css-fonts/Fonts.html

changeset 8430
330ae631cc35
parent 8397
8b231a13093c
child 8431
16e837f77ef5
     1.1 --- a/css-fonts/Fonts.html	Mon Jun 10 18:45:24 2013 -0700
     1.2 +++ b/css-fonts/Fonts.html	Tue Jun 11 14:34:10 2013 +0900
     1.3 @@ -115,7 +115,7 @@
     1.4  
     1.5     <h1>CSS Fonts Module Level 3</h1>
     1.6  
     1.7 -   <h2 class="no-num no-toc" id=editors-draft-10-june-2013>Editor's Draft 10
     1.8 +   <h2 class="no-num no-toc" id=editors-draft-11-june-2013>Editor's Draft 11
     1.9      June 2013</h2>
    1.10  
    1.11     <dl id=authors>
    1.12 @@ -123,7 +123,7 @@
    1.13  
    1.14      <dd><a
    1.15       href="http://dev.w3.org/csswg/css-fonts/">http://dev.w3.org/csswg/css-fonts/</a>
    1.16 -     <!-- <dd><a href="http://www.w3.org/TR/2013/ED-css3-fonts-20130610/">http://www.w3.org/TR/2013/ED-css3-fonts-20130610/</a> -->
    1.17 +     <!-- <dd><a href="http://www.w3.org/TR/2013/ED-css3-fonts-20130611/">http://www.w3.org/TR/2013/ED-css3-fonts-20130611/</a> -->
    1.18       
    1.19  
    1.20      <dt>Latest version:
    1.21 @@ -3342,6 +3342,12 @@
    1.22      Font</a>) or using the missing character glyph from a default font.
    1.23    </ol>
    1.24  
    1.25 +  <p>Optimizations of this process are allowed provided that an
    1.26 +   implementation behaves as if the algorithm had been followed exactly.
    1.27 +   Matching occurs in a well-defined order to insure that the results are as
    1.28 +   consistent as possible across user agents, given an identical set of
    1.29 +   available fonts and rendering technology.
    1.30 +
    1.31    <h3 id=cluster-matching><span class=secno>5.3 </span>Cluster matching</h3>
    1.32  
    1.33    <p>When text contains characters such as combining marks, ideally the base
    1.34 @@ -3402,8 +3408,8 @@
    1.35    <h3 id=char-handling-issues><span class=secno>5.4 </span>Character handling
    1.36     issues</h3>
    1.37  
    1.38 -  <p>The procedure above is always performed on text runs containing Unicode
    1.39 -   characters, documents using legacy encodings are assumed to have been
    1.40 +  <p>CSS font matching is always performed on text runs containing Unicode
    1.41 +   characters, so documents using legacy encodings are assumed to have been
    1.42     transcoded before matching fonts. For fonts containing <a
    1.43     href="#character-map"><em title="character map">character maps</em></a>
    1.44     for both legacy encodings and Unicode, the contents of the legacy encoding
    1.45 @@ -3413,22 +3419,23 @@
    1.46    <p>The font matching process does not assume that text runs are in either
    1.47     normalized or denormalized form (see <a href="#CHARMOD-NORM"
    1.48     rel=biblioentry>[CHARMOD-NORM]<!--{{CHARMOD-NORM}}--></a> for more
    1.49 -   details). Layout engines often convert base character plus combining
    1.50 -   character sequences into precomposed characters if they exist. The font
    1.51 -   matching algorithm outlined here supports both ways and fonts can
    1.52 -   generally support either but variations can occur. Authors should always
    1.53 +   details). Fonts may only support precomposed forms and not the decomposed
    1.54 +   sequence of base character plus combining marks. Authors should always
    1.55     tailor their choice of fonts to their content, including whether that
    1.56     content contains normalized or denormalized character streams.
    1.57  
    1.58 -  <p>If a given character is a Private-Use Area Unicode codepoint and none of
    1.59 -   the fonts in the fontlist contain a glyph for that codepoint, user agents
    1.60 -   must display some form of missing glyph symbol for that character rather
    1.61 -   than attempting <a href="#system-font-fallback"><em>system font
    1.62 -   fallback</em></a> for that codepoint. When matching the replacement
    1.63 -   character U+FFFD, user agents may skip the font matching process and
    1.64 -   immediately display some form of missing glyph symbol, they are not
    1.65 -   required to display the glyph from the font that would be selected by the
    1.66 -   font matching process.
    1.67 +  <p>If a given character is a Private-Use Area Unicode codepoint, user
    1.68 +   agents must only match font families named in the ‘<code
    1.69 +   class=property>font-family</code>’ list that are not generic families.
    1.70 +   If none of the families named in the ‘<code
    1.71 +   class=property>font-family</code>’ list contain a glyph for that
    1.72 +   codepoint, user agents must display some form of missing glyph symbol for
    1.73 +   that character rather than attempting <a
    1.74 +   href="#system-font-fallback"><em>system font fallback</em></a> for that
    1.75 +   codepoint. When matching the replacement character U+FFFD, user agents may
    1.76 +   skip the font matching process and immediately display some form of
    1.77 +   missing glyph symbol, they are not required to display the glyph from the
    1.78 +   font that would be selected by the font matching process.
    1.79  
    1.80    <p>In general, the fonts for a given family will all have the same or
    1.81     similar <a href="#character-map"><em title="character map">character
    1.82 @@ -3437,12 +3444,6 @@
    1.83     title="character map">character maps</em></a>. However, authors are
    1.84     cautioned that the use of such families can lead to unexpected results.
    1.85  
    1.86 -  <p>Optimizations of this process are allowed provided that an
    1.87 -   implementation behaves as if the algorithm had been followed exactly.
    1.88 -   Matching occurs in a well-defined order to insure that the results are as
    1.89 -   consistent as possible across user agents, given an identical set of
    1.90 -   available fonts and rendering technology.
    1.91 -
    1.92    <h3 id=font-matching-changes><span class=secno>5.5 </span>Font matching
    1.93     changes since CSS 2.1</h3>
    1.94  
    1.95 @@ -3478,8 +3479,8 @@
    1.96  *:lang(zh-tw) { font: 800 14pt/16.5pt &quot;Li Sung&quot;, serif; }
    1.97  </pre>
    1.98  
    1.99 -   <p>This selects any element that has the given language - Japanese or
   1.100 -    Traditional Chinese - and uses the appropriate font.
   1.101 +   <p>This selects any element that has the given language — Japanese or
   1.102 +    Traditional Chinese — and uses the appropriate font.
   1.103    </div>
   1.104  
   1.105    <h2 id=font-rend-props><span class=secno>6 </span>Font Feature Properties</h2>

mercurial