]> HyperText Markup Language Request For Comments Format Cisco Systems, Inc.
jhildebr@cisco.com
ICANN
paul.hoffman@icann.org
html css rfc In order to meet the evolving needs of the Internet community, the format for RFCs is changing from a plain-text, ASCII-only format to a canonical XML format that will in turn be rendered into several publication formats. This document defines the HTML format that will be rendered for an RFC or Internet-Draft.
As described in , the RFC Series is changing. One of those changes includes the RFC Editor publishing a non-canonical HTML version of RFCs. This document describes the HTML format that will be used as one of the publication formats for the RFC Series. It defines a strict subset of HTML appropriate for RFC Series documents. The visual layout of the document will be defined through a cascading style sheet (CSS) . The CSS will be included in the HTML file but will be described in a separate document. The details (particularly any vocabularies) described in this document are expected to change based on experience gained in implementing the RFC production center's toolset. Revised documents will be published capturing those changes as the toolset is completed. Other implementers must not expect those changes to remain backwards-compatible with the details described this document.
This section lists the design requirements used to create the HTML format described in this document. These requirements build on those found in . Many of these requirements are naturally fulfilled by using the output of the prep tool . The HTML has to render correctly on a list of browser versions that the RFC Editor will keep up to date outside of this document. The format will consist of a subset of HTML deemed to be widely implemented by common browsers at the time the specification is created, likely to continue to be widely-implemented, and unlikely to cause security issues. This will maximize the chances that future HTML renderers (such as new web browsers) will continue to produce readable text from the HTML format without the format needing to be changed frequently. These requirements are expected to change in the future to reflect the expectation that HTML rendering will be required for current versions of browsers and platforms, while ideally continuing to render correctly on recent versions of those browsers. The HTML documents from the RFC Editor or Internet-Drafts directory may be re-rendered from the canonical XML format in the future to ensure the ongoing readability of the documents. The intent is that any re-rendering would be due to exceptional circumstances rather than for minor annoyances. The HTML must display adequately in at least one text-based browser. Some consumers of the RFC series can only access the series on text-based terminals. The HTML document will be self-contained, without requiring external files for images, CSS, JavaScript, or the like. This will allow the HTML file to be moved over various non-HTTP transports (such as e-mail, FTP, and rsync) without breakage. Any use of JavaScript in the HTML document must not be required for comprehensive reading of the document, because some consumers of the RFC series routinely disable JavaScript for security purposes. The HTML document will allow easy local override of the default CSS formatting. This will allow users who have a different visual style that they prefer to make RFCs display with that style without having to alter the contents of the HTML document. This might also be valuable for allowing people with specific accessibility needs to use a customized CSS. HTML tags in documents will rarely have attributes whose only purpose is to affect the rendered styling, and those will only be used if it would not be possible to specify that styling in CSS. No such attributes are known at this time. Both user-defined and auto-generated anchors must be supported and linkable, with user-defined anchors appearing in an "id" attribute. Auto-generated anchors will be generated for every heading, paragraph, and so on, not just those that do not have user-defined anchors. User-defined anchors may, and auto-generated anchors will, appear next to paragraphs, figures, tables, blockquotes, and section titles. All section, subsections, figures, and paragraphs should have stable numbered link anchors. Additionally, anchors expressed in the source XML should be exposed as anchors in the HTML output as well. The HTML must make it easy to separate sections along with all of their subsections into separate files. This will make creating EPUB documents easier in the future. The HTML produced for Internet-Drafts will differ fromt that produced by the RFC Editor due to differences in the output from the prep tool. The abstract must be marked up or tagged in a way that popular search engines will extract it as a summary.
Normative information must be easily accessible to the following consumers: People with impaired vision, including those that use large fonts and those that use screen readers People with difficulty distinguishing between colors People who use devices with small screens, such as cell phones Other groups to be determined later Specific instances where goals for accessibility are important in the design choices of the format have been called out in the text. NOTE: designing for these consumers does not preclude the use of features they cannot use, but does require that key semantic data is not lost when read using the tools and settings that are required by a given constituency.
The RFC Editor will periodically determine which version of the HTML specification will be referenced for tools generating the format defined in this document. The starting version will be that defined in , commonly known as "HTML5". Although the HTML specification mandates several of the syntax and structure rules described in this document, they are called out here for emphasis.
The processor emitting HTML from the XML source will follow these rules: The HTML output is encoded as UTF-8, as specified in . The document is valid HTML. Double quotes (U+0022 QUOTATION MARK: ") are used to quote attribute values unless the HTML specification forbids quoting a particular attribute. Each logical line is terminated solely with a \n (U+000A: LINE FEED), otherwise known as "Unix-style" line endings. Other than \n (U+000A: LINE FEED), code points less than " " (U+0020: SPACE) (otherwise known as "control characters") are not used. Any character references that would generate these code points (such as 	) may not be used. NOTE: this rule explicitly forbids \t (U+0009: CHARACTER TABULATION), \f (U+000C: FORM FEED), and \r (U+000D: CARRIAGE RETURN) from appearing in the output. Comments in the source XML, if any, will not be copied into the HTML. The HTML output will be pretty-printed, using whatever consistent rules deemed best in the HTML production tool. NOTE: none of these rules affect the rendered output of the HTML, but are intended to increase the chance that difference tools that operate on the HTML output easier to write.
This section lists items that are common across multiple parts of the HTML document.
HTML elements that are generated from XML elements that include an "anchor" attribute will use the value of the "anchor" attribute as the value of the "id" attribute of the corresponding HTML element. The prep tool produces XML with "anchor" attributes in all elements that need them. Some HTML constructs (such as <section>) will use multiple instances of these identifiers.
Each paragraph, artwork, or sourcecode segment outside of a <figure> or <table> element will be appended with a space and a "pilcrow" (U+00B6: PILCROW SIGN), otherwise known as a "paragraph sign". For the purposes of clarity, in this document pilcrows are rendered as "&para;". The pilcrow will linked to the "id" attribute on the XML entity to which it is associated. The pilcrow will normally be invisible unless the element it is attached to is moused over. The pilcrow will be surrounded by a link that points to the element it is attached to. Pilcrows are never included inside a <table> or <figure> elements, since the figure number or table number serve as adequate link targets. Elements that might otherwise contain a pilcrow do not get marked with a pilcrow if they contain one or more child elements that are marked with a pilcrow. For example:

Four score and seven years ago our fathers brought forth on this continent, a new nation, conceived in Liberty, and dedicated to the proposition that all men are created equal.

]]>
Many elements in the v3 schema in contain attributes for ASCII equivalents of the Unicode text contained in the element or the Unicode attribute value. These alternatives are included in the HTML as in a <span> tag with the class "ascii". The <span> is included inside the tag that is showing the text to the user. For example:
HILDEBRAND Hildebrand ]]>
A few HTML elements are added to ensure particular syntax items can be styled appropriately using CSS. Commas that are not in running text (e.g., that might have white-space added before them by the HTML indentation step) are wrapped by an HTML <span> tag of CSS class "comma". Periods that are not in running text (e.g., that might have white-space added before them by the HTML indentation step) are wrapped by an HTML <span> tag of CSS class "fullStop". Open and close parentheses that are not running text (e.g., that might have white-space added before them by the HTML indentation step) are wrapped by an HTML <span> tag of CSS class "openParen" or "closeParen" respectively.
The front matter of the HTML format contains processing information, metadata of various types, and styling information that applies to the document as a whole. This section describes HTML that is not necessarily a direct transform from the XML format. For more details on each of the tags that generate content in this section, see .
The DOCTYPE of the document is "html", which declares that the document is compliant with HTML5. The document will start with exactly this string:
]]>
The root element of the document is <html>. This element includes a lang attribute, whose value is a language tag describing the natural language of the document. The language tag to be included is "en". The class of the <html> element will be copied verbatim from the XML <rfc> element's "mode" attribute, allowing CSS to style RFCs and Internet-Drafts differently from one another (if needed):
]]>
The root <html> will contain a <head> element that contains the following elements, as needed.
In order to be correctly processed by browsers that load the HTML using a mechanism that does not provide a valid MIME content-type or charset (such as from a local file system using a "file:" URL), the HTML <head> element contains a <meta> element, with charset attribute with value "utf-8":
]]>
The contents of the <title> element from the XML source will be placed inside an HTML <title> element in the header.
The following <meta> elements will be included: author - comma-separated <fullname>s of all of the <author>s from the XML source description - the <abstract> from the XML source generator - the name and version number of the software used to create the HTML keywords - comma-separated <keyword>s from the XML source For example:
]]>
The <head> element contains an embedded CSS style sheet in a <style> element. The styles in the style sheet are to be set consistently between documents by the RFC Editor, according to the best practices of the day. To ensure consistent formatting, individual style attributes are not used in the main portion of the document except in highly exceptional circumstances; each use of such attributes will be individually justified. Different readers of a specification will desire different formatting when reading the HTML versions of RFCs. To facilitate this, the <head> element also includes a <link> to a style sheet in the same directory as the HTML file, named "rfc-local.css". Any formatting in the linked style sheet will override the formatting in the included style sheet. For example:
body {} ... ]]>
Each <link> element from the XML source is copied into the HTML header.
Information about the document as a whole will appear as the first child of the HTML <body> element, embedded in an HTML <dl> element with id="identifiers". The defined terms in the definition list are "Workgroup:", "Series:", "Status:", "Published:", and "Authors:". For example:
Workgroup:
rfc-interest
Series:
Internet-Draft
Status:
Informational
Published:
Authors:
J. Hildebrand (Cisco Systems, Inc.)
H. Flanagan (RFC Editor)
]]>
The table of contents will follow the boilerplate if the XML's <rfc> element's tocInclude attribute has the value "true". An HTML <h2> heading containing the text "Table of Contents" will be followed by a <nav> element that contains a <ul> element for each depth of the section hierarchy. Each section will be represented by a <li> element containing links by the section number (from the "pn" attribute) and by the name (from the "slugifiedName" attribute of the <name> child element). Each <nav>, <ul>, and <li> element will have the class "toc". For example:
Table of Contents
The main body of the HTML document is processed according to the rules in .
The back matter of the HTML document includes an index (if generated), information about the authors, and further information about the document itself.
The index will be produced at the end of the document (before the author information) if and only if the XML document's <rfc> element has a indexInclude attribute with the value "true", and there is one or more <iref> elements in the document.
The index section will start with an <h2> heading containing the text "Index", followed by links to each of the lettered portions of the index. Links are not generated for letters that do not occur as the first letter of an index item. For example:
Index
C P
...]]>
The index index is followed by a <ul> tag that contains a <li> tag for each first letter represented in the index. This <li> tag has the class "indexChar", and contains an <a> tag with the id pointed to by the index index, as well as an href to itself. The <li> tag also includes a <ul> tag that will contain the index items. For example:
  • C
    • ...]]>
    Each index item can have multiple <iref> elements to point to, all with the same item attribute. Each index item is represented by a <li> tag of class "indexItem" containing a <span> of class "irefItem" for the item text and one of class "irefRefs" for the generated references (if there is at least one reference to the item not having a subitem). Each generated reference contains an <a> tag containing the section number where the <iref> was found, with an "href" attribute pointing to the "irefid" attribute of the <iref> element from the XML document. If the primary attribute of the <iref> element has the value "true", the <a> element in the HTML document will have the class "indexPrimary". Commas may be used to separate the generated references, surrounded by a <span> tag with class "comma". For example:
    Bullets 2 , 2 ...]]>
    If an index item has at least one subitem, the <li> of that item will contain a <ul>, with one <li> for each subitem, of class "indexSubItem". Each subitem is formatted similarly to items, except the class of the first <span> tag is "irefSubItem". For example:
  • Ordered 2
    • ...]]>
    At the end of the document, author information will be included inside an HTML <section> element. The class names have been chosen to match the class names in . Note: The following example shows several ASCII equivalents that are the same as their nominal equivalents for clarity; normally the ASCII equivalents would not be included for these cases.

    Authors' Addresses

    Joe Hildebrand Joe Hildebrand ( editor )
    Cisco Systems, Inc. Cisco Systems, Inc.
    1 Main Street
    Suite 1
    Denver Denver , CO 80202 80202
    US
    Phone: +1-720-555-1212 VOICE
    Fax: +1-303-555-1212 fax
    Email: jhildebr@example.com
    ]]>
    A few bits of metadata about the document that are less important to most readers are included after the author information. The style sheet might de-emphasize their display, or hide them altogether. These are gathered together into a <div> of class "docInfo". The finalized time is copied from the <rfc> element's prepTime attribute. The rendered time is the time that this HTML was generated. For example:
    Finalized: Rendered: ]]>
    At the very end of the document, the XML source that was used to produce this document will be included within a comment. This comment may be preceded by another comment that describes the source. Any instances of "--" in the XML will be modified to use U+002D characters: "--". Note that if the dashes in the original XML were in a comment start- (<!--) or end-delimiter (-->), the XML will not parse correctly without reversing this transformation.