMode</feature> <feature value="required">#color</feature> <feature value="required">#content</feature> <feature value="required">#core</feature> <feature value="required">#direction</feature> <feature value="required">#display</feature> <feature value="required">#display-block</feature> <feature value="required">#display-inline</feature> <feature value="required">#display-region</feature> <feature value="required">#displayAlign</feature> <feature value="required">#dropMode</feature> <feature value="required">#dropMode-dropNTSC</feature> <feature value="required">#dropMode-dropPAL</feature> <feature value="required">#dropMode-nonDrop</feature> <feature value="required">#extent</feature> <feature value="required">#extent-region</feature> <feature value="required">#extent-root</feature> <feature value="required">#fontFamily</feature> <feature value="required">#fontFamily-generic</feature> <feature value="required">#fontFamily-non-generic</feature> <feature value="required">#fontSize</feature> <feature value="required">#fontSize-anamorphic</feature> <feature value="required">#fontSize-isomorphic</feature> <feature value="required">#fontStyle</feature> <feature value="required">#fontStyle-italic</feature> <feature value="required">#fontStyle-oblique</feature> <feature value="required">#fontWeight</feature> <feature value="required">#fontWeight-bold</feature> <feature value="required">#frameRate</feature> <feature value="required">#frameRateMultiplier</feature> <feature value="required">#layout</feature> <feature value="required">#length</feature> <feature value="required">#length-cell</feature> <feature value="required">#length-em</feature> <feature value="required">#length-integer</feature> <feature value="required">#length-negative</feature> <feature value="required">#length-percentage</feature> <feature value="required">#length-pixel</feature> <feature value="required">#length-positive</feature> <feature value="required">#length-real</feature> <feature value="required">#lineBreak-uax14</feature> <feature value="required">#lineHeight</feature> <feature value="required">#markerMode</feature> <feature value="required">#markerMode-continuous</feature> <feature value="required">#markerMode-discontinuous</feature> <feature value="required">#metadata</feature> <feature value="required">#nested-div</feature> <feature value="required">#nested-span</feature> <feature value="required">#opacity</feature> <feature value="required">#origin</feature> <feature value="required">#overflow</feature> <feature value="required">#overflow-visible</feature> <feature value="required">#padding</feature> <feature value="required">#padding-1</feature> <feature value="required">#padding-2</feature> <feature value="required">#padding-3</feature> <feature value="required">#padding-4</feature> <feature value="required">#pixelAspectRatio</feature> <feature value="required">#presentation</feature> <feature value="required">#profile</feature> <feature value="required">#showBackground</feature> <feature value="required">#structure</feature> <feature value="required">#styling</feature> <feature value="required">#styling-chained</feature> <feature value="required">#styling-inheritance-content</feature> <feature value="required">#styling-inheritance-region</feature> <feature value="required">#styling-inline</feature> <feature value="required">#styling-nested</feature> <feature value="required">#styling-referential</feature> <feature value="required">#subFrameRate</feature> <feature value="required">#textAlign</feature> <feature value="required">#textAlign-absolute</feature> <feature value="required">#textAlign-relative</feature> <feature value="required">#textDecoration</feature> <feature value="required">#textDecoration-over</feature> <feature value="required">#textDecoration-through</feature> <feature value="required">#textDecoration-under</feature> <feature value="required">#textOutline</feature> <feature value="required">#textOutline-blurred</feature> <feature value="required">#textOutline-unblurred</feature> <feature value="required">#tickRate</feature> <feature value="required">#time-clock</feature> <feature value="required">#time-clock-with-frames</feature> <feature value="required">#time-offset</feature> <feature value="required">#time-offset-with-frames</feature> <feature value="required">#time-offset-with-ticks</feature> <feature value="required">#timeBase-clock</feature> <feature value="required">#timeBase-media</feature> <feature value="required">#timeBase-smpte</feature> <feature value="required">#timeContainer</feature> <feature value="required">#timing</feature> <feature value="required">#transformation</feature> <feature value="required">#unicodeBidi</feature> <feature value="required">#visibility</feature> <feature value="required">#visibility-block</feature> <feature value="required">#visibility-inline</feature> <feature value="required">#visibility-region</feature> <feature value="required">#wrapOption</feature> <feature value="required">#writingMode-horizontal</feature> <feature value="required">#writingMode-horizontal-lr</feature> <feature value="required">#writingMode-horizontal-rl</feature> <feature value="required">#writingMode-vertical</feature> <feature value="required">#writingMode</feature> <feature value="required">#zIndex</feature> <!-- optional (voluntary) feature support --> </features> <extensions xml:base="http://www.w3.org/ns/ttml/extension/"> <!-- required (mandatory) extension support --> <!-- optional (voluntary) extension support --> </extensions> </profile>
This appendix is normative.
The Timed Text Markup Language (TTML), also known as the Distribution Format Exchange Profile (DFXP) of the Timed Text Authoring Format (TTAF), satisfies a subset of the requirements established by [TTAF1-REQ]. The following table enumerates these requirements and indicates the extent to which they are satisfied by this specification, where S denotes a requirement is satisfied, P denotes a requirement is partially satisfied, and N denotes a requirement is not satisfied.
ID | Name | Status | Comments |
---|---|---|---|
R100 | Specification Format | S | |
R101 | Specification Modularity | S | |
R102 | Specification Organization | S | |
R103 | Core and Periphery | S | TT extension namespaces |
R104 | Evolution of Core | S | TT extension namespaces |
R105 | Ownership of Core | S | TT namespaces |
R106 | Surjection of Core | S | |
R107 | Evolution of Periphery | S | TT extension namespaces |
R108 | Ownership of Periphery | S | Non-TT namespaces |
R109 | Transformation | S | Supports 3GPP, QText, RealText, SAMI |
R110 | Streamable Transformation | S | Progressive decoding |
R111 | Accessibility – Content | S | Alternative document instances |
R112 | Accessibility – Authoring System | S | |
R200 | Authorability | S | |
R201 | Multiple Natural Languages | S | Alternative document instances |
R202 | Natural Language Coverage | S | Unicode 4.0 |
R203 | Natural Language Association Granularity | S | See xml:lang |
R204 | Minimum Character Representability | S | Unicode 4.0 |
R205 | Intrinsic and Extrinsic Text Content | P | Intrinsic only |
R206 | Markup Association | P | Intrinsic only |
R207 | Conditional Content | N | |
R208 | Flowed Text | S | |
R209 | Logical Flowed Text Vocabulary | S | |
R210 | Presentational Flowed Text Vocabulary | S | Implied mapping from logical flowed text. |
R211 | Flowed Text Vocabulary Relationship | S | |
R212 | Flowed Text Vocabulary Separation | N | |
R213 | Non-Flowed Text | N | |
R214 | Non-Flowed Text Vocabulary | N | |
R215 | Hybrid Flowed and Non-Flowed Text | N | |
R216 | Hyperlinking | N | Can support via XLink |
R217 | Embedded Graphics | N | |
R218 | Non-Embedded Graphics | N | |
R219 | Embedded Fonts | N | |
R220 | Non-Embedded Fonts | N | |
R221 | Descriptive Vocabulary | S | See ttm:agent , ttm:role |
R222 | Embedded Audio | N | |
R223 | Non-Embedded Audio | N | |
R290 | Markup Format | S | |
R291 | Markup Format and Unicode Interaction | S | |
R292 | Extrinsic Resource References | N | No extrinsic references |
R293 | Schema Validity Specification | S | |
R300 | Inline Styling | S | |
R301 | Inline Styling Form | P | Inline and referential styling |
R301 | Out-of-Line Styling | N | |
R301 | Out-of-Line Styling Form | N | |
R304 | Styling Prioritization | S | |
R305 | Style Parameters – Aural | N | |
R306 | Style Parameters – Visual | P | Supports absolute position, background color, color, display none, display alignment, font family, font size, font style, font weight, height, line height, origin, opacity, overflow, padding (before, after, start, end), text alignment, text shadow (as outline), visibility, width, writing mode, z-index |
R390 | Style Parameter Symmetry | S | |
R391 | Style Parameter Definitions | S | |
R392 | Style Parameter Shorthands | S | |
R401 | Inline Timing | S | |
R402 | Out-of-Line Timing | N | |
R403 | Synchronization Parameters | P | Supports begin, end, dur |
R404 | Synchronization Parameter Value Spaces | P | Supports offset values, media marker values (SMPTE 12M), wall-clock values |
R405 | Time Containment Semantics | P | Supports sequential, parallel |
R500 | Animation Modes | P | Supports discrete |
R502 | Highlight Animation | S | <set tts:backgroundColor="..."/> |
R503 | Fade Transition Animation | S | <set tts:opacity="..."/> |
R504 | Animated Style Parameters – Aural | N | |
R505 | Animated Style Parameters – Visual | P | Supports animating background color, color, display, opacity, origin, visibility |
N506 | Animated Content | S | |
R600 | Metadata Item Association | S | See metadata , Metadata.class |
R601 | Metadata Item Constituents | P | Supports name, value |
R602 | Metadata Item Value Representation | P | See metadata |
R603 | Metadata Item Extensibility | S | See metadata |
R604 | Metadata Item Validation | S | See metadata |
R690 | Dublin Core Preference | N | Uses ttm:copyright , ttm:desc , ttm:title |
This appendix provides information about the derivation of TTML vocabulary, separately describing derivation of elements and attributes.
The first column of Table J-1 – Elements specifies a TTML element vocabulary item; the second column specifies the syntactic and/or semantic model on which the vocabulary item is based; the third column specifies the reference that defines the model (if a model is indicated); the fourth column specifies details about the derivation; the last column refers to additional notes describing the nature of the derivation.
In the fourth column, which describes details of derivation, a
notation is use to indicate the addition or removal of an attribute.
For example, in the derivation of the tt:div
element, the
details column includes "-@class", which denotes that the
class
attribute that is specified for use with the
xhtml:div
model element is not specified for use with the
corresponding TTML element; in contrast, the details column includes
"+@begin", which denotes that a begin
attribute is added
that is not specified for use with the xhtml:div
model
element.
Element | Model | Reference | Details | Notes |
tt:body | xhtml:body | [XHTML 1.0] | -@class, -@dir, -@lang, -@on*, -@title; +@begin, +@dur, +@end, +@region, +@timeContainer, +@ttm:*, +@tts:*; content model subsetted to zero or more division (div) children, and supersetted by optional metadata and animation children | 1,2 |
tt:br | xhtml:br | [XHTML 1.0] | -@class, -@title; +@ttm:*, +@tts:*, +@xml:lang, +@xml:space; content model supersetted by optional metadata and animation children for congruity with other content vocabulary | 1,2 |
tt:div | xhtml:div | [XHTML 1.0] | -@class, -@dir, -@lang, -@on*, -@title; +@begin, +@dur, +@end, +@region, +@timeContainer, +@ttm:*, +@tts:*, +@xml:space; content model subsetted to zero or more paragraph (p) children, and supersetted by optional metadata and animation children | 1,2,3 |
tt:head | xhtml:head | [XHTML 1.0] | -@dir, -@lang, -@profile; +@id, +@xml:space; content model changed to optional metadata children, followed by optional styling child, followed by optional layout child | 1,3 |
tt:layout | fo:simple-page-master | [XSL 1.1] | conceptual derivation | 4 |
tt:metadata | svg:metadata | [SVG 1.1] | -@xml:base; +@ttm:*, +@xml:lang, +@xml:space; content model subsetted to foreign namespace element content only (no #PCDATA) | 3,5 |
tt:p | xhtml:p | [XHTML 1.0] | -@class, -@dir, -@lang, -@on*, -@title; +@begin, +@dur, +@end, +@region, +@timeContainer, +@ttm:*, +@tts:*, +@xml:space; content model subsetted to zero or more span children, and supersetted by optional metadata and animation children | 1,2,3 |
tt:region | fo:region-* | [XSL 1.1] | conceptual derivation | 4 |
tt:set | svg:set | [SVG 1.1] | -@* except begin, dur, end; +@tts:*, +@xml:lang, +@xml:space | 3,6 |
tt:span | xhtml:span | [XHTML 1.0] | -@class, -@dir, -@lang, -@on*, -@title; +@begin, +@dur, +@end, +@region, +@timeContainer, +@ttm:*, +@tts:*, +@xml:space; content model subsetted to zero or more #PCDATA or break (br) children, and supersetted by optional metadata and animation children | 1,2,3 |
tt:style | style specification | [CSS2] | XML representation of identified set of pairs of style property name and value, with optional inclusion of other styles by reference to other style elements | 7 |
tt:styling | xhtml:style | [XHTML 1.0] | XML representation of a set of style specifications sets, each represented by a style child element | 1,7 |
tt:tt | xhtml:html | [XHTML 1.0] | -@dir, -@lang; +@id, +@ttp:*, +@xml:space; content model subsetted by permitting body and/or head to be optional | 1,8 |
ttm:actor | mpeg7:Creator | [MPEG7-5] | conceptual derivation | 4 |
ttm:agent | mpeg7:Agent | [MPEG7-5] | conceptual derivation | 4 |
ttm:copyright | mpeg7:CopyrightString | [MPEG7-5] | conceptual derivation | 4 |
ttm:desc | svg:desc | [SVG 1.1] | -@class, -@style, -@xml:base | 2,5,9 |
ttm:name | mpeg7:Name | [MPEG7-5] | conceptual derivation | 4 |
ttm:title | svg:title | [SVG 1.1] | -@class, -@style, -@xml:base | 2,5,9 |
ttp:extension | @requiredExtensions | [SVG 1.1] | conceptual derivation | 10 |
ttp:extensions | @requiredExtensions | [SVG 1.1] | conceptual derivation | 10 |
ttp:feature | @requiredFeatures | [SVG 1.1] | conceptual derivation | 10 |
ttp:features | @requiredFeatures | [SVG 1.1] | conceptual derivation | 10 |
ttp:profile | @baseProfile | [SVG 1.1] | conceptual derivation | 11 |
Note:
Derivation is indicated with respect to the strict DTD defined by [XHTML 1.0], §A.1.
The class
attribute is effectively replaced by the
style
attribute, which, instead of specifying an inline style,
refers indirectly to one or more style
elements that define a set of style specification sets.
The xml:lang
and xml:space
attributes are defined for all element
types in order to support their inheritance semantics to operate in
the context of foreign namespace elements.
Derivation is conceptual (notional) only.
The xml:base
attribute is not used since there are no external
references from core vocabulary.
The attributeName
and to
attributes of
svg:set
are replaced by the
direct expression of the target attribute name and value by use of a
tts:*
attribute.
CSS style specification syntax is mapped to XML by use of attributes defined in the TT Style Namespace.
The xml:id
attribute is defined for use on all element types.
The style
attribute is supported only on
Content elements.
Derived from the use of @requiredExtensions
and @requiredFeatures
on the svg:svg
element,
but extended to support distinct specification of optionality.
Derived from the use of @baseProfile
and @version
on the svg:svg
element.
The first column of Table J-2 – Attributes specifies a TTML attribute vocabulary item; the second column specifies the syntactic and/or semantic model on which the vocabulary item is based; the third column specifies the reference that defines the model (if a model is indicated); the fourth column specifies details about the derivation; the last column refers to additional notes describing the nature of the derivation.
In the fourth column, which describes details of derivation, a
notation is use to indicate the addition or removal of an attribute
value. For example, in the derivation of the timeContainer
attribute,
the details column includes "-excl", which denotes that the
excl
value that is specified for use with the
timeContainer
model attribute is not specified for use with the
corresponding TTML attribute; similarly, an "+value"
in the details column indicates that the attribute's values have been
extended to include value.
Only those attributes that are specified for use on more than one TTML element type are listed below. Those per-element namespace attributes that are uniquely defined for a specific TTML element type are not listed below, but are considered to be part of the specific element type's derivation described in Table J-1 – Elements above.
Attribute | Model | Reference | Details | Notes |
begin | begin | [SMIL 2.1] | see notes | 2,3,4 |
dur | dur | [SMIL 2.1] | see notes | 2,3,4 |
end | end | [SMIL 2.1] | see notes | 2,3,4 |
region | master-reference | [XSL 1.1] | conceptual derivation | |
style | class | [CSS2] | dereferences style specification(s) directly | |
timeContainer | timeContainer | [SMIL 2.1] | -excl, -none; no default attribute value | 5 |
ttm:agent | none | used to attribute agent of content | ||
ttm:role | none | used to attribute role of content | ||
ttp:cellResolution | none | expresses uniform grid resolution for cell based coordinates | ||
ttp:clockMode | none | determines how to interpret time expressions | ||
ttp:frameRate | none | expresses integral frame rate | ||
ttp:frameRateMultiplier | none | used to express non-integral, rational frame rates | ||
ttp:markerMode | none | expresses marker continuity semantics | ||
ttp:pixelAspectRatio | none | expresses pixel aspect ratio of related media | ||
ttp:profile | none | expresses profile of TTML used by a Document Instance | ||
ttp:dropMode | none | expresses frame counting (drop) modes | ||
ttp:subFrameRate | none | expresses sub-frame rate | ||
ttp:tickRate | none | used to interpret tick based time expressions | ||
ttp:timeBase | none | used to interpret semantics of time expressions | ||
tts:backgroundColor | background-color | [XSL 1.1] | -inherit | 1,6 |
tts:color | color | [XSL 1.1] | -inherit | 6 |
tts:direction | direction | [XSL 1.1] | -inherit | |
tts:display | display | [CSS2] | only auto , none | |
tts:displayAlign | display-align | [XSL 1.1] | -inherit | 1 |
tts:extent | width , height | [XSL 1.1] | shorthand property | |
tts:fontFamily | font-family | [XSL 1.1] | -inherit , extends generic family names | 1 |
tts:fontSize | font-size | [XSL 1.1] | -inherit | 1,7 |
tts:fontStyle | font-style | [XSL 1.1] | -inherit , -backslant | 1 |
tts:fontWeight | font-weight | [XSL 1.1] | -inherit , -bolder , -lighter , -<number> | 1 |
tts:lineHeight | line-height | [XSL 1.1] | -inherit , -<number> , -<space> | 1 |
tts:opacity | opacity | [CSS3 Color] | -inherit | |
tts:origin | top , left | [XSL 1.1] | shorthand property | |
tts:overflow | overflow | [XSL 1.1] | -inherit , -auto , -error-if-overflow | |
tts:padding | padding | [XSL 1.1] | -inherit | 8 |
tts:showBackground | showBackground | [SMIL 2.1] | -inherit | |
tts:textAlign | text-align | [XSL 1.1] | -inherit | 1 |
tts:textDecoration | text-decoration | [XSL 1.1] | -inherit | 1,9,12 |
tts:textOutline | text-shadow | [XSL 1.1] | -inherit | 10,12 |
tts:unicodeBidi | unicode-bidi | [XSL 1.1] | -inherit | 1 |
tts:visibility | visibility | [XSL 1.1] | -inherit , -collapse | |
tts:wrapOption | wrap-option | [XSL 1.1] | -inherit | 1 |
tts:writingMode | writing-mode | [XSL 1.1] | -inherit , +tblr | 1 |
tts:zIndex | z-index | [XSL 1.1] | -inherit | 1 |
xml:id | xml:id | [XML ID] | complies with model | |
xml:lang | xml:lang | [XML 1.0] | complies with model | |
xml:space | xml:space | [XML 1.0] | see notes | 11 |
Note:
Attribute name and/or value(s) are normalized to use lowerCamelCase naming convention.
Restricted to expressing a clock value that denotes one of the
following in accordance to whether the parameter expressed by the
ttp:timeBase
attribute is media
, smpte
,
or clock
, respectively:
(1) an offset from an implicit syncbase that is linked to a media time
line, (2) an event time
that represents the occurrence of an implicit media marker, or (3) a wall-clock
time.
Syntactically subsets and supersets the [SMIL 2.1] Clock-value
syntax as follows:
(1) requires non-negative Full-clock-value
or
Timecount-value
;
(2) if Full-clock-value
then hours must be two or more digits;
(3) if Timecount-value
, then metric must be
specified;
(4) uses m
as alias for min
metric to denote
minutes;
(5) adds f
and t
metrics denoting frames and
ticks, respectively;
(6) adds alternative expression of optional Fraction
in
Full-clock-value
by specifying frame count or frame
count with subframe count.
Interpretation of time expression is further constrained by
parameters expressed by
ttp:clockMode
,
ttp:dropMode
,
ttp:frameRate
,
ttp:frameRateMultiplier
,
ttp:markerMode
,
ttp:subFrameRate
,
ttp:tickRate
, and
ttp:timeBase
attributes.
Uses subset of named colors from model to which two aliases are
added as follows: magenta
as fuchsia
, and cyan
as aqua
.
If not specified, then parallel (par) container semantics apply to the element types specified by 10.2.4 timeContainer.
Restricts size to length specification which can be a percentage; adds optional second length (or percentage) for specifying separate horizontal and vertical scaling of glyph's EM square.
Expressed in terms of writing mode relative padding properties rather than absolute padding properties.
Excludes blink
and no-blink
values.
Uses only one length specification instead of two, where one length defines distance of outline effect from nominal edge of glyph contour outline perpendicular to point of glyph contour. Percentage lengths are also added to express outline effect in relative to font size. Outline effects are intended to be drawn both outside of outer closed contours and inside of inner closed contours.
On root element, default attribute value specified as
default
, which is defined in terms of whitespace
normalization. Semantics of preservation and default normalization are
defined in terms of presentation semantics by 7.2.3 xml:space.
Defined to be inheritable.
This appendix specifies the compliance of this specification with the requirements and guidelines defined by QA Framework Specifications Guidelines [QAF SG].
Note:
No feature is deprecated by this version of this specification.
Note:
When making normative references to external specifications, specific clauses or sections are cited.
See also J Vocabulary Derivation.
Test assertions and test suites will be provided prior to entering Proposed Recommendation (PR) phase.
See criterion #3 in 3.2 Processor Conformance and definition of TTML Abstract Document Instance.
No feature is deprecated or obsoleted by this version of this specification.
TTML Content can be authored to meet the following characteristics that may be useful in streaming scenarios:
can be progressively encoded (i.e., does not require computing subsequent data prior to sending current data);
can be progressively decoded (if it does not require forward references, but uses only reverse references when necessary);
does not require dereferencing (and subsequent loading) of any resources other than TTML Content;
has timing structure compiled into inline format that makes possible a temporal ordering of content that follows temporal presentation order;
One possible means by which TTML Content may be streamed is to partition a Document Instance's information set into non-overlapping fragments, where one particular fragment, call it the root fragment, represents the front matter (head) of the Document Instance as well as its top level structural elements, and other fragments represent content whose time intervals are expected to be active in parallel.
In applications that require arbitrary (random) entry into a stream, i.e., the property of being able to start reading data at an arbitrary data access unit, the root fragment will be repetitively transmitted (inserted) into the stream in order to permit a decoder to resynchronize and acquire sufficient structural information in the information set in order to interpret subsequent content fragments.
An example of such a fragmentation of a Document Instance is shown in Figure 3 – Fragment Streaming.
Note:
This specification does not define a transport buffer model or a decoder capabilities model.
Another means by which TTML Content may be streamed is to partition a Document Instance's information set into temporally bound fragments, each of which is itself a document instance that contains all the front matter (head) and content required to present it, where the temporal interval for each fragment is constrained, either by the Document Interchange Context or by the Document Processing Context or by both, such that temporal overlaps with other fragments are resolved.
For example, some document processing contexts resolve temporal overlap by specifying that a maximum of one document instance can be active at any moment in the timeline, and additionally specify precedence rules to establish which document instance that is, in case there are multiple candidates.
In order to ensure that all the content is correctly displayed, any content that falls into an Intermediate Synchronic Document (ISD) whose interval overlaps with more than one fragmentation interval might need to be duplicated within each fragment that is required to generate that ISD. It is possible for implementations to identify adjacent ISDs that are identical except in their intervals and replace them with a single ISD covering the combined interval.
The intent of this approach is that presentation processors are able to generate an equivalent sequence of ISDs to that which would be generated by a non-streamed version of the document instance, while taking advantage of the ability to avoid duplicating some controllable amount of front matter, by varying the segment duration. Reconstruction of the original source is not guaranteed to result in an identical document instance.
Note:
This technique is used for example in [ISOBMFF] and in [EBU-TT Live].
An example of such a fragmentation of a Document Instance is shown in Figure 4 – Temporal fragmentation.
Note:
In the above example, a single document instance has content visible during intervals T1 through T6 and is fragmented into two document instances (segments), the first including all content and referenced styles and regions required during T1 through T3, the second including all content and referenced styles and regions required during T4 through T6.
In the absence of other requirements, it is recommended that a Document Instance be concretely encoded as a well-formed XML 1.0 [XML 1.0] document using the UTF-8 character encoding.
Note:
When using XML 1.0 [XML 1.0] as the concrete encoding of TTML, only the following named character entities are defined:
&
, '
, <
, >
, and "
.
This appendix describes the intended semantics for interpreting time expressions in Document Instances.
Note:
It is expected that the information in this appendix will be elevated to normative status in a future revision of this specification.
Note:
The phrase local real time as used below is intended to model a virtual real time clock in the document processing context, where local means in the immediate proximity of the implementation of this processing context. The intent of defining relationships with this virtual clock is to establish a locally valid physical realization of time for didactic purposes.
Note:
The phrase play rate as used below is intended to model a (possibly variable) parameter in the document processing context wherein the rate of playback (or interpretation) of time may artificially dilated or narrowed, for example, when slowing down or speeding up the rate of playback of a Related Media Object. Without loss of generality, the following discussion assumes a fixed play(back) rate. In the case of variable play rates, appropriate adjustments may need to be made to the resulting computations.
The Document Processing Context defines the applicable epoch and any epoch-related offsets to be used when establishing the Synthetic Document Syncbase.
When operating with the clock
time base, the following semantics should be applied for interpreting time expressions,
as defined by <timeExpression>, and their relationship to media time and local real time.
The clock time base C
is related to local real time R
expressed in the epoch E
(defined by the Document Processing Context) as follows:
|
Time value expressions, as denoted by a <timeExpression>, are related to
clock time C
as follows:
|
Note:
The frames and sub-frames terms and the frames (f
) metric of time expressions do not apply when using
the clock
time base.
The clock time base C
is independent of media time M
:
|
Note:
That is to say, timing is disconnected from (not necessarily proportional to) media time when
the clock
time base is used. For example, if the media play rate is zero (0), media playback is suspended; however, timing coordinates
will continue to advance according to the natural progression of clock time in direct proportion to the reference clock base. Furthermore, if the media play rate changes
during playback, presentation timing is not affected.
When operating with the media
time base, the following semantics should be applied for interpreting time expressions,
as defined by <timeExpression>, and their relationship to media time, document time, and local real time.
The media time base M
is related to local real time R
expressed in the epoch E
(defined by the Document Processing Context) as follows:
|
Time value expressions, as denoted by a <timeExpression>, are related to
media time M
in accordance to the
ttp:frameRate
, ttp:subFrameRate
, and ttp:frameRateMultiplier
parameters as follows:
|
Note:
If the computed frameRateMultiplier
ratio is not integral, then effectiveFrameRate
will be a non-integral rational.
Note:
The above formalisms assumes that the Root Temporal Extent corresponds with the beginning of a related media object. If
this assumption doesn't hold, then an additional offset that accounts for the difference may be introduced when computing media time M
.
The computed document times are used as the equivalent media times with no offset.
Note:
Any additional media time processing imposed by e.g. a wrapper format is out of scope of the TTML specification.
For example [ISOBMFF] and [MPEG DASH] provide mechanisms for wrapping TTML documents as samples or segments and defining the offset of their media timelines relative to other media such as audio or video.
The following document fragments are equivalent in timing:
<tt ttp:timeBase="media" ...> ... <body> <div xml:id="d1" begin="1s"> <p xml:id="p1" begin="0s" end="1s">First paragraph</p> <p xml:id="p2" begin="2s" end="3s">Second paragraph</p> </div> </body> </tt> |
<tt ttp:timeBase="media" ...> ... <body> <div xml:id="d1"> <p xml:id="p1" begin="1s" end="2s">First paragraph</p> <p xml:id="p2" begin="3s" end="4s">Second paragraph</p> </div> </body> </tt> |
In this example, using the the terminology of [SMIL 2.1], the implicit duration of the body element is identical to its computed duration which is also the Root Temporal Extent. |
However if the Document Processing Context specifies a range of applicable media times, those limit the resolved begin and end times and therefore the Root Temporal Extent.
In the following example the processing context defines that the media time range over which the document is active is from 1.5s to 3.5s, and this defines the Root Temporal Extent since timed content is present at both of those times.
The grey background indicates the periods when the Document Processing Context defines the document to be inactive. |
If the document processing context were to define that the media time range is 0-3s then the Root Temporal Extent would be from 1s to 2s as defined by p1 and p2 would never be shown. |
When operating with the smpte
time base, the following semantics should be applied for interpreting time expressions,
as defined by <timeExpression>, and their relationship to media time and local real time.
If the computed value of the ttp:markerMode
parameter is discontinuous
, then there is no direct relationship between
time expressions and media time M
or local real time R
. In
this case, time expressions refer to synchronization events (markers) emitted by the Document Processing Context when SMPTE Time
Codes are encountered in the Related Media Object.
Otherwise, if the computed value of the ttp:markerMode
parameter is continuous
, then the relationships between time expressions and
local real time and media time are as described below in terms of a Synthetic SMPTE Document Syncbase, here referred to as the
SMPTE time base S
.
|
The SMPTE time base S
is related to the media time base M
as follows:
|
Given the derived media time base as described above, then
media time base M
is related to the local real time R
as described in N.2 Media Time Base above.
This section provides examples of the following common caption styles using TTML Content to obtain the desired behavior:
Pop-On Captions
Roll-Up Captions
Paint-On Captions
An example of paint-on captions. In this example, two regions are targeted with alternating, paint-on content, where content is timed using explicit sequential time containment rules. Each paragraph is non-overlapping in time and appears in the same single row of its targeted region.
<tt ttp:cellResolution="60 20" xml:lang="en" xmlns="http://www.w3.org/ns/ttml" xmlns:ttp="http://www.w3.org/ns/ttml#parameter" xmlns:tts="http://www.w3.org/ns/ttml#styling"> <head> <layout> <region xml:id="r1" tts:color="white" tts:origin="10c 4c" tts:extent="40c 1c"/> <region xml:id="r2" tts:color="yellow" tts:origin="10c 8c" tts:extent="40c 1c"/> </layout> </head> <body> <div timeContainer="seq"> <p region="r1" dur="4s">Lorem ipsum dolor sit</p> <p region="r2" dur="4s">Amet consectetur adipiscing elit</p> <p region="r1" dur="6s">Sed do eiusmod tempor incididunt labore</p> <p region="r2" dur="4s">et dolore magna aliqua</p> <p region="r1" dur="7s">Ut enim ad minim veniam quis, nostrud</p> </div> </body> </tt> |
An example of roll-up captions. Roll-up effects are achieved by using overlapped time intervals, where zero, one, or two paragraphs appear in the region at a given time. Each paragraph consumes a single row (line) of the region since no wrapping occurs. Depending on whether a presentation processor supports smooth scrolling between adjacent synchronic intermediate document instances, the transitions, i.e., moving an old paragraph (line) out and a new paragraph (line) in, will be either smooth or discrete.
<tt ttp:cellResolution="60 20" xml:lang="en" xmlns="http://www.w3.org/ns/ttml" xmlns:ttp="http://www.w3.org/ns/ttml#parameter" xmlns:tts="http://www.w3.org/ns/ttml#styling"> <head> <layout> <region xml:id="r1" tts:color="white" tts:origin="10c 4c" tts:extent="40c 2c" tts:displayAlign="after"/> </layout> </head> <body> <div region="r1"> <p begin="0s" end="8s">Lorem ipsum dolor sit</p> <p begin="4s" end="12s">Amet consectetur adipiscing elit</p> <p begin="8s" end="18s">Sed do eiusmod tempor incididunt labore</p> <p begin="14s" end="25s">et dolore magna aliqua</p> <p begin="18s" end="29s">Ut enim ad minim veniam quis, nostrud</p> </div> </body> </tt> |
An example of paint-on captions. Paint-on effects are achieved by using timed span
elements in order to expose (paint) inline text units, e.g., words,
over some time interval. Here, five paragraphs have their individual words sequentially timed in order to paint one new word every second. The end of the active duration of
each inline element coincides with the end of the div
element's parallel time container, so that once a word is painted, it remains in the region (on its rendered line) until the div
element's active time interval lapses.
<tt ttp:cellResolution="60 20" xml:lang="en" xmlns="http://www.w3.org/ns/ttml" xmlns:ttp="http://www.w3.org/ns/ttml#parameter" xmlns:tts="http://www.w3.org/ns/ttml#styling"> <head> <layout> <region xml:id="r1" tts:color="white" tts:origin="10c 4c" tts:extent="40c 5c" tts:displayAlign="after"/> </layout> </head> <body> <div region="r1" begin="0s" end="25s"> <p> <span begin="0s">Lorem</span> <span begin="1s">ipsum</span> <span begin="2s">dolor</span> <span begin="3s">sit</span> </p> <p> <span begin="4s">Amet</span> <span begin="5s">consectetur</span> <span begin="6s">adipiscing</span> <span begin="7s">elit</span> </p> <p> <span begin="8s">Sed</span> <span begin="9s">do</span> <span begin="10s">eiusmod</span> <span begin="11s">tempor</span> <span begin="12s">incididunt </span> <span begin="13s">labore</span> </p> <p> <span begin="14s">et</span> <span begin="15s">dolore</span> <span begin="16s">magna</span> <span begin="17s">aliqua</span> </p> <p> <span begin="18s">Ut</span> <span begin="19s">enim</span> <span begin="20s">ad</span> <span begin="21s">minim</span> <span begin="22s">veniam</span> <span begin="23s">quis,</span> <span begin="24s">nostrud</span> </p> </div> </body> </tt> |
The editor acknowledges the current and former members of the Timed Text Working Group, the members of other W3C Working Groups, and industry experts in other forums who have contributed directly or indirectly to the process or content of this document as follows:
Glenn Adams, Thomas Bause-Mason, John Birch, Kees Blom, Bert Bos, Brad Botkin, Dick Bulterman, Frans de Jong, Mike Dolan, Martin Dürst, Donald Evans, Geoff Freed, Al Gilman, Giles Godart-Brown, Markus Gylling, Markku Hakkinen, Sean Hayes, Erik Hodge, Philipp Hoschka, Suzi Hyun, Michael Jordan, Masahiko Kaneko, Courtney Kennedy, George Kerscher, David Kirby, Andrew Kirkpatrick, Philippe Le Hégaret, Pierre-Anthony Lemieux, Chris Lilley, Jason Livingston, Monica Martin, Matthew May, Nigel Megitt, Thierry Michel, Frank Olivier, Soohong Daniel Park, Silvia Pfeiffer, Brian Raymor, David Ronca, Patrick Schmitz, David Singer, Craig Smithpeters, Andreas Tai, and Mohamed Zergaoui.
The editor wishes to especially acknowledge the following contributions by members: Micheal Dolan (SMPTE time codes, streaming; SMPTE liaison), David Kirby (introductory example document; SMPTE time codes, descriptive metadata; EBU/AAF liaison), Geoff Freed (styling and example images of style properties), Sean Hayes (advanced profile concepts, including applicative timing), Eric Hodge (timing), Thierry Michel (metadata), and Dave Singer (animation, scrolling).
The Working Group dedicates this specification to our colleague David Kirby.