Fri, 13 Sep 2013 13:34:40 -0700
[css-ruby] Add more explanation for ruby-merge: auto, and check in missing image.
1 <!--
3 Issues:
4 bidi
5 box layout/sizing
6 clean up inter-character vs. parallel layout requirements
8 Redo all examples with consistent font. (M+ 2p?)
10 -->
12 <!DOCTYPE html>
13 <html lang="en">
14 <head>
15 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
16 <title>CSS Ruby Module Level 1</title>
17 <link rel=contents href="#contents">
18 <link rel=index href="#index">
19 <link rel="stylesheet" type="text/css" href="../default.css">
20 <link href="../csslogo.ico" rel="shortcut icon" type="image/x-icon">
21 <link rel="stylesheet" type="text/css" href="http://www.w3.org/StyleSheets/TR/W3C-[STATUS].css">
22 </head>
24 <body class="h-entry">
26 <div class="head">
27 <!--logo-->
29 <h1 class="p-name">CSS Ruby Module Level 1</h1>
31 <h2 class="no-num no-toc">[LONGSTATUS] <time class="dt-updated" datetime="[CDATE]">[DATE]</time> <!-- for HTML4 doctype: <span class="value-title" title="[CDATE]">[DATE]</span></span> --> </h2>
32 <dl>
33 <dt>This version:
34 <dd><a class="u-url" href="[VERSION]">[VERSION]</a>
36 <dt>Latest version:
37 <dd><a href="http://www.w3.org/TR/[SHORTNAME]/">http://www.w3.org/TR/css3-ruby/</a>
39 <dt>Editor's draft:
40 <dd><a href="http://dev.w3.org/csswg/[SHORTNAME]/">http://dev.w3.org/csswg/[SHORTNAME]/</a>
41 (<a href="https://dvcs.w3.org/hg/csswg/log/tip/[SHORTNAME]/Overview.src.html">change log</a>)
43 <dt>Previous version:
44 <dd><a href="http://www.w3.org/TR/2011/WD-css3-ruby-20110630/">
45 http://www.w3.org/TR/2011/WD-css3-ruby-20110630/</a>
47 <dt>Issue Tracking:</dt>
48 <dd><a rel="issues" href="http://www.w3.org/Style/CSS/Tracker/products/FIXME">http://www.w3.org/Style/CSS/Tracker/products/FIXME</a>
50 <dt>Feedback:</dt>
51 <dd><a href="mailto:www-style@w3.org?subject=%5BSHORTNAME%5D%20feedback"
52 >www-style@w3.org</a>
53 with subject line “<kbd>[[SHORTNAME]]
54 <var>… message topic …</var></kbd>”
55 (<a rel="discussion" href="http://lists.w3.org/Archives/Public/www-style/"
56 >archives</a>)
58 <dt>Editors:
59 <dd class="p-author h-card vcard">
60 <a class="p-name fn u-url url" rel="author"
61 href="http://fantasai.inkedblade.net/contact">Elika J. Etemad</a>,
62 <a class="p-org org h-org" href="http://www.mozilla.org/">Mozilla</a>
63 <dd class="p-author h-card vcard">
64 <a class="p-name fn u-url url" rel="author"
65 href="mailto:koji.a.ishii@mail.rakuten.com">Koji Ishii</a>,
66 <span class="p-org org">Rakuten, Inc.</span>
67 <dd class="p-author h-card vcard">
68 <a class="p-name fn u-url url" rel="author"
69 href="mailto:ishida@w3.org">Richard Ishida</a>,
70 <span class="p-org org">W3C</span>
72 <dt>Former editors:
73 <dd>Michel Suignard, Microsoft
74 <dd>Marcin Sawicki, Microsoft
75 </dl>
77 <!--copyright-->
79 <hr title="Separator for header">
80 </div>
82 <h2 class="no-num no-toc" id="abstract">Abstract</h2>
84 <p>
85 <span class="p-summary">
86 “Ruby” are short runs of text alongside the base text,
87 typically used in East Asian documents to indicate pronunciation
88 or to provide a short annotation.
89 This module describes the rendering model and formatting controls
90 related to displaying ruby annotations in CSS.
91 </span>
93 <a href="http://www.w3.org/TR/CSS/">CSS</a> is a language for describing
94 the rendering of structured documents (such as HTML and XML) on screen, on
95 paper, in speech, etc.
97 <h2 class="no-num no-toc" id="status">Status of this document</h2>
99 <!--status-->
101 <p>The following features are at risk: …
103 <h2 class="no-num no-toc" id="contents">
104 Table of Contents</h2>
106 <!--toc-->
108 <h2 id="intro">
109 Introduction</h2>
111 <p><em>This section is not normative.</em>
113 <h3 id="placement">
114 Module interactions</h3>
116 <p>This module extends the inline box model of CSS Level 2 [[!CSS21]]
117 to support ruby.
119 <p>None of the properties in this module apply to the <code>::first-line</code> or
120 <code>::first-letter</code> pseudo-elements.
122 <h3 id="values">
123 Values</h3>
125 <p>This specification follows the
126 <a href="http://www.w3.org/TR/CSS21/about.html#property-defs">CSS property
127 definition conventions</a> from [[!CSS21]]. Value types not defined in
128 this specification are defined in CSS Level 2 Revision 1 [[!CSS21]].
129 Other CSS modules may expand the definitions of these value types: for
130 example [[CSS3VAL]], when combined with this module, expands the
131 definition of the <var><length></var> value type as used in this specification.</p>
133 <p>In addition to the property-specific values listed in their definitions,
134 all properties defined in this specification also accept the
135 <a href="http://www.w3.org/TR/CSS21/cascade.html#value-def-inherit">inherit</a>
136 keyword as their property value. For readability it has not been repeated
137 explicitly.
139 <h3 id="conventions">
140 Document conventions</h3>
142 <p>Many typographical conventions in East Asian typography depend
143 on whether the character rendered is wide (CJK) or narrow (non-CJK).
144 There are a number of illustrations in this document
145 for which the following legend is used:
147 <dl>
148 <dt><img alt="Symbolic wide-cell glyph representation" width="39" height="39" src="images/fullwidth.gif">
149 <dd>Wide-cell glyph (e.g. Han) that is the <var>n</var>th character in the text run.
150 They are typically sized to 50% when used as annotations.
151 <dt><img alt="Symbolic narrow-cell glyph representation" width="19" height="39" src="images/halfwidth.gif">
152 <dd>Narrow-cell glyph (e.g. Roman) which is the <var>n</var>th glyph in the text run.
153 </dl>
155 <p>The orientation which the above symbols assume in the diagrams
156 corresponds to the orientation that the glyphs they represent
157 are intended to assume when rendered by the user agent.
158 Spacing between these characters in the diagrams is incidental,
159 unless intentionally changed to make a point.
161 <h3 id="ruby-def">
162 What is ruby?</h3>
164 <p><dfn>Ruby</dfn> is the commonly-used name for a run of text
165 that appears alongside another run of text (referred to as the “base”)
166 and serves as an annotation or a pronunciation guide associated with that run of text.
168 <p>The following figures show two examples of Ruby,
169 a simple case and one with more complicated structure.
171 <div class="example">
172 <p>In this first example, a single annotation is used to annotate the base text.
173 <div class="figure">
174 <p><img src="images/licence.png"
175 alt="Example of ruby applied on top of a Japanese expression">
176 <p class="caption">Example of ruby used in Japanese (simple case)
177 </div>
178 <p>In Japanese typography, this case is sometimes called
179 <i lang="ja">taigo</i> ruby or group-ruby (per-word ruby),
180 because the annotation as a whole is associated
181 with multi-character word (as a whole).
182 </div>
184 <div class="example">
185 <p>In this second example,
186 two levels of annotations are attached to a base sequence:
187 the hiragana characters on top refer to the pronunciation of each of the base kanji characters,
188 while the words “Keio” and “University” on the bottom are annotations describing the English translation.
189 <div class="figure">
190 <p><img src="images/ruby-univ.gif"
191 alt="Example showing complex ruby with annotation text over and under the base characters">
192 <p class="caption">Complex ruby with annotation text over and under the base characters
193 </div>
194 <p>
195 <p>Notice that to allow correct association between the hiragana characters and
196 their corresponding Kanji base characters,
197 the spacing between these Kanji characters is adjusted.
198 (This happens around the fourth Kanji character in the figure above.)
199 To avoid variable spacing between the Kanji characters in the example above
200 the hiragana annotations can be styled as a <i>collapsed annotation</i>,
201 which will look more like the group-ruby example earlier.
202 However because the base-annotation pairings are recorded in the ruby structure,
203 if the text breaks across lines, the annotation characters will stay
204 correctly paired with their respective base characters.
205 </div>
207 <p><i>Ruby</i> formatting as used in Japanese is described in JIS X-4051 [[JIS4051]] (in Japanese)
208 and in Requirements for Japanese Text Layout [[JLREQ]] (in English and Japanese)].
209 In HTML, ruby structure and markup to represent it is described
210 in the Ruby Markup Extension specification.
211 This module describes the CSS rendering model
212 and formatting controls relevant to ruby layout of such markup.
214 <h2 id="ruby-model">
215 Ruby Formatting Model</h2>
217 <p>The CSS ruby model is based on
218 the <a href="http://darobin.github.io/html-ruby/">HTML Ruby Markup Extension</a>
219 and <a href="http://www.w3.org/TR/ruby/">XHTML Ruby Annotation Recommendation</a> [[RUBY]].
220 In this model, a ruby structure consists of
221 one or more <dfn>ruby base</dfn> elements representing the base (annotated) text,
222 associated with one or more levels of <dfn>ruby annotation</dfn> elements representing the annotations.
223 The structure of ruby is similar to that of a table:
224 there are “rows” (the base text level, each annotation level)
225 and “columns” (each <i>ruby base</i> and its corresponding <i>ruby annotations</i>).
227 <p>Consecutive bases and annotations are grouped together into <dfn>ruby segments</dfn>.
228 Within a <i>ruby segment</i>, a <i>ruby annotation</i> may span multiple <i>ruby bases</i>.
230 <p class="note">In HTML, a single <code><ruby></code> element may contain multiple <i>ruby segments</i>.
231 (In the XHTML Ruby model, a single <code><ruby></code> element can only contain one <i>ruby segment</i>.)
233 <h3 id="ruby-display">
234 Ruby-specific 'display' property values</h3>
236 <p>For document languages (such as XML applications) that do not have pre-defined ruby elements,
237 authors must map document language elements to ruby elements;
238 this is done with the 'display' property.
240 <table class="propdef">
241 <tr>
242 <th>Name:
243 <td>display
244 <tr>
245 <th><a href="#values">New Values</a>:
246 <td>ruby | ruby-base | ruby-text | ruby-base-container | ruby-text-container
247 </table>
249 <p>The following new 'display' values assign ruby layout roles to an arbitrary element:
251 <dl>
252 <dt>''ruby''
253 <dd>Specifies that an element generates a <dfn title="ruby container | ruby container box">ruby container box</dfn>.
254 (Corresponds to HTML/XHTML <code><ruby></code> elements.)
255 <dt>''ruby-base''
256 <dd>Specifies that an element generates a <dfn title="ruby base box | ruby base">ruby base box</dfn>.
257 (Corresponds to HTML/XHTML <code><rb></code> elements.)
258 <dt>''ruby-text''
259 <dd>Specifies that an element generates a <dfn title="ruby annotation box | ruby annotation">ruby annotation box</dfn>.
260 (Corresponds to HTML/XHTML <code><rt></code> elements.)
261 <dt>''ruby-base-container''
262 <dd>Specifies that an element generates a <dfn title="ruby base container box | ruby base container">ruby base container box</dfn>.
263 (Corresponds to XHTML <code><rbc></code> elements; always implied in HTML.)
264 <dt>''ruby-text-container''
265 <dd>Specifies that an element generates a <dfn title="ruby annotation container box | ruby annotation container">ruby annotation container box</dfn>.
266 (Corresponds to HTML/XHTML <code><ruby></code> elements.)
267 </dl>
269 <h3 id="box-fixup">
270 Anonymous Ruby Box Generation</h3>
272 <p>The CSS model does not require that the document language
273 include elements that correspond to each of these components.
274 Missing parts of the structure are implied through the anonymous box generation rules
275 <a href="http://www.w3.org/TR/CSS21/tables.html#anonymous-boxes">similar to those used to normalize tables</a>. [[!CSS21]]
277 <ol>
278 <li>Any in-flow block-level boxes directly contained by a
279 <i>ruby container</i>,
280 <i>ruby base container</i>,
281 <i>ruby annotation container</i>,
282 <i>ruby base box</i>,
283 or <i>ruby annotation box</i>
284 are forced to be inline-level boxes,
285 and their 'display' value computed accordingly.
286 For example,
287 the 'display' property of an in-flow element with ''display: block''
288 parented by an element with ''display: ruby-text''
289 computes to ''inline-block''.
290 This computation occurs after any intermediary anonymous-box fixup
291 (such as that required by internal table elements).
293 <li>Any consecutive sequence of <i>ruby bases</i> not parented by a <i>ruby base container</i>
294 is wrapped in an anonymous <i>ruby base container</i>.
295 Similarly, any consecutive sequence of <i>ruby annotations</i> not parented by a <i>ruby annotation container</i>
296 is wrapped in an anonymous <i>ruby annotation container</i>.
298 <li>Within each <i>ruby base container</i>,
299 each sequence of inline-level boxes is wrapped in an anonymous <i>ruby base box</i>.
300 Similarly, within each <i>ruby annotation container</i>,
301 each sequence of inline-level boxes is wrapped in an anonymous <i>ruby annotation box</i>.
303 <li>A sequence of <i>ruby base containers</i> and/or <i>ruby annotation containers</i>
304 not parented by a <i>ruby container</i>
305 is wrapped in an anonymous <i>ruby container</i>.
306 </ol>
308 <p>At this point, all ruby layout structures are properly parented,
309 and the UA can start to associate bases with their annotations.
311 <p class="note">
312 Note that the UA is not required to create any of these anonymous boxes in its internal structures,
313 as long as pairing and layout behaves as if they existed.
315 <h3 id="pairing">
316 Ruby Pairing and Annotation Levels</h3>
318 <p>Within a ruby structure,
319 each <i>ruby base</i> is associated with <i>ruby annotations</i>
320 and vice versa.
321 A <i>ruby base</i> can be associated with at most one <i>ruby annotation</i> per annotation level.
322 If there are multiple annotation levels, it can therefore be associated with multiple <i>ruby annotations</i>.
323 A <i>ruby annotation</i> is associated with one or more <i>ruby bases</i>;
324 annotations can span multiple bases.
326 <p><dfn>Annotation pairing</dfn> is the process of associating
327 <i>ruby annotations</i> with <i>ruby bases</i>.
329 <ol>
330 <li>
331 <p>First, the ruby structure is divided into <i>ruby segments</i>,
332 each consisting of a single <i>ruby base container</i>
333 followed by one or more <i>ruby annotation containers</i>.
334 If the first child of a <i>ruby container</i> is a <i>ruby annotation container</i>,
335 an anonymous, empty <i>ruby base container</i> is assumed to exist before it.
336 Similarly, if the <i>ruby container</i> contains consecutive <i>ruby base containers</i>,
337 anonymous, empty <i>ruby annotation containers</i> are assumed to exist between them.
338 The <i>ruby base container</i> in each segment is thus associated
339 with each of the <i>ruby annotation containers</i> in that segment.
341 <p>Each <i>ruby annotation containers</i> in a <i>ruby segment</i>
342 represents one <dfn title="annotation level | level">level</dfn> of annotation:
343 the first one represents the first level of annotation,
344 the second one represents the second level of annotation,
345 and so on.
347 <li>Within each <i>ruby segment</i>,
348 each <i>ruby base box</i> in the <i>ruby base container</i>
349 is paired with one <i>ruby annotation box</i>
350 from each <i>ruby annotation container</i> in its <i>ruby segment</i>.
351 If there are not enough <i>ruby annotations</i> in a <i>ruby annotation container</i>,
352 the last one is associated with any excess <i>ruby bases</i>.
353 (If there are not any in the <i>ruby annotation container</i>, an anonymous empty one is assumed to exist.)
354 If there are not enough <i>ruby bases</i>,
355 any remaining <i>ruby annotations</i> are assumed to be associated
356 with empty, anonymous bases inserted at the end of the <i>ruby base container</i>.
358 <p>If an implementation supports ruby markup with explicit spanning
359 (e.g. XHTML Complex Ruby Annotations),
360 it must adjust the pairing rules to pair spanning annotations to multiple bases
361 appropriately.
362 </ol>
364 <p>A this point, ruby “columns” are defined,
365 each represented by a single <i>ruby base</i>
366 and associated with one <i>ruby annotation</i> (possibly an empty, anonymous one)
367 from each <i>annotation level</i>.
369 <h4 id="nested-pairing">
370 Nested Ruby</h4>
372 <p>When <i>ruby containers</i> are nested,
373 pairing begins with the deepest <i>ruby container</i>,
374 then expands out,
375 treating each <i>ruby container</i> nested within another <i>ruby container</i>
376 essentially as a single <i>ruby base</i> in the outer <i>ruby container</i>,
377 and associating each <i>ruby annotation</i>
378 paired with the nested <i>ruby container</i>
379 as being associated with (spanning) all of its <i>ruby bases</i>.
381 <p>Using nested <i>ruby containers</i> thus allows the representation
382 of complex spanning relationships.
384 <p class="issue">This has to be Level 1 because HTML5 allows it, so we have to handle it. Yay HTML5.
386 <h3 id="autohide">
387 Autohiding Annotations</h3>
389 <p>If a <i>ruby annotation</i> has the exact same content as its base,
390 it is <dfn title="hidden ruby annotation | hidden annotation">hidden</dfn>.
391 Hiding a <i>ruby annotation</i> does not affect annotation pairing
392 or the block-axis positioning of boxes in other <i>levels</i>.
393 However the <i>hidden annotation</i> is not visible,
394 and it has no impact on layout
395 other than to separate adjacent sequences of <i>ruby annotation boxes</i> within its level,
396 as if they belonged to separate segments
397 and the <i>hidden annotation</i>’s base were not a <i>ruby base</i> but an intervening inline.
399 <div class="example">
400 <p>This is to allow correct inlined display of annotations
401 for Japanese words that are a mix of kanji and hirangana.
402 For example, the word <i>振り仮名</i> should be inlined as
403 <p class="figure">振り仮名(ふりがな)
404 <p>and therefore marked up as
405 <pre>
406 <!-- --><ruby>
407 <!-- --> <rb>振</rb><rb>り</rb><rb>仮</rb><rb>名</rb>
408 <!-- --> <rp>(</rp><rt>ふ</rt><rt>り</rt><rt>が</rt><rt>な</rt><rp>)</rp>
409 <!-- --><ruby></pre>
410 <p>However, when displayed as ruby, the “り” should be hidden
411 <div class="figure">
412 <p><img src="images/furigana-separate.png"
413 alt="Hiragana annotations for 振り仮名 appear, each above its base character.">
414 <p class="caption">Hiragana ruby for 振り仮名
415 </div>
416 </div>
418 <p class="note">
419 Future levels of CSS Ruby may add controls for this,
420 however in this level it is always forced.
422 <p>The content comparison for this auto-hiding behavior
423 takes place prior to white space collapsing.
424 <span class="issue">Is this easier? Or after collapsing is easier? We should do whatever is easier, as it really doesn't matter much which way to go.
426 <h3 id="white-space">
427 White Space</h3>
429 <p class="issue">I'm unsure exactly where space should be trimmed. :/
430 But pretty sure we need to keep spaces between things,
431 otherwise ruby only works for CJK.
433 <p><i>Collapsible</i> white space within a ruby structure is discarded
434 at the beginning and end of a <i>ruby container</i>, <i>ruby annotation container</i>, or <i>ruby base container</i>,
435 and at the beginning/end of a <i>ruby annotation box</i> or <i>ruby base box</i> if white space is not its only contents.
436 Between <i>ruby segments</i>, between <i>ruby bases</i>, and between <i>ruby annotations</i>, however,
437 white space is not discarded.
438 If such white space is <i>collapsible</i>, it will collapse
439 following the standard <a href="http://www.w3.org/TR/css3-text/#white-space-rules">white space processing rules</a>. [[!CSS3TEXT]]
440 Between <i>ruby segments</i>, however,
441 the contextual text for determining collapsing behavior is given by the <i>ruby bases</i> on either side,
442 not the text on either side of the white space in the source document.
444 <div class="note">
445 <p>Note that the white space processing rules
446 cause a white space sequence containing a <i>segment break</i> (such as a line feed)
447 to <a href="http://www.w3.org/TR/css3-text/#line-break-transform">collapse to nothing</a> between CJK characters.
448 This means that CJK ruby can safely use white space for indentation of the ruby markup.
449 For example, the following markup will display without any spaces:
450 <pre>
451 <!-- --><ruby>
452 <!-- --> <rb>東</rb><rb>京</rb>
453 <!-- --> <rt>とう</rt><rt>きょう</rt>
454 <!-- --></ruby></pre>
455 <p>However, this markup will:
456 <pre>
457 <!-- --><ruby>
458 <!-- --> <rb>東</rb> <rb>京</rb>
459 <!-- --> <rt>とう</rt> <rt>きょう</rt>
460 <!-- --></ruby></pre>
461 </div>
463 <p>Any preserved white space is then wrapped in an anonymous box belonging to
464 the <i>ruby base container</i> (if between <i>ruby bases</i>),
465 <i>ruby annotation container</i> (if between <i>ruby annotations</i>),
466 or <i>ruby container</i> (if between <i>ruby segments</i>).
467 In the latter case, the text is considered part of the <i>base level</i>.
468 This box does not take part in pairing.
469 It merely ensures separation between adjacent bases/annotations.
471 <div class="example">
472 <p>These rules allow ruby to be used with space-separated scripts such as Latin.
473 For example,
474 <pre>
475 <!-- --><ruby>
476 <!-- --> <rb>W</rb><rb>W</rb><rb>W</rb>
477 <!-- --> <rt>World</rt> <rt>Wide</rt> <rt>Web</rt>
478 <!-- --></ruby></pre>
479 <p>They also ensure that annotated white space is preserved. For example,
480 <pre>
481 <!-- --><ruby>
482 <!-- --> <rb>Aerith</rb><rb> </rb><rb>Gainsboro</rb>
483 <!-- --> <rt>エアリス</rt><rt>・</rt><rt>ゲインズブール</rt>
484 <!-- --></ruby></pre>
485 </div>
487 <p class="issue">Specify how this impacts layout, or not.
489 <h3 id="ruby-layout">
490 Ruby layout</h3>
492 <p>When a ruby structure is laid out,
493 its base level is laid out on the line,
494 aligned according to its 'vertical-align' property
495 exactly as if its <i>bases</i> were a regular sequence of <i>inline</i> boxes.
496 Each <i>ruby base container</i> is sized and positioned
497 to contain exactly the full height of its <i>ruby bases</i>.
499 <p><i>Ruby annotations</i> associated with the base level
500 are then positioned with respect to their <i>ruby base boxes</i>
501 according to the applicable 'ruby-position' values.
502 <i>Ruby annotations</i> within a level (within a single <i>ruby container</i>)
503 are aligned to each other as if they were inline boxes
504 participating in the same inline formatting context.
505 Each <i>ruby annotation container</i> is sized and positioned
506 to contain exactly the full height of its <i>ruby annotations</i>.
508 <p>A ruby container (or fragment thereof)
509 measures as wide as the content of its widest level.
510 Similarly, <i>ruby base boxes</i> and <i>ruby annotation boxes</i>
511 within a ruby “column” have the measure of the widest content in that “column”.
512 In the case of spanning <i>annotations</i>
513 (whether actually spanning or pretending to span per 'ruby-collapse'),
514 the measures of the <i>ruby annotation box</i> and
515 the sum of its associated <i>ruby base boxes</i> must match.
517 <p>How the extra space is distributed
518 when ruby content is narrower than the measure of its box
519 is specified by the 'ruby-align' property.
521 <h4 id="inter-character-layout">
522 Inter-character ruby layout</h4>
524 <p>Inter-character annotations have special layout.
525 When 'ruby-position' indicates ''inter-character'' annotations,
526 the affected <i>annotation boxes</i>
527 are spliced into and measured as part of the layout of the base level.
528 The <i>ruby base container</i> must be sized to include both the <i>base boxes</i>
529 as well as the ''inter-character'' <i>annotation boxes</i>.
530 The affected <i>ruby annotation container</i> is similarly sized
531 so that its content box coincides with that of the <i>ruby base container</i>.
533 <p>For the purpose of laying out other levels of annotations,
534 an ''inter-character'' annotation effectively becomes part of its base.
535 <span class="issue">Or should it become a quasi-base between two bases?</span>
536 A spanning ''inter-character'' annotation is placed after
537 all the bases that it spans.
539 <h3 id="box-style">
540 Styling Ruby Boxes</h4>
542 <p>In most respects, ruby boxes can be styled similar to inline boxes.
543 However, the UA is not required to support
544 any of the box properties (borders, margins, padding),
545 any of the background properties or outline properties,
546 or any other property that illustrates the bounds of the box
547 on <i>ruby base container boxes</i>, <i>ruby annotation container boxes</i>,
548 or <a href="#nested-pairing">ruby-internal <i>ruby container boxes</i></a>.
549 The UA may implement these boxes simply as abstractions for inheritance
550 and control over the layout of their contents.
552 <p class="issue">
553 Alternatively... use margins to control offsets?
554 Or could line-height be adequate? Its centering behavior can be awkward.
556 <h3 id="line-breaks">
557 Ruby box and line breaking</h3>
559 <p>When there is not enough space for an entire <i>ruby container</i> to fit on the line,
560 the ruby may be broken wherever all levels simultaneously allow a break.
561 Ruby most often breaks between base-annotation sets,
562 but if the line-breaking rules allow it, can also break within a <i>ruby base</i>
563 (and, in parallel, its associated <i>annotation boxes</i>).
565 <p>Whenever ruby breaks across lines, <i>ruby annotations</i> must stay
566 with their respective <i>bases</i>.
567 The line <em>must not</em> break between a <i>ruby base</i> and its <i>annotations</i>,
568 even in the case of ''inter-character'' <i>annotations</i>.
570 <div class="figure">
571 <img src="images/r-break-b.gif"
572 alt='Diagram showing the line breaking opportunity in a "Bopomofo" ruby'>
573 <p class="caption">''inter-character'' ruby line breaking opportunity
574 </div>
576 <h4 id="break-between">
577 Breaking between bases</h4>
579 <p>In typical cases, <i>ruby base boxes</i> and <i>ruby annotation boxes</i>
580 are styled to forbid internal line wrapping and do not contain forced breaks.
581 (See <a href="#default-stylesheet">Appendix A</a>.)
582 In such cases the <i>ruby container</i> can only break between adjacent <i>ruby bases</i>,
583 and only if no <i>ruby annotations</i> span those <i>ruby bases</i>.
585 <div class="figure">
586 <p><img src="images/r-break-a.gif"
587 alt="Diagram showing the line breaking opportunity in a complex ruby">
588 <p class="caption">Ruby line breaking opportunity
589 </div>
591 <p>Whether ruby can break between two adjacent <i>ruby bases</i>
592 is controlled by normal line-breaking rules for the base text,
593 exactly as if the <i>ruby bases</i> were adjacent <i>inline</i> boxes.
594 (The annotations are ignored when determining soft wrap opportunities for the base level.)
596 <div class="example">
597 <p>For example, if two adjacent ruby bases are “蝴” and “蝶”,
598 the line may break between them,
599 because lines are normally allowed to break between two Han characters.
600 However, if 'word-break' is ''keep-all'', that line break is forbidden.
601 <pre><ruby>蝴<rt>hú</rt>蝶<rt>dié</rt></pre>
602 </div>
604 <p>Inter-base white space is significant for evaluating line break opportunities between <i>ruby bases</i>.
605 As with white space between inlines, it collapses when the line breaks there.
606 Similarly, annotation white space is also trimmed at a line break.
608 <div class="example">
609 <p>For example, given the following markup:
610 <pre><ruby><rb>one</rb> <rb>two</rb> <rt>1</rt> <rt>2</rt></ruby></pre>
611 <p>Due to the space, the line may break between “one” and “two“.
612 If the line breaks there, that space—and the space between “1” and “2”—disappears,
613 in accordance with standard CSS white space processing rules. [[CSS3TEXT]]
614 </div>
616 <h4 id="break-within">
617 Breaking within bases</h4>
619 <p>For longer base texts, it is sometimes appropriate to allow breaking within a base-annotation pair.
620 For example, if an English sentence is annotated with its Japanese translation,
621 allowing the text to wrap allows for reasonable line breaking behavior in the paragraph.
623 <p class="issue">
624 Insert scanned example so people don't think this is just the ramblings of an insane spec-writer.
626 <p>Line-breaking within a <i>ruby base</i> is only allowed if the 'white-space' property
627 of the <i>ruby base</i> and all its parallel <i>annotations</i> allow it,
628 and there exists a <i>soft wrap opportunity</i> <em>within</em> (i.e. not at the start or end)
629 the content of each base/annotation box.
630 Since there is no structural correspondance between fragments of content
631 within <i>ruby bases</i> and <i>annotations</i>,
632 the UA may break at any set of opportunities;
633 but it is recommended that the UA attempt to proportionally balance
634 the amount of content inside each fragment.
636 <p>There are no line breaking opportunities within ''inter-character'' <i>annotations</i>.
638 <p>Ruby alignment takes place within each fragment, after line-breaking.
640 <h3 id="ruby-bidi">
641 Bidi Reordering</h3>
643 <p class="issue">The easiest thing would be to treat each <i>ruby base</i> as an embedding of its 'direction'.
644 But this means that implicit bidi reordering is scoped to within a single base.
646 <!--
648 <p>The Unicode bidirectional algorithm orders logically-stored text for visual presentation
649 when characters from scripts of opposing directionalities are mixed
650 within a single paragraph.
651 (See [[CSS3-WRITING-MODES]] for a more in-depth discussion of bidirectional text in CSS.)
652 Bidi reordering of ruby-annotation pairs is controlled by the ordering of the base text, as follows:
654 <h4 id="bidi-B">Proposal A</h4>
656 <p class="note">This proposal is simpler, but will require more frequent tagging of mixed-directionality content.
658 <p>To avoid the interference of the <i>annotations</i> in the ordering of base text,
659 all annotations are ignored for the purpose of resolving neutral characters in the base level.
661 <p>To preserve the correspondance of <i>ruby annotations</i>
662 to their respective <i>ruby bases</i>,
663 a few restrictions are imposed:
664 <ul>
665 <li>The contents of a <i>ruby base</i> or <i>ruby annotation</i> must remain contiguous.
666 To this end, the contents of each <i>ruby base</i> and <i>ruby annotation</i> are treated as if within a <i>bidi isolation</i>.
667 <li>For the purpose of ordering <i>ruby bases</i> within a <i>ruby container</i>,
668 each <i>ruby base</i> is treated as a strong character of its specified 'direction'.
669 (<i>Ruby annotations</i> are positioned with respect to their <i>bases</i>,
670 so are affected by reordering, but do not themselves participate in reordering.)
671 <li>All <i>bases</i> spanned by a single <i>annotation</i> must remain contiguous.
672 To this end, the embedding level of all bases spanned by a spanning <i>annotation</i>
673 is increased by two prior to reordering.
674 </ul>
676 <h4 id="bidi-B">Proposal B</h4>
678 <p class="note">This preserves some aspects of implicit bidi.
679 For example, annotating each half of "first-second" would not cause the word to reverse itself to "second-first" within an opposite-order paragraph.
681 <p>To avoid the interference of the <i>annotations</i> in the ordering of the base text,
682 all annotations are ignored for the purpose of resolving neutral characters in the base text;
683 and the base text is resolved exactly as if each <i>ruby base</i> were
684 just a normal <i>inline box</i> embedded in the <i>ruby container</i>'s inline formatting context.
686 <p>Furthermore, to preserve the correspondance of <i>ruby annotations</i>
687 to their respective <i>ruby bases</i>,
688 a few restrictions are imposed:
689 <ul>
690 <li>The contents of a <i>ruby base</i> must remain contiguous.
691 <li>All <i>bases</i> spanned by a single <i>annotation</i> must remain contiguous.
692 </ul>
693 <p>To this end, the text of <i>ruby base</i> boxes that are not <i>bidi-isolated</i>
694 must have its embedding level increased by two;
695 and if an <i>annotation</i> spans more than one <i>ruby base</i>,
696 the spanned text's embedding level
697 (or the embedding level assigned to the isolation, if the <i>ruby base</i> is <i>bidi-isolated</i>)
698 is further increased by two.
700 -->
702 <h3 id="line-height">
703 Ruby box and line stacking</h3>
705 <p>The 'line-height' property controls spacing between lines in CSS.
706 When inline content on line is shorter than the 'line-height',
707 half-leading is added on either side of the content,
708 as specificed in <a href="http://www.w3.org/TR/CSS21/visudet.html#line-height">CSS2.1§10.8</a>. [[!CSS21]]
710 <p>In order to ensure consistent spacing of lines,
711 documents with ruby typically ensure that the 'line-height' is large enough
712 to accommodate ruby between lines of text.
713 Therefore, ordinarily, <i>ruby annotation containers</i> and <i>ruby annotation boxes</i>
714 do not contribute to the measured height of a line's inline contents;
715 any alignment (see 'vertical-align') and line-height calculations
716 are performed using only the <i>ruby base container</i>,
717 exactly as if it were a normal inline.
719 <p>However, if the 'line-height' specified on the <i>ruby container</i>
720 is less than the distance between
721 the top of the top <i>ruby annotation container</i>
722 and the bottom of the bottom <i>ruby annotation container</i>,
723 then additional leading is added
724 on the appropriate side of the <i>ruby base container</i>
725 such that if a block consisted of three lines
726 each containing ruby identical to this,
727 none of the <i>ruby containers</i> would overlap.
729 <p class="note">Note that this does not ensure that the <i>ruby annotations</i> remain within the line box.
730 It merely ensures that <em>if all lines had equal spacing</em>
731 and equivalent amounts and positioning of <i>ruby annotations</i>,
732 there would be enough room to avoid overlap.
734 <p>Authors should ensure appropriate 'line-height' and 'padding' to accommodate ruby,
735 and be particularly careful at the beginning or end of a block
736 and when a line contains inline-level content
737 (such as images, inline blocks, or elements shifted with 'vertical-align')
738 taller than the paragraph's default font size.
740 <div class="figure">
741 <p><img src="images/rlh-a.gif"
742 alt="The content of each line sits in the middle of its line height;
743 the additional space on each side is called half-leading.
744 Ruby fits between lines if it is smaller than twice the half-leading,
745 but this means that it occupies space belonging to the half-leading of the previous line.">
746 <p class="caption">Ruby annotations will often overflow the line;
747 authors should ensure content over/under a ruby-annotated line
748 is adequately spaced to leave room for the ruby.
749 </div>
751 <p class="note">More control over how ruby affects alignment and line layout
752 will be part of the CSS Line Layout Module Level 3.
753 Note, it is currently in the process of being rewritten;
754 the current drafts should not be relied upon.
756 <h2 id="ruby-props">
757 Ruby Properties</h2>
759 <p>The following properties are introduced to control ruby positioning and alignment.
761 <h3 id="rubypos">
762 Ruby positioning: the 'ruby-position' property</h3>
764 <table class="propdef">
765 <tr>
766 <th>Name:
767 <td><dfn>ruby-position</dfn>
768 <tr>
769 <th><a href="#values">Value</a>:
770 <td>[ over | under | inter-character ] && [ right | left ]
771 <tr>
772 <th>Initial:
773 <td>over right
774 <tr>
775 <th>Applies to:
776 <td>ruby annotation containers
777 <tr>
778 <th>Inherited:
779 <td>yes
780 <tr>
781 <th>Percentages:
782 <td>N/A
783 <tr>
784 <th>Media:
785 <td>visual
786 <tr>
787 <th>Computed value:
788 <td>specified value
789 <tr>
790 <th>Animatable:
791 <td>no
792 <tr>
793 <th>Canonical order:
794 <td><abbr title="follows order of property value definition">per grammar</abbr>
795 </table>
797 <p>This property controls position of the ruby text with respect to its base.
798 Values have the following meanings:
800 <p class="issue"><span class="issuehead">Issue-107: </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=&hdr-1-name=subject&hdr-1-query=ruby-position%3A+undesirable+default+value+%27before%27+for+complex+ruby&hdr-2-name=from&hdr-2-query=&hdr-3-name=message-id&hdr-3-query=&period_month=&period_year=&index-grp=Public__FULL&index-type=t&type-index=www-style&resultsperpage=20&sortby=date">this thread</a> and <a href="http://www.w3.org/Search/Mail/Public/advanced_search?keywords=&hdr-1-name=subject&hdr-1-query=Styling+of+complex+Ruby&hdr-2-name=from&hdr-2-query=&hdr-3-name=message-id&hdr-3-query=&period_month=&period_year=&index-grp=Public__FULL&index-type=t&type-index=public-i18n-core&resultsperpage=20&sortby=date">this one</a>.</p>
801 <dl>
802 <dt><dfn title="ruby-position:over">''over''</dfn>
803 <dd>The ruby text appears <i>over</i> the base in horizontal text.
805 <div class="figure">
806 <p><img src="images/shinkansen-top.gif"
807 alt="Diagram of ruby glyph layout in horizontal mode with ruby text appearing above the base">
808 <p class="caption">Ruby over Japanese base text in horizontal layout
809 </div>
810 </dd>
812 <dt><dfn title="ruby-position:right">''right''</dfn>
813 <dd>The ruby text appears on the right side of the base in vertical text.
814 <div class="figure">
815 <p><img src="images/shinkansen-right.gif" width="33"
816 alt="Diagram of ruby glyph layout in vertical mode with ruby text apearing vertically on the right of the base">
817 <p class="caption">Ruby to the right of Japanese base text in vertical layout
818 </div>
819 </dd>
821 <dt><dfn title="ruby-position:under">''under''</dfn>
822 <dd>The ruby text appears under the base in horizontal text.
823 This is a relatively rare setting used in ideographic East Asian writing systems,
824 most easily found in educational text.
826 <div class="figure">
827 <p><img src="images/shinkansen-bottom.gif"
828 alt="Diagram of ruby glyph layout in horizontal mode with ruby text appearing below the base">
829 <p class="caption">Ruby under Japanese base text in horizontal layout
830 </div>
831 </dd>
833 <dt><dfn title="ruby-position:left">''left''</dfn>
834 <dd>The ruby text appears on the left side of the base in vertical text.
836 <div class="figure">
837 <p><img src="images/shinkansen-left.gif"
838 alt="Diagram of ruby glyph layout in vertical mode with ruby text apearing vertically on the left of the base">
839 <p class="caption">Ruby to the left of Japanese base text in vertical layout
840 </div>
841 </dd>
843 <dt><dfn title="ruby-position:inter-character">''inter-character''</dfn></dt>
844 <dd>
845 <p>The ruby text appears on the right of the base in horizontal text.
846 This value forces the 'writing-mode' of the <i>ruby annotation</i> to be vertical.
848 <p>This value is provided for the special case of traditional Chinese
849 as used especially in Taiwan:
850 ruby (made of <a href="#g-bopomofo">bopomofo</a> glyphs) in that context
851 appears vertically along the right side of the base glyph,
852 even when the layout of the base characters is horizontal:
854 <div class="figure">
855 <p><img src="images/bopomofo.gif"
856 alt="Example of Taiwanese-style ruby">
857 <p class="caption">“Bopomofo” ruby in traditional Chinese
858 (ruby text shown in blue for clarity) in horizontal layout
859 </div>
860 <p class="note">
861 Note that the user agent is responsible for ensuring the correct relative alignment and positioning of the glyphs,
862 including those corresponding to the tone marks, when displaying.
863 Tone marks are spacing characters that occur (in memory) at the end of the ruby text for each base character.
864 They are usually displayed in a separate column to the right of the bopomofo characters,
865 and the height of the tone mark depends on the number of characters in the syllable.
866 One tone mark, however, is placed above the bopomofo, not to the right of it.
867 <!-- See Taiwanese requirements doc for EPUB at http://epub-revision.googlecode.com/files/EGLS_TW_eng.ppt -->
868 </dd>
869 </dl>
871 <p>If multiple <i>ruby annotation containers</i> have the same 'ruby-position',
872 they stack along the block axis,
873 with lower levels of annotation closer to the base text.
875 <h3 id="collapsed-ruby">
876 Collapsed Ruby Annotations: the 'ruby-merge' property</h3>
878 <table class="propdef">
879 <tr>
880 <th>Name:
881 <td><dfn>ruby-merge</dfn>
882 <tr>
883 <th><a href="#values">Value</a>:
884 <td>separate | collapse | auto
885 <tr>
886 <th>Initial:
887 <td>separate
888 <tr>
889 <th>Applies to:
890 <td>ruby annotation containers
891 <tr>
892 <th>Inherited:
893 <td>yes
894 <tr>
895 <th>Percentages:
896 <td>N/A
897 <tr>
898 <th>Media:
899 <td>visual
900 <tr>
901 <th>Computed value:
902 <td>specified value
903 <tr>
904 <th>Animatable:
905 <td>no
906 <tr>
907 <th>Canonical order:
908 <td><abbr title="follows order of property value definition">per grammar</abbr>
909 </table>
911 <p>
912 This property controls how ruby annotation boxes should be rendered
913 when there are more than one in a ruby container box.
915 <p>Possible values:</p>
916 <dl>
917 <dt><dfn title="ruby-merge:separate">''separate''</dfn>
918 <dd>
919 <p>
920 Each ruby annotation box is rendered in the same column(s) as its corresponding base box(es).
921 This style is called “mono ruby” in [[JLREQ]].
923 <div class="example">
924 <p>For example, the following two markups render the same:
925 <pre><ruby>無<rt>む</ruby><ruby>常<rt>じょう</ruby></pre>
926 <p>and:
927 <pre><ruby style="ruby-merge:separate"><rb>無<rb>常<rt>む<rt>じょう</ruby></pre>
928 </div>
929 </dd>
931 <dt><dfn title="ruby-merge:collapse">''collapse''</dfn>
932 <dd>
933 <p>
934 All <i>ruby annotation boxes</i> within the same <i>ruby segment</i> on the same line are concatenated,
935 and laid out as if their contents belonged to a single <i>ruby annotation box</i>
936 spanning all their associated <i>ruby base boxes</i>.
937 This style renders similar to “group ruby” in [[JLREQ]],
938 except that <i>ruby annotations</i> are kept together with their respective <i>ruby bases</i> when breaking lines.
939 </p>
941 <div class="example">
942 <p>The following two markups render the same both characters fit on one line:
943 <pre><ruby>無常<rt>むじょう</ruby></pre>
944 <p>and:
945 <pre><ruby style="ruby-merge:collapse"><rb>無<rb>常<rt>む<rt>じょう</ruby></pre>
946 <p>However, the second one renders the same as ''ruby-position: separate''
947 when the two bases are split across lines.
948 </div>
949 </dd>
951 <dt><dfn title="ruby-merge:auto">''auto''</dfn></dt>
952 <dd>
953 <p>
954 The user agent may use any algorithm to determine how each ruby annotation box
955 is rendered to its corresponding base box,
956 with the intention that if all annotations fit over their respective bases,
957 the result is identical to “mono ruby”,
958 but if some annotations are wider than their bases
959 the space is shared in some way
960 to avoid forcing space between bases.
961 <div class="example">
962 <p>
963 One possible algorithm is described as “jukugo ruby” in [[JLREQ]].
964 <p>
965 Another, more simplified algorithm of “jukugo ruby” is
966 to render as ''separate'' if all ruby annotation boxes fit
967 within the advances of their corresponding base boxes,
968 and render as ''collapse'' otherwise.
969 </p>
970 </div>
971 </dd>
972 </dl>
974 <h3 id="ruby-align-property"><a name="rubyalign"></a>
975 Ruby Text Distribution: the 'ruby-align' property</h3>
977 <table class="propdef">
978 <tr>
979 <th>Name:
980 <td><dfn>ruby-align</dfn>
981 <tr>
982 <th><a href="#values">Value</a>:
983 <td>start | center | space-between | space-around
984 <tr>
985 <th>Initial:
986 <td>space-around
987 <tr>
988 <th>Applies to:
989 <td>ruby bases, ruby annotations, ruby base containers, ruby annotation containers
990 <tr>
991 <th>Inherited:
992 <td>yes
993 <tr>
994 <th>Percentages:
995 <td>N/A
996 <tr>
997 <th>Media:
998 <td>visual
999 <tr>
1000 <th>Computed value:
1001 <td>specified value (except for initial and inherit)
1002 </table>
1004 <p>This property specifies how text is distributed within the various ruby boxes
1005 when their contents do not exactly fill their respective boxes.
1006 Note that space distributed by 'ruby-align' is unrelated to, and independent of,
1007 any space distributed due to justification.
1009 <p>Values have the following meanings:
1010 <p class="issue"><span class="issuehead">Issue: </span> Tony Graham has <a href="http://www.w3.org/Style/XSL/Group/FO/wiki/Ruby#Treat_CSS3_.22ruby-align.22_As_Shorthand.3F">suggested </a>that distribute-letter and distribute-space be values of a ruby-group-distribution property, and line-edge be moved to a ruby-alignment-edge property, and that the rest be gathered under a ruby-alignment property. And that ruby-align become a shorthand.</p>
1011 <dl>
1012 <dt><dfn title="ruby-align:start">''start''</dfn></dt>
1013 <dd>The ruby content is aligned with the start edge of its box.
1014 <div class="figure">
1015 <p><img
1016 alt="Diagram of glyph layout in left aligned ruby when ruby text is shorter than base"
1017 width="145" height="91" src="images/ra-l.gif" /><img
1018 width="145" height="91"
1019 alt="Diagram of glyph layout in left aligned ruby when ruby text is longer than base"
1020 src="images/ra-l-rb.gif" />
1021 <p class="caption">''start'' ruby distribution
1022 </div>
1023 </dd>
1025 <dt><dfn title="ruby-align:center">''center''</dfn></dt>
1026 <dd>The ruby content is centered within its box.
1027 <div class="figure">
1028 <p><img width="145" height="91"
1029 alt="Diagram of glyph layout in center aligned ruby when ruby text is shorter than base"
1030 src="images/ra-c.gif" /><img width="145" height="91"
1031 alt="Diagram of glyph layout in center aligned ruby when ruby text is longer than base"
1032 src="images/ra-c-rb.gif" />
1033 <p class="caption">''center'' ruby distribution
1034 </div>
1035 </dd>
1037 <dt><dfn title="ruby-align:space-between">''space-between''</dfn></dt>
1038 <dd>
1039 <p>The ruby content expands as defined for normal text justification
1040 (as defined by 'text-justify'),
1041 except that if there are no <a href="http://www.w3.org/TR/css3-text/#expansion-opportunity"><i>expansion opportunities</i></a>
1042 the content is centered.
1043 <div class="figure">
1044 <p><img width="145" height="91"
1045 alt="Diagram of glyph layout in distribute-letter aligned ruby when ruby text is shorter than base"
1046 src="images/ra-dl.gif" /><img width="145" height="91"
1047 alt="Diagram of glyph layout in distribute-letter aligned ruby when ruby text is longer than base"
1048 src="images/ra-dl-rb.gif" />
1049 <p class="caption">''space-between'' ruby distribution
1050 </div>
1051 </dd>
1053 <dt><dfn title="ruby-align:space-around">''space-around''</dfn></dt>
1054 <dd>
1055 <p>As for ''space-between''
1056 except that there exists an extra <a href="http://www.w3.org/TR/css3-text/#expansion-opportunity"><i>expansion opportunity</i></a>
1057 whose space is distributed half before and half after the ruby content.
1058 <div class="example">
1059 <p>Since a typical implementation will by default define <i>expansion opportunities</i>
1060 between every adjacent pair of CJK <i>characters</i>
1061 and not between adjacent pairs of Latin <i>characters</i>,
1062 this should result in the behavior recommended by [[JLREQ]]:
1063 for wide-cell ruby content to be distributed...
1064 <div class="figure">
1065 <p><img width="145" height="91"
1066 alt="Diagram of glyph layout in auto aligned ruby when ruby text is shorter than base"
1067 src="images/ra-ds.gif" /><img width="145" height="91"
1068 alt="Diagram of glyph layout in auto aligned ruby when ruby text is longer than base"
1069 src="images/ra-ds-rb.gif" />
1070 <p class="caption">Wide-cell text in ''space-around'' ruby distribution is spaced apart
1071 </div>
1072 <p>... and narrow-cell glyph ruby to be centered.
1073 <div class="figure">
1074 <p><img
1075 alt="Diagram of glyph layout in auto aligned ruby when halfwidth ruby text is shorter than base"
1076 width="145" height="91"
1077 src="images/ra-c-h.gif" /><img
1078 alt="Diagram of character layout in auto aligned ruby when ruby text is longer than narrow-width base"
1079 width="145" height="91"
1080 src="images/ra-c-rb-h.gif" />
1081 <p class="caption">Narrow-width ruby text in ''space-around'' ruby distribution is centered
1082 </div>
1083 </div>
1084 </dd>
1085 </dl>
1087 <p class="issue">Add a paragraph explaining how to distribute space in situations with spanning annotations.
1088 <!--
1089 <p>For a complex ruby with spanning elements, one additional consideration is
1090 required. If the spanning element spans multiple 'rows' (other rbc or rtc
1091 elements), and the ruby alignment requires space distribution among the
1092 'spanned' elements, a ratio must be determined among the 'columns' of spanned
1093 elements. This ratio is computed by taking into consideration the widest
1094 element within each column.</p>
1095 -->
1096 <h2 id="edge-effects">
1097 Edge Effects</h2>
1099 <h3 id="ruby-overhang">
1100 Overhanging Ruby</h3>
1102 <p>
1103 When <i>ruby annotation box</i> is longer than its corresponding <i>ruby base box</i>,
1104 the <i>ruby annotation box</i> may partially overhang adjacent boxes.
1105 </p>
1106 <p>
1107 This level of the specification does not define
1108 how much the overhang may be allowed, and under what conditions.
1109 </p>
1111 <p>If the ruby text is not allowed to overhang,
1112 then the ruby behaves like a traditional inline box,
1113 i.e. only its own contents are rendered within its boundaries
1114 and adjacent elements do not cross the box boundary:
1116 <div class="figure">
1117 <p><img src="images/ro-n.gif"
1118 alt="Diagram showing the ruby boxes interacting with adjacent text">
1119 <p class="caption">Simple ruby whose text is not allowed to overhang adjacent text
1120 </div>
1122 <p>However, if <i>ruby annotation</i> content is allowed to overhang adjacent elements
1123 and it happens to be wider than its base,
1124 then the adjacent content is partially rendered within the area of the <i>ruby container box</i>,
1125 while the <i>ruby annotation</i> may partially overlap the upper blank parts of the adjacent content:
1127 <div class="figure">
1128 <p><img src="images/ro-a.gif"
1129 alt="Diagram showing the ruby boxes interacting with adjacent text">
1130 <p class="caption">Simple ruby whose text is allowed to overhang adjacent text
1131 </div>
1133 <p>The <i>ruby annotations</i> related to a <i>ruby base</i>
1134 must never overhang another <i>ruby base</i>.
1136 <p>The alignment of the contents of the base or the ruby text
1137 is not affected by overhanging behavior.
1138 The alignment is achieved the same way regardless of the overhang behavior setting
1139 and it is computed before the space available for overlap is determined.
1140 It is controlled by the 'ruby-align' property.
1142 <p class="issue">
1143 I suspect overhanging interacts with alignment in some cases;
1144 might need to look into this later.
1146 <p>This entire logic applies the same way in vertical ideographic layout,
1147 only the dimension in which it works in such a layout is vertical,
1148 instead of horizontal.
1150 <div class="example">
1151 <p>
1152 The user agent may use [[JIS4051]] recommendation of
1153 using one ruby text character length as the maximum overhang length.
1154 Detailed rules for how ruby text can overhang adjacent characters for Japanese are described by [[JLREQ]].
1155 </p>
1156 </div>
1158 <h3 id="line-edge">
1159 Line-edge Alignment</h3>
1161 <p>
1162 When a <i>ruby annotation box</i> that is longer than its <i>ruby base</i>
1163 is at the start or end edge of a line,
1164 the user agent <em>may</em> force the side of the <i>ruby annotation</i> that touches the edge of the line
1165 to align to the corresponding edge of the base.
1166 This type of alignment is described by [[JLREQ]].
1167 </p>
1168 <p>
1169 This level of the specification does not provide a mechanism to control this behavior.
1170 </p>
1171 <div class="figure">
1172 <p><img src="images/ra-le-l.gif"
1173 alt="Diagram of glyph layout in line-edge aligned ruby when ruby text is shorter than base">
1174 <img src="images/ra-le-r.gif"
1175 alt="Diagram of glyph layout in line-edge aligned ruby when ruby text is longer than base">
1176 <p class="caption">Line-edge alignment
1177 </div>
1179 <!--
1180 <h3 id="rubyover">
1181 Ruby overhanging: the 'ruby-overhang' property</h3>
1183 <table class="propdef">
1184 <tr>
1185 <th>Name:
1186 <td><dfn>ruby-overhang</dfn>
1187 <tr>
1188 <th>Value:
1189 <td>auto | start | end | none
1190 <tr>
1191 <th>Initial:
1192 <td>none
1193 <tr>
1194 <th>Applies to:
1195 <td>the parent of elements with display: ruby-text
1196 <tr>
1197 <th>Inherited:
1198 <td>yes
1199 <tr>
1200 <th>Percentages:
1201 <td>N/A
1202 <tr>
1203 <th>Media:
1204 <td>visual
1205 <tr>
1206 <th>Computed value:
1207 <td>specified value (except for initial and inherit)
1208 </table>
1210 <p>This property determines whether, and on which side, ruby text is allowed
1211 to partially overhang any adjacent text in addition to its own base, when the
1212 ruby text is wider than the ruby base. Note that ruby text is never allowed to
1213 overhang glyphs belonging to another ruby base. <span class="issue"><span class="issuehead">Issue: </span> This rule must be broken if we are to allow support for jukugo ruby.</span> Also the user agent is free to assume
1214 a maximum amount by which ruby text may overhang adjacent text. The user agent may use
1215 the [[JIS4051]] recommendation of using one ruby text character
1216 length as the maximum overhang length. Detailed rules for how ruby text can overhang adjacent characters for Japanese are described by [[JLREQ]].</p>
1218 <p>Possible values:</p>
1219 <dl>
1220 <dt><strong>auto</strong></dt>
1221 <dd>The ruby text can overhang text adjacent to the base on either side. [[JLREQ]] and [[JIS4051]] specify the categories of characters that
1222 ruby text can overhang. The user agent is free to follow those recommendations or specify its own classes of
1223 characters to overhang. This is the initial value.
1224 <div class="figure">
1225 <p><img class="example" width="177" height="91"
1226 alt="Diagram of glyph layout in overhanging ruby" src="images/ro-a.gif" /></p>
1227 <p><b>Figure 4.3.1</b>: Ruby overhanging adjacent text</p>
1228 </div>
1229 </dd>
1230 <dt><strong>start</strong></dt>
1231 <dd>The ruby text can only overhang the text that precedes it. That means, for
1232 example, that ruby cannot overhang text that is to the right of it in
1233 horizontal LTR layout, and it cannot overhang text that is below it in
1234 vertical-ideographic layout.
1235 <div class="figure">
1236 <p><img class="example" width="199" height="91"
1237 alt="Diagram of glyph layout when ruby overhangs the preceding glyphs only"
1238 src="images/ro-s.gif" /></p>
1239 <p><b>Figure 4.3.2</b>: Ruby overhanging preceding text only</p>
1240 </div>
1241 </dd>
1242 <dt><strong>end</strong></dt>
1243 <dd>The ruby text can only overhang the text that follows it. That means, for
1244 example, that ruby cannot overhang text that is to the left of it in
1245 horizontal LTR layout, and it cannot overhang text that is above it in
1246 vertical-ideographic layout.
1247 <div class="figure">
1248 <p><img class="example" width="198" height="91"
1249 alt="Diagram of glyph layout when ruby overhangs the following characters only"
1250 src="images/ro-e.gif" /></p>
1251 <p><b>Figure 4.3.3</b>: Ruby overhanging following text only</p>
1252 </div>
1253 </dd>
1254 <dt><strong>none</strong></dt>
1255 <dd>The ruby text cannot overhang any text adjacent to its base, only its
1256 own base.
1258 <div class="figure">
1259 <p><img class="example" width="220" height="91"
1260 alt="Diagram of glyph layout in non-overhanging ruby"
1261 src="images/ro-n.gif" /></p>
1262 <p><b>Figure 4.3.4</b>: Ruby not allowed to overhang adjacent text</p>
1263 </div>
1264 </dd>
1265 </dl>
1267 <h3 id="rubyspan">
1268 Ruby annotation spanning: the 'ruby-span' property</h3>
1270 <table class="propdef">
1271 <tr>
1272 <th>Name:
1273 <td><dfn>ruby-span</dfn>
1274 <tr>
1275 <th>Value:
1276 <td>attr(x) | none
1277 <tr>
1278 <th>Initial:
1279 <td>none
1280 <tr>
1281 <th>Applies to:
1282 <td>elements with display: ruby-text
1283 <tr>
1284 <th>Inherited:
1285 <td>no
1286 <tr>
1287 <th>Percentages:
1288 <td>N/A
1289 <tr>
1290 <th>Media:
1291 <td>visual
1292 <tr>
1293 <th>Computed value:
1294 <td><number>
1295 </table>
1297 <p>This property controls the spanning behavior of annotation elements. </p>
1299 <p class="note"><span class="note-label">Note:</span> A XHTML user agent may also use the <samp>rbspan</samp>
1300 attribute to get the same effect.</p>
1302 <p>Possible values:</p>
1304 <dl>
1305 <dt><strong>attr(x)</strong></dt>
1306 <dd>The value of attribute 'x' as a string value. The string value is
1307 evaluated as a <number> to determine the number of ruby base elements to be
1308 spanned by the annotation element. If the <number> is '0', it is replaced by
1309 '1'.The <number> is the computed value. </dd>
1310 <dt>none</dt>
1311 <dd>No spanning. The computed value is '1'.</dd>
1312 </dl>
1314 <p>The following example shows an XML example using the 'display' property
1315 values associated with the 'ruby structure and the 'ruby-span' property</p>
1316 <pre class="xml">myruby { display: ruby; }
1317 myrbc { display: ruby-base-container; }
1318 myrb { display: ruby-base; }
1319 myrtc.before { display: ruby-text-container; ruby-position: before}
1320 myrtc.after { display: ruby-text-container; ruby-position: after}
1321 myrt { display: ruby-text; ruby-span: attr(rbspan); }
1322 ...
1323 <myruby>
1324 <myrbc>
1325 <myrb>10</myrb>
1326 <myrb>31</myrb>
1327 <myrb>2002</myrb>
1328 </myrbc>
1329 <myrtc class="before">
1330 <myrt>Month</myrt>
1331 <myrt>Day</myrt>
1332 <myrt>Year</myrt>
1333 </myrtc>
1334 <myrtc class="after">
1335 <myrt rbspan="3">Expiration Date</myrt>
1336 </myrtc>
1337 </myruby></pre>
1338 -->
1340 <h2 id="default-stylesheet" class="no-num">
1341 Appendix A: Default Style Sheet</h2>
1343 <p><em>This section is informative.</em>
1345 <h3 id="default-ua-ruby" class="no-num">
1346 <span class="secno">A.1</span> Supporting Ruby Layout</h3>
1348 <p>The following represents a default UA style sheet
1349 for rendering HTML and XHTML ruby markup as ruby layout:
1351 <pre>
1352 <!-- -->ruby { display: ruby; }
1353 <!-- -->rb { display: ruby-base; white-space: nowrap; }
1354 <!-- -->rt { display: ruby-text; white-space: nowrap; font-size: 50%; }
1355 <!-- -->rbc { display: ruby-base-container; }
1356 <!-- -->rtc { display: ruby-text-container; }
1357 <!-- -->ruby, rb, rt, rbc, rtc { unicode-bidi: isolate; }</pre>
1359 <p>Additional rules for UAs supporting the relevant features of [[CSS3-TEXT-DECOR]] and [[CSS3-FONTS]]:
1360 <pre>rt { font-variant-east-asian: ruby; text-emphasis: none; }</pre>
1362 <p class="note">Authors should not use the above rules;
1363 a UA that supports ruby layout should provide these by default.
1365 <h3 id="default-inline" class="no-num">
1366 <span class="secno">A.2</span> Inlining Ruby Annotations</h3>
1368 <p>The following represents a sample style sheet
1369 for rendering HTML and XHTML ruby markup as inline annotations:
1371 <pre>ruby, rb, rt, rbc, rtc, rp {
1372 <!-- --> display: inline; white-space: inherit;
1373 <!-- --> font-variant-east-asian: inherit; text-emphasis: inherit; }</pre>
1375 <h3 id="default-parens" class="no-num">
1376 <span class="secno">A.3</span> Generating Parentheses</h3>
1378 <p>Unfortunately, because Selectors cannot match against text nodes,
1379 it's not possible with CSS to express rules that will automatically and correctly
1380 add parentheses to unparenthesized ruby annotations in HTML.
1381 (This is because HTML ruby allows implying the <i>ruby base</i> from raw text, without a corresponding element.)
1382 However, these rules will handle cases where either <code><rb></code>
1383 or <code><rtc></code> is used rigorously.
1385 <pre>
1386 <!-- -->/* Parens around <rtc> */
1387 <!-- -->rtc::before { content: "("; }
1388 <!-- -->rtc::after { content: ")"; }
1390 <!-- -->/* Parens before first <rt> not inside <rtc> */
1391 <!-- -->rb + rt::before,
1392 <!-- -->rtc + rt::before { content: "("; }
1394 <!-- -->/* Parens after <rt> not inside <rtc> */
1395 <!-- -->rb ~ rt:last-child::after,
1396 <!-- -->rt + rb::before { content: ")"; }
1397 <!-- -->rt + rtc::before { content: ")("; }</pre>
1399 <h2 id="glossary">
1400 Glossary</h2>
1401 <dl>
1402 <dt><a id="g-bopomofo"><strong><span
1403 lang="zh">Bopomofo</span></strong></a></dt>
1404 <dd>37 characters and 4 tone markings used as phonetics in Chinese,
1405 especially standard Mandarin.</dd>
1406 <dt><a id="g-hanja"><strong><span
1407 lang="ko">Hanja</span></strong></a></dt>
1408 <dd>Subset of the Korean writing system that utilizes ideographic
1409 characters borrowed or adapted from the Chinese writing system. Also see
1410 <a href="#g-kanji"><span lang="ja">Kanji</span></a>.</dd>
1411 <dt><a id="g-hiragana"><strong><span
1412 lang="ja">Hiragana</span></strong></a></dt>
1413 <dd>Japanese syllabic script, or character of that script. Rounded and
1414 cursive in appearance. Subset of the Japanese writing system, used together
1415 with kanji and katakana. In recent times, mostly used to write Japanese
1416 words when kanji are not available or appropriate, and word endings and
1417 particles. Also see <a
1418 href="#g-katakana"><span lang="ja">Katakana</span></a>.</dd>
1419 <dt><a id="g-ideogram"><strong>Ideograph</strong></a></dt>
1420 <dd>A character that is used to represent an idea, word, or word component,
1421 in contrast to a character from an alphabetic or syllabic script. The most
1422 well-known ideographic script is used (with some variation) in East Asia
1423 (China, Japan, Korea,...).</dd>
1424 <dt><a id="g-kana"><strong><span lang="ja">Kana</span></strong></a></dt>
1425 <dd>Collective term for hiragana and katakana.</dd>
1426 <dt><a id="g-kanji"><strong>Kanji</strong></a></dt>
1427 <dd>Japanese term for ideographs; ideographs used in Japanese. Subset of the
1428 Japanese writing system, used together with hiragana and katakana. Also see <a
1429 href="#g-hanja"><span lang="ko">Hanja</span></a>.</dd>
1430 <dt><a id="g-katakana"><strong><span
1431 lang="ja">Katakana</span></strong></a></dt>
1432 <dd>Japanese syllabic script, or character of that script. Angular in
1433 appearance. Subset of the Japanese writing system, used together with
1434 kanji and hiragana. In recent times, mainly used to write foreign words. Also see <a
1435 href="#g-hiragana"><span lang="ja">Hiragana</span></a>.</dd>
1436 <dt><a id="g-monoruby" name="g-monoruby"><strong>Mono-ruby</strong></a></dt>
1437 <dd>In Japanese typography: Ruby associated with a single character of
1438 the base text.</dd>
1439 <dt><a id="g-ruby"><strong>Ruby</strong></a></dt>
1440 <dd>A run of text that appears in the vicinity of another run of text and
1441 serves as an annotation or a pronunciation guide for that text.</dd>
1442 </dl>
1444 <h2 id="conformance">
1445 Conformance</h2>
1447 <h3 id="conventions">
1448 Document conventions</h3>
1450 <p>Conformance requirements are expressed with a combination of
1451 descriptive assertions and RFC 2119 terminology. The key words “MUST”,
1452 “MUST NOT”, “REQUIRED”, “SHALL”, “SHALL NOT”, “SHOULD”, “SHOULD NOT”,
1453 “RECOMMENDED”, “MAY”, and “OPTIONAL” in the normative parts of this
1454 document are to be interpreted as described in RFC 2119.
1455 However, for readability, these words do not appear in all uppercase
1456 letters in this specification.
1458 <p>All of the text of this specification is normative except sections
1459 explicitly marked as non-normative, examples, and notes. [[!RFC2119]]</p>
1461 <p>Examples in this specification are introduced with the words “for example”
1462 or are set apart from the normative text with <code>class="example"</code>,
1463 like this:
1465 <div class="example">
1466 <p>This is an example of an informative example.</p>
1467 </div>
1469 <p>Informative notes begin with the word “Note” and are set apart from the
1470 normative text with <code>class="note"</code>, like this:
1472 <p class="note">Note, this is an informative note.</p>
1474 <h3 id="conformance-classes">
1475 Conformance classes</h3>
1477 <p>Conformance to CSS Ruby Module
1478 is defined for three conformance classes:
1479 <dl>
1480 <dt><dfn title="style sheet!!as conformance class">style sheet</dfn>
1481 <dd>A <a href="http://www.w3.org/TR/CSS21/conform.html#style-sheet">CSS
1482 style sheet</a>.
1483 <dt><dfn>renderer</dfn></dt>
1484 <dd>A <a href="http://www.w3.org/TR/CSS21/conform.html#user-agent">UA</a>
1485 that interprets the semantics of a style sheet and renders
1486 documents that use them.
1487 <dt><dfn id="authoring-tool">authoring tool</dfn></dt>
1488 <dd>A <a href="http://www.w3.org/TR/CSS21/conform.html#user-agent">UA</a>
1489 that writes a style sheet.
1490 </dl>
1492 <p>A style sheet is conformant to CSS Ruby Module
1493 if all of its statements that use syntax defined in this module are valid
1494 according to the generic CSS grammar and the individual grammars of each
1495 feature defined in this module.
1497 <p>A renderer is conformant to CSS Ruby Module
1498 if, in addition to interpreting the style sheet as defined by the
1499 appropriate specifications, it supports all the features defined
1500 by CSS Ruby Module by parsing them correctly
1501 and rendering the document accordingly. However, the inability of a
1502 UA to correctly render a document due to limitations of the device
1503 does not make the UA non-conformant. (For example, a UA is not
1504 required to render color on a monochrome monitor.)
1506 <p>An authoring tool is conformant to CSS Ruby Module
1507 if it writes style sheets that are syntactically correct according to the
1508 generic CSS grammar and the individual grammars of each feature in
1509 this module, and meet all other conformance requirements of style sheets
1510 as described in this module.
1512 <h3 id="partial">
1513 Partial implementations</h3>
1515 <p>So that authors can exploit the forward-compatible parsing rules to
1516 assign fallback values, CSS renderers <strong>must</strong>
1517 treat as invalid (and <a href="http://www.w3.org/TR/CSS21/conform.html#ignore">ignore
1518 as appropriate</a>) any at-rules, properties, property values, keywords,
1519 and other syntactic constructs for which they have no usable level of
1520 support. In particular, user agents <strong>must not</strong> selectively
1521 ignore unsupported component values and honor supported values in a single
1522 multi-value property declaration: if any value is considered invalid
1523 (as unsupported values must be), CSS requires that the entire declaration
1524 be ignored.</p>
1526 <h3 id="experimental">
1527 Experimental implementations</h3>
1529 <p>To avoid clashes with future CSS features, the CSS2.1 specification
1530 reserves a <a href="http://www.w3.org/TR/CSS21/syndata.html#vendor-keywords">prefixed
1531 syntax</a> for proprietary and experimental extensions to CSS.
1533 <p>Prior to a specification reaching the Candidate Recommendation stage
1534 in the W3C process, all implementations of a CSS feature are considered
1535 experimental. The CSS Working Group recommends that implementations
1536 use a vendor-prefixed syntax for such features, including those in
1537 W3C Working Drafts. This avoids incompatibilities with future changes
1538 in the draft.
1539 </p>
1541 <h3 id="testing">
1542 Non-experimental implementations</h3>
1544 <p>Once a specification reaches the Candidate Recommendation stage,
1545 non-experimental implementations are possible, and implementors should
1546 release an unprefixed implementation of any CR-level feature they
1547 can demonstrate to be correctly implemented according to spec.
1549 <p>To establish and maintain the interoperability of CSS across
1550 implementations, the CSS Working Group requests that non-experimental
1551 CSS renderers submit an implementation report (and, if necessary, the
1552 testcases used for that implementation report) to the W3C before
1553 releasing an unprefixed implementation of any CSS features. Testcases
1554 submitted to W3C are subject to review and correction by the CSS
1555 Working Group.
1557 <p>Further information on submitting testcases and implementation reports
1558 can be found from on the CSS Working Group's website at
1559 <a href="http://www.w3.org/Style/CSS/Test/">http://www.w3.org/Style/CSS/Test/</a>.
1560 Questions should be directed to the
1561 <a href="http://lists.w3.org/Archives/Public/public-css-testsuite">public-css-testsuite@w3.org</a>
1562 mailing list.
1564 <h2 class=no-num id="acknowledgments">
1565 Acknowledgments</h2>
1567 <p>This specification would not have been possible without the help from:</p>
1569 <p>Stephen Deach, Martin Dürst, Hideki Hiura(<span lang="ja">樋浦 秀樹</span>), Masayasu Ishikawa(<span lang="ja">石川
1570 雅康</span>), Chris
1571 Pratley, Takao Suzuki(<span lang="ja">鈴木 孝雄</span>), Frank Yung-Fong Tang, Chris Thrasher, Masafumi Yabe<span lang="ja">家辺
1572 勝文</span>), Steve Zilles.</p>
1574 <h2 class="no-num" id="changes">
1575 Changes</h2>
1577 <p>The following major changes have been made since the previous Working Draft:
1578 <dl>
1579 <dt>Remove 'ruby-span' and mentions of <code>rbspan</code>.
1580 <dd>
1581 Explicit spanning is not used in HTML ruby in favor of implicit spanning.
1582 This can't handle some pathological double-sided spanning cases,
1583 but there seems to be no requirement for these at the moment.
1584 (For implementations that support full complex XHTML Ruby,
1585 they can imply spanning from the markup the same magic way
1586 that we handle cell spanning from tables. It doesn't seem
1587 necessary to include controls this in Level 1.)
1589 <dt>Defer 'ruby-overhang' and ''ruby-align: line-end'' to Level 2.
1590 <dd>
1591 It's somewhat complicated, advanced feature.
1592 Proposal is to make this behavior UA-defined
1593 and provide some examples of acceptable options.
1595 <dt>Close issue requesting 'display: rp': use ''display: none''.
1596 <dd>
1597 The i18nwg added an issue requesting a display value for <rp> elements.
1598 They're supposed to be hidden when &tl;ruby> is displayed as ruby.
1599 But this is easily accomplished already with ''display: none''.
1601 <dt>Change 'ruby-position' values to match 'text-emphasis-position'.
1602 <dd>
1603 Other than ''inter-character'', which we need to keep,
1604 it makes more sense to align ruby positions with 'text-emphasis-position',
1605 which can correctly handle various combinations of horizontal/vertical preferences.
1607 <dt>Remove unused values of 'ruby-align'.
1608 <dd>
1609 ''left'', ''right'', and ''end'' are not needed.
1611 <dt>Replace ''auto'', ''distribute-letter'', and ''distribute-space'' with ''space-between'' and ''space-around''.
1612 <dd>
1613 The ''auto'' value relied on inspecting content to determine behavior;
1614 this can be avoided by just using ''space-around'' with standard justification rules
1615 (which allow spacing between CJK but not between Latin).
1616 Replaced ''distribute-letter'' and ''distribute-space'' with
1617 ''space-between'' and ''space-around'' for consistency with Flexbox distribution keywords
1618 and to avoid links to the definition of ''text-justify: distribute''.
1620 <dt>Added 'ruby-merge' property to control jukugo rendering.
1621 <dd>
1622 This is a stylistic effect, not a structural one;
1623 the previous model assumed that it was structural and suggested handling it by changing markup. :(
1625 <dt>Remove ''inline'' from 'ruby-position'.
1626 <dd>
1627 This is do-able via ''display: inline'' on all the ruby-related elements,
1628 see <a href="#default-inline">Appendix A</a>
1630 <dt>Added <a href="#default-style">Default Style</a> rules
1631 <dd>
1632 As requested by i18nwg.
1634 <dt>Wrote anonymous box generation rules
1635 <dd>
1636 And defined pairing of bases and annotations.
1637 Should now handle all the crazy proposed permutations of HTML ruby markup.
1638 </dl>
1640 <h2 class=no-num id="references">
1641 References</h2>
1643 <h3 class="no-num" id="normative-references">
1644 Normative references</h3>
1645 <!--normative-->
1647 <h3 class="no-num" id="other-references">
1648 Other references</h3>
1649 <!--informative-->
1651 <h2 class="no-num" id="index">
1652 Index</h2>
1653 <!--index-->
1655 <h2 class="no-num" id="property-index">
1656 Property index</h2>
1657 <!-- properties -->
1659 </body>
1660 </html>
1661 <!-- Keep this comment at the end of the file
1662 Local variables:
1663 mode: sgml
1664 sgml-declaration:"~/SGML/HTML4.decl"
1665 sgml-default-doctype-name:"html"
1666 sgml-minimize-attributes:t
1667 sgml-nofill-elements:("pre" "style" "br")
1668 sgml-live-element-indicator:t
1669 sgml-omittag:nil
1670 sgml-shorttag:nil
1671 sgml-namecase-general:t
1672 sgml-general-insert-case:lower
1673 sgml-always-quote-attributes:t
1674 sgml-indent-step:nil
1675 sgml-indent-data:t
1676 sgml-parent-document:nil
1677 sgml-exposed-tags:nil
1678 sgml-local-catalogs:nil
1679 sgml-local-ecat-files:nil
1680 End:
1681 -->