W3C

Widget Packaging and XML Configuration

W3C Working Draft 07 June 2011

This version:
http://www.w3.org/TR/2011/WD-widgets-20110607/
Latest version:
http://www.w3.org/TR/widgets/
Previous version:
http://www.w3.org/TR/2011/WD-widgets-20110322/
Latest Editor's draft:
http://dev.w3.org/2006/waf/widgets/
Test suite:
http://dev.w3.org/2006/waf/widgets/test-suite/
Implementation report:
http://dev.w3.org/2006/waf/widgets/imp-report/
Differences history:
Differences between the last published draft and this document.
Differences between latest editor's draft and latest version.
Editor:
Marcos Cáceres, W3C Invited Expert

Abstract

This specification standardizes a packaging format and metadata for a class of software known as widgets. Unlike traditional user interface widgets (e.g., buttons, input boxes, toolbars, etc.), widgets as specified in this document are full-fledged client-side applications that are authored using technologies such as HTML and then packaged for distribution. Examples range from simple clocks, stock tickers, news casters, games and weather forecasters, to complex applications that pull data from multiple sources to be "mashed-up" and presented to a user in some interesting and useful way.

The specification relies on PKWare's Zip specification as the archive format, XML as a configuration document format, and a series of steps that runtimes follow when processing and verifying various aspects of a package. The packaging format acts as a container for files used by a widget. The configuration document is an XML vocabulary that declares metadata and configuration parameters for a widget. The steps for processing a widget package describe the expected behavior and means of error handling for runtimes while processing the packaging format, configuration document, and other relevant files.

This specification is part of the Widgets family of specifications, which together standardize widgets as a whole.

Status of this Document

This section describes the status of this document at the time of its publication. Other documents may supersede this document. A list of current W3C publications and the latest revision of this technical report can be found in the W3C technical reports index at http://www.w3.org/TR/.

Publication as a Working Draft does not imply endorsement by the W3C Membership. This is a draft document and may be updated, replaced or obsoleted by other documents at any time. It is inappropriate to cite this document as other than work in progress. You can find the latest Editor's Draft of this document in the W3C's CVS repository.

This is the 07 June 2011 Last Call Working Draft of this specification.  The Last Call period ends on 28 June 2011. The public is encouraged to send comments to the WebApps Working Group's public mailing list public-webapps@w3.org (archive). See W3C mailing list and archive usage guidelines. A detailed list of changes from the previous version is also available from the W3C's CVS server.

IMPORTANT: By the time the Last Call comment period ends (June 28), the Working Group expects to have data that at least two independent implementations pass 100% of this spec's test suite. As such, if no substantive changes must be made as a result of the Last Call review, we will not publish a Candidate Recommendation for this spec and the next publication will be Proposed Recommendation.

This document is produced by the Web Applications WG, part of the Rich Web Client Activity in the W3C Interaction Domain. It is expected that this document will progress along the W3C's Recommendation track.

This document was produced by a group operating under the 5 February 2004 W3C Patent Policy. W3C maintains a public list of any patent disclosures made in connection with the deliverables of the group; that page also includes instructions for disclosing a patent. An individual who has actual knowledge of a patent which the individual believes contains Essential Claim(s) must disclose the information in accordance with section 6 of the W3C Patent Policy.

The public is encouraged to send comments to the Web Apps Working Group's public mailing list public-Webapps@w3.org (archive). See W3C mailing list and archive usage guidelines. A detailed list of changes from the previous version is also available from the W3C's CVS server.

Table of Contents

1. Introduction

This section is non-normative.

Widgets are full-fledged client-side applications that are authored using Web standards such as [HTML] and packaged for distribution. They are typically downloaded and installed on a client machine or device where they run as stand-alone applications, but they can also be embedded into Web pages and run in a Web browser. Examples range from simple clocks, stock tickers, news casters, games and weather forecasters, to complex applications that pull data from multiple sources to be "mashed-up" and presented to a user in some interesting and useful way (see [Widgets-Landscape] for more information).

This specification is intended to specify a part of the Web platform closely related to [HTML].

1.1. Design Goals and Requirements

This section is non-normative.

The design goals and requirements for this specification are documented in the [Widgets-Requirements] document.

This document addresses the 25 requirements relating to "Packaging" and "Configuration Document" of the 30 April 2009 Working Draft of the Widgets Requirements Document:

  1. Packaging Format: see packaging format.
  2. Media Type: see the valid widget media type.
  3. File Extension: see widget file extension.
  4. Internal Abstract Structure: see Zip archive and widget package.
  5. Reserved Resource Names: see reserved file names table.
  6. Addressing Scheme: see valid path.
  7. Multilingual File Names: see Zip relative path, particularly in respect to support for [UTF-8].
  8. Localization Guidelines: see element-based localization, folder-based localization.
  9. Automatic Localization: see element-based localization, folder-based localization.
  10. Device Independent Delivery: all aspects of this document where developed with this requirement in mind.
  11. Data Compression: see the valid compression methods.
  12. Derive the Media Type of Resources: see the rule for identifying the media type of a file.
  13. Format and Schema: see configuration document, table of configuration defaults, and the [Widgets-Relax NG Schema] for the configuration document.
  14. Widget Metadata: see configuration document (particularly the elements).
  15. Authorship Metadata: see the author element.
  16. Copyright Notice and License Metadata: see the license element.
  17. Visual Rendering Dimensions: see the widget element.
  18. Declarative Bootstrap see the content element.
  19. Automated Bootstrap see the default start file.
  20. Iconic Representations: see the icon element, default icons table and custom icons table.
  21. Configuration Parameters: the param element (used in conjunction with the feature element).
  22. Author-defined Start-up Values (Preferences): see the preference element.
  23. Feature Access Declarations: see the feature element.
  24. Configuration Document Independence: see configuration document.
  25. Preferred Display Mode: see the viewmodes attribute of the widget element.

1.2. How This Document is Organized

This section is non-normative.

This document is organized into two halves, but not explicitly marked as such. The first half defines the various aspects of what constitutes the packaging format, the configuration document, and reserved files, such as default icons and locale folders. Where possible, the first half avoids describing aspects related to processing, which are described in detail in the second half of the document.

The second half, which starts with the section titled "Steps for Processing a widget package", defines the steps required to process a widget package as well as the expected behavior of a user agent as it processes the packaging format, the configuration document, and attempts to find localized content. The second half of this document also deals with error handling in the event that a user agent encounters unsupported or missing files, or DOM nodes that are in error in the configuration document. Wherever processing is relevant, sections in the first half of the document link to sections in the second half of the document.

1.3. Typographic Conventions

This section is non-normative.

This section defines the typographical conventions used by this specification. Some text in this specification is non-normative. Non-normative text includes:

Everything else in this specification is normative.

Defined terms appear as this sample defined term. Such terms are referenced as sample defined term, providing a link back to the term definition.

Words that denote a conformance clause or testable assertion use keywords from [RFC2119]: must, must not, should, recommended, may and optional. The keywords must, must not, should, recommended, may and optional in this specification are to be interpreted as described in [RFC2119].

Variables are formatted specially, e.g. variable. Code is also specially formatted, such as code.

Words in italics denote a formal definition given in an external specification.

This is an example. Examples are used to explain concepts or demonstrate how to use a feature. Examples are non-normative.

Note: This is a note, it usually contains useful supplementary information in a non-normative form.

Authoring Guidelines: This is an Authoring Guideline. Its purpose is to provide authors with best-practice authoring techniques. Authoring guidelines are non-normative.

1.4. The Widget Family of Specifications

This section is non-normative.

This specification is part of the Widgets family of specifications, which together standardize widgets as a whole. The list of specifications that make up the Widgets Family of Specifications can be found on the Working Group's wiki.

2. Conformance

There is only one class of product that can claim conformance to this specification: a user agent.

Note: Implementers can partially check their level of conformance to this specification by successfully passing the test cases of the [P&C-Test-Suite]. Note, however, that passing all the tests in the test suite does not imply complete conformance to this specification; It only implies that the implementation conforms to aspects tested by the test suite (the test suite does not provide tests for any optional conformance clauses).

3. Definitions

The following terms are used throughout this specification so they are gathered here for the readers convenience. The following list of terms is not exhaustive; other terms are defined throughout this specification.

Arbitrary means that a character, or text string, or file-name, or folder-name is not reserved for the purpose of this specification.

An author is a person who created a widget package or an authoring tool that generated a widget package.

Initialization means a user agent procedurally stepping through the steps for processing a widget package.

A language tag is a text string that matches the production of a Language-Tag defined in the [BCP47] specifications (see the IANA Language Subtag Registry for an authoritative list of possible values, see also the Maintenance Agency for ISO 3166 country codes).

A media type is defined in the [MIME] specification.

Reserved means that a character, or text string, or file-name, or folder-name has a specified purpose and semantics in this specification or in some other specification or system. The intended purpose for any reserved thing is given when the term is used.

Supported means that a user agent implements a mentioned specification, or conformance clause, or is able to process or otherwise render mentioned media type.

Unsupported means the user agent does not implement a mentioned specification, or feature, or is unable to render or otherwise process a mentioned media type.

A widget is defined by the [Widgets-Landscape] as an end-user's conceptualization of an interactive single purpose application for displaying and/or updating local data or data on the Web, packaged in a way to allow a single download and installation on a user's machine, mobile phone, or Internet-enabled device. Because widgets are packaged, they can be shared by users without relying on [HTTP].

3.1. Character Definitions

This section groups common sets of [Unicode] code points into definitions for the purpose processing in this specification.

The space characters are code points:

The Unicode white space characters are code points marked in the [Unicode] specification with the property "White_Space", including (but not limited to - see [Unicode] for the authoritive list):

The Zip forbidden characters are code points:

4. User Agents

A user agent is an implementation of this specification that also supports [XML], [XMLNS], [UTF-8], [Unicode], [DOM3CORE], [SNIFF], and [ZIP] (see optional aspects of the Zip specification).

In addition to widget packages, a user agent may support other legacy and proprietary application packaging formats.

It is optional for a user agent to support the optional aspects of the Zip specification.

Note: The user agent described in this specification does not necessarily denote a "widget user agent" at large: that is, a user agent that implements all the specifications, and dependencies, defined in the Widgets Family of Specifications. The user agent described is this specification is only concerned with how to processes Zip archives and configuration documents.

4.1. Optional Aspects of the Zip Specification

The optional aspects of the Zip specification are as follows. These aspects represent general features defined in the [ZIP] specification that this specification does not make use of:

5. Zip Archive

The Zip archive file format, defined in the [ZIP] specification, is the packaging format for widget packages.

A file entry is the data held by a local file header, file data, and (optional) data descriptor, as defined in the [ZIP] specification, for each physical file or folder contained in a Zip archive.

A potential Zip archive is a data object claiming to be a Zip archive, that has not been verified to be a valid Zip archive.

A valid Zip archive is a data object that the user agent has verified as conforming to the production of a .Zip file as defined by the Zip File Format Specification [ZIP] and meets the requirements of this specification (See Step 2).

The magic numbers for a Zip archive is the byte sequence: 50 4B 03 04.

5.1. Compression Methods

A compression method is the compression algorithm or storage method that was used to encode the file data of a file entry when the zip archive was created by the author. The compression method that encoded the file data of a file entry is identified by the numeric value derived from the compression method field defined in the [ZIP] specification.

The valid compression methods, as indicated by the compression method field, for a file entry are:

8
Data is compressed using [Deflate].
0
Data is Stored (no compression), as defined in the [ZIP] specification.

Authoring Guidelines: To ensure interoperability, compress file entries in Zip archives with [Deflate] or Stored (no compression); other compression methods can result in in the Zip archive being treated as an invalid widget package. Of the valid compression methods, [Deflate] is the preferred compression method.

5.2. Version of Zip Needed to Extract a File Entry

The version needed to extract is the 2-byte sequence in the local file header of a file entry that indicates the minimum supported version of the [ZIP] specification needed to extract the file data.

The valid versions needed to extract values are as follows. Each value is assigned one or more meanings by the [ZIP] specification:

1.0

Default value specified in the [ZIP] specification.

2.0

The file data is compressed using [Deflate], or the file data is a folder, or the file has been encrypted using traditional PKWARE encryption.

Note: If the Zip archive has been encrypted using traditional PKWARE encryption, then the user agent will treat the Zip archive as an invalid widget package in Step 2.

5.3. Zip Relative Paths

A Zip relative path is the variable-length string derived from the file name field of the local file header of a file entry.

Note: A Zip relative path is said to be relative as it stores the string that represents file and folder names relative to where the Zip archive was created on a file system (e.g. images/bg.png), as opposed to storing an absolute path (e.g. c:\images\bg.png). The value of a Zip relative path will generally resemble the string value of a name of the file or folder(s) on the device on which the Zip archive was created, but with the exception of the path delimiter being a U+002F SOLIDUS "/" character. Note also that a Zip relative path is not a URI reference; Zip relative paths need to be converted to URI references before they can be used in context that make use of URIs.

A valid Zip relative path is one that matches the production of Zip-rel-path in the following [ABNF]:

Zip-rel-path   = [locale-folder] *folder-name file-name /
                 [locale-folder] 1*folder-name
locale-folder  = %x6C %x6F %x63 %x61 %x6C %x65 %x73
                 "/" lang-tag "/"
folder-name    = file-name "/"
file-name      = 1*allowed-char
allowed-char   = safe-char / zip-UTF8-char
zip-UTF8-char  = UTF8-2 / UTF8-3 / UTF8-4
safe-char      = ALPHA  / DIGIT / SP  / "$" / "%" / 
                 "'"    / "-"   / "_" / "@" / "~" /
                 "("    / ")"   / "&" / "+" / "," /
                 "="    / "["   / "]" / "."
UTF8-2         = %xC2-DF UTF8-tail
UTF8-3         = %xE0 %xA0-BF UTF8-tail / %xE1-EC 2( UTF8-tail ) /
                 %xED %x80-9F UTF8-tail / %xEE-EF 2( UTF8-tail )
UTF8-4         = %xF0 %x90-BF 2( UTF8-tail ) / %xF1-F3 3( UTF8-tail ) /
                 %xF4 %x80-8F 2( UTF8-tail )
UTF8-tail      = %x80-BF
lang-tag       = primary-subtag *( "-" subtag )
primary-subtag = 1*8low-alpha
subtag = 1*8(alphanum) alphanum = low-alpha / DIGIT
low-alpha = %x61-7a

ALPHA, DIGIT, and SP are defined in the [ABNF] specification (but essentially represent alphanumerical characters and the U+0020 SPACE code point respectively).

5.4. Interoperability Considerations

This section is non-normative.

Some issues can arise with regards to character encodings of file names, the length of zip relative paths, and the use of certain strings as file names. This sections is intended to help authors avoid potential interoperability issues.

Authoring Guideline

5.4.1. Paths Lengths

This section is non-normative.

Authors need to be aware that having excessively long path names (e.g. over 120 characters) can also result in interoperability issues on some operating systems. This is because some operating systems have restrictions on how long a path length can be, so authors should try to keep the lengths of paths at less than 250 bytes. In addition, Unicode code points may require more than one byte to encode a character, which can result in a path whose length is less than 250 characters but whose size is greater than 250 bytes!

5.4.2. Character sets

This section is non-normative.

Authors need to be aware that, at the time of publication, there are interoperability issues with regards to using characters outside the safe-chars range for file or folder names in a Zip archive when using Zipping tools bundled with operating systems. The interoperability issues have arisen from non-conforming implementations of the [ZIP] specification across operating systems: very few, if any, correctly support encoding file names in Unicode.

In the case where the Zip relative path is encoded using [UTF-8], the language encoding flag (EFS) needs to be set.

If an author chooses to use the utf8-chars, they need to thoroughly test their widgets on various platforms prior to distribution; otherwise it is suggested that authors restrict file and folder names to the safe-chars (characters in the US-ASCII range).

5.4.3. File Names

This section is non-normative.

Authors need to avoid using the Zip forbidden characters when naming the files used by a widget. These characters are reserved to maintain interoperability across various file systems and with [URI]s.

Authors need to avoid using the following words as either a folder or a file-name in a Zip relative path as they are reserved by some operating systems (case-insensitive): CON, PRN, AUX, NUL, COM1, COM2, COM3, COM4, COM5, COM6, COM7, COM8, COM9, LPT1, LPT2, LPT3, LPT4, LPT5, LPT6, LPT7, LPT8, LPT9, CLOCKS$. For example, the following names are ok: "CON-tact.txt", "printer.lpt1", "DCOM1.pdf". However, "com3.txt" "Lpt1", "CoM9.gif" would not be.

In addition, authors need to avoid having a "." U+002E FULL STOP as the last character of a file or folder name as some operating systems will remove the character when the file is extracted from the Zip archive onto the device. Furthermore, avoid having the space character (SP) at the start or end of a file name; and take caution when using the "+" U+002B PLUS SIGN, as it might cause issues on some operating systems.

6. Widget Packages

A widget package is a valid Zip archive that contains the following:

Note: See step 1 - Acquire a Potential Zip Archive for instructions on how to process a widget package.

6.1. Invalid Widget Package

During the steps for processing a widget package, certain error conditions can result in a Zip archive being treated as an invalid widget package. An invalid Widget package is a condition whereby a Zip archive, or a file within the Zip archive, is deemed to be corrupt beyond recovery or is non-conforming to, or unsupported by, this specification in such a way that it would not be possible for the user agent to continue processing.

6.2. Files and Folders

The root of the widget package is the top-most path level of the Zip archive. The root of the widget package contains files and folders, some of which are reserved (see reserved file names table).

A file is the decompressed physical representation of a file entry (i.e., a file extracted into its physical form as it was prior to being added to the Zip archive).

A folder is a file entry whose file name field matches the production of folder-name in a valid Zip relative path (the last character of the file name field is a U+002F SOLIDUS) and whose version needed to extract is 2.0.

A processable file is a file that:

6.3. Reserved File and Folder Names

The reserved file names table, below, contains a list of file names that are reserved for some purpose by this specification. The first column of the reserved file names table contains a case-sensitive list of file names. The second column of the table denotes the purpose for which the file name is reserved.

Reserved File Names Table
file name Type reserved for purpose
config.xml file Configuration document
icon.png file Default icon
icon.gif file Default icon
icon.jpg file Default icon
icon.ico file Default icon
icon.svg file Default icon
index.html file Default start file
index.htm file Default start file
index.svg file Default start file
index.xhtml file Default start file
index.xht file Default start file
locales folder Container for localized content

Files named using the naming conventions for distributor signatures and the naming convention for an author signature, as defined in the [Widgets-DigSig] specification, are also reserved in this specification.

Authoring Guideline: Authors are strongly encourage to package all files, except the reserved files and the container for localized content, within a single subdirectory named, for instance, after the widget. This is to avoid unzipping several files into the end-user's current working directory. Future versions of this specification may include rules for locating index.html and config.xml within such directories.

For example, best-practice for packaging a widget would look something like this:

  • widget file: boat.wgt
    • file: index.html
    • configuration document: config.xml
    • folder: boat/
      • folder: scripts/
        • script: engine.js
      • folder: images/
        • icon: header.png
    • folder: locales/
      • folder: en-gb/
        • folder: boat/
          • folder: images/
            • icon: hearder.png

6.4. Digital Signatures

A widget package contains a digital signature, and hence is digitally signed, if the widget package contains one or more files that conform to the [Widgets-DigSig] specification.

6.5. Start Files

A start file designates a file from the widget package to be loaded by the user agent when it instantiates the widget. This specification defines two kinds of start file: custom start file and default start file.

6.5.1. Custom Start File

A custom start file is a processable file inside the widget package identified by a content element's src attribute.

6.5.2. Default Start Files

A default start file is a reserved start file at the root of the widget package or at the root of a locale folder whose file name case-sensitively matches a file name given in the file name column of the default start files table, and whose media type matches the media type given in the media type column of the table.

It is optional for a user agent to support the media types listed in the default start files table.

If a user agent encounters a file matching a file name given in the file name column of the default start files table in an arbitrary folder, then user agent must treat that file as an arbitrary file.

For example, "foo/bar/index.html" would be treated as an arbitrary file.

Default Start Files Table
file name media type
index.htm text/html
index.html text/html
index.svg image/svg+xml
index.xhtml application/xhtml+xml
index.xht application/xhtml+xml

Note: See Step 8 for instructions on finding a default start file.

Authoring Guidelines: Always include at least one start file in a widget package.

6.6. Icons

An icon is a file that is used to represent the widget in various application contexts (e.g. the icon that the user activates to instantiate a widget, or an icon in a dock or task bar or some other visual context). The icon is intended to help users of visual browsers to recognize the widget at a glance. There are two kinds of icons defined by this specification, custom icons and default icons.

6.6.1. Custom Icons

A custom icon is an icon explicitly declared by an author via an icon element in a configuration document. A custom icon can be located at the root of the widget package, or at the root of a locale folder, or in an arbitrary folder.

6.6.2. Default Icons

A default icon is a reserved icon, either at the root of the widget package or at the root of a locale folder, whose file name case-sensitively and exactly matches a file name given in the file name column of the default icons table.

Default Icons Table
file name media type
icon.svg image/svg+xml
icon.ico image/vnd.microsoft.icon
icon.png image/png
icon.gif image/gif
icon.jpg image/jpeg

It is optional for a user agent to support the media types listed in the default icons table.

6.7. Media Type

The valid widget media type is the string application/widget.

Authoring Guidelines: If the protocol over which the widget package is transferred supports the [MIME] specification (e.g. [HTTP]), then make sure that the widget is labeled with an application/widget media type. Failure to correctly label a widget package can result in the widget package being treated as an invalid widget package.

6.8. File Extension

A widget file extension is the text string that case-insensitively matches the string ".wgt" (e.g. .wgt, .WGt, .WgT, etc. are all valid).

For example in widget.WGT, the ".WGT" component is the file extension.

Authoring Guidelines: If it is anticipated that the widget will be distributed by means lacking MIME support, then include the widget file extension. The widget file extension is not necessary if the widget package is labeled as a application/widget when served over HTTP. The widget file extension is required for widget packages on systems where it is customary for file names to include a file extension that symbolizes (or is associated with) a media type.

7. Configuration Document

A configuration document is an [XML] document that has a widget element at its root that is in the widget namespace. A widget package has exactly one configuration document located at the root of the widget package.

Note: Please see Step 7 for details of how the elements of the configuration document are processed by a user agent.

A valid configuration document file name is the string config.xml.

A user agent must treat any file in an arbitrary folder or locale folders that uses the file name config.xml as an arbitrary file.

Authoring Guidelines: Be sure to always include a configuration document at the root of the widget package and that the config.xml file name is in lowercase form. To ensure interoperability, encode the configuration document as [UTF-8].

7.1. Example Configuration Document

The following is an example of a typical configuration document:

<?xml version="1.0" encoding="UTF-8"?>
<widget xmlns       = "http://www.w3.org/ns/widgets"
        id          = "http://example.org/exampleWidget"
        version     = "2.0 Beta"
        height      = "200"
        width       = "200"
        viewmodes   = "fullscreen">
 
  <name short="Example 2.0">
    The example Widget!
  </name>

  <feature name="http://example.com/camera">
    <param name="autofocus" value="true"/>

  </feature>

  <preference name     = "apikey"
              value    = "ea31ad3a23fd2f"
              readonly = "true" />

  <description>
    A sample widget to demonstrate some of the possibilities.
  </description>

  <author href  = "http://foo-bar.example.org/"
          email = "foo-bar@example.org">Foo Bar Corp</author>

  <icon src="icons/example.png"/>
  <icon src="icons/boo.png"/>
  <content src="myWidget.html"/>

  <license>
Example license (based on MIT License)
Copyright (c) 2008 The Foo Bar Corp.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
INSULT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
  </license>
</widget>

Note: Implementers are encouraged to expose relevant information provided by the configuration document to the user. Having "visual metadata" encourages authors to make full use of the configuration document format. See Step 7 for instructions on how to process a configuration document.

Authoring Guidelines: The only mandatory element in a configuration document is the widget element. All other elements and their respective attributes are optional. The following example shows the smallest possible configuration document that a user agent will be able to process. The reason to include this sole element is to explicitly inform a user agent or conformance checker that this zip file attempts to conform to this specification.



<!-- example of the smallest possible conforming configuration document -->
<widget xmlns="http://www.w3.org/ns/widgets"/>

7.2. Namespace

The widget namespace URI for a configuration document is http://www.w3.org/ns/widgets [XMLNS].

No provision is made for an explicit version number in this specification. If a future version of this specification requires explicit versioning of the document format, a different namespace will be used.

Authoring Guidelines: Be sure to declare the widget namespace as part of the widget element. If the namespace is absent, then the widget will be treated by the user agent as an invalid widget package.

7.3. Proprietary Extensions

This section is non-normative.

Implementers or authors intending to extend the configuration document format with their own [XML] elements and attributes (or those defined in other specifications) can do so by using a separate [XMLNS] namespace. This specification does not define a model for processing [XML] elements outside the widget namespace (they are simply ignored during processing).

Example of extending the configuration document format:

<widget xmlns="http://www.w3.org/ns/widgets"
   xmlns:ex="http://example.org/">
       <icon src="idle.png"  ex:role="inactive"/>

       <icon src="big.png" ex:role="big"/>
       <ex:datasource>{'a':'b','c':'d'}</ex:datasource>

       <content src="widget.html"/>
</widget>

7.4. Types of Attributes

This section defines the different attribute types used in the configuration document and what constitutes valid values for those attribute types.

An attribute is invalid if its value does not conform to its said attribute type; that is, if the value of the attribute is in error given the processing rules for that type of attribute.

Boolean attribute

A boolean attribute is a keyword attribute that can only be used with a valid boolean value. A valid boolean value is a keyword that case-sensitively matches true or false. Unless specified otherwise, the default behavior, which is used when the attribute is omitted or has a value other than the two allowed values, is false. The way a user agent interprets a boolean attribute is defined as part of an attribute's definition (see, for example, the feature element's required attribute).

String attribute

The value of a string attribute is any string that conforms to [XML] as a valid string for an XML attribute. The purpose of this attribute type is to classify strings that are not affected by the dir attribute, such as the email attribute (i.e., these attributes will not treated as displayable-strings during Step 7).

Displayable-string attribute

An attribute whose primary purpose is to convey human readable information, such as the name element's short attribute and the widget element's version attribute.

Keyword attribute

A keyword is a string that is reserved for the purpose of this specification. The value of a keyword attribute is a keyword that is one of a finite set specified in the attribute's definition in the case given in this specification.

Keyword list attribute

An attribute defined as taking one or more keywords as a value, which are separated by space characters.

Media type attribute

An attribute whose value is defined as containing a valid media type. A valid media type is string that matches the production for valid-MIME-type in the following [ABNF]:


valid-MIME-type = type "/" subtype *(";" parameter)

The type, subtype, and parameter tokens are defined in the [MIME] specification.

Language attribute

An attribute whose value is defined as containing a valid language tag (see the IANA Language Subtag Registry for an authoritative list of possible values, see also the Maintenance Agency for ISO 3166 country codes). A valid language tag is a string that conforms to the production of a Language-Tag, as defined in the [BCP47] specification.

Numeric attribute

The value of a numeric attribute is a string containing a valid non-negative integer. A valid non-negative integer is a string that consists of one or more code points in the range U+0030 DIGIT ZERO (0) to U+0039 DIGIT NINE (9). For example, the strings "002", "323", "23214", and so on.

Path attribute

An attribute defined as containing a valid path. A valid path is one that matches the production of a Zip-rel-path or a Zip-abs-path.

Authoring Guidelines: A valid path is not a URI: a valid path represents a hierarchical path to a file inside a Zip archive, which exactly matches the value of a file name field of a local file header of a file entry. This means that valid paths need not be URL encoded.

IRI attribute

An attribute defined as containing a valid IRI. A valid IRI is one that matches the IRI token of the [IRI] specification.

Authoring Guidelines: Because of the risk of confusion between IRIs that would be equivalent if dereferenced, the use of %-escaped characters in feature names is strongly discouraged.

Version attribute

A displayable-string attribute whose value is any arbitrary string value (possibly empty) within the constraints allowed for [XML] attributes. This specification does not mandate any specific format, semantics, or special processing rule for the format of a version attribute.

Authoring Guidelines: For the purpose of this specification, the structure of version tags has no semantics; they are just treated as arbitrary strings (e.g. '1.0' is not less than '2.0', but is simply different). However, for the sake of consistency, one can choose to use the following [ABNF]:

rec-version-tag = 1*DIGIT "."  1*DIGIT [ "." 1*DIGIT]
*[ 1*ALPHA / SP / 1*DIGIT ]
Example version tags:
  • Version 1.0 Beta
  • 1.0 RC1
  • 1.0-Build/1580
  • Joey the dog [5.1.2100]
Example of rec-version-tag:
  • 1.0
  • 1.10.1 beta1
  • 1.02.12 RC1

7.5. Global Attributes

This section describes the behavior and expected usage of other relevant attributes that are part of the [XML] specification and this specification. In this specification, these attributes are referred to as global attributes because they can be used on any element in a configuration document.

Although global attributes can be used on any element in this specification, they sometimes have no effect on the element on which they are used. For example, applying dir attribute on an icon element will have no effect on the icon elements or any of its attributes. What effect specifying a global attribute has on an elements is determined by Step 7 of this specification.

Authoring Guidelines: Although it is optional for an author to use any global attributes, their usage is recommended when appropriate (e.g., when declaring the language will help with legibility and when directional information will assist the user agent render text correctly).

7.5.1. The xml:lang Attribute

A language attribute that specifies, through a language tag, the language of the contents and attribute values of XML elements (see the IANA Language Subtag Registry). The [XML] specification specifies the xml:lang attribute and its influence on child nodes.

Authoring Guidelines: Although [BCP47] recommends that language tags be casefolded in a particular way for presentation, case has no meaning in a language tag. As a reminder to authors that user-agents map all language tags to lowercase, all examples in this document use lowercase. See also folder-based localization, which also requires authors to use language tags in lowercase form as the names of folders.

Avoid subtags from the IANA Language Subtag Registry marked as deprecated, grandfathered, or redundant. The intended purpose of the xml:lang attribute is to declare the primary language of an element, its attributes, and its descendent nodes; as such, it has no implication with regards to the directionality of the text in the user agent. To specify the directionality of text, see the dir attribute.

7.5.2. The dir Attribute

A keyword attribute used to specify the directionality in which human-readable text is to be represented by a user agent (e.g., the text content of the name element, the description element, and the license element). The directionality set by the dir attribute applies to the text content and any displayable string attributes of the element where it is used, and to child elements in its content unless overridden with another instance of dir (i.e., in this specification, the dir attribute only affects the short attribute of the name element and to the version attribute of the widget element).

The possible value of a dir attribute is one of the valid directional indicators:

ltr

Left-to-right text. Request that the Unicode [BIDI] algorithm treat characters within an element as embedded left-to-right.

rtl

Right-to-left text. Request that the Unicode [BIDI] algorithm treat characters within an element as embedded right-to-left.

lro

Left-to-right override. Forces the Unicode [BIDI] algorithm to treat characters within an element as strong left-to-right characters.

rlo

Right-to-left override. Forces the Unicode [BIDI] algorithm to treat characters within an element as strong right-to-left characters.

Note:For security reasons, implementations intending to display IRIs and IDNA addresses found in the configuration document are strongly encouraged to follow the security advice given in [UTR36]. This could include, for example, behaving as if the dir attribute had no effect on any IRI attributes, path attributes, and the author element's email attribute.

The base direction of a dir attribute is either set explicitly by the nearest parent element that uses the dir attribute; or, in the absence of such an attribute, the base direction is inherited from the default direction of the document, which is left-to-right ("ltr").

7.5.3. Examples of Usage

The following example demonstrates the dir attribute being applied globally to a configuration document.

<widget xmlns="http://www.w3.org/ns/widgets" dir="rtl" xml:lang="fa">
<name short="آب و هوا">‎
آب و هوا برنامه</name>
   <description> 
این نرم افزار به شما اجازه می دهد تا برای دیدن آب و هوا در منطقه محلی تان.
   </description>
</widget>

The following example shows the dir attribute applied to localized content.

<widget xmlns="http://www.w3.org/ns/widgets">

   <name short="Weather">
    Weather Application
   </name>

   <name short="آب و هوا" xml:lang="fa" dir="rtl"> 
    آب و هوا برنامه
   </name>

</widget>

The following example shows the dir attribute used with mixed language content:


<widget xmlns="http://www.w3.org/ns/widgets">
  <name short="Weather">
   Weather! a totally awesome application!
  </name>

  <name short="آب و هوا" xml:lang="fa" dir="rtl">
  <span dir="ltr" xml:lang="en">Weather!</span> برنامه واقعا بزرگ
  </name>

</widget>

7.6. The widget Element and its Attributes

The widget element serves as a container for the other elements of the configuration document.

Context in which this element is used:
The widget element is the root element of a configuration document.
Occurrences:
Exactly one, at the root element of the [XML] document.
Expected children (in any order):
name: zero or more (one element is allowed per language).
description: zero or more (one element is allowed per language).
author: zero or one.
license: zero or more (one element is allowed per language).
icon: zero or more.
content: zero or one.
feature: zero or more.
preference: zero or more.
Localizable via xml:lang:
No. Inheritance of the value of this attribute by author, preference, icon, and content will have no effect during processing in Step 7.
Attributes:
Global attributes, id, version, height, width, viewmodes.

7.6.1. The id Attribute

An IRI attribute that denotes an identifier for the widget.

Authoring Guidelines: It is optional for authors to use the id attribute with a widget element.

7.6.2. The version Attribute

A version attribute that specifies the version of the widget.

Authoring Guidelines: It is optional for authors to use the version attribute with a widget element.

7.6.3. The height Attribute

A numeric attribute greater than 0 that indicates the preferred viewport height of the instantiated custom start file in CSS pixels [CSS].

Authoring Guidelines: It is optional for authors to use the height attribute with a widget element.

7.6.4. The width Attribute

A numeric attribute greater than 0 that indicates the preferred viewport width of the instantiated custom start file in CSS pixels [CSS].

Authoring Guidelines: It is optional for authors to use the width attribute with a widget element.

7.6.5. The viewmodes Attribute

A keyword list attribute that denotes the author's preferred view mode, followed by the next most preferred view mode and so forth. When the attribute is missing, or is left empty, it implies that the author expects the user agent to select an appropriate viewmode for the widget.

The concept of a viewport is defined in [CSS], but is essentially a window or other viewing area on the screen (see section 9.1.1 The viewport of [CSS2]). The concept of a view mode is defined in the [View-Modes] specification.

Authoring Guidelines: It is optional for authors to use the viewmodes attribute with a widget element.

7.6.6. The defaultlocale attribute

A language attribute that specifies, through a language tag, the author's preferred locale for the widget. Its intended use is to provide a fallback in case the user agent cannot match any of the widget's localized content to the user agent locales list or in case the author has not provided any unlocalized content.

Authoring Guidelines: It is optional for authors to use the defaultlocale attribute with a widget element.

7.6.7. Example of Usage

The following example shows how the widget element can be used.

<widget xmlns     = "http://www.w3.org/ns/widgets"
        id        = "http://example.org/exampleWidget"
        version   = "2.0 Beta"
        height    = "200"
        width     = "200"
        viewmodes = "windowed floating"/>

7.6.8. Example of Usage of the defaultlocale attribute

The following example shows how the widget element's defaultlocale attribute can be used:

<widget xmlns = "http://www.w3.org/ns/widgets"
        defaultlocale = "en-us">

   <name short="Weather" xml:lang="en-us">
    The Ultimate Weather Widget
   </name>

   <name short="Boletim" xml:lang="pt">
    Boletim Metereológico
   </name>

</widget>

7.7. The name Element and its Attributes

The name element represents the full human-readable name for a widget that is used, for example, in an application menu or in other contexts.

Context in which this element is used:
In a widget element.
Content model:
Any.
Occurrences:
Zero or more (one element is allowed per language).
Expected children:
span: zero or more.
Localizable via xml:lang:
Yes.

Authoring Guidelines: The value of the xml:lang attribute needs to be unique for any subsequent element of this type.

Attributes:
Global attributes, short.

7.7.1. The short Attribute

A displayable-string attribute intended to represent a condensed name for a widget (e.g., a name that could be used in context were only limited space is available, such as underneath an icon).

Authoring Guidelines: It is optional for authors to use the short attribute with an name element.

7.7.2. Example of Usage

The following example shows the usage of the name element.

<widget xmlns="http://www.w3.org/ns/widgets">

   <name short="Weather">
    The Ultimate Weather Widget
   </name>

   <name short="Boletim" xml:lang="pt">
    Boletim Metereológico
   </name>

</widget>

7.8. The description Element and its Attributes

The description element represents a human-readable description of the widget.

Context in which this element is used:
In a widget element.
Content model:
Any.
Occurrences:
Zero or more (one element is allowed per language).
Expected children:
span: zero or more.
Localizable via xml:lang:
Yes.

Authoring Guidelines: The value of the xml:lang attribute needs to be unique for any subsequent element of this type. If two or more elements with the same xml:lang attribute value are encountered, the user agent will ignore all but the matching first element. See Step 7 for more details.

Attributes:
Global attributes.

7.8.1. Example of Usage

An example usage of the description element.

<widget xmlns="http://www.w3.org/ns/widgets">
  <name>Dahut Chaser</name>
  <description>
Combining the latest weather info with your GPS position,
this widget alerts you of any significant dahut activity in your
area. When a big one walks by, the widget plots the best route on a map based
on the dahut's trajectory so you can chase it! With support for
built-in cameras, you can quickly upload all the Alpine action to
your blog or to the insane dahut chaser web site! Awesome!
  </description>
</widget>

7.9. The author Element and its Attributes

An author element represents people or an organization attributed with the creation of the widget.

Context in which this element is used:
As a child of the widget element.
Content model:
Any.
Occurrences:
Zero or one.
Expected children:
span: zero or more.
Localizable via xml:lang:
No. Only the first instance of this element in document order will be used, regardless of the value of xml:lang (if any).
Attributes:
Global attributes, href, email.

7.9.1. The href Attribute

An IRI attribute whose value represents an IRI that the author associates with himself or herself (e.g., a homepage, a profile on a social network, etc.).

Authoring Guidelines: It is optional for authors to use the href attribute with an author element.

7.9.2. The email Attribute

A string attribute that represents an email address associated with the author.

Authoring Guidelines: It is optional for authors to use the email attribute with an author element.

7.9.3. Example of Usage

The following example shows the expected usage of the author element.

<widget xmlns="http://www.w3.org/ns/widgets">
    <name>Café Finder</name>
    <author href  = "http://dahut.example.org/developers/"
            email = "cafefinder@example.org">
      Mr. Jo and Julia Bacalhau
    </author>
</widget>

7.10. The license Element and its Attributes

The license element represents a software license, which includes, for example, a usage agreement, redistribution statement, and/or a copyright license terms under which the content of the widget package is provided.

Context in which this element is used:
As a child of the widget element.
Content model:
Any.
Expected children:
span: zero or more.
Occurrences:
Zero or more (one element is allowed per language).
Localizable via xml:lang:
Yes.

Authoring Guidelines: The value of the xml:lang attribute needs to be unique for any subsequent element of this type. The content of localized license elements shouldn't be used to present different versions of a license, just translations of the same license.

Attributes:
Global attributes, href.

7.10.1. The href Attribute

A valid IRI or a valid path that points to a representation of a software and/or content license.

Authoring Guidelines: It is optional for authors to use the href attribute with a license element.

7.10.2. Example of Usage

This example shows the expected usage of the license element's href attribute.

<widget xmlns="http://www.w3.org/ns/widgets">
  <license href="http://www.w3.org/Consortium/Legal/2002/copyright-software-20021231">
Distributed under the W3C Software License.
  </license>
</widget>

This example shows the expected usage of the license element when the href attribute is omitted.

<widget xmlns="http://www.w3.org/ns/widgets">
  <license>
	... 
   3.3.1 Widgets can use any APIs or libraries, prescribed by anyone.
   Widgets a are a free and open Web technology, so can be produced for free 
   and sold anywhere. Widgets can be written in JavaScript 
   so can run on any conforming engine (without the annoying restrictions of 
   C, C++, or Objective-C). You can even "cross-compile" them, if you want.  
	...  
  </license>
</widget>

7.11. The icon Element and its Attributes

The icon element represents a custom icon for the widget.

Context in which this element is used:
As a child of the widget element.
Content model:
Empty.
Occurrences:
Zero or more.
Localizable via xml:lang:
No. Relies on folder-based localization.
Attributes:
Global attributes, src, width, height.

7.11.1. The src Attribute

A path attribute that points to a file inside the widget package.

Authoring Guidelines: When an icon element is used, it is required for authors to use the src attribute.

7.11.2. The width Attribute

A numeric attribute greater than 0 that represents, in CSS pixels [CSS2], the author's preferred width for the icon. A user agent may ignore this value when changing the height icon to fit a rendering context or for accessibility reasons.

Authoring Guidelines: It is optional for authors to use the width attribute of an icon element.

7.11.3. The height Attribute

A numeric attribute greater than 0 that represents, in CSS pixels [CSS2], the author's preferred height for the icon. A user agent may ignore this value when changing the height icon to fit a rendering context or for accessibility reasons.

Authoring Guidelines: It is optional for authors to use the height attribute of an icon element.

7.11.4. Example of Usage

This example shows the expected usage of the icon element. The example declares three icon elements, two of which are raster and one of which is an [SVGTiny] file. The raster graphics would be used for display contexts smaller than 256x256 pixels. Document order of the elements is irrelevant.


<widget xmlns="http://www.w3.org/ns/widgets">
  <icon src="icons/medium.png"/>
  <icon src="icons/big.svg" width="256" height="256"/>
  <icon src="icons/tiny.png"/>
</widget>

7.12. The content Element and its Attributes

The content element is used by an author to declare which custom start file the user agent is expected to use when it instantiates the widget.

Context in which this element is used:
As a child of the widget element.
Content model:
Empty.
Occurrences:
Zero or one.
Localizable via xml:lang:
No. Relies on folder-based localization.
Attributes:
Global attributes, src, type, encoding.

7.12.1. The src Attribute

A path attribute that allows an author to point to a file within the widget package.

Authoring Guidelines: When a content element is used, it is required for authors to use the src attribute.

7.12.2. The type Attribute

A media type attribute that indicates the media type of the file referenced by the src attribute.

Authoring Guidelines: It is optional for authors to use the type attribute with a content element. When the value is absent, the user agent will assume the value text/html.

7.12.3. The encoding Attribute

A keyword attribute that denotes the character encoding of the file identified by the src attribute. The value is the "name" or "alias" of any "Character Set" listed in [IANA-Charsets].

Authoring Guidelines: It is optional for authors to use the encoding attribute with a content element. Where aliases are given by the [IANA-Charsets] registry, authors are encouraged to use the value of the "preferred MIME name" (if any) from the registry.

The default encoding is [UTF-8], which has the name "UTF-8" in the [IANA-Charsets] registry.

Aside from the default encoding, it is optional for a user agent to support other character encodings.

7.12.4. Example of Usage

This example shows the expected usage of the content element:

<widget xmlns="http://www.w3.org/ns/widgets">
  <content src="myWidget.html"/>

</widget>

This example shows the content element being used with a encoding attribute to override the default value of the encoding attribute (UTF-8) with the GB2312 character set, which the author has used to encode simplified Chinese characters:

<widget xmlns="http://www.w3.org/ns/widgets">
  <name xml:lang="zh-cn">古老瓷地图</name>
  <name>Ancient Chinese Maps</name>
  <content src="china-maps.html" encoding="GB2312"/>
</widget>

This example shows the content element being used with a type attribute to instantiate a widget created with a proprietary media type:


<widget xmlns="http://www.w3.org/ns/widgets">
  <name>Location-Based Games Finder</name>
  <content src="lbg-maps.swf" type="application/x-shockwave-flash"/>
  <feature name="http://example.org/api.geolocation"
           required="false"/>
</widget>

7.13. The feature Element and its Attributes

A feature is a URI identifiable runtime component (e.g. an Application Programming Interface or video decoder). The act of a an author requesting the availability of a feature through a feature element is referred to as a feature request. The feature element serves as a standardized means to request the binding of an IRI identifiable runtime component to a widget for use at runtime. Using a feature element denotes that, at runtime, a widget can attempt to access the feature identified by the feature element's name attribute. How a user agent makes use of features depends on the user agent's security policy, hence activation and authorization requirements for features are beyond the scope of this specification. A feature has zero or more parameters associated with it.

Context in which this element is used:
As a child of the widget element.
Content model:
Zero or more param elements.
Occurrences:
Zero or more.
Localizable via xml:lang:
No.
Attributes:
Global attributes, name, required.

7.13.1. The name Attribute

An IRI attribute that identifies a feature that is needed by the widget at runtime (such as an API).

Authoring Guidelines: When the feature element is declared, it is required for authors to use the name attribute.

7.13.2. The required Attribute

A boolean attribute that indicates whether or not this feature has to be available to the widget at runtime.

When set to true, the required attribute denotes that a feature is absolutely needed by the widget to function correctly, and without the availability of this feature the widget serves no useful purpose or won't execute properly.

When set to false, the required attribute denotes that a widget can function correctly without the feature being supported or otherwise made available by the user agent.

Authoring Guidelines: It is optional for authors to use the required attribute with an feature element. However, authors need to be aware that a user agent will behave as if the required attribute had been set to true when the required attribute is absent, meaning that the named feature needs to be available at runtime or the widget will be treated as an invalid widget package.

7.13.3. Example of Usage

This example demonstrates a widget that would like to use a fictional geo-location API feature, but would still be able to function if the feature cannot be made available to the widget by the user agent.

<widget xmlns="http://www.w3.org/ns/widgets">
   <feature name     = "http://example.org/api/geolocation"
            required = "false"/> 
</widget>

7.14. The param Element and its Attributes

The param element defines a parameter for a feature. A parameter is a name-value pair that is associated with the corresponding feature for which the parameter is declared for. A author establishes the relationship between a parameter and feature by having a param element as a direct child of a feature element in document order.

Context in which this element is used:
As a child of the feature element.
Content model:
Empty.
Occurrences:
Zero or more.
Localizable via xml:lang:
No.
Attributes:
Global attributes, name, value.

7.14.1. The name Attribute

A string attribute that denotes the name of this parameter.

Authoring Guideline:When a param element is declared, it is required for authors to use the name attribute.

7.14.2. The value Attribute

A string attribute that denotes the value of this parameter.

Authoring Guidelines: When a param element is used, it is required for authors to use the value attribute.

7.14.3. Example of Usage

This example demonstrates a widget that makes use of a fictional geo-location feature where by its accuracy is set to "low" via a param element.


<widget xmlns="http://www.w3.org/ns/widgets">
   <feature name="http://example.org/api/geolocation">
      <param name="accuracy" value="low"/> 
   </feature>
</widget>

7.15. The preference Element and its Attributes

The preference element allows authors to declare one or more preferences: a preference is a persistently stored name-value pair that is associated with the widget the first time the widget is initiated.

Note: A user agent that supports the [Widgets-APIs] specification will expose any declared preference at runtime in the manner described in the [Widgets-APIs] specification.

Context in which this element is used:
As a child of the widget element.
Occurrences:
Zero or more.
Expected children:
none.
Localizable via xml:lang:
No.
Attributes:
Global attributes, name, value, readonly.

7.15.1. The name Attribute

A string that denotes the name of this preference.

Authoring Guidelines: When a preference element is used, it is required for authors to use the name attribute.

7.15.2. The value Attribute

A string that denotes the value of this preference.

Authoring Guidelines: It is optional for authors to use the value attribute with a preference element.

7.15.3. The readonly Attribute

A boolean attribute indicating whether this preference can, or cannot, be overwritten at runtime (e.g., via an author script). When set to true, it means that the preference cannot be overwritten. When set to false, it means that the preference can be overwritten.

Authoring Guidelines: It is optional for authors to use the readonly attribute with a preference element. If the readonly attribute is absent, the user agent will act as if the readonly attribute had been set to false (meaning that the preference can be overwritten at runtime).

7.15.4. Example of Usage

This example shows a widget where two preferences are set. The second preference is set as "read only" via the readonly attribute, which means the values of that preference cannot be changed at runtime.

<widget xmlns="http://www.w3.org/ns/widgets">
   <preference name     = "skin"
               value    = "alien"/>

   <!-- The preference below will be protected
        from modification and deletion at runtime -->

   <preference name     = "api-key"
               value    = "f6d3a312f9d742"
               readonly = "true"/>
</widget>

7.16. The span Element and its Attributes

The span element is a wrapper for text content; on its own it serves no useful function.

It is expected that authors will use the span element with the global attributes. When combined with the dir attribute, the span element can indicate the textual directionality of text content. In other words, it allows authors to set the base direction for the Unicode bidirectional algorithm [BIDI]. When combined with the xml:lang attribute, the span element allows the author to indicate the particular language used for a subset of text content within another element.

Context in which this element is used:
As a child of the name, author, license, and/or description element.
Occurrences:
Zero or more.
Expected children:
Any.
Localizable via xml:lang:

No, meaning that declaring xml:lang on a span element within a parent element (e.g., a name element) will not affect the behavior of element-based localization (see Example of Usage below to see how this works).

Authoring Guidelines: Authors are encouraged to use xml:lang to indicate the language information for text content contained within a span element. See Example of Usage for more information.

Attributes:
Global attributes.

7.16.1. Example of Usage

This section is informative.

This example shows the span elements being used to indicate directionality of text as well as language information. Note that the name element's xml:lang attribute is set to an empty string to allow it to be used as default content in the process of element-based localization:

<widget
 xmlns="http://www.w3.org/ns/widgets"
 xml:lang="he" dir="rtl">
 <name xml:lang="" dir="ltr">
  <span xml:lang="en">GPS Weather!</span>
 </name>
 <description>
  יישומון ה-
  <span dir="ltr" xml:lang="en">GPS Weather!</span> מאפשר
 לך לבדוק את מזג הא
וויר בכל נקודת GPS ברחבי העולם.
 </description>

<description xml:lang="" dir="ltr">
   <span xml:lang="en">The GPS Weather! widget lets you check
   the weather at any point around the world with GPS.</span>
</description>
</widget>

8. Internationalization and localization

Internationalization, or i18n, is the design and development of a product, application or document content that enables localization for target audiences that vary in culture, region, or language. Localization refers to the adaptation of a product, application or document content to meet the language, cultural and other requirements of a specific target market (a "locale").

Note: See also the Web Services Internationalization Usage Scenarios and the Unicode Locale Data Markup Language for an informative discussion on the term locale.

Localized content is content an author has explicitly localized: that is, a widget package that contains content localized using folder-based localization or a configuration document that contains content localized via element-based localization.

Default content is content included in the widget package or in the configuration document that has not been explicitly localized or content explicitly indicated by the author to be used as default content via the defaultlocale attribute of the widget element. In the case of a widget package, this means content outside the container for localized content. In the case of a configuration document, this means any element without an explicitly declared or inherited xml:lang attribute.

A localized file is any file that has been placed inside a locale folder (i.e., localized content that makes use of folder-based localization). A widget package contains zero or more localized files. All files and folders, except for the digital signatures and the configuration document, can be localized via folder-based localization.

For example, a locale folder "locales/ja/" (Japanese) might contain an HTML document translated into Japanese.

Authoring Guidelines: Locale folders need to be placed in the container for localized content (a locale folder not in a container for localized content will be treated as an arbitrary folder). In addition, authors making use of localization features provided by this specification should translate, localize, or alter localized content for a the given locale, and test their widgets thoroughly.

8.1. Bidirectional text

In conjunction to mandating that user agents support [Unicode], this specification provides the dir attribute and span element as a markup-based means of influencing the directionality for bidirectional Unicode text [BIDI]. See the definitions of the dir attribute and span element for usage examples.

8.2. Localization Model

This specification provides two means that authors can use to explicitly localize the content of a widget:

  1. By placing localized file content in locale folders, a process referred to as folder-based localization.

  2. By explicitly marking XML elements in the configuration document as localized, a process referred to as element-based localization.

Both forms of localization are described below. Folder-based and element-based localization rely on the user agent to match the language ranges held by the user agent locales to the appropriate locale folders and/or localized XML elements in the widget's configuration document.

8.3. Folder-based localization

This specification defines the concept of folder-based localization, which is the process whereby an author places files inside folders that are named in a manner that conforms to a language-range ABNF rule of this specification. That is, by naming folders in lower-case using values derived from the IANA Language Subtag Registry such as "en-us", "en-gb", and so on, but avoiding subtags marked as deprecated, grandfathered, or redundant in the IANA Language Subtag Registry. These locale folders are then placed inside the container for localized content.

Authoring Guidelines Although BCP 47 recommends a particular case-folding convention, the use of upper or lowercase letters has no meaning in a language tag. Because folders inside a widget package are treated by the user-agent in a case-sensitive manner, the names of the folders inside a 'locale' folder must be all lowercase. All language tags are mapped to lowercase for matching purposes (although they can appear in any form in the configuration file or elsewhere).

The container for localized content is a reserved folder at the root of the widget package whose folder-name case-sensitively matches the string 'locales'. A container for localized content contains zero or more locale folders.

A locale folder is a folder whose file name field matches the production of locale-folder and is a direct descendant of the container for localized content (e.g., "/locales/en-us", "/locales/fr", etc). A locale folder contains zero or more arbitrary folders and/or files.

Authoring Guidelines Authors need to avoid region, script or other subtags except where they add useful distinguishing information to a locale folder. In addition, avoid including empty locale folders in a widget package (unless there is a good reason to include them).

An example of a widget that uses folder-based localization:

widget.wgt
 locales/zh-hans-cn/a.gif
 locales/zh-hans-cn/f.gif
 locales/zh-hans/a.gif
 locales/zh-hans/b.gif
 locales/zh/a.gif
 locales/zh/b.gif
 locales/zh/c.gif
 a.gif
 b.gif
 c.gif
 d.gif
 index.html
 config.xml

Authors can further facilitate the localization process by grouping files into folder hierarchies made up of matching subtags, as is shown in the example.

Assuming the widget's locale is "zh-hans-cn", a reference to:

This works at all sub-levels, so long as the parent subtag matches the child subtags. So, for example, the "cn" region can make use of the localized files in the "zh-hans" folder level, the "zh" folder level, and the unlocalized files at the root of the widget package. The user agent always prioritizes files in sub-folders over files in locale folders closer to the root of the widget package. Conversely, if the widget's locale were "en-us", references to a.gif, b.gif, c.gif and d.gif would all resolve to the files in the root of the widget package.

Note also that the user agent treats any file or folder outside the container for localized content as default content.

8.4. Element-Based Localization

This specification defines the concept of element-based localization, which allows authors to use the xml:lang attribute to explicitly indicate that an [XML] element in the configuration document has been localized.

The following is an example of element-based localization:

<widget xmlns="http://www.w3.org/ns/widgets">
   <name short="Weather">
   The Ultimate Weather Widget
   </name>

   <name short="Boletim" xml:lang="pt">
    Boletim Metereológico
   </name>
</widget>

Some of the elements in the widgets namespace are defined to be localizable via xml:lang as part of the element's definition (with either "yes" or "no"). See, for example, the name element. When "yes", it means that an author can utilize xml:lang to achieve element-based localization either directly or indirectly through the inheritance of the value of xml:lang. How element-based localization is handled is specified in detail in Step 7.

Note: The xml:lang attribute can be used on any element in order to indicate which language is used in the content and attribute values of that element. As specified in the XML Specification, its value is inherited, such that if an element has an xml:lang attribute, all of its descendants are considered to be in that language as well, unless they specify their own xml:lang attribute. Note that an element can indicate that it is in no specific language by setting xml:lang to the empty string, irrespective of whether any of its ancestors has an xml:lang attribute.

For example, if an author uses the xml:lang attribute on the widget element, then all child elements inherit the value xml:lang. This means that the first name element below behaves as xml:lang="en" had been explicitly used. However, in the second name element, the declaration of xml:lang="pt" overrides xml:lang="en" inherited from the widget element. Finally, the last name element overrides xml:lang with an empty string, so that element will be treated by the user agent as default content.

<widget xmlns="http://www.w3.org/ns/widgets"
 xml:lang="en">

   <name short="I'm in english, though not explicitly marked as such!">
    Behaves as if xml:lang="en"
   </name>

   <name xml:lang="pt">
    The declaration of xml:lang="pt" overrides
    xml:lang="en" inherited from the widget element. 
   </name>
   
   <name xml:lang="">
    The user agent will treat this as unlocalized content. 
   </name>

</widget>

If an element is marked as being localizable via xml:lang with "yes", the specification indicates that only one element is allowed per language:

One element is allowed per language means that only one element of a type is allowed to be used per language (e.g., although many name elements can be present in a configuration document, only one name element will be selected by the user agent for the English language). During processing (Step 7), the user agent will only match the first element, in document order, that matches a language range in the user agent locales and ignore any subsequent repetitions of the element that contain a matching xml:lang value (even if that element's content is different).

For example, assume the user agent locales only contains the following language range: "en-us" (English as used in the United States). As only one instance of the description element is allowed per language, in the following code the user agent would match the first description element but would ignore the second and third description elements.


<widget xmlns="http://www.w3.org/ns/widgets">

   <description xml:lang="en">
      This element would be used.
   </description>

   <description xml:lang="en">
      This element would be ignored because there is already
      a description element that uses xml:lang="en".
   </description>

   <description>
      This element is unlocalized, and would be used if the user agent's
      locale does not match any localized description elements.
   </description>

	 <description xml:lang="">
		This element would be ignored because there is already an unlocalized 
      description element! Using xml:lang="" makes this element behave as if
      it is unlocalized.
	 </description>

</widget>

However, if the user agent locales only contained "*", or did not match any of the localized description elements, then the user agent would match the third description element above.

In the case whereby the author does not use an xml:lang attribute, and no element of a particular type with xml:lang matches the user agent locales, the user agent will use the first element that is default content, in document order, that matches the element type being sought.

For example, now assume that the user agent locales only contains the following language range: "jp" (Japanese). As only one instance of the description element is allowed per language, in the following code the user will agent ignore the first two description elements, but would match the third (unlocalized) description element.


<widget xmlns="http://www.w3.org/ns/widgets">

   <description xml:lang="en">
       This element would be ignored.
   </description>

   <description xml:lang="en">
      This element would be ignored.
   </description>

   <description>
      In this case, this unlocalized element would be used.
   </description>

</widget>
  

8.5. Localization Examples

This section is non-normative.

This section presents three examples of how widgets can be localized. Each example is intended to showcase how the localization algorithm is intended to work.

8.5.1. Simple Example

This example shows a widget that displays the days of the week based on the language ranges held by the user agent locales. If the user agent is unable to match a language range to any locale folder, the widget displays /index.html at the root of the widget package.

config: config.xml

<widget xmlns="http://www.w3.org/ns/widgets">
<name>What day is it?</name>
<description>
 This widget highlights the current day
 of the week.
</description>
</widget> 

file: locale/es/index.html

<!doctype html>
<title>¿Qué día es?</title>
<script src="scripts/dayfinder.js"></script>
<body>
<p>Hoy es: lunes, martes, miércoles,
   jueves, viernes, sábado,
   domingo

8.5.2. Complex Example

The following is an example of a localized widget with multiple localized icons, start files and configuration documents. Some relevant things to note from the example:

config: /config.xml

<widget xmlns="http://www.w3.org/ns/widgets">
  <name xml:lang="ko">웃기는 고양이</name>
  <content src="cats.html"/>
</widget> 

file: /locales/en-au/cats.html

<!doctype html>
<title>G'day! LOL Cats!</title>
<script src="scripts/engine.js">
...

file: /locales/es/cats.html

<!doctype html>
<title>Gatos Graciosos!</title>
<script src="scripts/engine.js">
...

8.5.3. Fallback Behavior Example

This specification allows authors to place files and folders they don't wish to localize at the root of the widget package. At runtime, if the user agent fails to find a file in a locale folder, it will always search at the root of the widget package for that missing file. The purpose of this 'fallback' model is to reduce the number of files that need to be created in order to localize a widget package.

The example below demonstrates how a user agent attempts to locate a file that is absent in a localized scenario. Assume the user agent's locale is 'en-gb' and the zip relative path being sought is "images/mast.png":

  1. Firstly, the user agent will search for a folder that matches the user agent's locale as closely as possible for the desired file. In this case, the user agent would attempt to locate the relative path 'images/mast.png' in '/locales/en-gb/', but would fail.

  2. Secondly, if the file is absent, the user agent will search for the absent file in any other sub-folder that is in the language range of the current locale folder. So the next place the user agent would look is in the 'en/' folder, where it would match the zip relative path 'images/mast.png'.

  3. Lastly, if the above fails, the user agent will search for the absent file at the root of the widget package. So, if the user agent's locale did not find the zip relative path in one of the locale folders, then 'images/mast.png' file at the root of the widget would be matched and this default content would be used.

Now consider the for various Chinese variants: 'zh-hans-cn', 'zh-hans', and 'zh' below. In this case, to find the 'flag.png' file for Mainland Chinese in simplified script 'zh-hans-cn', the user agent would first look in 'zh-hans-cn', followed by 'zh-hans', then in 'zh' where the file is located.

To find the 'mast.png' file, the user agent would look in 'zh-hans-cn', followed by 'zh-hans', followed 'zh', and finally at the root of the widget package where the absent file is actually located.

file: /index.html


<!doctype html>
<title>Patriotic Boat</title>
<script src="scripts/engine.js">
</script>
<body>
  <img src="flag.png">
  <img src="mast.png">

9. Steps for Processing a Widget Package

The steps for processing a widget package involves nine steps that a user agent follows in sequential order, responding accordingly if any of the steps result in an error or if the specification asks for the user agent to skip a step. The procedures for what to do when an error is encountered are described as part of each step; however, there are times when it will not be possible for the user agent to recover from an error and it will be forced to treat the widget as an invalid widget package.

In the event that a user agent encounters an invalid widget package during the steps for processing a widget package, a user agent must abort all processing of the steps for processing a widget package.

Note:A user agent can optimize steps for processing a widget package and associated processing rules , or perform the steps in a different order, but the end result needs to be indistinguishable from the result that would be obtained by following the specification.

9.1. Processing Rules

This section defines various processing rules, which are algorithms used during the steps for processing a widget package.

These algorithm makes use of a few special concepts defined below:

Text node

Any Text node, including CDATASection nodes (any Node with node type TEXT_NODE or CDATA_SECTION_NODE) as defined in the [DOM3Core] specification. For example, the worlds "hello world!" in the following name element: <name>hello world!</name>.

Localizable string

A data structure containing a sequence of one or more strings, each having some associated directional information and language information (if any). The purpose of an localizable string is to assist user agent in correctly applying the Unicode [BIDI] algorithm when displaying text.

For example, the string "Internationalization نشاط التدويل! Activity." The string contains both Arabic and English that mixes left-to-right text, right-to-left text, and a directionality-neutral punctuation mark that could not be correctly displayed without directional information.

null

A special symbol to indicate that a value has no data. For example, "let x be null" or "if y is empty, then return null".

Note: Note: Although languages such as ECMA Script and Java support null as a native value type, there are some programming languages that have no notion of null or where null is problematic (e.g. C++). Implementations in these languages need to substitute a language-specific value or symbol which is functionally equivalent to null, or if no equivalent exists, to have no value at all. For example, the value 0 may represent null for the height of a widget, since the height of a widget is defined as a non-negative integer greater than 0. In such a case, 0 could be treated as if it were null.

9.1.1. Rule for Verifying a Zip Archive

The rule for verifying a zip archive is described in this section. The algorithm returns either true or an error.

This specification does not provide the technical details of how to actually perform the checks, for which implementers need to refer to the [ZIP] specification.

  1. If the Zip archive is split into multiple files or spans multiple volumes, as defined in the [ZIP] specification, then return an error and terminate this algorithm.

  2. If the Zip archive is encrypted, as defined in [Zip], return an error and terminate this algorithm.

  3. Otherwise, return true.

9.1.2. Rule for Extracting File Data from a File Entry

The rule for extracting file data from a file entry is as follows:

  1. Let path be the zip relative path that identifies the file entry being sought.

  2. Let file entry be the file entry identified by the path.

  3. Let file be the result of decompressing (or extracting) the file data from file entry using [Zip].

  4. Return file.

Note: For efficiency, a user agent can extract specific files as they are needed for processing rather than extracting all the files at once. As a security precaution, implementations are discouraged from extracting file entries from un-trusted widgets directly onto the file system. Instead, implementations could use, for example, a virtual file system or mapping to access files inside a widget package.

9.1.3. Rule for Finding a File Within a Widget Package

The rule for finding a file within a widget package is given in the following algorithm. The algorithm returns either a processable file, null, or an error.

For the sake of comparison and matching, it is recommended that a user agent treat all Zip relative paths as [UTF-8].

Note: This specification does not define how links in documents other than the configuration document are to be dereferenced. For handling links in other documents, such as (X)HTML, CSS, SVG, etc., please refer to the [Widgets-URI] specification.

  1. Let path be the path to the file entry being sought by the user agent.

  2. If path is not a valid path, return an error and terminate this algorithm.

  3. If the path starts with a U+002F SOLIDUS (e.g., "/style/master.css"), then remove the first U+002F SOLIDUS from path.

  4. Let path-components be the result of splitting path at each occurrence of a U+002F SOLIDUS character, removing that U+002F SOLIDUS character in the process.

  5. if the first item in path-components case-sensitively matches the string "locales", then:

    1. If the path-components does not contain a second item, then return null.

    2. If the second item in path-components is not a valid language-range, then return null and terminate this algorithm.

    3. Otherwise, continue.

  6. For each lang-range in the user agent locales:

    1. Let path be the concatenation of the string "locales/", the lang-range, a U+002F SOLIDUS character, and the path (e.g., locales/en-us/cats.png, where "en-us" is the lang-range and "cats.png" is the path).

    2. If path case-sensitively matches the file name field of a file entry within the widget package that is a folder, then return an error and terminate this algorithm.

    3. If path case-sensitively matches the file name field of a file entry within the widget package that is a file, let file be the result of applying the rule for extracting file data from a file entry to path.

    4. If file is a processable file, then return file and terminate this algorithm.

    5. If the path points to a file entry that is not a processable file, then return an error and terminate this algorithm.

  7. If every lang-range in the user agent locales have been searched, then search for a file entry whose file name field matches path from the root of the widget package:

    1. If path points to a file entry within the widget package that is a folder, then return an error and terminate this algorithm.

    2. If path points to a file entry within the widget package that is a file, let file be the result of applying the rule for extracting file data from a file entry to path.

    3. If file is a processable file, then return file and terminate this algorithm.

    4. If the path points to a file entry that is not a processable file, then return an error and terminate this algorithm.

  8. Otherwise, return null.

9.1.4. Rule for determining directionality

The rule for determining directionality is given in the following algorithm. The algorithm always returns one of the valid directional indicators as a string.

  1. Let element be the element to be processed.

  2. If element is the root element of the configuration document:

    1. If it does not contain a dir attribute, return "ltr" and terminate this algorithm.

    2. if it does contain a dir attribute, and the value of the attribute case-sensitively matches one of the valid directional indicators, return the value of the attribute and terminate this algorithm. if the value does not case-sensitively matches one of the valid directional indicators, return the "ltr" and terminate this algorithm.

  3. If element does not contain a dir attribute, recursively apply rule for determining directionality to the direct parent element of element and return the result.

  4. If element contains a dir attribute, let direction be the result of applying the rule for getting a single attribute value to the dir attribute of element:

    1. If direction case-sensitively matches one of the valid directional indicators, return direction.

    2. If direction did not case-sensitively match one of the valid directional indicators, apply the rule for determining directionality to the direct parent element of element and return the result .

9.1.5. Rule for Getting a Single Attribute Value

The rule for getting a single attribute value is given in the following algorithm. The algorithm always returns either a string or a localizable string, which can be empty.

  1. Let attribute be the attribute to be processed.

  2. Let value be the value of the attribute to be processed.

  3. In value, replace any sequences of space characters (in any order) with a single U+0020 SPACE character.

  4. In value, remove any leading or trailing U+0020 SPACE characters.

  5. If the attribute is not a displayable-string attribute, then let result be a string that contains the value of value.

  6. Otherwise, if and only if the attribute is a displayable-string attribute:

    1. Let result be a localizable string that contains the value of value.

    2. Let element be the element that owns attribute.

    3. Let direction be the result of applying the rule for determining directionality to element.

    4. Associate direction with result.

    5. Let lang be the language tag derived from having processed the xml:lang attribute on either element, or in element's ancestor chain as per [XML]. If xml:lang was not used anywhere in the ancestor chain, then let lang be an empty string.

    6. Associate lang with result.

  7. Return result.

9.1.6. Rule for Getting a List of Keywords From an Attribute

The rule for getting a list of keywords from an attribute is given by the following algorithm. The algorithm takes a string as input, and returns a list of strings which can be empty.

  1. Let result be the value of the attribute to be processed.

  2. In result, replace any sequences of space characters (in any order) with a single U+0020 SPACE character.

  3. In result, remove any leading or trailing U+0020 SPACE character.

  4. In result, split the string at each occurrence of a U+0020 character, removing that U+0020 character in the process.

  5. Return result.

9.1.7. Rule for Verifying a File Entry

The rule for verifying a file entry is given in the following algorithm. The algorithm always returns a boolean value.

For the file entry, check the following data in the local file header.

  1. If the value of the CRC-32 field (defined in the [ZIP] specification) fails a CRC-32 check, return false and terminate this algorithm.

  2. The file name field is an empty string, return false and terminate this algorithm.

  3. The file name field contains Zip forbidden characters, return false and terminate this algorithm.

  4. The file name field is a sequence exclusively composed of (one or more) space characters or a mixed sequence of space characters and U+002E FULL STOP (".") (e.g. " . .   "), return false and terminate this algorithm.

  5. The file name field is an invalid Zip relative path, return false and terminate this algorithm.

  6. return true.

9.1.8. Rule for Getting Text Content

The rule for getting text content is given in the following algorithm. The algorithm always returns a localizable string, which can be empty.

  1. Let input be the Element to be processed.

  2. Let bidi-text be an empty localizable string.

  3. If input has no child nodes, return an bidi-text and terminate this algorithm.

  4. Let lang be the language tag derived from having processed the xml:lang attribute on either input, or in input's ancestor chain as per [XML]. If xml:lang was not used anywhere in the ancestor chain, then let lang be an empty string.

  5. Let direction be the result of applying the rule for determining directionality to input.

  6. For each child node of input:

    1. If the current child node is a text node, the associate direction and lang with the current child node.

    2. If the current child node is an Element, recursively apply the rule for getting text content:

      1. Take all the text nodes, in order, from the returned localizable string and and associate them with bidi-text.
  7. Associate lang and direction with bidi-text.

  8. return bidi-text.

For example, the following configuration document:

<?xml version="1.0" encoding="UTF-8"?>
<widget xmlns="http://www.w3.org/ns/widgets" dir="rtl" xml:lang="fr"> <name dir="rlo"> <span xml:lang="jp">Hello <span dir="rtl"><span dir="rlo"/>backwards</span></span> </name> </widget>

When the rule for getting text content is applied could be logically represented as pseudo code:


{dir: rlo,
 lang: fr,
 textNodes:
	[{data: "Hello ",
     direction: "rlo",
     lang: "jp"},
              
     {data: "backwards",
      direction: "rtl",
      lang: "jp"}]
}

9.1.9. Rule for Getting Text Content with Normalized White Space

The rule for getting text content with normalized white space is given in the following algorithm. The algorithm always returns a string, which can be empty.

  1. Let input be the Element to be processed.

  2. Let result be the result of applying the rule for getting text content to input.

  3. In result, convert any sequence of one or more Unicode white space characters into a single U+0020 SPACE.

  4. In result, remove any leading or trailing U+0020 SPACE character.

  5. Return result.

For example, the user agent would ignore the author and blink elements, but their Text nodes would be extracted (together with directional and language information):

<widget xmlns="http://www.w3.org/ns/widgets"
 		 xmlns:x="http://x.x.example/x" xml:lang="en">
   <name>
     The <blink xml:lang="en-us">Awesome</blink>
     <author email="dude@example.com">Super <x:blink dir="rtl">Dude</x:blink></author>
     Widget</name>
</widget>

The resulting widget name would be "The Awesome Super Dude Widget" (please note that Dude is rendered right-to-left).

9.1.10. Rule for Parsing a Non-negative Integer

The rule for parsing a non-negative integer is given in the following algorithm. This algorithm returns the number zero, a positive integer, or an error.

  1. Let input be the string being parsed.

  2. Let result have the value 0.

  3. If the length of input is 0, return an error.

  4. Let position be a pointer into input, initially pointing at first character of the string.

  5. Let nextchar be the character in input at position.

  6. If the nextchar is one of the space characters, increment position. If position is past the end of input, return an error and terminate this algorithm. Otherwise, go to step 5 in this algorithm.

  7. If the nextchar is not one of U+0030 (0) .. U+0039 (9), then return result.

  8. If the nextchar is one of U+0030 (0) .. U+0039 (9):

    1. Multiply result by ten.

    2. Add the value of the nextchar to result.

    3. Increment position.

    4. If position is not past the end of input, go to 5 in this algorithm.

  9. Return result.

9.1.11. Rule for Identifying the Media Type of a File

The rule for identifying the media type of a file is given by the following algorithm. The algorithm always returns a string.

Note:This rule is only to be applied when explicitly instructed to by the specification (e.g., during Step 7). There are situations where alternative means are defined by the specification to identify the media type of a file (e.g., by the presence of the content element's type attribute or deriving the media type of a default start file from the default start files table).

  1. Let file be the file to be processed.

  2. Let content-type be an empty string.

  3. Let extension be an empty string.

  4. Let name be the file-name component of the zip relative path that identifies the file.

    For example, the name for "some/zip/rel/path/hello.png" would be "hello.png".

  5. If the first character of the name is a U+002E 'FULL STOP' character, and the file name contains no other U+002E 'FULL STOP' character, then go to step 10 of this algorithm.

    For example, if the name is ".htaccess", jump to step 10 and derive the media type using the [SNIFF] specification.

  6. If the first character or last character of the name is not a U+002E 'FULL STOP' character, but name contains one or more U+002E 'FULL STOP' characters, then let extension be the sequence of characters from the last U+002E 'FULL STOP' (inclusive) to the end of name.

    The value of extension for the file name "cat.html" would be ".html". The value of extension for "...html" would be ".html". And the value of extension for "hello." would be an empty string.

  7. If the first character of the name is a U+002E 'FULL STOP' character, and the file name contains another U+002E 'FULL STOP' character, then let extension be the sequence of characters from the last U+002E 'FULL STOP' (inclusive) to the end of name (if any).

    For example, if the name is ".myhidden.html", then the extension would be ".html".

  8. If extension is an empty string, go to step 10 in this algorithm.

  9. Check that each character in the extension is either in the U+0041-U+005A range or in the U+0061-U+007A range (ASCII uppercase and lowercase characters, respectively) or in the U+0030-U+0039 range (ASCII numbers 0 to 9):

    1. If any character in the extension is outside the U+0041-U+005A range or the U+0061-U+007A range or the U+0030-U+0039 range, then go to step 10 in this algorithm.

      For example, if the extension is ".pñg", the go to step 10 in this algorithm.

    2. If all characters in the extension are in any of the U+0041-U+005A range or in the U+0061-U+007A range or the U+0030-U+0039 range (e.g., "Mp3"), then attempt to case-insensitive match the value of extension to one of the values in the file extension column in the file identification table. If there is a match, then return let content-type be the corresponding value from the media type column.

    3. Go to step 11.

  10. Let content-type be the result of processing file through the [SNIFF] specification.

  11. Return the value of content-type.
File Identification Table
file extension media type
.html text/html
.htm text/html
.css text/css
.js application/javascript
.xml application/xml
.txt text/plain
.wav audio/x-wav
.xhtml application/xhtml+xml
.xht application/xhtml+xml
.gif image/gif
.png image/png
.ico image/vnd.microsoft.icon
.svg image/svg+xml
.jpg image/jpeg
.mp3 audio/mpeg

It is optional for a user agent to support the media types given in the file identification table.

9.1.12. Rule for Deriving the user agent locales

The rule for deriving the user agent locales is as follows:

  1. Let unprocessed locales list be a comma-separated list that contains the end-user's language ranges.

  2. For each range in the unprocessed locales list:

    1. If this range begins with the subtag '*' (e.g. "*-us" or just "*"), or contains any space characters, skip all the steps in this algorithm below, and move onto the next range.

    2. If this range begins with the subtag "i" or the range is marked as "deprecated" in the IANA Language Subtag Registry, skip all the steps in this algorithm below, and move onto the next range.

    3. If this range contains any subtag '*', remove the '*' and its preceding hyphen (U+002F) (e.g., 'en-*-us' becomes 'en-us').

    4. While range contains subtags:

      1. Add the value of the range to the user agent locales.

      2. Remove the right most subtag from range and append the resulting value to user agent locales. Continue removing the right most subtag and adding the result to user agent locales until there are no subtags left in range.

        For example, if the range was "zh-hans-cn", then the user agent locales become "zh-hans-cn,zh-hans,zh".

      3. Move onto the next range and go to step 1 in this algorithm.

  3. Append the value "*" to the end of user agent locales.

For example, an unprocessed locales list that contains "en-us,en-au,en,fr-ca,zh-hans-cn" would result in a user agent locales that contains "en-us,en,en-au,en,en,fr-ca,fr,zh-hans-cn,zh-hans,zh,*".

For example, an unprocessed locales list that contains "en-us,en,fr-ca,en,en-ca" would result in a user agent locales that contains "en-us,en,en,fr-ca,fr,en,en-ca,en,*".

9.1.13. Rule for Determining if a Potential Zip Archive is a Zip Archive

The rule for determining if a potential Zip archive is a Zip archive is given by the following algorithm.

  1. Let potential archive be the acquired resource.

  2. Check if the first four bytes of potential archive match the magic numbers for a Zip archive (50 4B 03 04).

  3. If the first four bytes do not match the magic numbers for a Zip archive, then return an error.

  4. Otherwise, return true.

Note: A user agent can inspect the potential archive once it has acquired the first four bytes of the potential Zip archive or can wait until all the data of the potential Zip archive has been completely acquired.

Step 1 - Acquire a Potential Zip Archive

Step 1 involves acquiring a potential Zip archive and confirming that it is a Zip archive by applying the rule for determining if a potential Zip archive is a Zip archive. A user agent will acquire a potential Zip archive from a data transfer protocol that either labels resources with a media type (e.g. [HTTP]) or from a data transfer protocol that does not label resources with a media type (e.g., BitTorrent or Bluetooth).

9.1.14. Acquisition of a Potential Zip archive Labeled with a Media Type

It is recommended that a user agent support acquisition of a potential Zip archive from a protocol that labels resources with a media type (e.g., getting a widget package over [HTTP]). If a user agent supports acquisition of a potential Zip archive from a protocol that labels resources with a media type, then the user agent needs to process resources labeled with the valid widget media type (application/widget), regardless of whether the resource contains a file extension or not, by applying the rule for determining if a potential Zip archive is a Zip archive. During the acquisition of a potential Zip archive labeled with a media type, unless the user agent supports legacy or proprietary media types, unsupported media types are in error and the user agent must treat the potential Zip archive as an invalid widget package.

If the result of the user agent applying the rule for determining if a potential Zip archive is a Zip archive is true, meaning that the potential Zip archive is a Zip archive, then the user agent must proceed to Step 2. Otherwise, if an error is returned, the user agent must treat the potential Zip archive as an invalid widget package.

For example, in [HTTP], where the Content-Type header matches application/widget.

If the protocol used for acquisition of a potential Zip archive does not provide, or otherwise include, a media type, then a user agent should treat the acquired potential Zip archive as if it has been acquired from a protocol that does not label resources with a media type.

In this example, the media type of the Content-Type is not one supported by the user agent, so the user agent would treat the potential Zip archive as an invalid widget package:

Request
GET /foo.wgt HTTP/1.1
Host: www.example.com
Accept: application/widget
Response
HTTP/1.1 200 OK
Date: Tue, 04 Sep 2007 00:00:38 GMT
Last-Modified: Mon, 03 Sep 2007 06:47:19 GMT
Content-Length: 1337
Content-Type: application/x-gadget

9.1.15. Acquisition of Potential Zip Archive not Labeled with a Media Type

When acquiring a potential Zip archive that has not been labeled with a media type (e.g., from a file system), a user agent should attempt to process the resource regardless of the file extension (including situations when the file extension is absent) by applying the rule for determining if a potential Zip archive is a Zip archive. If the rule for determining if a potential Zip archive is a Zip archive return true, proceed to Step 2. Otherwise, if an error was returned, the user agent must treat the potential Zip archive as an invalid widget package.

Step 2 - Verify the Zip Archive

To verify that a Zip archive and its file entries conform to this specification, a user agent must apply the rule for verifying a zip archive. If the rule for verifying a zip archive returns true, then the user agent must go to Step 3. Otherwise, if the rule for verifying a zip archive returns an error, then the user agent must treat the Zip archive as an invalid widget package.

Step 3 - Set the Configuration Defaults

In Step 3, a user agent must set the following variables and default values as defined in the table of configuration defaults.

Note: When a null value is assigned to a variable in the table of configuration defaults, a user agent needs to treats the value as null (i.e., not as an empty string and not as the text string "null").

Table of Configuration Defaults
Variable Type Overridden in Description
author email String Step 7 The value of the author element's email attribute (if any).
author href IRI Step 7 The value of the author element's href attribute (if any).
author name String Step 7 The text content of the author element (if any).
feature list List Step 7 A list of features that correspond to features that were requested via feature elements (if any). Each item in the list corresponds to a feature element's name attribute, whether it is required, and any associated parameters (if any).
icons List of file entries Step 7, Step 9 The icons of the widget as they correspond to the default icons and to the occurrence of custom icons that are supported by the widget package (if any).
start file encoding String Step 7 The character encoding of the custom start file, corresponding to either the content element's encoding attribute (if any), or the default encoding.
start file content-type String Step 7 The media type of the start file, corresponding to the content element's type attribute or to a media type derived from the default start files table.
widget config doc File Step 6 The file that is the configuration document for the widget package.
widget description String Step 7 The text content of the description element in the configuration document.
widget height positive number Step 7 The value of the widget element's height attribute in the configuration document (if any).
widget id String Step 7

The value of the widget element's id attribute in the configuration document (if any).

widget license String Step 7 The text content of the license element in the configuration document (if any).
widget license file File Step 7 A file derived if the value of the license element's href is a Zip relative path to a file within the widget package.
widget license href IRI Step 7 The value of the license element's href attribute in the configuration document (if any).
widget name Localizable String Step 7 The text content of the name element in the configuration document (if any).
widget preferences List Step 7

The widget's preferences, corresponding to the preference elements in the configuration document (if any).

Unless an end-user explicitly requests that these values be reverted to the values as declared in the configuration document, a user agent must not reset the value of the widget preferences variable on subsequent initializations of the widget.

widget short name Localizable String Step 7 The value of the name element's short attribute in the configuration document (if any).
Variable Type Overridden by Description
widget version Localizable String Step 7 The value of the widget element's version attribute in the configuration document (if any).
widget width positive number Step 7 The value of the widget element's width attribute in the configuration document (if any).
widget window modes List of strings Step 7 The value of the widget element's viewmodes attribute in the configuration document (if any).
widget start file File entry Step 7, Step 8 The start file for the widget package, corresponding to either one of the default start files table or the file identified by the content element's src attribute.
user agent locales

List of strings

Step 5 A list of language tags.

Step 4 - Locate and Process the Digital Signature

If the user agent does not support [Widgets-DigSig], then the user agent must skip Step 4 and go to Step 5. Otherwise, the user agent must apply the algorithm to locate digital signatures, which is defined in the [Widgets-DigSig] specification under the section named Locating and Processing Digital Signatures.

Step 5 - Derive the User Agent's Locales

The end-user's language ranges represents the end-user's preferred languages and regional settings, which are derived from the operating system or directly from the user agent. As there are numerous ways a user agent can derive the end-user's preferred languages and regional settings, the means by which those values are derived are beyond the scope of this specification and left up to the implementation.

During the rule for deriving the user agent locales defined below, the user agent will need to construct a list unprocessed locales. Each item in the unprocessed locales is a string in lowercase form, that conforms to the production of a Language-Tag, as defined in the [BCP47] specification. A string that conforms to the production of a Language-Tag is referred to as a language range [BCP47] (e.g. 'en-au', which is the range of English as spoken in Australia, and 'fr-ca', which is the range of French as spoken in Canada, etc.). A language range is composed of one or more subtags that are delimited by a U+002D HYPHEN-MINUS ("-").

The first item of the unprocessed locales represents the user's most preferred language range (i.e., the language/region combination the user agent assumes the end-user most wants to see content in), followed by the next most preferred language range, and so forth.

For example, in an unprocessed locales list that contains 'en-us,en,fr,es', English as spoken in the United States is preferred over English, and English is preferred over French, and French is preferred over Spanish, and Spanish is preferred over default content.

For example, the end-user may have specified her preferred languages and regional settings at install time by selecting a preferred language, or languages from a list, or a list of preferred languages and regional settings could have been dynamically derived from the end-user's geographical location, etc.

In Step 5, the user agent must apply the rule for deriving the user agent locales.

Step 6 - Locate the Configuration Document

This step involves searching within the Zip archive for a configuration document.

In Step 6, a user agent must apply the algorithm to locate the configuration document.

The algorithm to locate the configuration document is as follows:

  1. Search at the root of the widget package for a file entry whose file name field case-sensitively matches the valid configuration document file name (config.xml).

  2. If a match is made, then let widget config doc to the result of applying rule for extracting file data from a file entry to the matching file entry. If no match is made (meaning that widget config doc is null), then treat the Zip archive as an invalid widget package.

Step 7 - Process the Configuration Document

The purpose of processing the configuration document is to override the values in the table of configuration defaults, which are used during initialization and at runtime, and to select the appropriate localized content (if any) to be presented to the end user.

In conjunction to the algorithm for processing a configuration document given below, this section firstly defines some terminology used by the processing algorithm and describes how localized elements are processed.

During Step 7, a user agent must apply the algorithm to process a configuration document.

9.1.16. Terminology Used in Processing Algorithm

In the algorithm to process a configuration document, the term in error is used to mean that an element, or attribute, or file in a configuration document is non-conforming to the rules of this specification. How an element or an attribute is to be treated when it is in error is always given when the term is used; but will generally require the user agent to ignore any element, attribute, or file that is in error.

To ignore means to act as if the element, attribute, or file that is in error is absent (i.e., not declared or included by the author) in the widget package or configuration document. A user agent must, however, keep a record of all element types it has attempted to process even if they were ignored (this is to determine if the user agent has attempted to process an element of a given type already).

In the case the user agent is asked to ignore an [XML] element or node, a user agent must :

  1. Stop processing the current element, ignoring all of the element's attributes and child nodes (if any), and proceed to the next element in the elements list.

  2. Make a record that it has attempted to process an element of that type.

In the following example, the user agent ignores both content elements. The user agent ignores the first because it lacks a src attribute. The user agent ignores the second because it is not the first content element to be encountered by the user agent.

<widget xmlns="http://www.w3.org/ns/widgets">
	<!-- User agent ignores the first, but records 
       that it has attempted to process a content element 
   -->

  <content/>
  
  <!-- The use agent knows that it has previously attempted 
       to process a content element, hence this content element 
       is ignored. 
   -->
 
  <content src="cats.html"/>

</widget> 

To associate means that two or more pieces of information are bound and stored together for the purpose of later processing (e.g., the name of a feature, and if it is required, and any associated parameters). How associated data is represented is left up-to the implementation (e.g., a user agent could use an array, an object, a hash map, etc.).

The lookup algorithm is defined in [BCP47] (see RFC 5646). It is used in this Step to match localized content in the configuration document to the language ranges held by the user agent locales (if any).

9.1.17. Algorithm to Process a Configuration Document

The algorithm to process a configuration document is as follows:

  1. Let doc be the result of loading the widget config doc as a [DOM3Core] Document using an [XML] parser that is both [XMLNS]-aware and xml:lang aware.

  2. If doc is not namespace well-formed [XML], then the user agent must terminate this algorithm and treat this widget package as an invalid widget package.

  3. Let root element be the documentElement of doc.

  4. If the root element is not a widget element in the widget namespace, then the user agent must terminate this algorithm and treat this widget package as an invalid widget package.

  5. Otherwise, the element is a widget element:

    1. If the defaultlocale attribute is used, then let default locale be the result of applying the rule for getting a single attribute value to the defaultlocale attribute:

      1. If the default locale is in error or an empty string or already contained by the user agent locales list, then the user agent must ignore the defaultlocale attribute.

      2. If potential default locale is a valid language tag and the user agent locales does not contain the value of default locale, the user agent must prepend the value of potential default locale into the the user agent locales list as the second-last item (i.e., at position length - 1).

        For example, if the default locale is the value "fr", and the user agent locales contains the values "jp,us,*", then the user agent locales list becomes "jp,us,fr,*".

        For example, if the default locale is the value "en", and the user agent locales only contains the value "*", then the user agent locales list becomes "en,*".

        For example, if the default locale is the value "en", and the user agent locales already contains the values "en,*", then the user agent would ignore the default locale because it is already contained by the user agent locales list.

    2. If the id attribute is used, then let id be the result of applying the rule for getting a single attribute value to the id attribute. If id is a valid IRI, then let widget id be the value of the id. If the id is in error, then the user agent must ignore the attribute.

    3. If the version attribute is used, then let version value be the result of applying the rule for getting a single attribute value to the version attribute. If the version is an empty string, then the user agent must ignore the attribute; otherwise, let widget version be the value of version value.

    4. If the height attribute is used, then let normalized height be the result of applying the rule for parsing a non-negative integer to the value of the attribute. If the normalized height is not in error and greater than 0, then let widget height be the value of normalized height. If the height attribute is in error, then the user agent must ignore the attribute.

    5. If the width attribute is used, then let normalized width be the result of applying the rule for parsing a non-negative integer to the value of the attribute. If the normalized width is not in error and greater than 0, then let widget width be the value of normalized width. If the width attribute is in error, then the user agent must ignore the attribute.

    6. If the viewmodes attribute is used, then the user agent must let viewmodes list be the result of applying the rule for getting a list of keywords from an attribute:

      1. From the viewmode list, remove any unsupported items.

      2. From the viewmode list, remove any duplicated items from right to left.

        For example, viewmode list with a value of "windowed fullscreen windowed floating fullscreen windowed" would become "windowed fullscreen floating".

      3. Let widget window modes be the value of viewmodes list.

  6. If the widget element does not contain any child elements, then the user agent must terminate this algorithm and go to Step 8.

  7. Otherwise, let element list be an empty list.

  8. For each range in the user agent locales, starting from the first and moving to the last:

    1. If the value of range is not "*", then retaining document order, let matching elements be the result of applying lookup to the child elements that are defined as being localizable via xml:lang that are direct descendents of the root element and whose xml:lang attribute matches the current range. Append matching elements to the element list.

      Note: In the context of this specification, the above conformance requirement is intended to match the name, description, and license elements. However, it is written in an abstract manner to provide a hook for future specifications that want to define elements that also support being localizable via xml:lang.

    2. If the value of range is "*", retaining document order, let unlocalized elements be all child elements that are direct descendents of the root element that do not have an implicit or explicit xml:lang attribute (i.e., match default content). Append unlocalized elements to the element list.

      For example, consider the following configuration document.

      <widget xmlns="http://www.w3.org/ns/widgets">
      <name>El Widget!</name>
      <name xml:lang="fr">Le Widget</name>
      <name xml:lang="en">The Widget</name>
      </widget> 

      For a use agent whose user agent locales contains "en,fr,*", the matching elements would be in the following order:

      1. <name xml:lang="en">The Widget</name>
      2. <name xml:lang="fr">Le Widget</name>
      3. <name>El Widget!</name>

      For a use agent whose user agent locales contains "en,*", the matching elements would be in the following order:

      1. <name xml:lang="en">The Widget</name>
      2. <name>El Widget!</name>

      For a use agent whose user agent locales contains "jp,*", the matching elements would be in the following order:

      1. <name>El Widget!</name>
  9. For each element in the elements list, if the element is one of the following:

    A name element:

    If this is not the first name element encountered by the user agent, then the user agent must ignore this element.

    If this is the first name element encountered by the user agent, then the user agent must:

    1. Record that an attempt has been made by the user agent to process a name element.

    2. Let widget name be the result of applying the rule for getting text content with normalized white space to this element.

    3. If the short attribute is used, then let widget short name be the result of applying the rule for getting a single attribute value to the short attribute.

    A description element:

    If this is not the first description element encountered by the user agent, then the user agent must ignore this element.

    If this is the first description element encountered by the user agent, then the user agent must:

    1. Record that an attempt has been made by the user agent to process a description element.

    2. let widget description be the result of applying the rule for getting text content to this element.

    A license element:

    If this is not the first license element encountered by the user agent, then the user agent must ignore this element.

    If this is the first license element used, then the user agent must:

    1. Record that an attempt has been made by the user agent to process a license element.

    2. Let license text be the result of applying the rule for getting text content to this element. Associate license text with widget license.

    3. If the href attribute is used, then let potential license href be the result of applying the rule for getting a single attribute value to the href attribute.

    4. If potential license href is not a valid IRI or a valid path, then the href attribute is in error and the user agent must ignore the attribute.

    5. If potential license href is a valid IRI, then let widget license href be the value of potential license href.

    6. If license href is a valid path, then let file be the result of applying the rule for finding a file within a widget package to license href.

    7. If file is not a processable file, as determined by applying the rule for identifying the media type of a file, then ignore this element.

    8. Otherwise, let widget license file be the value of file.

    An icon element:

    If the src attribute of this icon element is absent, then the user agent must ignore this element.

    Let path be the result of applying the rule for getting a single attribute value to the src attribute of this icon element. If path is not a valid path or is an empty string, then the user agent must ignore this element.

    Let file be the result of applying the rule for finding a file within a widget package to path. If file is not a processable file, as determined by applying the rule for identifying the media type of a file, or already exists in the icons list, then the user agent must ignore this element.

    Otherwise,

    1. If the height attribute is used, then let potential height be the result of applying the rule for parsing a non-negative integer to the attribute's value. If the potential height is not in error and greater than 0, then associate the potential height with file. Otherwise, the height attribute is in error and the user agent must ignore the attribute.

    2. If the width attribute is used, then let potential width be the result of applying the rule for parsing a non-negative integer to the attribute's value. If the potential width is not in error and greater than 0, then associate the potential width with file. Otherwise, the width attribute is in error and the user agent must ignore the attribute.

    3. Add file and any associated potential width and/or potential height to the list of icons.

    An author element:

    If this is not the first author element encountered by the user agent, then the user agent must ignore this element.

    If this is the first author element used, then the user agent must:

    1. Record that an attempt has been made by the user agent to process a author element.

    2. If the href attribute is used, then let href-value be the value of applying the rule for getting a single attribute value to the href attribute.

    3. If href-value is a valid IRI, then let author href be the value of the href attribute. Otherwise, if href-value is not a valid IRI, then ignore the href attribute.

    4. If the email attribute is used, then let author email be the result of applying the rule for getting a single attribute value to the email attribute.

    5. Let author name be the result of applying the rule for getting text content with normalized white space to this element.

    A preference element:

    If a value attribute of the preference element is used, but the name attribute is absent, then this preference element is in error and the user agent must ignore this element. Otherwise, the user agent must:

    1. Let name be the result of applying the rule for getting a single attribute value to the name attribute.

    2. If the name is an empty string, then this element is in error; ignore this element.

    3. If widget preferences already contains a preference whose name case-sensitively matches the value of name, then this element is in error; ignore this element.

    4. If name was not in error, let preference be an empty object.

    5. Associate name with preference.

    6. Let value be the result of applying the rule for getting a single attribute value to the value attribute.

    7. Associate value with preference.

    8. If a readonly attribute is used, then let readonly be the result of applying the rule for getting a single attribute value to the readonly attribute. If readonly is not a valid boolean value, then let the value of readonly be the value 'false'.

    9. Associate readonly with the preference.

    10. Add the preference and the associated name, value and readonly variables the list of widget preferences.

    A content element:

    If this is not the first content element encountered by the user agent, then the user agent must ignore this element.

    If this is the first content element, then the user agent must:

    1. Record that an attempt has been made by the user agent to process a content element.

    2. If the src attribute of the content element is absent or an empty string, then the user agent must ignore this element.

    3. Let path be the result of applying the rule for getting a single attribute value to the value of the src attribute.

    4. If path is not a valid path, then the user agent must ignore this element.

    5. If path is a valid path, then let file be the result of applying the rule for finding a file within a widget package to path. If file is null or in error, then the user agent must ignore this element.

    6. If the type attribute of the content element is absent, then check if file is supported by the user agent by applying the rule for identifying the media type of a file. If the file is supported, then let the widget start file be the file referenced by the src attribute and let start file content-type be the supported media type as was derived by applying the rule for identifying the media type of a file.

    7. If the encoding attribute is used, then let content-encoding be the result of applying the rule for getting a single attribute value to the value of the encoding attribute. If the character encoding represented by the value of content-encoding is supported by the user agent, then let the start file encoding be the value of content-encoding. If content-encoding is an empty string or unsupported by the user agent, then a user agent must ignore the encoding attribute.

    8. If the type attribute is used, then let content-type be the result of applying the rule for getting a single attribute value to the value of the type attribute. If the value of content-type is a media type supported by the user agent, then let the start file content-type be the value of content type. If value of content-type is invalid or unsupported by the user agent, then a user agent must treat the widget package as an invalid widget package.

    9. If the start file encoding was set in step 7 of this algorithm as a result of processing a valid and supported value for the content element's encoding attribute, then the user agent must skip this step in this algorithm. Otherwise, if the value of content-type is a media type supported by the user agent and if content-type contains one or more [MIME] parameter components whose purpose is to declare the character encoding of the start file (e.g., the value "text/html;charset=Windows-1252", where charset is a parameter component whose purpose is to declare the character encoding of the start file), then let start file encoding be the value of the last supported parameter components whose purpose is to declare the character encoding of the start file.

      In the following example, the user agent would set the start file encoding to ISO-8859-1 and would ignore the charset parameter used in the type attribute.

      <widget xmlns="http://www.w3.org/ns/widgets"> 
      <content src = "start.php" type = "text/html;charset=Windows-1252" encoding = "ISO-8859-1" />
      </widget>

      In the following example the user agent would set the start file encoding to Windows-1252, if the user agent supports that character encoding.

      <widget xmlns="http://www.w3.org/ns/widgets"> 
        <content src  = "start.php" 
                 type = "text/html;charset=Windows-1252"/>
      </widget>
    A param element:

    If this param element is not a direct child of a feature element, then the user agent must ignore this param element.

    Note: How a param element is to be processed when it is inside a feature element is defined below.

    A feature element:

    The user agent must process a feature element in the following manner:

    1. If the name attribute of this feature element is absent, then the user agent must ignore this element.

    2. Let feature-name be the result of applying the rule for getting a single attribute value to the value of the name attribute.

      Note: This specification allows feature elements with the same name attribute value to be declared more than once. Handling of duplicate feature requests is left up to the implementation or the specification that defines the feature.

    3. If a required attribute is used, then let required-feature be the result of applying the rule for getting a single attribute value to the required attribute. If the required attribute is not used or if required-feature is not a valid boolean value, then let the value of required-feature be the value 'true'.

    4. If feature-name is not a valid IRI, and required-feature is true, then the user agent must treat this widget as an invalid widget package.

    5. If feature-name is not a valid IRI, and required-feature is false, then the user agent must ignore this element.

    6. If feature-name is not supported by the user agent, and required-feature is true, then the user agent must treat this widget as an invalid widget package.

    7. If feature-name is not supported by the user agent, and required-feature is false, then the user agent must ignore this element.

    8. Associate the value of required-feature with feature-name.

    9. If the feature element contains any param elements as direct descendants, then, for each child param element that is a direct descendent of this feature element, starting from the first moving to the last in document order:

      1. If a value attribute is used, but the name attribute is absent, then this param element is in error and the user agent must ignore this element.

      2. If a name attribute is used, but the value attribute is absent, then this param element is in error and the user agent must ignore this element.

      3. Let param-name be the result of applying the rule for getting a single attribute value to the name attribute. If the param-name is an empty string, then this param element is in error and the user agent must ignore this element.

      4. If, and only if, param-name is not in error or an empty string, then let param-value be the result of applying the rule for getting a single attribute value to the value attribute.

      5. Associate param-name and param-value with feature-name.

    10. Append feature-name, and any associated required-feature, and associated parameters, to the feature list.

    Any other type of element:

    If the user agent supports the element, then the user agent must process it in accordance with whatever specification defines that element (if any). Otherwise, the user agent must ignore the element.

Step 8 - Locate the Start File

If widget start file of the table of configuration defaults contains a file (i.e. widget start file is not null), then a user agent must skip Step 8 and go to Step 9.

If widget start file does not contain a file, the user agent must apply the algorithm to locate a default start file.

The algorithm to locate a default start file is as follows:

  1. For each file name in the default start files table (from top to bottom) that has a media type that is supported by the user agent:

    1. Let potential-start-file be the result of applying the rule for finding a file within a widget package to file name.

    2. If potential-start-file is null or in error, ignore this file name and move onto the next file name in the default start files table.

    3. If potential-start-file is a file, then:

      1. Let widget start file be the value of potential-start-file.

      2. Let start file content-type be the media type given in the media type column of the default start files table.

      3. Terminate this algorithm and go to Step 9.

  2. If after searching for every file in the default start files table no default start file is found, then treat this widget as an invalid widget package.

Step 9 - Process the Default Icons

This step describes how to locate the default icons.

In Step 9, a user agent must apply the algorithm to locate the default icons.

The algorithm to locate the default icons is as follows:

  1. For each file name in the default icons table (from top to bottom) that has a media type that is supported by the user agent:

    1. Let potential-icon be the result of applying the rule for finding a file within a widget package to file name.

    2. If the potential-icon is a processable file, determined by the media type given in the media type column of the default icons table, and the potential-icon does not already exist in the icons list of the table of configuration defaults, then append the value of potential-icon to the icons list of the table of configuration defaults.

    3. Move onto the next file name in the default icons table.

Appendix

Media Type Registration for application/widget

This appendix is the MIME media type registration for "application/widget". This registration has been approved by the Internet Assigned Numbers Authority (IANA) .

Registration with IANA was conducted in conformance with BCP 13 and W3CRegMedia.

Type name:

application

subtype name:

widget

Required parameters:

None.

Optional parameters:

None.

Encoding considerations:

Widget packages are binary data and thus are encoded for MIME transmission. As a widget package is binary, it requires encoding on transports not capable of handling binary. The same guidelines that apply to application/octet-stream apply to widget packages (see [MIME]).

Security considerations:

In addition to the security considerations specified for Zip files in the [Zip-MIME] registration, there are a number of security considerations that need to be taken into account when dealing with widget packages and configuration documents.

As the configuration document format is [XML] and will commonly be encoded using [Unicode], the security considerations described in [XML-MIME] and [UTR36] apply. In addition, implementers need to impose their own implementation-specific limits on the values of otherwise unconstrained attribute types, e.g. to prevent denial of service attacks, to guard against running out of memory, or to work around platform-specific limitations.

The configuration document allows authors, through the feature element, to request permission to enable third-party runtime components and APIs. As these features are outside the scope of this specification, significant caution needs to be taken when granting a widget the capability to use a feature. Features themselves define their own security considerations.

Widget packages will generally contain ECMAscript, HTML, CSS files, and other media, which are executed in a sand boxed environment. As such, implementers need to be aware of the security implications for the types they support. Specifically, implementers need to consider the security implications outlined in the [CSS-MIME] specification, the [ECMAScript-MIME] specification, and the [HTML-MIME] specification.

As widget packages can contain content that is able to simultaneously interact with the local device and a remote host, implementers need to consider the privacy implications resulting from exposing private information to a remote host. Mitigation and in-depth defensive measures are an implementation responsibility and not prescribed by this specification. However, in designing these measures, implementers are advised to enable user awareness of information sharing, and to provide easy access to interfaces that enable revocation of permissions.

As this specification relies on the standardized heuristics for determining the content type of files defined in the [SNIFF] specification, implementers need to consider the security considerations discussed in the [SNIFF] specification.

As this specification allows for the declaration of IRIs within certain elements of a configuration documents, implementers need to consider the security considerations discussed in the [IRI] specification. Implementations intending to display IRIs and IDNA addresses found in the configuration document are strongly encouraged to follow the security advice given in [UTR36]. This could include, for example, behaving as if the dir attribute had no effect on any IRI attributes, path attributes, and the author element's email attribute.

In addition, user agents need to be careful about trusting path components found in the widget package. Such path components might be interpreted by operating systems as pointing at security critical files outside the widget environment proper, and naive unpacking of widget packages into the file system might lead to undesirable and security relevant effects, such as overwriting of system files.

Interoperability considerations:

Some issues can arise with regards to character encodings of file names, the length of zip relative paths, and the use of certain strings as file names.

This specification does not put a restriction on the byte length of a Zip relative path, so a user agent should to be able to deal with Zip relative paths that have lengths longer than 250 bytes. As some operating systems have restrictions on how long a path length can be, authors need to keep the lengths of relative paths at less than 250 bytes. Unicode code points may require more than one byte to encode a character, which can result in a path whose length is less than 250 characters but whose size is greater than 250 bytes.

Authors need to be aware that, at the time of publication, there are interoperability issues with regards to using characters outside the safe-chars range for file or folder names in a Zip archive when using Zipping tools bundled with operating systems. The interoperability issues have arisen from non-conforming implementations of the [ZIP] specification across operating systems: very few, if any, correctly support encoding file names in Unicode.

In the case where the Zip relative path is encoded using [UTF-8], the language encoding flag (EFS) needs to be set.

If an author chooses to use the utf8-chars, they need to thoroughly test their widgets on various platforms prior to distribution; otherwise it is suggested that authors restrict file and folder names to the safe-chars (characters in the US-ASCII range). In addition, having excessively long path names (e.g. over 120 characters) can also result in interoperability issues on some operating systems.

Authors need to avoid using Zip forbidden characters when naming the files used by a widget. These characters are reserved to maintain interoperability across various file systems and with [URI]s.

Authors need to avoid using the following words as either a folder or a file-name in a Zip relative path as they are reserved by some operating systems (case-insensitive): CON, PRN, AUX, NUL, COM1, COM2, COM3, COM4, COM5, COM6, COM7, COM8, COM9, LPT1, LPT2, LPT3, LPT4, LPT5, LPT6, LPT7, LPT8, LPT9, CLOCKS$. For example, the following names are ok: "CON-tact.txt", "printer.lpt1", "DCOM1.pdf". However, "com3.txt" "Lpt1", "CoM9.gif" would not be.

In addition, authors need to avoid having a "." U+002E FULL STOP as the last character of a file or folder name as some operating systems will remove the character when the file is extracted from the Zip archive onto the device. Furthermore, avoid having the space character (SP) at the start or end of a file name; and take caution when using the "+" U+002B PLUS SIGN, as it might cause issues on some operating systems.

Published specification:

http://www.w3.org/TR/widgets/

Applications that use this media type:

User agents that claim conformance to this specification.

Magic number(s):

50 4B 03 04

File extension(s):

wgt

Macintosh file type code(s):

None.

Person & email address to contact for further information:

Steven Pemberton, member-webapps@w3.org

Intended usage:

Common.

Restrictions on usage:

None.

Author:

The W3C's Web Applications Working Group

Linking To a Widget Package From a HTML Document

This section is non-normative.

This section only applies to HTML user agents [HTML][HTML4] [XHTML1.1].

Auto-discovery enables a user agent to identify and install a widget package that is associated with an HTML page. When a page points to a widget package, user agents should expose the presence of the widget package to the end-user and allow the end-user to install the widget.

The link type "widget" indicates that a link of this type references a document that is a widget package. In HTML, it may be specified for the a, area and link elements to create a hyperlink.

For example:

<a rel="widget"
   href="http://example.org/exampleWidget">
   The Example Widget
</a>

Table of Elements and Their Attributes

This section is non-normative.

This table lists all elements and respective attributes, as well as child-parent relationships, that make up the language of the configuration document format defined in this specification.

Element Description Parent Expected Children Attributes Type
widget The root element of a configuration document. none. xml:lang language
dir keyword
id IRI
version version
height numeric
width numeric
viewmodes Keyword list
defaultlocale language
name The name of the widget. widget xml:lang language
dir keyword
short displayable-string
description Some text that describes the purpose of the widget. widget xml:lang language
dir keyword
author The person or person that created the widget. widget xml:lang language
dir keyword
href IRI
email string
license The license under which the widget is distributed. widget xml:lang language
dir keyword
href IRI or path
icon An iconic representation of the widget. widget none. xml:lang language
dir keyword
src path
width numeric
height numeric
content The means to point to the "main file" of a widget; serves as a boot-strapping mechanism. widget none. xml:lang language
dir keyword
src numeric
type media type
encoding keyword
feature A means to request the availability of a feature, such as an API, that would not normally be part of the default set of features provided by the user agent at runtime. widget xml:lang language
dir keyword
name keyword
required boolean
preference A means to declare a name-value pair that is made available to the widget at runtime. widget none. xml:lang language
dir keyword
name keyword
value keyword
readonly boolean
param A means of declaring a parameter that can be used with a feature. feature none. xml:lang language
dir keyword
name string
value string
span A generic text container which is mainly used for internationalization purposes. xml:lang language
dir keyword

Acknowledgements

Huge thanks to everyone who contributed their time and sent feedback to our public mailing, particularly the i18n WG.

This specification would not exist without the contribution of the following individuals:

Aaron Boodman, Adam Barth, Addison Phillips, Alexander Dreiling, Andrew Sledd, Andrew Welch, Arun Ranganathan, Arthur Barstow, Bárbara Barbosa Neves, Bil Corry, Brian Wilson, Bjoern Hoehrmann, Benoit Suzanne, Bert Bos, Boris Zbarsky, Bradford Lassey, Bryan Sullivan, Cameron McCormack, Cliff Schmidt, Claudio Venezia, Coach Wei, Corin Edwards, Cynthia Shelly, Cyril Concolato, Dan Brickley, Dan Connolly, Daniel Silva, David Clarke, Dean Jackson, David Poehlman, David Pollington, David Rogers, Dominique Hazael-Massieux, Doug Schepers, Ed Voas, Felix Sasaki, Francois Daoust, Frederick Hirsch, Gautam Chandna, Geir Pedersen, Gene Vayngrib, Gorm Haug Eriksen, Guido Grassel, Guenter Klas, Hans S. Tømmerholt, Hari Kumar G, Henri Sivonen, Henry Story, Ian Hickson, Ivan Demarino, Jay Sweeney, Jean-Claude Dufourd, Jeff Decker, Jere Käpyaho, Jim Ley, Jo Rabin, Jon Ferraiolo, Jonas Sicking, Jose Manuel Cantera Fonseca, Josh Soref, Jouni Hakala, Joey Bacalhau, Julian Reschke, Kevin Lawver, Kai Hendry, Krzysztof Maczyński, Lachlan Hunt, Larry Masinter, Laurens Holst, Mark Priestley, Marc Silbey, Marcin Hanclik, Mark Baker, Martin J. Dürst, Michael Cooper, Max Froumentin, Mikko Pohja, Mohamed Zergaoui, Najib Tounsi, Noah Mendelsohn, Oguz Kupusoglu, Ola Andersson, Olli Immonen, Paddy Byers, Paul Libbrecht , Philipp Heltewig, Philip Taylor, Rainer Hillebrand, Robert Sayre, Rune F. Halvorsen, Samuel Santos, Scott Wilson, Sean Mullan, Sigbjorn Finne, Simon Pieters, Stephen Paul Weber, Stephen Jolly, Stephane Sire, Steven Faulkner, Thomas Landspurg, Thomas Roessler, Tiago Neves, William Edney, Yoan Blanc, Yves Savourel.

Special thanks go to Arve Bersvendsen, Robin Berjon, and Charles McCathieNevile who helped edit various versions of this specification.

A big thanks to Anne van Kesteren for being such a great supporter of this work! Also, huge thanks to the folks at Apple, specially Maciej Stachowiak, who contributed immensely to make all the widgets specs free and open!

Special thanks also to David Håsäther for creating and maintaining the [Widgets-Relax NG Schema] for the configuration document format.

Parts of this document reproduce text and behavior from the [HTML] specification and from the XBL 2.0 specification (as permitted by both specifications by their copyright statements).

Graphic icons used some examples of this specification were created by Yusuke Kamiyamane and are available for use under a Creative Commons Attribution 3.0 license.

This specification is dedicated to the children of India.

Normative References

[ABNF]
Augmented BNF for Syntax Specifications: ABNF. RFC5234. D. Crocker and P. Overell. January 2008.
[BIDI]
Unicode Standard Annex #9: Unicode Bidirectional Algorithm. M. Davis.
[BCP47]
Tags for Identifying Languages, A. Phillips and M. Davis.
[CP437]
IBM CPGID 00437. Code Page 47, Character Encoding Scheme. 1984.
[CSS2]
Cascading Style Sheets Level 2 Specification. B. Bos, I. Hickson, T. Çelik, H. Wium Lie. (Work in Progress).
[CSS-MIME]
The text/css Media Type . RFC 2318. H. Lie, B. Bos, and C. Lilley. IETF. March 1998.
[Deflate]
DEFLATE Compressed Data Format Specification version 1.3. P. Deutsch, The Internet Society, May 1996.
[DOM3Core]
Document Object Model (DOM) Level 3 Core Specification. A. Le Hors, P. Le Hégaret, L. Wood, G. Nicol, J. Robie, M. Champion, S. Byrne, editors. W3C Recommendation 07 April 2004.
[IANA-Charsets]
IANA Character Set Registry.
[IRI]
Internationalized Resource Identifiers (IRIs). RFC3987, M. Duerst, M. Suignard. January 2005.
[MIME]
Multipurpose Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies. RFC2045, N. Freed and N. Borenstein, IETF, November 1996.
Multipurpose Internet Mail Extensions(MIME) Part Two: Media Types. RFC2046, N. Freed and N. Borenstein, IETF, November 1996.
[RFC2119]
Key words for use in RFCs to Indicate Requirement Levels. RFC2119, S. Bradner. March 1997.
[SNIFF]
Media Type Sniffing. A. Barth and I. Hickson. IETF (Work in Progress).
[SVGTiny]
Scalable Vector Graphics (SVG) Tiny 1.2 Specification. O. Andersson, R. Berjon, E. Dahlström, A. Emmons, J. Ferraiolo, A. Grasso, V. Hardy, S. Hayman, D. Jackson, C. Lilley, C. McCormack, A. Neumann, C. Northway, A. Quint, N. Ramani, D. Schepers, A. Shellshear. W3C Recommendation, 22 December 2008.
[Unicode]
The Unicode Standard.
[URI]
Uniform Resource Identifier (URI): Generic Syntax. RFC 3986, T. Berners-Lee, R. Fielding and L. Masinter. January 2005.
[UTF-8]
UTF-8: A Transformation format of ISO 1064. RFC 3629, F. Yergeau. IETF, November 2003.
[UTR36]
UTR #36: Unicode Security Considerations, M. Davis, M. Suignard. Unicode Consortium.
[Widgets-DigSig]
XML Digital Signatures for Widgets. M. Cáceres, F. Hirsch, Stuart Knightley, and M. Priestley. (Work in Progress).
[View-Modes]
The 'view-mode' Media Feature. R. Berjon. W3C. (Work in Progress).
[XML]
Extensible Markup Language (XML) 1.0 (Fifth Edition). T. Bray, J. Paoli, C. M. Sperberg-McQueen, E. Maler, F. Yergeau. W3C, November 2008.
[XML-MIME]
XML Media Types. RFC3023. M. Murata, S. St. Laurent, D. Kohn. IETF. January 2001.
[XMLNS]
Namespaces in XML (Second Edition). T. Bray, D. Hollander, A. Layman, R. Tobin. W3C Recommendation, August 2006.
[ZIP]
.Zip File Format Specification. PKWare Inc.
[ZIP-MIME]
IANA Media Type Assignment.

Informative References

[ECMAScript-MIME]
Scripting Media Types. RFC4329. B. Hoehrmann. IETF. April 2006.
[HTML4]
HTML 4.01 Specification, D. Raggett, A. Le Hors, I. Jacobs, 24 December 1999.
[HTML]
HTML5, I. Hickson and D. Hyatt. W3C Working Draft. (Work in Progress)
[HTML-MIME]
The 'text/html' Media Type. D. Connolly and L. Masinter. IETF. June 2000.
[HTTP]
Hypertext Transfer Protocol -- HTTP/1.1. RFC 2616, R. Fielding, et al. June 1999.
[Widgets-Landscape]
The Widget Landscape (Q1 2008). M. Cáceres. (Work in progress).
[Widgets-Requirements]
Widgets Requirements, M. Cáceres and M. Priestley. W3C Working Draft (Work in progress).
[Widgets-Relax NG Schema]
Relax NG Schema for the Widgets Family of Specifications. D. Håsäther, R. Berjon, and M. Cáceres. (Work in Progress).
[P&C-Test-Suite]
Test Suite for Widget Packaging and XML Configuration. Work in Progress.
[Widgets-APIs]
The Widget Interface. M. Cáceres. (Work in progress).
[Widgets-URI]
Widget URI Scheme. R. Berjon. (Work in progress).
[XHTML1.1]
XHTML™ 1.1 - Module-based XHTML - Second Edition. S. McCarron and M. Ishikawa. W3C Recommendation 23 November 2010.