Overview.bs

<pre class='metadata'>
Title: CSS Ruby Layout Module Level 1
Shortname: css-ruby
Level: 1
Status: ED
Work Status: Exploring
Group: csswg
ED: https://drafts.csswg.org/css-ruby-1/
TR: https://www.w3.org/TR/css-ruby-1/
Previous Version: https://www.w3.org/TR/2014/WD-css-ruby-1-20140805/
Previous Version: https://www.w3.org/TR/2013/WD-css3-ruby-20130919/
Editor: Elika J. Etemad / fantasai, Invited Expert, http://fantasai.inkedblade.net/contact, w3cid 35400
Editor: Koji Ishii, Google, kojiishi@gmail.com, w3cid 45369
Editor: Xidorn Quan, Mozilla https://www.mozilla.org/, https://www.upsuper.org/, w3cid 73936
Abstract: “Ruby”, a form of interlinear annotation, are short runs of text alongside the base text.
Abstract: They are typically used in East Asian documents to indicate pronunciation or to provide a short annotation.
Abstract: This module describes the rendering model and formatting controls related to displaying ruby annotations in CSS.
Ignored terms: internal table elements
Ignored vars: any
</pre>

<!--

Issues:
	box layout/sizing
	clean up inter-character vs. parallel layout requirements

Redo all examples with consistent font. (M+ 2p?)

-->

<h2 id="intro">
Introduction</h2>

	<p><em>This section is not normative.</em>

<h3 id="placement">
Module interactions</h3>

	<p>This module extends the inline box model of CSS Level 2 [[!CSS2]]
	to support ruby.

	<p>None of the properties in this module apply to the <code>::first-line</code> or
	<code>::first-letter</code> pseudo-elements.

<h3 id="values">
Values</h3>

	This specification follows the <a href="https://www.w3.org/TR/CSS2/about.html#property-defs">CSS property definition conventions</a> from [[!CSS2]].
	Value types not defined in this specification are defined in CSS Values & Units [[!CSS-VALUES-3]].
	Other CSS modules may expand the definitions of these value types.

	In addition to the property-specific values listed in their definitions,
	all properties defined in this specification
	also accept the <a>CSS-wide keywords</a> keywords as their property value.
	For readability they have not been repeated explicitly.

<h3 id="diagram-conventions">
Diagram conventions</h3>

	<p>Many typographical conventions in East Asian typography depend
	on whether the character rendered is wide (CJK) or narrow (non-CJK).
	There are a number of illustrations in this document
	for which the following legend is used:

	<dl>
		<dt><img alt="Symbolic wide-cell glyph representation" width="39" height="39" src="images/fullwidth.gif">
		<dd>Wide-cell glyph (e.g. Han) that is the <var>n</var>th character in the text run.
		They are typically sized to 50% when used as annotations.
		<dt><img alt="Symbolic narrow-cell glyph representation" width="19" height="39" src="images/halfwidth.gif">
		<dd>Narrow-cell glyph (e.g. Roman) which is the <var>n</var>th glyph in the text run.
	</dl>

	<p>The orientation which the above symbols assume in the diagrams
	corresponds to the orientation that the glyphs they represent
	are intended to assume when rendered by the user agent.
	Spacing between these characters in the diagrams is incidental,
	unless intentionally changed to make a point.

<h3 id="ruby-def">
What is ruby?</h3>

	<p><dfn export>Ruby</dfn> is the commonly-used name for a run of text
	that appears alongside another run of text (referred to as the “base”)
	and serves as an annotation or a pronunciation guide associated with that run of text.

	<p>The following figures show two examples of Ruby,
	a simple case and one with more complicated structure.

	<div class="example">
		<p>In this first example, a single annotation is used to annotate the base text.
		<div class="figure">
			<p><img src="images/licence.png"
			        alt="Example of ruby applied on top of a Japanese expression">
			<p class="caption">Example of ruby used in Japanese (simple case)
		</div>
		<p>In Japanese typography, this case is sometimes called
		<span lang="ja">taigo</span> ruby or group-ruby (per-word ruby),
		because the annotation as a whole is associated
		with multi-character word (as a whole).
	</div>

	<div class="example">
		<p>In this second example,
		two levels of annotations are attached to a base sequence:
		the hiragana characters on top refer to the pronunciation of each of the base kanji characters,
		while the words “Keio” and “University” on the bottom are annotations describing the English translation.
		<div class="figure">
			<p><img src="images/ruby-univ.gif"
			        alt="Example showing complex ruby with annotation text over and under the base characters">
			<p class="caption">Complex ruby with annotation text over and under the base characters
		</div>
		<p>
		<p>Notice that to allow correct association between the hiragana characters and
		their corresponding Kanji base characters,
		the spacing between these Kanji characters is adjusted.
		(This happens around the fourth Kanji character in the figure above.)
		To avoid variable spacing between the Kanji characters in the example above
		the hiragana annotations can be styled as a <i>collapsed annotation</i>,
		which will look more like the group-ruby example earlier.
		However because the base-annotation pairings are recorded in the ruby structure,
		if the text breaks across lines, the annotation characters will stay
		correctly paired with their respective base characters.
	</div>

	<p><i>Ruby</i> formatting as used in Japanese is described in JIS X-4051 [[JIS4051]] (in Japanese)
	and in Requirements for Japanese Text Layout [[JLREQ]] (in English and Japanese)].
	In HTML, ruby structure and markup to represent it is described
	in the Ruby Markup Extension specification.
	This module describes the CSS rendering model
	and formatting controls relevant to ruby layout of such markup.

<h2 id="ruby-model">
Ruby Box Model</h2>

	<p>The CSS ruby model is based on
	the <a href="https://www.w3.org/TR/html5/text-level-semantics.html#the-ruby-element">W3C HTML5 Ruby Markup</a> model
	and the <a href="https://www.w3.org/TR/ruby/">XHTML Ruby Annotation Recommendation</a> [[RUBY]].
	In this model, a ruby structure consists of
	one or more <i>ruby base</i> elements representing the base (annotated) text,
	associated with one or more levels of <i>ruby annotation</i> elements representing the annotations.
	The structure of ruby is similar to that of a table:
	there are “rows” (the <i lt="base level">base text level</i>, each <i>annotation level</i>)
	and “columns” (each <i>ruby base</i> and its corresponding <i>ruby annotations</i>).

	<p>Consecutive bases and annotations are grouped together into <i>ruby segments</i>.
	Within a <i>ruby segment</i>, a <i>ruby annotation</i> may span multiple <i>ruby bases</i>.

	<p class="note">In HTML, a single <code>&lt;ruby&gt;</code> element may contain multiple <i>ruby segments</i>.
	(In the XHTML Ruby model, a single <code>&lt;ruby&gt;</code> element can only contain one <i>ruby segment</i>.)

<h3 id="ruby-display">
Ruby-specific 'display' Values</h3>

	<p>For document languages (such as XML applications) that do not have pre-defined ruby elements,
	authors must map document language elements to ruby elements;
	this is done with the 'display' property.

	<pre class=propdef partial>
	Name: display
	New values: ruby | ruby-base | ruby-text | ruby-base-container | ruby-text-container
	</pre>

	<p>The following new 'display' values assign ruby layout roles to an arbitrary element:

	<dl dfn-type=value export dfn-for=display>
		<dt><dfn>ruby</dfn>
			<dd>Specifies that an element generates a <dfn dfn for lt="ruby container | ruby container box">ruby container box</dfn>.
			(Corresponds to HTML/XHTML <code>&lt;ruby&gt;</code> elements.)
		<dt><dfn>ruby-base</dfn>
			<dd>Specifies that an element generates a <dfn dfn for lt="ruby base box | ruby base">ruby base box</dfn>.
			(Corresponds to HTML/XHTML <code>&lt;rb&gt;</code> elements.)
		<dt><dfn>ruby-text</dfn>
			<dd>Specifies that an element generates a <dfn dfn for lt="ruby annotation box | ruby annotation | annotation">ruby annotation box</dfn>.
			(Corresponds to HTML/XHTML <code>&lt;rt&gt;</code> elements.)
		<dt><dfn>ruby-base-container</dfn>
			<dd>Specifies that an element generates a <dfn dfn for lt="ruby base container box | ruby base container">ruby base container box</dfn>.
			(Corresponds to XHTML <code>&lt;rbc&gt;</code> elements; generated as an anonymous box in HTML.)
		<dt><dfn>ruby-text-container</dfn>
			<dd>Specifies that an element generates a <dfn dfn for lt="ruby annotation container box | ruby annotation container">ruby annotation container box</dfn>.
			(Corresponds to HTML/XHTML <code>&lt;rtc&gt;</code> elements.)
	</dl>

	<p class="advisement">Authors using a language (such as HTML)
	that supports dedicated ruby markup
	should use that markup rather than
	styling arbitrary elements (like <code>&lt;span&gt;</code>)
	with ruby 'display' values.
	Using the correct markup ensures that screen readers
	and non-CSS renderers can interpret the ruby structures.

<h4 id="formatting-context">
The Ruby Formatting Context</h4>

	<p><i>Ruby containers</i> are non-atomic inline-level boxes.
	Like inline boxes, they break across lines,
	and their containing block is the nearest block container ancestor.
	And just as the contents of an inline box
	participate in the same inline formatting context that contains the inline box itself,
	a <i>ruby container</i> and its base-level contents
	participate in the same inline formatting context that contains the <i>ruby container</i> itself.

	<p>However <i>ruby containers</i> also establish a <dfn export>ruby formatting context</dfn>
	that builds further structure around their segment of the inline formatting context.
	<i>Ruby bases</i>, <i>ruby annotations</i>, <i>ruby base containers</i>, and <i>ruby annotation containers</i>
	are <dfn export lt="internal ruby boxes|internal ruby display types">internal ruby boxes</dfn>:
	like <i>internal table elements</i>,
	they have specific roles in ruby layout,
	and participate in their <i>ruby container</i>’s <i>ruby formatting context</i>.

	<p>As with the contents of inline boxes,
	the containing block for the contents of a <i>ruby container</i> (and all its <i>internal ruby boxes</i>)
	is the containing block of the <i>ruby container</i>.
	So floats, for example, are trapped by the <i>ruby container</i>’s containing block,
	not any of the ruby box types.

	<p class="issue">Are internal ruby boxes inline-level?

<h4 id="block-ruby">
Non-Inline Ruby</h4>

	<p>If an element has an <i>inner display type</i> of ''ruby''
	and an <i>outer display type</i> other than ''display/inline'',
	then it generates two boxes:
	a principal box of the required <i>outer display type</i> type,
	and an inline-level <i>ruby container</i>.
	All properties specified on the element apply to the principal box
	(and if inheritable, inherit to the <i>ruby container box</i>).
	This allows styling the element as a block,
	while correctly maintaining the internal ruby structure.

	<p class="note">
	Note that absolute positioning or floating an element causes its 'display' value
	to compute to a block-level equivalent. (See [[!CSS-DISPLAY-3]] or [[!CSS2]] section 9.7.)
	For the <i>internal ruby display types</i>,
	this causes their 'display' value to compute to ''display/block''.

<h3 id="box-fixup">
Anonymous Ruby Box Generation</h3>

	<p>The CSS model does not require that the document language
	include elements that correspond to each of these components.
	Missing parts of the structure are implied through the anonymous box generation rules
	<a href="https://www.w3.org/TR/CSS2/tables.html#anonymous-boxes">similar to those used to normalize tables</a>. [[!CSS2]]

	<ol>

		<li id="anon-gen-inlinize"><strong><a>Inlinify</a> block-level boxes:</strong>
			Any in-flow boxes directly contained by a
			<i>ruby container</i>,
			<i>ruby base container</i>,
			<i>ruby annotation container</i>,
			<i>ruby base box</i>,
			or <i>ruby annotation box</i>
			are “<a>inlinified</a>” per [[!CSS-DISPLAY-3]]),
			and their 'display' value computed accordingly,
			so that they contain only inline-level content.
			For example,
			the 'display' property of an in-flow element with ''display: block''
			parented by an element with ''display: ruby-text''
			computes to ''inline-block''.

		<li id="anon-gen-anon-ruby"><strong>Generate anonymous ruby containers</strong>:
			Any <a href="https://www.w3.org/TR/CSS2/tables.html#anonymous-boxes">consecutive</a> sequence of
			improperly-contained
			<i>ruby base containers</i>,
			<i>ruby annotation containers</i>,
			<i>ruby bases</i>,
			and/or
			<i>ruby annotations</i>
			(and any intervening <i>white space</i>)
			is wrapped in an anonymous <i>ruby container</i>.
			For the purpose of this step:
			<ul>
				<li>an improperly-contained <i>ruby base</i> is one not parented by a <i>ruby base container</i> or <i>ruby container</i>
				<li>an improperly-contained <i>ruby annotation</i> is one not parented by a <i>ruby annotation container</i> or <i>ruby container</i>
				<li>an improperly-contained <i>ruby base container</i> or <i>ruby annotation container</i>
				    is one not parented by a <i>ruby container</i>
			</ul>

		<li id="anon-gen-bare-inlines"><strong>Wrap misparented inline-level content:</strong>
			Any consecutive sequence of text and inline-level boxes
			directly parented by a <i>ruby container</i> or <i>ruby base container</i>
			is wrapped in an anonymous <i>ruby base</i>.
			Similarly, any consecutive sequence of text and inline-level boxes
			directly parented by a <i>ruby annotation container</i>
			is wrapped in an anonymous <i>ruby annotation</i>.
			(For this purpose, misparented <i>internal table elements</i>
			are treated as <a>inline-level content</a>
			since, being parented by ruby boxes,
			they will be ultimately wrapped by an <a>inline table</a>.)

			However, if an anonymous box so constructed contains only <i>white space</i>,
			it is considered <dfn>intra-ruby white space</dfn>
			and is either discarded
			or preserved
			as described below.

		<li id="anony-gen-trim-space"><strong>Trim leading/trailing white space:</strong>
			Any <i>intra-ruby white space</i>
			that is not the sole child of its parent
			and occurs at the beginning or end of
			a <i>ruby container</i>, <i>ruby annotation container</i>, or <i>ruby base container</i>
			is removed, as if it had ''display: none''

		<li id="anon-gen-discard-space"><strong>Remove inter-level white space:</strong>
			Any <i>intra-ruby white space</i>
			whose immediately adjacent siblings match one of the patterns below
			is <dfn>inter-level white space</dfn>
			and is removed, as if it had ''display: none''.
			<table class="data">
			<thead>
				<tr><th>Previous box
				    <th>Next box
			<tbody>
				<tr><td><var>any</var>
				    <td><i>ruby annotation container</i>
				<tr><td>not <i>ruby annotation</i>
				    <td><i>ruby annotation</i>
<!--
				<tr><td><i>ruby base</i> or <i>ruby base container</i>
				    <td><i>ruby annotation</i> or <i>ruby annotation container</i>
				<tr><td><i>ruby annotation container</i>
				    <td><i>ruby annotation container</i>
				<tr><td><i>ruby annotation</i>
				    <td><i>ruby annotation container</i>
				<tr><td><i>ruby annotation container</i>
				    <td><i>ruby annotation</i>
-->
			</table>

		<li id="anon-gen-interpret-space"><strong>Interpret intra-level white space:</strong>
			Any <i>intra-ruby white space</i> box
			whose immediately adjacent siblings match one of the patterns below
			is assigned the box type and subtype defined in the table below:
			<table class="data">
			<colgroup span=2></colgroup>
			<colgroup span=2></colgroup>
			<thead>
				<tr><th>Previous box
				    <th>Next box
				    <th>Box type
				    <th>Subtype
			<tbody>
				<tr><td><i>ruby base</i>
				    <td><i>ruby base</i>
				    <td><i>ruby base</i>
				    <td><dfn>inter-base white space</dfn>
				<tr><td><i>ruby annotation</i>
				    <td><i>ruby annotation</i>
				    <td><i>ruby annotation</i>
				    <td><dfn>inter-annotation white space</dfn>
				<tr><td><i>ruby annotation</i> or <i>ruby annotation container</i>
				    <td><i>ruby base</i> or <i>ruby base container</i>
				    <td rowspan=3><i>ruby base</i>
				    <td rowspan=3><dfn>inter-segment white space</dfn>
				<tr><td><i>ruby base</i> or <i>ruby base container</i>
				    <td><i>ruby base container</i>
				<tr><td><i>ruby base container</i>
				    <td><i>ruby base</i> or <i>ruby base container</i>
			</table>
			The <dfn>intra-level white space</dfn> boxes defined above are
			treated specially for pairing and layout. See below.

		<li id="anon-gen-unbreak"><strong>Suppress line breaks:</strong>
			Convert all forced line breaks inside <i>ruby annotations</i> (regardless of 'white-space' value)
			as defined for <i>collapsible</i> segment breaks in <a href="https://www.w3.org/TR/css-text-3/#line-break-transform">CSS Text Level 3 &sect; 4.1.2</a>.
			<p class="issue">The goal of this is to simplify the layout model by suppressing any line breaks within ruby annotations.
			Alternatively we could try to define some kind of acceptable behavior for them.

		<li id="anon-gen-anon-containers"><strong>Generate anonymous level containers:</strong>
			Any consecutive sequence of <i>ruby bases</i> and <i>inter-base white space</i>
			(and not <i>inter-segment white space</i>)
			not parented by a <i>ruby base container</i>
			is wrapped in an anonymous <i>ruby base container</i>.
			Similarly, any consecutive sequence of <i>ruby annotations</i> and <i>inter-annotation white space</i>
			not parented by a <i>ruby annotation container</i>
			is wrapped in an anonymous <i>ruby annotation container</i>.
	</ol>

	<p class="issue">Make <a href="http://lists.w3.org/Archives/Public/www-archive/2014Jun/att-0027/PastedGraphic-1.png">this diagram</a> into an example.

	<p>Once all ruby layout structures are properly parented,
	the UA can start to associate bases with their annotations.

	<p class="note">
	Note that the UA is not required to create any of these anonymous boxes
	(or the anonymous empty <i>intra-level white space</i> boxes in the pairing section)
	in its internal structures,
	as long as pairing and layout behaves as if they existed.

<h3 id="ruby-pairing">
Annotation Pairing</h3>

	<p>Annotation <dfn>pairing</dfn> is the process of associating
	<i>ruby annotations</i> with <i>ruby bases</i>.
	Each <i>ruby annotation</i> is associated with one or more <i>ruby bases</i>,
	and is said to <dfn>span</dfn> those bases.
	(A <i>ruby annotation</i> that <i>spans</i> multiple bases is called
	a <dfn>spanning annotation</dfn>.)

	A <i>ruby base</i> is can be associated with
	only one <i>ruby annotation</i> per <i>annotation level</i>.
	However, if there are multiple <i>annotation levels</i>,
	it can be associated with multiple <i>ruby annotations</i>.

	Once pairing is complete, ruby “column” units are defined,
	each represented by a single <i>ruby base</i>
	and one <i>ruby annotation</i> (possibly an empty, anonymous one)
	from each <i>annotation level</i> in its <i>ruby segment</i>.


<h4 id="segment-pairing">
Segment Pairing and Annotation Levels</h4>

	<p>A ruby structure is divided into <dfn>ruby segments</dfn>,
	each consisting of a single <i>ruby base container</i>
	followed by one or more <i>ruby annotation containers</i>.
	Each <i>ruby annotation container</i> in a <i>ruby segment</i>
	represents one <dfn lt="annotation level | level">level</dfn> of annotation for the base text:
	the first one represents the first level of annotation,
	the second one represents the second level of annotation,
	and so on.
	The <i>ruby base container</i> represents the <dfn>base level</dfn>.
	The <i>ruby base container</i> in each segment is thus paired
	with each of the <i>ruby annotation containers</i> in that segment.

	<p>In order to handle degenerate cases, some empty anonymous containers are assumed:
	<ul>
		<li>If the first child of a <i>ruby container</i> is a <i>ruby annotation container</i>,
		an anonymous, empty <i>ruby base container</i> is assumed to exist before it.
		<li>Similarly, if the <i>ruby container</i> contains consecutive <i>ruby base containers</i>,
		anonymous, empty <i>ruby annotation containers</i> are assumed to exist between them.
	</ul>

	<p><i>Inter-segment white space</i> is effectively a <i>ruby segment</i> of its own.

<h4 id="base-annotation-pairing">
Unit Pairing and Spanning Annotations</h4>

	<p>Within a <i>ruby segment</i>,
	each <i>ruby base</i> in the <i>ruby base container</i>
	is paired with one <i>ruby annotation</i>
	from each <i>ruby annotation container</i> in its <i>ruby segment</i>.

	<p>If a <i>ruby annotation container</i> contains only
	a single, anonymous <i>ruby annotation</i>,
	then that <i>ruby annotation</i> is paired with (i.e. <i>spans</i> across)
	all of the <i>ruby bases</i> in its <i>ruby segment</i>.

	<p>Otherwise, each <i>ruby annotation</i> is paired,
	in order, with the corresponding <i>ruby base</i> in that segment</i>.
	If there are not enough <i>ruby annotations</i> in a <i>ruby annotation container</i>,
	the remaining <i>ruby bases</i> are paired with anonymous empty annotations
	inserted at the end of the <i>ruby annotation container</i>.
	If there are not enough <i>ruby bases</i>,
	any remaining <i>ruby annotations</i> pair with empty, anonymous bases
	inserted at the end of the <i>ruby base container</i>.

	<p>If an implementation supports ruby markup with explicit spanning
	(e.g. XHTML Complex Ruby Annotations),
	it must adjust the pairing rules to pair <i>spanning annotations</i>
	to their bases appropriately.

	<p><i>Intra-level white space</i> does not participate in standard annotation <i>pairing</i>.
	However, if the immediately-adjacent <i>ruby bases</i> or <i>ruby annotations</i>
	are paired
	<ul>
		<li>with two <i>ruby bases</i> or <i lt="ruby annotations">annotations</i>
		that surround corresponding <i>intra-level white space</i> in another level,
		then the so-corresponding <i>intra-level white space</i> boxes are also paired.
		<li>with a single spanning <i>ruby annotation</i>,
		then the <i>intra-level white space</i> is also paired to that <i>ruby annotation</i>
		<li>with two <i>ruby bases</i> or <i lt="ruby annotations">annotations</i>
		with no intervening <i>intra-level white space</i>,
		then the <i>intra-level white space</i> box pairs with
		an anonymous empty <i>intra-level white space</i> box assumed to exist between them.
	</ul>

	<p class="issue">Insert diagram</p>

<h4 id="nested-pairing">
Complex Spanning with Nested Ruby</h4>

	<p>When <i>ruby containers</i> are <dfn lt="nested ruby">nested</dfn>,
	pairing begins with the deepest <i>ruby container</i>,
	then expands out.
	From the pairing perspective of the outer <i>ruby container</i>,
	each <i>ruby container</i> nested within another <i>ruby container</i>
	counts as representing a single <i>ruby base</i>/<i>annotation</i> per level.
	The outer <i>ruby container</i>’s <i>ruby annotations</i> paired to the <i>nested ruby</i>
	are therefore paired with (and <i>span</i>) all of the nested <i>ruby container</i>’s <i>ruby bases</i>.
	Each <i>ruby annotation container</i> in the nested <i>ruby container</i>
	occupies the same <i>annotation level</i> in the outer <i>ruby container</i>
	as it does in the inner one
	and participates in its layout as if it were directly contained in the outer <i>ruby container</i>.

	<p>This process is recursive.
	Thus, using nested <i>ruby containers</i> allows the representation
	of complex spanning relationships.

	<p class="issue">It's not clear whether this falls out of layout handling of ruby containers inside ruby bases
	or needs to be handled specially.
	Waiting until layout is better-defined to find out...

<h3 id="autohide">
Autohiding Base-identical Annotations</h3>

	<p>If a <i>ruby annotation</i> has the exact same text content as its base,
	it is <dfn lt="hidden ruby annotation | hidden annotation">hidden</dfn>.
	Hiding a <i>ruby annotation</i> does not affect annotation pairing
	or the block-axis positioning of boxes in other <i>levels</i>.
	However the <i>hidden annotation</i> is not visible,
	and it has no impact on layout
	other than to separate adjacent sequences of <i>ruby annotation boxes</i> within its level,
	as if they belonged to separate segments
	and the <i>hidden annotation</i>’s base were not a <i>ruby base</i> but an intervening inline.

	<div class="example">
		<p>This is to allow correct inlined display of annotations
		for Japanese words that are a mix of kanji and hiragana.
		For example, the word <span lang=ja>振り仮名</span> should be inlined as
		<p class="figure">振り仮名(ふりがな)
		<p>and therefore marked up as
		<pre>
<!--		-->&lt;ruby>
<!--		-->  &lt;rb>振&lt;/rb>&lt;rb>り&lt;/rb>&lt;rb>仮&lt;/rb>&lt;rb>名&lt;/rb>
<!--		-->  &lt;rp>(&lt;/rp>&lt;rt>ふ&lt;/rt>&lt;rt>り&lt;/rt>&lt;rt>が&lt;/rt>&lt;rt>な&lt;/rt>&lt;rp>)&lt;/rp>
<!--		-->&lt;/ruby></pre>
		<p>However, when displayed as ruby, the “り” should be hidden
		<div class="figure">
			<p><img src="images/furigana-separate.png"
			        alt="Hiragana annotations for 振り仮名 appear, each pronunciation above its kanji base character.">
			<p class="caption">Hiragana ruby for 振り仮名. Notice there is no hiragana annotation above り,  since it is already in hiragana.
		</div>
	</div>

	<p>When the computed value of 'ruby-merge' is ''collapse'',
	the autohiding is disabled.
	When the computed value of 'ruby-merge' is ''auto'',
	the user agent may decide whether to autohide or not,
	but it is recommended to autohide if the algorithm the user agent chose
	produces the results similar to ''separate'' would produce.

	<p>The content comparison for this auto-hiding behavior
	takes place prior to white space collapsing ('white-space') and text transformation ('text-transform')
	and ignores elements (considers only the <code>textContent</code> of the boxes).

	<p class="note">
		Future levels of CSS Ruby may add controls for auto-hiding,
		but in this level it is always forced.

<h3 id="white-space">
White Space Collapsing</h3>

	<p>White space within a ruby structure is <a href="#anon-gen-discard-space">discarded</a>
	<ul>
		<li>at the beginning and end of a <i>ruby container</i>, <i>ruby annotation container</i>, or <i>ruby base container</i>,
		<li>between a <i>ruby base container</i> and its following <i>ruby annotation container</i>,
		<li>between <i>ruby annotation containers</i>.
	</ul>

	<div class="example">
		For example, the following markup will display without any spaces:
		<pre>
<!--		-->&lt;ruby>
<!--		-->  &lt;rb>東&lt;/rb>&lt;rb>京&lt;/rb>
<!--		-->  &lt;rt>とう&lt;/rt>&lt;rt>きょう&lt;/rt>
<!--		-->  &lt;rt>Tō&lt;/rt>&lt;rt>kyō&lt;/rt>
<!--		-->&lt;/ruby></pre>
	</div>

	<p>Between <i>ruby segments</i>, between <i>ruby bases</i>, and between <i>ruby annotations</i>, however,
	white space is not discarded,
	and is maintained for rendering
	as <i lt="inter-base white space">inter-base</i>,
	<i lt="inter-annotation white space">inter-annotation</i>,
	or <i>inter-segment white space</i>.
	(See <a href="#box-fixup">Anonymous Ruby Box Generation</a>, above.)

	<div class="example">
		<p>The rules preserving white space allow ruby to be used with space-separated scripts such as Latin.
		For example,
		<pre>
<!--		-->&lt;ruby>
<!--		-->  &lt;rb>W&lt;/rb>&lt;rb>W&lt;/rb>&lt;rb>W&lt;/rb>
<!--		-->  &lt;rt>World&lt;/rt> &lt;rt>Wide&lt;/rt> &lt;rt>Web&lt;/rt>
<!--		-->&lt;/ruby></pre>
		<p>They also ensure that annotated white space is preserved. For example,
		<pre>
<!--		-->&lt;ruby>
<!--		-->  &lt;rb>Aerith&lt;/rb>&lt;rb> &lt;/rb>&lt;rb>Gainsboro&lt;/rb>
<!--		-->  &lt;rt>エアリス&lt;/rt>&lt;rt>・&lt;/rt>&lt;rt>ゲインズブール&lt;/rt>
<!--		-->&lt;/ruby></pre>
	</div>

	<p>Where undiscarded white space is <i>collapsible</i>, it will collapse
	following the standard <a href="https://www.w3.org/TR/css3-text/#white-space-rules">white space processing rules</a>. [[!CSS3TEXT]]
	For <i>collapsible white space</i> between <i>ruby segments</i> (<i>inter-segment white space</i>), however,
	the contextual text for determining collapsing behavior is given by the <i>ruby bases</i> on either side,
	not the text on either side of the white space in source document order.

	<div class="note">
		<p>Note that the white space processing rules
		cause a white space sequence containing a <i>segment break</i> (such as a line feed)
		to <a href="https://www.w3.org/TR/css3-text/#line-break-transform">collapse to nothing</a> between Han and Kana characters.
		This means that Chinese and Japanese ruby can safely use white space for indentation of ruby markup.
		For example, the following markup will display without any spaces:
		<pre>
<!--		-->&lt;ruby>
<!--		-->  屋&lt;rt>おく&lt;/rt>内&lt;rt>ない&lt;/rt>
<!--		-->  禁&lt;rt>きん&lt;/rt>煙&lt;rt>えん&lt;/rt>
<!--		-->&lt;/ruby></pre>
		<p>However, white space that does not contain a <i>segment break</i> does not collapse completely away,
		so this markup will display with a space between the first and second ruby pairs:
		<pre>
<!--		-->&lt;ruby>
<!--		-->  屋&lt;rt>おく&lt;/rt>	内&lt;rt>ない&lt;/rt>
<!--		-->  禁&lt;rt>きん&lt;/rt>	煙&lt;rt>えん&lt;/rt>
<!--		-->&lt;/ruby></pre>
	</div>

<h2 id="ruby-layout">
Ruby Layout</h2>

	<p>When a ruby structure is laid out,
	its <i>base level</i> is laid out on the line,
	aligned according to its 'vertical-align' property
	exactly as if its <i>ruby bases</i> were a regular sequence of inline boxes.
	Each <i>ruby base container</i> is sized and positioned
	to contain exactly all of its <i>ruby bases</i>’ margin boxes.

	<p><i>Ruby annotations</i> associated with the <i>base level</i>
	are then positioned with respect to their <i>ruby base boxes</i>
	according to the applicable 'ruby-position' values.
	<i>Ruby annotations</i> within a level (within a single <i>ruby container</i>)
	are aligned to each other as if they were inline boxes
	participating in the same inline formatting context.
	Each <i>ruby annotation container</i> is sized and positioned
	to contain exactly all of its <i>ruby annotations</i>’ margin boxes.

	<p><i>Ruby annotation containers</i> are stacked outward
	over or under their corresponding <i>ruby base container</i>,
	without any intervening space.

	<p class="issue">Should block-axis margins collapse?
	This makes layout more robust,
	but is inconsistent with how inlines behave along the inline-axis.

	<p>A ruby container (or fragment thereof)
	measures as wide as the content of its widest level.
	Similarly, <i>ruby base boxes</i> and <i>ruby annotation boxes</i>
	within a ruby “column” have the measure of the widest content in that “column”.
	In the case of <i>spanning annotations</i>
	(whether actually spanning or pretending to span per 'ruby-merge'),
	the measures of the <i>ruby annotation box</i> and
	the sum of its associated <i>ruby base boxes</i> must match.

	<p>How the extra space is distributed
	when ruby content is narrower than the measure of its box
	is specified by the 'ruby-align' property.

	<p class="issue">Should the ruby bases and annotations size to the column,
	or size to the content?

<h3 id="inter-character-layout">
Inter-character Ruby Layout</h3>

	<p>Inter-character annotations have special layout.
	When 'ruby-position' indicates ''inter-character'' annotations,
	the affected <i>ruby annotation boxes</i>
	are spliced into and measured as part of the layout of the <i>base level</i>.
	The <i>ruby base container</i> must be sized to include both the <i>ruby base boxes</i>
	as well as the ''inter-character'' <i>ruby annotation boxes</i>.
	The affected <i>ruby annotation container</i> is similarly sized
	so that its content box coincides with that of the <i>ruby base container</i>.

	<p>For the purpose of laying out other levels of annotations,
	an ''inter-character'' annotation effectively becomes part of its base.
	<span class="issue">Or should it become a quasi-base between two bases?</span>
	A spanning ''inter-character'' annotation is placed after
	all the bases that it spans.

<h3 id="box-style">
Styling Ruby Boxes</h3>

	<p>In most respects, ruby boxes can be styled similar to inline boxes.
	However, the UA is not required to support
	any of the box properties (borders, margins, padding),
	any of the background properties or outline properties,
	or any other property that illustrates the bounds of the box
	on <i>ruby base container boxes</i>, <i>ruby annotation container boxes</i>,
	or <a href="#nested-pairing">ruby-internal ruby container boxes</a>.
	The UA may implement these boxes simply as abstractions for inheritance
	and control over the layout of their contents.

<h3 id="line-breaks">
Breaking Across Lines</h3>

	<p>When there is not enough space for an entire <i>ruby container</i> to fit on the line,
	the ruby may be broken wherever all levels simultaneously allow a break.
	Ruby most often breaks between base-annotation sets,
	but if the line-breaking rules allow it, can also break within a <i>ruby base</i>
	(and, in parallel, its associated <i>ruby annotation boxes</i>).

	<p>Whenever ruby breaks across lines, <i>ruby annotations</i> must stay
	with their respective <i>ruby bases</i>.
	The line <em>must not</em> break between a <i>ruby base</i> and its <i>annotations</i>,
	even in the case of ''inter-character'' <i>annotations</i>.

	<div class="figure">
		<img src="images/r-break-b.gif"
		     alt='Diagram showing the line breaking opportunity in a "Bopomofo" ruby'>
		<p class="caption">''inter-character'' ruby line breaking opportunity
	</div>

<h4 id="break-between">
Breaking Between Bases</h4>

	<p>In typical cases, <i>ruby base boxes</i> and <i>ruby annotation boxes</i>
	are styled to forbid internal line wrapping and do not contain forced breaks.
	(See <a href="#default-stylesheet">Appendix A</a>.)
	In such cases the <i>ruby container</i> can only break between adjacent <i>ruby bases</i>,
	and only if no <i>ruby annotations</i> span those <i>ruby bases</i>.

	<div class="figure">
		<p><img src="images/r-break-a.gif"
		     alt="Diagram showing the line breaking opportunity in a complex ruby">
		<p class="caption">Ruby line breaking opportunity
	</div>

	<p>Whether ruby can break between two adjacent <i>ruby bases</i>
	is controlled by normal line-breaking rules for the base text,
	exactly as if the <i>ruby bases</i> were adjacent inline boxes.
	(The annotations are ignored when determining soft wrap opportunities for the <i>base level</i>.)

	<div class="example">
		<p>For example, if two adjacent ruby bases are “蝴” and “蝶”,
		the line may break between them,
		because lines are normally allowed to break between two Han characters.
		However, if 'word-break' is ''keep-all'', that line break is forbidden.
		<pre>&lt;ruby>蝴&lt;rt>hú&lt;/rt>蝶&lt;rt>dié&lt;/rt></pre>
	</div>

	<p>Inter-base white space is significant for evaluating line break opportunities between <i>ruby bases</i>.
	As with white space between inlines, it collapses when the line breaks there.
	Similarly, annotation white space is also trimmed at a line break.

	<div class="example">
		<p>For example, given the following markup:
		<pre>&lt;ruby>&lt;rb>one&lt;/rb> &lt;rb>two&lt;/rb> &lt;rt>1&lt;/rt> &lt;rt>2&lt;/rt>&lt;/ruby></pre>
		<p>Due to the space, the line may break between “one” and “two“.
		If the line breaks there, that space&mdash;and the space between “1” and “2”&mdash;disappears,
		in accordance with standard CSS white space processing rules. [[CSS3TEXT]]
	</div>

<h4 id="break-within">
Breaking Within Bases</h4>

	<p>For longer base texts, it is sometimes appropriate to allow breaking within a base-annotation pair.
	For example, if an English sentence is annotated with its Japanese translation,
	allowing the text to wrap allows for reasonable line breaking behavior in the paragraph.

	<p class="issue">
	Insert scanned example so people don't think this is just the ramblings of an insane spec-writer.

	<p>Line-breaking within a <i>ruby base</i> is only allowed if the 'white-space' property
	of the <i>ruby base</i> and all its parallel <i>annotations</i> allow it,
	and there exists a <i>soft wrap opportunity</i> <em>within</em> (i.e. not at the start or end)
	the content of each base/annotation box.
	Since there is no structural correspondence between fragments of content
	within <i>ruby bases</i> and <i>annotations</i>,
	the UA may break at any set of opportunities;
	but it is recommended that the UA attempt to proportionally balance
	the amount of content inside each fragment.

	<p>There are no line breaking opportunities within ''inter-character'' <i>annotations</i>.

	<p>Ruby alignment takes place within each fragment, after line-breaking.

<h3 id="bidi">
Bidi Reordering</h3>

	<p>The Unicode bidirectional algorithm reorders logically-stored text for visual presentation
	when characters from scripts of opposing directionalities are mixed
	within a single paragraph.

	<p>To preserve the correspondance of <i>ruby annotations</i>
	to their respective <i>ruby bases</i>,
	a few restrictions must be imposed:
	<ul>
		<li>The contents of a <i>ruby base</i> or <i>ruby annotation</i> must remain contiguous.
		<li><i>Ruby annotations</i> must be reordered together with their <i>ruby bases</i>.
		<li>All <i>ruby bases</i> spanned by a single <i>ruby annotation</i> must remain contiguous.
	</ul>

	<p>To this end,
	<ul>
		<li>
			Bidi isolation is forced on all <i>internal ruby boxes</i> and the <i>ruby container</i>:
			the ''unicode-bidi/normal'' and ''embed'' values of 'unicode-bidi' compute to ''isolate'',
			and ''bidi-override'' computes to ''isolate-override''.
			<p class="note">
				Note this means that implicit bidi reordering does not work across ruby bases,
				so authors will need to ensure that the <i>ruby container</i>’s declared directionality
				does indeed match its contents.
		<li>
			During layout, <i>ruby segments</i> are ordered within the <i>ruby container</i>
			by the 'direction' property of their <i>ruby container</i>.
		<li>
			Within a segment, <i>ruby bases</i> and <i>ruby annotations</i>
			are ordered within their respective containers
			by the 'direction' property of the segment’s <i>ruby base container</i>.
			<span class="note">
				Note this means the 'direction' property on <i>ruby annotation containers</i>
				is ignored for the purpose of layout.
				However, it can still inherit into the container's children
				and thereby affect the <i>inline base direction</i>
				of any <i>ruby annotations</i> it contains.
			</span>
	</ul>

	<p>As with other inline-level content,
	the bidi reordering of <i>internal ruby boxes</i> happens after line-breaking
	so that content is divided across lines according to its logical order.

	<p>See [[CSS3-WRITING-MODES]] for a more in-depth discussion of bidirectional text in CSS.

	<!-- Some alternate proposals exist in the 2013 draft's comments section. -->

<h3 id="line-height">
Line Spacing</h3>

	<p>The 'line-height' property controls spacing between lines in CSS.
	When inline content on line is shorter than the 'line-height',
	half-leading is added on either side of the content,
	as specified in <a href="https://www.w3.org/TR/CSS2/visudet.html#line-height">CSS2.1&sect;10.8</a>. [[!CSS2]]

	<p>In order to ensure consistent spacing of lines,
	documents with ruby typically ensure that the 'line-height' is large enough
	to accommodate ruby between lines of text.
	Therefore, ordinarily, <i>ruby annotation containers</i> and <i>ruby annotation boxes</i>
	do not contribute to the measured height of a line's inline contents;
	any alignment (see 'vertical-align') and line-height calculations
	are performed using only the <i>ruby base container</i>,
	exactly as if it were a normal inline.

	<p>However, if the 'line-height' specified on the <i>ruby container</i>
	is less than the distance between
	the top of the top <i>ruby annotation container</i>
	and the bottom of the bottom <i>ruby annotation container</i>,
	then additional leading is added
	on the appropriate side of the <i>ruby base container</i>
	such that if a block consisted of three lines
	each containing ruby identical to this,
	none of the <i>ruby containers</i> would overlap.

	<p class="note">Note that this does not ensure that the <i>ruby annotations</i> remain within the line box.
	It merely ensures that <em>if all lines had equal spacing</em>
	and equivalent amounts and positioning of <i>ruby annotations</i>,
	there would be enough room to avoid overlap.

	<p>Authors should ensure appropriate 'line-height' and 'padding' to accommodate ruby,
	and be particularly careful at the beginning or end of a block
	and when a line contains inline-level content
	(such as images, inline blocks, or elements shifted with 'vertical-align')
	taller than the paragraph's default font size.

	<div class="figure">
		<p><img src="images/rlh-a.gif"
		        alt="The content of each line sits in the middle of its line height;
		             the additional space on each side is called half-leading.
		             Ruby fits between lines if it is smaller than twice the half-leading,
		             but this means that it occupies space belonging to the half-leading of the previous line.">
		<p class="caption">Ruby annotations will often overflow the line;
		authors should ensure content over/under a ruby-annotated line
		is adequately spaced to leave room for the ruby.
	</div>

	<p class="note">More control over how ruby affects alignment and line layout
	will be part of the CSS Line Layout Module Level 3.
	Note, it is currently in the process of being rewritten;
	the current drafts should not be relied upon.

<h2 id="ruby-props">
Ruby Formatting Properties</h2>

	<p>The following properties are introduced to control ruby
	<a href="#rubypos">positioning</a>,
	<a href="#collapsed-ruby">text distribution</a>,
	and <a href="#rubyalign">alignment</a>.

<h3 id="rubypos">
Ruby Positioning: the 'ruby-position' property</h3>

	<pre class="propdef">
	Name: ruby-position
	Value: over | under | inter-character
	Initial: over
	Applies to: ruby annotation containers
	Inherited: yes
	Computed value: specified keyword
	Animation type: discrete
	</pre>

	<p>This property controls position of the ruby text with respect to its base.
	Values have the following meanings:

	<p class="issue"><span class="issuehead">Issue-107:&nbsp;</span> Roland Steiner has requested the addition of an auto value as default. See <a href="http://www.w3.org/Search/Mail/Public/advanced_search?keywords=&amp;hdr-1-name=subject&amp;hdr-1-query=ruby-position%3A+undesirable+default+value+%27before%27+for+complex+ruby&amp;hdr-2-name=from&amp;hdr-2-query=&amp;hdr-3-name=message-id&amp;hdr-3-query=&amp;period_month=&amp;period_year=&amp;index-grp=Public__FULL&amp;index-type=t&amp;type-index=www-style&amp;resultsperpage=20&amp;sortby=date">this thread</a> and <a href="http://www.w3.org/Search/Mail/Public/advanced_search?keywords=&amp;hdr-1-name=subject&amp;hdr-1-query=Styling+of+complex+Ruby&amp;hdr-2-name=from&amp;hdr-2-query=&amp;hdr-3-name=message-id&amp;hdr-3-query=&amp;period_month=&amp;period_year=&amp;index-grp=Public__FULL&amp;index-type=t&amp;type-index=public-i18n-core&amp;resultsperpage=20&amp;sortby=date">this one</a>. Current proposal is to add an optional ''alternate?'' keyword.</p>
	<dl dfn-for=ruby-position dfn-type=value>
		<dt><dfn>over</dfn>
		<dd>The ruby text appears <a>line-over</a> the base.

			<div class="figure">
				<p><img src="images/shinkansen-top.gif"
				        alt="Diagram of ruby glyph layout in horizontal mode with ruby text appearing above the base">
				<p class="caption">Ruby over Japanese base text in horizontal layout
			</div>
			<div class="figure">
				<p><img src="images/shinkansen-right.gif" width="33"
				        alt="Diagram of ruby glyph layout in vertical mode with ruby text appearing vertically on the right of the base">
				<p class="caption">Ruby to the right of Japanese base text in vertical layout
			</div>
		</dd>

		<dt><dfn>under</dfn>
		<dd>The ruby text appears <a>line-under</a> the base.
			This is a relatively rare setting used in ideographic East Asian writing systems,
			most easily found in educational text.

			<div class="figure">
				<p><img src="images/shinkansen-bottom.gif"
				        alt="Diagram of ruby glyph layout in horizontal mode with ruby text appearing below the base">
				<p class="caption">Ruby under Japanese base text in horizontal layout
			</div>
			<div class="figure">
				<p><img src="images/shinkansen-left.gif"
				        alt="Diagram of ruby glyph layout in vertical mode with ruby text appearing vertically on the left of the base">
				<p class="caption">Ruby to the left of Japanese base text in vertical layout
			</div>
		</dd>

		<dt><dfn>inter-character</dfn></dt>
		<dd>
			<p>The ruby text appears on the right of the base in horizontal text.
			This value forces the computed value of 'writing-mode' of the <i>ruby annotation container</i> to be ''vertical-rl''.

			<p>This value is provided for the special case of traditional Chinese
			as used especially in Taiwan:
			ruby (made of <a href="#g-bopomofo">bopomofo</a> glyphs) in that context
			appears vertically along the right side of the base glyph,
			even when the layout of the base characters is horizontal:

				<div class="figure">
					<p><img src="images/bopomofo.gif"
					        alt="Example of Taiwanese-style ruby">
					<p class="caption">“Bopomofo” ruby in traditional Chinese
					(ruby text shown in blue for clarity) in horizontal layout
				</div>
		</dd>
	</dl>

	<p>If multiple <i>ruby annotation containers</i> have the same 'ruby-position',
	they stack along the block axis,
	with lower levels of annotation closer to the base text.

<h3 id="collapsed-ruby">
Sharing Annotation Space: the 'ruby-merge' property</h3>

	<pre class="propdef">
	Name: ruby-merge
	Value: separate | collapse | auto
	Initial: separate
	Applies to: ruby annotation containers
	Inherited: yes
	Computed value: specified keyword
	Animation type: by computed value type
	</pre>

	<p>
		This property controls how ruby annotation boxes should be rendered
		when there are more than one in a ruby container box:
		whether each pair should be kept separate,
		the annotations should be <dfn lt="collapsed annotation">collapsed</dfn> and rendered as a group,
		or the separation should be determined based on the space available.

	<p>Possible values:</p>
	<dl dfn-for=ruby-merge dfn-type=value>
		<dt><dfn>separate</dfn>
		<dd>
			<p>
				Each ruby annotation box is rendered in the same column(s) as its corresponding base box(es).
				This style is called “mono ruby” in [[JLREQ]].

			<div class="example">
				<p>For example, the following two markups render the same:
				<pre>&lt;ruby&gt;無&lt;rt&gt;む&lt;/ruby&gt;&lt;ruby&gt;常&lt;rt&gt;じょう&lt;/ruby&gt;</pre>
				<p>and:
				<pre>&lt;ruby style="ruby-merge:separate"&gt;&lt;rb&gt;無&lt;rb&gt;常&lt;rt&gt;む&lt;rt&gt;じょう&lt;/ruby&gt;</pre>
			</div>
		</dd>

		<dt><dfn>collapse</dfn>
		<dd>
			<p>
				All <i>ruby annotation boxes</i> within the same <i>ruby segment</i> on the same line are concatenated,
				and laid out as if their contents belonged to a single <i>ruby annotation box</i>
				spanning all their associated <i>ruby base boxes</i>.
				This style renders similar to “group ruby” in [[JLREQ]],
				except that <i>ruby annotations</i> are kept together with their respective <i>ruby bases</i> when breaking lines.
			</p>

			<div class="example">
				<p>The following two markups render the same both characters fit on one line:
				<pre>&lt;ruby&gt;無常&lt;rt&gt;むじょう&lt;/ruby&gt;</pre>
				<p>and:
				<pre>&lt;ruby style="ruby-merge:collapse"&gt;&lt;rb&gt;無&lt;rb&gt;常&lt;rt&gt;む&lt;rt&gt;じょう&lt;/ruby&gt;</pre>
				<p>However, the second one renders the same as ''ruby-position: separate''
				when the two bases are split across lines.
			</div>
		</dd>

		<dt><dfn>auto</dfn></dt>
		<dd>
			<p>
				The user agent may use any algorithm to determine how each ruby annotation box
				is rendered to its corresponding base box,
				with the intention that if all annotations fit over their respective bases,
				the result is identical to “mono ruby”,
				but if some annotations are wider than their bases
				the space is shared in some way
				to avoid imposing space between bases.
			<div class="example">
			<p>
				One possible algorithm is described as “jukugo ruby” in [[JLREQ]].
			<p>
				Another, more simplified algorithm of “jukugo ruby” is
				to render as ''separate'' if all ruby annotation boxes fit
				within the advances of their corresponding base boxes,
				and render as ''collapse'' otherwise.
			</p>
			</div>
		</dd>
	</dl>

<h3 id="ruby-align-property"><span id="rubyalign"></span>
Ruby Text Distribution: the 'ruby-align' property</h3>

	<pre class="propdef">
	Name: ruby-align
	Value: start | center | space-between | space-around
	Initial: space-around
	Applies to: ruby bases, ruby annotations, ruby base containers, ruby annotation containers
	Inherited: yes
	Computed value: specified keyword
	Animation type: by computed value type
	</pre>

	<p>This property specifies how text is distributed within the various ruby boxes
	when their contents do not exactly fill their respective boxes.
	Note that space distributed by 'ruby-align' is unrelated to, and independent of,
	any space distributed due to justification.

	<p>Values have the following meanings:
	<dl dfn-for=ruby-align dfn-type=value>
		<dt><dfn>start</dfn></dt>
		<dd>The ruby content is aligned with the start edge of its box.
			<div class="figure">
				<p><img
					alt="Diagram of glyph layout in left aligned ruby when ruby text is shorter than base"
					width="145" height="91" src="images/ra-l.gif" /><img
					width="145" height="91"
					alt="Diagram of glyph layout in left aligned ruby when ruby text is longer than base"
					src="images/ra-l-rb.gif" />
				<p class="caption">''ruby-align/start'' ruby distribution
			</div>
		</dd>

		<dt><dfn>center</dfn></dt>
		<dd>The ruby content is centered within its box.
			<div class="figure">
				<p><img width="145" height="91"
					alt="Diagram of glyph layout in center aligned ruby when ruby text is shorter than base"
					src="images/ra-c.gif" /><img width="145" height="91"
					alt="Diagram of glyph layout in center aligned ruby when ruby text is longer than base"
					src="images/ra-c-rb.gif" />
				<p class="caption">''ruby-align/center'' ruby distribution
			</div>
		</dd>

		<dt><dfn>space-between</dfn></dt>
		<dd>
			<p>The ruby content expands as defined for normal text justification
				(as defined by 'text-justify'),
				except that if there are no <i>justification opportunities</i>
				the content is centered.
			<div class="figure">
				<p><img width="145" height="91"
				alt="Diagram of glyph layout in distribute-letter aligned ruby when ruby text is shorter than base"
				src="images/ra-dl.gif" /><img width="145" height="91"
				alt="Diagram of glyph layout in distribute-letter aligned ruby when ruby text is longer than base"
				src="images/ra-dl-rb.gif" />
				<p class="caption">''ruby-align/space-between'' ruby distribution
			</div>
		</dd>

		<dt><dfn>space-around</dfn></dt>
		<dd>
			<p>As for ''ruby-align/space-between''
			except that there exists an extra <i>justification opportunities</i>
			whose space is distributed half before and half after the ruby content.
			<div class="example">
				<p>Since a typical implementation will by default define <i>justification opportunities</i>
					between every adjacent pair of CJK <i>characters</i>
					and not between adjacent pairs of Latin <i>characters</i>,
					this should result in the behavior recommended by [[JLREQ]]:
					for wide-cell ruby content to be distributed...
				<div class="figure">
					<p><img width="145" height="91"
					alt="Diagram of glyph layout in auto aligned ruby when ruby text is shorter than base"
					src="images/ra-ds.gif" /><img width="145" height="91"
					alt="Diagram of glyph layout in auto aligned ruby when ruby text is longer than base"
					src="images/ra-ds-rb.gif" />
					<p class="caption">Wide-cell text in ''ruby-align/space-around'' ruby distribution is spaced apart
				</div>
				<p>... and narrow-cell glyph ruby to be centered.
				<div class="figure">
					<p><img
					alt="Diagram of glyph layout in auto aligned ruby when halfwidth ruby text is shorter than base"
					width="145" height="91"
					src="images/ra-c-h.gif" /><img
					alt="Diagram of character layout in auto aligned ruby when ruby text is longer than narrow-width base"
					width="145" height="91"
					src="images/ra-c-rb-h.gif" />
					<p class="caption">Narrow-width ruby text in ''ruby-align/space-around'' ruby distribution is centered
				</div>
			</div>
		</dd>
	</dl>

	<p class="issue">Add a paragraph explaining how to distribute space in situations with spanning annotations.
<!--
	<p>For a complex ruby with spanning elements, one additional consideration is
		required. If the spanning element spans multiple 'rows' (other rbc or rtc
		elements), and the ruby alignment requires space distribution among the
		'spanned' elements, a ratio must be determined among the 'columns' of spanned
		elements. This ratio is computed by taking into consideration the widest
		element within each column.</p>
-->

<h3 id="ruby-text-decoration">
Ruby Text Decoration</h3>

	<p>Text decoration does not propagate from the base text to the annotations.
	<p>When text decoration is specified on an ancestor of the ruby,
	it is drawn across the entire content area of the ruby base container,
	including any extra space added on either side of the ruby base contents to accommodate long annotations.
	When text decoration is specified on a ruby base itself,
	this extra space is not decorated,
	similar to how a box's own padding is not decorated when text decoration is specified directly on that box. [[!CSS3-TEXT-DECOR]]

	<p>Text decoration may be specified directly on ruby base containers
	and ruby annotation containers:
	in such cases it is propagated to all of the container's bases or annotations (respectively),
	and is also drawn between them for continuity.

	<p>The positions of ruby annotations may be adjusted
	to avoid potential conflicts
	with overline and underline decorations applied to the base text.
	In the basic case of consistent font size and baseline alignment,
	an underline or overline is positioned
	between the base level and any annotations on that side.

	<p class="issue">This section needs some clarification about
	drawing decorations between the content of adjacent bases/annotations.
	Depends on if those boxes are as wide as their column or not.

<h2 id="edge-effects">
Edge Effects</h2>

<h3 id="ruby-overhang">
Overhanging Ruby</h3>

	<p>
		When <i>ruby annotation box</i> is longer than its corresponding <i>ruby base box</i>,
		the <i>ruby annotation box</i> may partially overhang adjacent boxes.
	</p>
	<p>
		This level of the specification does not define
		how much the overhang may be allowed, and under what conditions.
	</p>

	<p>If the ruby text is not allowed to overhang,
	then the ruby behaves like a traditional inline box,
	i.e. only its own contents are rendered within its boundaries
	and adjacent elements do not cross the box boundary:

	<div class="figure">
		<p><img src="images/ro-n.gif"
		        alt="Diagram showing the ruby boxes interacting with adjacent text">
		<p class="caption">Simple ruby whose text is not allowed to overhang adjacent text
	</div>

	<p>However, if <i>ruby annotation</i> content is allowed to overhang adjacent elements
	and it happens to be wider than its base,
	then the adjacent content is partially rendered within the area of the <i>ruby container box</i>,
	while the <i>ruby annotation</i> may partially overlap the upper blank parts of the adjacent content:

	<div class="figure">
	<p><img src="images/ro-a.gif"
		      alt="Diagram showing the ruby boxes interacting with adjacent text">
	<p class="caption">Simple ruby whose text is allowed to overhang adjacent text
	</div>

	<p>The <i>ruby annotations</i> related to a <i>ruby base</i>
	must never overhang another <i>ruby base</i>.

	<p>The alignment of the contents of the base or the ruby text
	is not affected by overhanging behavior.
	The alignment is achieved the same way regardless of the overhang behavior setting
	and it is computed before the space available for overlap is determined.
	It is controlled by the 'ruby-align' property.

	<p class="issue">
		I suspect overhanging interacts with alignment in some cases;
		might need to look into this later.

	<p>This entire logic applies the same way in vertical ideographic layout,
	only the dimension in which it works in such a layout is vertical,
	instead of horizontal.

	<div class="example">
	<p>
		The user agent may use [[JIS4051]] recommendation of
		using one ruby text character length as the maximum overhang length.
		Detailed rules for how ruby text can overhang adjacent characters for Japanese are described by [[JLREQ]].
	</p>
	</div>

<h3 id="line-edge">
Line-edge Alignment</h3>

	<p>
		When a <i>ruby annotation box</i> that is longer than its <i>ruby base</i>
		is at the start or end edge of a line,
		the user agent <em>may</em> force the side of the <i>ruby annotation</i> that touches the edge of the line
		to align to the corresponding edge of the base.
		This type of alignment is described by [[JLREQ]].
	</p>
	<p>
		This level of the specification does not provide a mechanism to control this behavior.
	</p>
	<div class="figure">
		<p><img src="images/ra-le-l.gif"
			alt="Diagram of glyph layout in line-edge aligned ruby when ruby text is shorter than base">
			<img src="images/ra-le-r.gif"
			alt="Diagram of glyph layout in line-edge aligned ruby when ruby text is longer than base">
		<p class="caption">Line-edge alignment
	</div>

	<!--
<h3 id="rubyover">
Ruby overhanging: the 'ruby-overhang' property</h3>

  <table class="propdef">
    <tr>
      <th>Name:
      <td><dfn>ruby-overhang</dfn>
    <tr>
      <th>Value:
      <td>auto | start | end | none
    <tr>
      <th>Initial:
      <td>none
    <tr>
      <th>Applies to:
      <td>the parent of elements with display: ruby-text
    <tr>
      <th>Inherited:
      <td>yes
    <tr>
      <th>Percentages:
      <td>N/A
    <tr>
      <th>Computed value:
      <td>specified value (except for initial and inherit)
  </table>

<p>This property determines whether, and on which side, ruby text is allowed
to partially overhang any adjacent text in addition to its own base, when the
ruby text is wider than the ruby base. Note that ruby text is never allowed to
overhang glyphs belonging to another ruby base. <span class="issue"><span class="issuehead">Issue:&nbsp;</span> This rule must be broken if we are to allow support for jukugo ruby.</span> Also the user agent is free to assume
a maximum amount by which ruby text may overhang adjacent text. The user agent may use
the [[JIS4051]] recommendation of using one ruby text character
length as the maximum overhang length. Detailed rules for how ruby text can overhang adjacent characters for Japanese are described by [[JLREQ]].</p>

<p>Possible values:</p>
<dl>
  <dt><strong>auto</strong></dt>
    <dd>The ruby text can overhang text adjacent to the base on either side.   	  [[JLREQ]] and [[JIS4051]] specify the categories of characters that
      ruby text can overhang. The user agent is free to follow those recommendations or specify its own classes of
      characters to overhang. This is the initial value.
		<div class="figure">
      <p><img class="example" width="177" height="91"
      alt="Diagram of glyph layout in overhanging ruby" src="images/ro-a.gif" /></p>
      <p><b>Figure 4.3.1</b>: Ruby overhanging adjacent text</p>
      </div>
    </dd>
  <dt><strong>start</strong></dt>
    <dd>The ruby text can only overhang the text that precedes it. That means, for
      example, that ruby cannot overhang text that is to the right of it in
      horizontal LTR layout, and it cannot overhang text that is below it in
      vertical-ideographic layout.
		<div class="figure">
      <p><img class="example" width="199" height="91"
      alt="Diagram of glyph layout when ruby overhangs the preceding glyphs only"
      src="images/ro-s.gif" /></p>
      <p><b>Figure 4.3.2</b>: Ruby overhanging preceding text only</p>
      </div>
    </dd>
  <dt><strong>end</strong></dt>
    <dd>The ruby text can only overhang the text that follows it. That means, for
      example, that ruby cannot overhang text that is to the left of it in
      horizontal LTR layout, and it cannot overhang text that is above it in
      vertical-ideographic layout.
		<div class="figure">
      <p><img class="example" width="198" height="91"
      alt="Diagram of glyph layout when ruby overhangs the following characters only"
      src="images/ro-e.gif" /></p>
      <p><b>Figure 4.3.3</b>: Ruby overhanging following text only</p>
      </div>
    </dd>
  <dt><strong>none</strong></dt>
    <dd>The ruby text cannot overhang any text adjacent to its base, only its
      own base.

      <div class="figure">
      <p><img class="example" width="220" height="91"
      alt="Diagram of glyph layout in non-overhanging ruby"
      src="images/ro-n.gif" /></p>
      <p><b>Figure 4.3.4</b>: Ruby not allowed to overhang adjacent text</p>
      </div>
    </dd>
</dl>

<h3 id="rubyspan">
Ruby annotation spanning: the 'ruby-span' property</h3>

  <table class="propdef">
    <tr>
      <th>Name:
      <td><dfn>ruby-span</dfn>
    <tr>
      <th>Value:
      <td>attr(x) |  none
    <tr>
      <th>Initial:
      <td>none
    <tr>
      <th>Applies to:
      <td>elements with display: ruby-text
    <tr>
      <th>Inherited:
      <td>no
    <tr>
      <th>Percentages:
      <td>N/A
    <tr>
      <th>Computed value:
      <td>&lt;number&gt;
  </table>

<p>This property controls the spanning behavior of annotation elements. </p>

<p class="note"><span class="note-label">Note:</span> A XHTML user agent may also use the <samp>rbspan</samp>
attribute to get the same effect.</p>

<p>Possible values:</p>

<dl>
  <dt><strong>attr(x)</strong></dt>
    <dd>The value of attribute 'x' as a string value. The string value is
    evaluated as a &lt;number&gt; to determine the number of ruby base elements to be
    spanned by the annotation element. If the &lt;number&gt; is &#39;0&#39;, it is replaced by
    &#39;1&#39;.The &lt;number&gt; is the computed value. </dd>
  <dt>none</dt>
  <dd>No spanning. The computed value is &#39;1&#39;.</dd>
</dl>

<p>The following example shows an XML example using the 'display' property
values associated with the 'ruby structure and the 'ruby-span' property</p>
<pre class="xml">myruby       { display: ruby; }
myrbc        { display: ruby-base-container; }
myrb         { display: ruby-base; }
myrtc.before { display: ruby-text-container; ruby-position: before}
myrtc.after  { display: ruby-text-container; ruby-position: after}
myrt         { display: ruby-text; ruby-span: attr(rbspan); }
...
&lt;myruby&gt;
  &lt;myrbc&gt;
    &lt;myrb&gt;10&lt;/myrb&gt;
    &lt;myrb&gt;31&lt;/myrb&gt;
    &lt;myrb&gt;2002&lt;/myrb&gt;
  &lt;/myrbc&gt;
  &lt;myrtc class=&quot;before&quot;&gt;
    &lt;myrt&gt;Month&lt;/myrt&gt;
    &lt;myrt&gt;Day&lt;/myrt&gt;
    &lt;myrt&gt;Year&lt;/myrt&gt;
  &lt;/myrtc&gt;
  &lt;myrtc class=&quot;after&quot;&gt;
    &lt;myrt rbspan=&quot;3&quot;&gt;Expiration Date&lt;/myrt&gt;
  &lt;/myrtc&gt;
&lt;/myruby&gt;</pre>
	-->

<h2 id="default-stylesheet" class="no-num">
Appendix A: Default Style Sheet</h2>

	<p><em>This section is informative.</em>

<h3 id="default-ua-ruby" class="no-num">
<span class="secno">A.1</span> Supporting Ruby Layout</h3>

	<p>The following represents a default UA style sheet
	for rendering HTML and XHTML ruby markup as ruby layout:

	<pre>
		ruby { display: ruby; }
		rp   { display: none; }
		rbc  { display: ruby-base-container; }
		rtc  { display: ruby-text-container; }
		rb   { display: ruby-base; white-space: nowrap; }
		rt   { display: ruby-text; }
		ruby, rb, rt, rbc, rtc { unicode-bidi: isolate; }

		rtc, rt {
			font-variant-east-asian: ruby;  /* See [[CSS3-FONTS]] */
			text-emphasis: none;            /* See [[CSS3-TEXT-DECOR]] */
			white-space: nowrap;
			line-height: 1; }
		rtc:lang(zh), rt:lang(zh) {
			ruby-align: center; }

		rtc, :not(rtc) > rt {
			font-size: 50%; }
		rtc:lang(zh-TW), :not(rtc) > rt:lang(zh-TW) {
			font-size: 30%; }               /* bopomofo */
	</pre>

	<p class="note">Authors should not use the above rules:
	a UA that supports ruby layout should provide these by default.

<h3 id="default-inline" class="no-num">
<span class="secno">A.2</span> Inlining Ruby Annotations</h3>

	<p>The following represents a sample style sheet
	for rendering HTML and XHTML ruby markup as inline annotations:

	<pre>ruby, rb, rt, rbc, rtc, rp {
<!--	-->  display: inline; white-space: inherit;
<!--	-->  font: inherit; text-emphasis: inherit; }</pre>

<h3 id="default-parens" class="no-num">
<span class="secno">A.3</span> Generating Parentheses</h3>

	<p>Unfortunately, because Selectors cannot match against text nodes,
	it's not possible with CSS to express rules that will automatically and correctly
	add parentheses to unparenthesized ruby annotations in HTML.
	(This is because HTML ruby allows implying the <i>ruby base</i> from raw text, without a corresponding element.)
	However, these rules will handle cases where either <code>&lt;rb&gt;</code>
	or <code>&lt;rtc&gt;</code> is used rigorously.

	<pre>
<!--	-->/* Parens around &lt;rtc> */
<!--	-->rtc::before { content: "("; }
<!--	-->rtc::after  { content: ")"; }

<!--	-->/* Parens before first &lt;rt> not inside &lt;rtc> */
<!--	-->rb  + rt::before,
<!--	-->rtc + rt::before { content: "("; }

<!--	-->/* Parens after &lt;rt> not inside &lt;rtc> */
<!--	-->rb ~ rt:last-child::after,
<!--	-->rt + rb::before  { content: ")"; }
<!--	-->rt + rtc::before { content: ")("; }</pre>

<h2 id="glossary">
Glossary</h2>
<dl>
  <dt id="g-bopomofo" lang="zh">Bopomofo
    <dd>37 characters and 4 tone markings used as phonetics in Chinese,
      especially standard Mandarin.
      <p class="note">
          Note that the user agent is responsible for ensuring the correct relative alignment and positioning of the glyphs,
          including bopomofo tone marks, when displaying text,
          whether it occurs in ruby annotations or as normal inline text.
          Bopomofo Tone marks are spacing characters that occur (in memory) at the end of the ruby text for each base character.
          They are usually displayed in a separate column to the right of or above the bopomofo characters,
          and the position of the tone mark depends on the number of characters in the syllable.
          One tone mark, however, is placed before the bopomofo, not over it.
      <!-- See Taiwanese requirements doc for EPUB at http://epub-revision.googlecode.com/files/EGLS_TW_eng.ppt -->
  <dt id="g-hanja" lang="ko">Hanja
    <dd>Subset of the Korean writing system that utilizes ideographic
      characters borrowed or adapted from the Chinese writing system. Also see
      <a href="#g-kanji"><span lang="ja">Kanji</span></a>.</dd>
  <dt id="g-hiragana" lang="ja">Hiragana
    <dd>Japanese syllabic script, or character of that script. Rounded and
    cursive in appearance. Subset of the Japanese writing system, used together
    with kanji and katakana. In recent times, mostly used to write Japanese
    words when kanji are not available or appropriate, and word endings and
    particles. Also see <a
      href="#g-katakana"><span lang="ja">Katakana</span></a>.</dd>
  <dt id="g-ideogram">Ideograph
    <dd>A character that is used to represent an idea, word, or word component,
    in contrast to a character from an alphabetic or syllabic script. The most
    well-known ideographic script is used (with some variation) in East Asia
    (China, Japan, Korea,...).</dd>
  <dt id="g-kana" lang="ja">Kana
    <dd>Collective term for hiragana and katakana.</dd>
  <dt id="g-kanji">Kanji
    <dd>Japanese term for ideographs; ideographs used in Japanese. Subset of the
    Japanese writing system, used together with hiragana and katakana. Also see <a
      href="#g-hanja"><span lang="ko">Hanja</span></a>.</dd>
  <dt id="g-katakana" lang="ja">Katakana
    <dd>Japanese syllabic script, or character of that script. Angular in
    appearance. Subset of the Japanese writing system,&nbsp; used together with
    kanji and hiragana. In recent times, mainly used to write foreign words. Also see <a
      href="#g-hiragana"><span lang="ja">Hiragana</span></a>.</dd>
</dl>

<h2 class=no-num id="acknowledgments">
Acknowledgments</h2>

	<p>This specification would not have been possible without the help from:</p>

	<p>
	David Baron,
	Robin Berjon,
	Stephen Deach,
	Martin Dürst,
	Hideki Hiura (<span lang="ja">樋浦 秀樹</span>),
	Masayasu Ishikawa (<span lang="ja">石川雅康</span>),
	Taichi Kawabata,
	Chris Pratley,
	Xidorn Quan,
	Takao Suzuki (<span lang="ja">鈴木 孝雄</span>),
	Frank Yung-Fong Tang,
	Chris Thrasher,
	Masafumi Yabe (<span lang="ja">家辺勝文</span>),
	Boris Zbarsky,
	Steve Zilles.

	<p>Special thanks goes to the previous editors:
	Michel Suignard and Marcin Sawicki of Microsoft,
	and Richard Ishida of W3C.

<h2 class="no-num" id="changes">
Changes</h2>

	<p>The following major changes have been made since the previous Working Draft:
	<ul>
		<li>Rewrote anonymous box generation rules and white space handling rules,
		defined specialized pairing of anonymous white space boxes.
		<li>Took nested ruby handling out of pairing.
		(Will be handling it via sizing/layout.)
		<li>Defined bidi layout of ruby structures.
	</ul>