mu/html/403unicode.mu.html

425 lines
44 KiB
HTML

<!DOCTYPE html>
<html>
<head>
<meta charset="UTF-8">
<title>Mu - 403unicode.mu</title>
<meta name="Generator" content="Vim/8.2">
<meta name="plugin-version" content="vim8.1_v2">
<meta name="syntax" content="none">
<meta name="settings" content="number_lines,use_css,pre_wrap,no_foldcolumn,expand_tabs,line_ids,prevent_copy=,use_input_for_pc=fallback">
<meta name="colorscheme" content="minimal-light">
<style>
<!--
pre { white-space: pre-wrap; font-family: monospace; color: #000000; background-color: #ffffd7; }
body { font-size:12pt; font-family: monospace; color: #000000; background-color: #ffffd7; }
a { color:inherit; }
* { font-size:12pt; font-size: 1em; }
.PreProc { color: #c000c0; }
.muRegEdx { color: #af5f00; }
.LineNr { }
.muRegEdi { color: #00af00; }
.muRegEsi { color: #005faf; }
.muRegEbx { color: #5f00ff; }
.Constant { color: #008787; }
.muFunction { color: #af5f00; text-decoration: underline; }
.muRegEcx { color: #870000; }
.Delimiter { color: #c000c0; }
.Special { color: #ff6060; }
.muTest { color: #5f8700; }
.muComment { color: #005faf; }
-->
</style>
<script>
<!--
/* function to open any folds containing a jumped-to line before jumping to it */
function JumpToLine()
{
var lineNum;
lineNum = window.location.hash;
lineNum = lineNum.substr(1); /* strip off '#' */
if (lineNum.indexOf('L') == -1) {
lineNum = 'L'+lineNum;
}
var lineElem = document.getElementById(lineNum);
/* Always jump to new location even if the line was hidden inside a fold, or
* we corrected the raw number to a line ID.
*/
if (lineElem) {
lineElem.scrollIntoView(true);
}
return true;
}
if ('onhashchange' in window) {
window.onhashchange = JumpToLine;
}
-->
</script>
</head>
<body onload='JumpToLine();'>
<a href='https://github.com/akkartik/mu/blob/main/403unicode.mu'>https://github.com/akkartik/mu/blob/main/403unicode.mu</a>
<pre id='vimCodeElement'>
<span id="L1" class="LineNr"> 1 </span><span class="muComment"># Helpers for Unicode.</span>
<span id="L2" class="LineNr"> 2 </span><span class="muComment">#</span>
<span id="L3" class="LineNr"> 3 </span><span class="muComment"># Mu has no characters, only code points and graphemes.</span>
<span id="L4" class="LineNr"> 4 </span><span class="muComment"># Code points are the indivisible atoms of text streams.</span>
<span id="L5" class="LineNr"> 5 </span><span class="muComment"># <a href="https://en.wikipedia.org/wiki/Code_point">https://en.wikipedia.org/wiki/Code_point</a></span>
<span id="L6" class="LineNr"> 6 </span><span class="muComment"># Graphemes are the smallest self-contained unit of text.</span>
<span id="L7" class="LineNr"> 7 </span><span class="muComment"># Graphemes may consist of multiple code points.</span>
<span id="L8" class="LineNr"> 8 </span><span class="muComment">#</span>
<span id="L9" class="LineNr"> 9 </span><span class="muComment"># Mu graphemes are always represented in utf-8, and they are required to fit</span>
<span id="L10" class="LineNr"> 10 </span><span class="muComment"># in 4 bytes. (This can be confusing if you focus just on ASCII, where Mu's</span>
<span id="L11" class="LineNr"> 11 </span><span class="muComment"># graphemes and code-points are identical.)</span>
<span id="L12" class="LineNr"> 12 </span><span class="muComment">#</span>
<span id="L13" class="LineNr"> 13 </span><span class="muComment"># Mu doesn't currently support combining code points, or graphemes made of</span>
<span id="L14" class="LineNr"> 14 </span><span class="muComment"># multiple code points. One day we will.</span>
<span id="L15" class="LineNr"> 15 </span><span class="muComment"># <a href="https://en.wikipedia.org/wiki/Combining_character">https://en.wikipedia.org/wiki/Combining_character</a></span>
<span id="L16" class="LineNr"> 16 </span>
<span id="L17" class="LineNr"> 17 </span><span class="PreProc">fn</span> <span class="muTest"><a href='403unicode.mu.html#L17'>test-unicode-serialization-and-deserialization</a></span> <span class="Delimiter">{</span>
<span id="L18" class="LineNr"> 18 </span> <span class="PreProc">var</span> i/<span class="muRegEbx">ebx</span>: int <span class="Special">&lt;-</span> copy <span class="Constant">0</span>
<span id="L19" class="LineNr"> 19 </span> <span class="PreProc">var</span> init?/<span class="muRegEsi">esi</span>: boolean <span class="Special">&lt;-</span> copy <span class="Constant">1</span>/true
<span id="L20" class="LineNr"> 20 </span> <span class="Delimiter">{</span>
<span id="L21" class="LineNr"> 21 </span> compare i, <span class="Constant">0x10000</span> <span class="muComment"># 32 bits of utf-8 are sufficient for <a href="https://en.wikipedia.org/wiki/Plane_(Unicode)#Basic_Multilingual_Plane">https://en.wikipedia.org/wiki/Plane_(Unicode)#Basic_Multilingual_Plane</a></span>
<span id="L22" class="LineNr"> 22 </span> <span class="muComment"># but not emoji</span>
<span id="L23" class="LineNr"> 23 </span> <span class="PreProc">break-if-&gt;=</span>
<span id="L24" class="LineNr"> 24 </span> <span class="PreProc">var</span> c/eax: code-point <span class="Special">&lt;-</span> copy i
<span id="L25" class="LineNr"> 25 </span> <span class="PreProc">var</span> _g/eax: grapheme <span class="Special">&lt;-</span> <a href='403unicode.mu.html#L104'>to-grapheme</a> c
<span id="L26" class="LineNr"> 26 </span> <span class="PreProc">var</span> g/<span class="muRegEcx">ecx</span>: grapheme <span class="Special">&lt;-</span> copy _g
<span id="L27" class="LineNr"> 27 </span> <span class="PreProc">var</span> c2/eax: code-point <span class="Special">&lt;-</span> <a href='403unicode.mu.html#L56'>to-code-point</a> g
<span id="L28" class="LineNr"> 28 </span> compare i, c2
<span id="L29" class="LineNr"> 29 </span> <span class="Delimiter">{</span>
<span id="L30" class="LineNr"> 30 </span> <span class="PreProc">break-if-=</span>
<span id="L31" class="LineNr"> 31 </span> <span class="Delimiter">{</span>
<span id="L32" class="LineNr"> 32 </span> compare init?, <span class="Constant">0</span>/false
<span id="L33" class="LineNr"> 33 </span> <span class="PreProc">break-if-=</span>
<span id="L34" class="LineNr"> 34 </span> <a href='501draw-text.mu.html#L339'>draw-text-wrapping-right-then-down-from-cursor-over-full-screen</a> <span class="Constant">0</span>/screen, <span class="Constant">&quot;F - <a href='403unicode.mu.html#L17'>test-unicode-serialization-and-deserialization</a>: &quot;</span>, <span class="Constant">3</span>/fg <span class="Constant">0</span>/bg
<span id="L35" class="LineNr"> 35 </span> <span class="Delimiter">}</span>
<span id="L36" class="LineNr"> 36 </span> init? <span class="Special">&lt;-</span> copy <span class="Constant">0</span>/false
<span id="L37" class="LineNr"> 37 </span> <a href='501draw-text.mu.html#L394'>draw-int32-hex-wrapping-right-then-down-from-cursor-over-full-screen</a> <span class="Constant">0</span>/screen, i, <span class="Constant">3</span>/fg <span class="Constant">0</span>/bg
<span id="L38" class="LineNr"> 38 </span> <a href='501draw-text.mu.html#L339'>draw-text-wrapping-right-then-down-from-cursor-over-full-screen</a> <span class="Constant">0</span>/screen, <span class="Constant">&quot;/&quot;</span>, <span class="Constant">3</span>/fg <span class="Constant">0</span>/bg
<span id="L39" class="LineNr"> 39 </span> <span class="Delimiter">{</span>
<span id="L40" class="LineNr"> 40 </span> <span class="PreProc">var</span> x/eax: int <span class="Special">&lt;-</span> copy g
<span id="L41" class="LineNr"> 41 </span> <a href='501draw-text.mu.html#L394'>draw-int32-hex-wrapping-right-then-down-from-cursor-over-full-screen</a> <span class="Constant">0</span>/screen, x, <span class="Constant">3</span>/fg <span class="Constant">0</span>/bg
<span id="L42" class="LineNr"> 42 </span> <span class="Delimiter">}</span>
<span id="L43" class="LineNr"> 43 </span> <a href='501draw-text.mu.html#L339'>draw-text-wrapping-right-then-down-from-cursor-over-full-screen</a> <span class="Constant">0</span>/screen, <span class="Constant">&quot;/&quot;</span>, <span class="Constant">3</span>/fg <span class="Constant">0</span>/bg
<span id="L44" class="LineNr"> 44 </span> <span class="Delimiter">{</span>
<span id="L45" class="LineNr"> 45 </span> <span class="PreProc">var</span> x2/eax: int <span class="Special">&lt;-</span> copy c2
<span id="L46" class="LineNr"> 46 </span> <a href='501draw-text.mu.html#L394'>draw-int32-hex-wrapping-right-then-down-from-cursor-over-full-screen</a> <span class="Constant">0</span>/screen, x2, <span class="Constant">3</span>/fg <span class="Constant">0</span>/bg
<span id="L47" class="LineNr"> 47 </span> <span class="Delimiter">}</span>
<span id="L48" class="LineNr"> 48 </span> <a href='501draw-text.mu.html#L339'>draw-text-wrapping-right-then-down-from-cursor-over-full-screen</a> <span class="Constant">0</span>/screen, <span class="Constant">&quot; &quot;</span>, <span class="Constant">3</span>/fg <span class="Constant">0</span>/bg
<span id="L49" class="LineNr"> 49 </span> <span class="Delimiter">}</span>
<span id="L50" class="LineNr"> 50 </span> i <span class="Special">&lt;-</span> add <span class="Constant">0xf</span> <span class="muComment"># to speed things up; ensure increment is not a power of 2</span>
<span id="L51" class="LineNr"> 51 </span> <span class="PreProc">loop</span>
<span id="L52" class="LineNr"> 52 </span> <span class="Delimiter">}</span>
<span id="L53" class="LineNr"> 53 </span><span class="Delimiter">}</span>
<span id="L54" class="LineNr"> 54 </span>
<span id="L55" class="LineNr"> 55 </span><span class="muComment"># transliterated from tb_utf8_char_to_unicode in <a href="https://github.com/nsf/termbox">https://github.com/nsf/termbox</a></span>
<span id="L56" class="LineNr"> 56 </span><span class="PreProc">fn</span> <span class="muFunction"><a href='403unicode.mu.html#L56'>to-code-point</a></span> in: grapheme<span class="PreProc"> -&gt; </span>_/eax: code-point <span class="Delimiter">{</span>
<span id="L57" class="LineNr"> 57 </span> <span class="PreProc">var</span> g/<span class="muRegEbx">ebx</span>: int <span class="Special">&lt;-</span> copy in
<span id="L58" class="LineNr"> 58 </span> <span class="muComment"># if single byte, just return it</span>
<span id="L59" class="LineNr"> 59 </span> <span class="Delimiter">{</span>
<span id="L60" class="LineNr"> 60 </span> compare g, <span class="Constant">0xff</span>
<span id="L61" class="LineNr"> 61 </span> <span class="PreProc">break-if-&gt;</span>
<span id="L62" class="LineNr"> 62 </span> <span class="PreProc">var</span> result/eax: code-point <span class="Special">&lt;-</span> copy g
<span id="L63" class="LineNr"> 63 </span> <span class="PreProc">return</span> result
<span id="L64" class="LineNr"> 64 </span> <span class="Delimiter">}</span>
<span id="L65" class="LineNr"> 65 </span> <span class="muComment">#</span>
<span id="L66" class="LineNr"> 66 </span> <span class="PreProc">var</span> len/<span class="muRegEdx">edx</span>: int <span class="Special">&lt;-</span> <a href='403unicode.mu.html#L302'>grapheme-length</a> in
<span id="L67" class="LineNr"> 67 </span> <span class="muComment"># extract bits from first byte</span>
<span id="L68" class="LineNr"> 68 </span> <span class="PreProc">var</span> b/eax: byte <span class="Special">&lt;-</span> copy-byte g
<span id="L69" class="LineNr"> 69 </span> <span class="PreProc">var</span> result/<span class="muRegEdi">edi</span>: code-point <span class="Special">&lt;-</span> copy b
<span id="L70" class="LineNr"> 70 </span> <span class="Delimiter">{</span>
<span id="L71" class="LineNr"> 71 </span> compare len, <span class="Constant">2</span>
<span id="L72" class="LineNr"> 72 </span> <span class="PreProc">break-if-!=</span>
<span id="L73" class="LineNr"> 73 </span> result <span class="Special">&lt;-</span> and <span class="Constant">0x1f</span>
<span id="L74" class="LineNr"> 74 </span> <span class="Delimiter">}</span>
<span id="L75" class="LineNr"> 75 </span> <span class="Delimiter">{</span>
<span id="L76" class="LineNr"> 76 </span> compare len, <span class="Constant">3</span>
<span id="L77" class="LineNr"> 77 </span> <span class="PreProc">break-if-!=</span>
<span id="L78" class="LineNr"> 78 </span> result <span class="Special">&lt;-</span> and <span class="Constant">0x0f</span>
<span id="L79" class="LineNr"> 79 </span> <span class="Delimiter">}</span>
<span id="L80" class="LineNr"> 80 </span> <span class="Delimiter">{</span>
<span id="L81" class="LineNr"> 81 </span> compare len, <span class="Constant">4</span>
<span id="L82" class="LineNr"> 82 </span> <span class="PreProc">break-if-!=</span>
<span id="L83" class="LineNr"> 83 </span> result <span class="Special">&lt;-</span> and <span class="Constant">0x07</span>
<span id="L84" class="LineNr"> 84 </span> <span class="Delimiter">}</span>
<span id="L85" class="LineNr"> 85 </span> <span class="muComment"># extract bits from remaining bytes</span>
<span id="L86" class="LineNr"> 86 </span> g <span class="Special">&lt;-</span> shift-right <span class="Constant">8</span>
<span id="L87" class="LineNr"> 87 </span> <span class="PreProc">var</span> i/<span class="muRegEcx">ecx</span>: int <span class="Special">&lt;-</span> copy <span class="Constant">1</span>
<span id="L88" class="LineNr"> 88 </span> <span class="Delimiter">{</span>
<span id="L89" class="LineNr"> 89 </span> compare i, len
<span id="L90" class="LineNr"> 90 </span> <span class="PreProc">break-if-&gt;=</span>
<span id="L91" class="LineNr"> 91 </span> <span class="PreProc">var</span> b/eax: byte <span class="Special">&lt;-</span> copy-byte g
<span id="L92" class="LineNr"> 92 </span> b <span class="Special">&lt;-</span> and <span class="Constant">0x3f</span>
<span id="L93" class="LineNr"> 93 </span> result <span class="Special">&lt;-</span> shift-left <span class="Constant">6</span>
<span id="L94" class="LineNr"> 94 </span> result <span class="Special">&lt;-</span> or b
<span id="L95" class="LineNr"> 95 </span> g <span class="Special">&lt;-</span> shift-right <span class="Constant">8</span>
<span id="L96" class="LineNr"> 96 </span> i <span class="Special">&lt;-</span> increment
<span id="L97" class="LineNr"> 97 </span> <span class="PreProc">loop</span>
<span id="L98" class="LineNr"> 98 </span> <span class="Delimiter">}</span>
<span id="L99" class="LineNr"> 99 </span> <span class="PreProc">return</span> result
<span id="L100" class="LineNr">100 </span><span class="Delimiter">}</span>
<span id="L101" class="LineNr">101 </span>
<span id="L102" class="LineNr">102 </span><span class="muComment"># transliterated from tb_utf8_unicode_to_char in <a href="https://github.com/nsf/termbox">https://github.com/nsf/termbox</a></span>
<span id="L103" class="LineNr">103 </span><span class="muComment"># <a href="https://wiki.tcl-lang.org/page/UTF%2D8+bit+by+bit">https://wiki.tcl-lang.org/page/UTF%2D8+bit+by+bit</a> explains the algorithm</span>
<span id="L104" class="LineNr">104 </span><span class="PreProc">fn</span> <span class="muFunction"><a href='403unicode.mu.html#L104'>to-grapheme</a></span> in: code-point<span class="PreProc"> -&gt; </span>_/eax: grapheme <span class="Delimiter">{</span>
<span id="L105" class="LineNr">105 </span> <span class="PreProc">var</span> c/eax: int <span class="Special">&lt;-</span> copy in
<span id="L106" class="LineNr">106 </span> <span class="PreProc">var</span> num-trailers/<span class="muRegEcx">ecx</span>: int <span class="Special">&lt;-</span> copy <span class="Constant">0</span>
<span id="L107" class="LineNr">107 </span> <span class="PreProc">var</span> first/<span class="muRegEdx">edx</span>: int <span class="Special">&lt;-</span> copy <span class="Constant">0</span>
<span id="L108" class="LineNr">108 </span> $to-grapheme:compute-length: <span class="Delimiter">{</span>
<span id="L109" class="LineNr">109 </span> <span class="muComment"># single byte: just return it</span>
<span id="L110" class="LineNr">110 </span> compare c, <span class="Constant">0x7f</span>
<span id="L111" class="LineNr">111 </span> <span class="Delimiter">{</span>
<span id="L112" class="LineNr">112 </span> <span class="PreProc">break-if-&gt;</span>
<span id="L113" class="LineNr">113 </span> <span class="PreProc">var</span> g/eax: grapheme <span class="Special">&lt;-</span> copy c
<span id="L114" class="LineNr">114 </span> <span class="PreProc">return</span> g
<span id="L115" class="LineNr">115 </span> <span class="Delimiter">}</span>
<span id="L116" class="LineNr">116 </span> <span class="muComment"># 2 bytes</span>
<span id="L117" class="LineNr">117 </span> compare c, <span class="Constant">0x7ff</span>
<span id="L118" class="LineNr">118 </span> <span class="Delimiter">{</span>
<span id="L119" class="LineNr">119 </span> <span class="PreProc">break-if-&gt;</span>
<span id="L120" class="LineNr">120 </span> num-trailers <span class="Special">&lt;-</span> copy <span class="Constant">1</span>
<span id="L121" class="LineNr">121 </span> first <span class="Special">&lt;-</span> copy <span class="Constant">0xc0</span>
<span id="L122" class="LineNr">122 </span> <span class="PreProc">break</span> $to-grapheme:compute-length
<span id="L123" class="LineNr">123 </span> <span class="Delimiter">}</span>
<span id="L124" class="LineNr">124 </span> <span class="muComment"># 3 bytes</span>
<span id="L125" class="LineNr">125 </span> compare c, <span class="Constant">0xffff</span>
<span id="L126" class="LineNr">126 </span> <span class="Delimiter">{</span>
<span id="L127" class="LineNr">127 </span> <span class="PreProc">break-if-&gt;</span>
<span id="L128" class="LineNr">128 </span> num-trailers <span class="Special">&lt;-</span> copy <span class="Constant">2</span>
<span id="L129" class="LineNr">129 </span> first <span class="Special">&lt;-</span> copy <span class="Constant">0xe0</span>
<span id="L130" class="LineNr">130 </span> <span class="PreProc">break</span> $to-grapheme:compute-length
<span id="L131" class="LineNr">131 </span> <span class="Delimiter">}</span>
<span id="L132" class="LineNr">132 </span> <span class="muComment"># 4 bytes</span>
<span id="L133" class="LineNr">133 </span> compare c, <span class="Constant">0x1fffff</span>
<span id="L134" class="LineNr">134 </span> <span class="Delimiter">{</span>
<span id="L135" class="LineNr">135 </span> <span class="PreProc">break-if-&gt;</span>
<span id="L136" class="LineNr">136 </span> num-trailers <span class="Special">&lt;-</span> copy <span class="Constant">3</span>
<span id="L137" class="LineNr">137 </span> first <span class="Special">&lt;-</span> copy <span class="Constant">0xf0</span>
<span id="L138" class="LineNr">138 </span> <span class="PreProc">break</span> $to-grapheme:compute-length
<span id="L139" class="LineNr">139 </span> <span class="Delimiter">}</span>
<span id="L140" class="LineNr">140 </span> <span class="muComment"># more than 4 bytes: unsupported</span>
<span id="L141" class="LineNr">141 </span> <span class="muComment"># TODO: print error message to stderr</span>
<span id="L142" class="LineNr">142 </span> compare c, <span class="Constant">0x1fffff</span>
<span id="L143" class="LineNr">143 </span> <span class="Delimiter">{</span>
<span id="L144" class="LineNr">144 </span> <span class="PreProc">break-if-&gt;</span>
<span id="L145" class="LineNr">145 </span> <span class="PreProc">return</span> <span class="Constant">0</span>
<span id="L146" class="LineNr">146 </span> <span class="Delimiter">}</span>
<span id="L147" class="LineNr">147 </span> <span class="Delimiter">}</span>
<span id="L148" class="LineNr">148 </span> <span class="muComment"># emit trailer bytes, 6 bits from 'in', first two bits '10'</span>
<span id="L149" class="LineNr">149 </span> <span class="PreProc">var</span> result/<span class="muRegEdi">edi</span>: grapheme <span class="Special">&lt;-</span> copy <span class="Constant">0</span>
<span id="L150" class="LineNr">150 </span> <span class="Delimiter">{</span>
<span id="L151" class="LineNr">151 </span> compare num-trailers, <span class="Constant">0</span>
<span id="L152" class="LineNr">152 </span> <span class="PreProc">break-if-&lt;=</span>
<span id="L153" class="LineNr">153 </span> <span class="PreProc">var</span> tmp/<span class="muRegEsi">esi</span>: int <span class="Special">&lt;-</span> copy c
<span id="L154" class="LineNr">154 </span> tmp <span class="Special">&lt;-</span> and <span class="Constant">0x3f</span>
<span id="L155" class="LineNr">155 </span> tmp <span class="Special">&lt;-</span> or <span class="Constant">0x80</span>
<span id="L156" class="LineNr">156 </span> result <span class="Special">&lt;-</span> shift-left <span class="Constant">8</span>
<span id="L157" class="LineNr">157 </span> result <span class="Special">&lt;-</span> or tmp
<span id="L158" class="LineNr">158 </span> <span class="muComment"># update loop state</span>
<span id="L159" class="LineNr">159 </span> c <span class="Special">&lt;-</span> shift-right <span class="Constant">6</span>
<span id="L160" class="LineNr">160 </span> num-trailers <span class="Special">&lt;-</span> decrement
<span id="L161" class="LineNr">161 </span> <span class="PreProc">loop</span>
<span id="L162" class="LineNr">162 </span> <span class="Delimiter">}</span>
<span id="L163" class="LineNr">163 </span> <span class="muComment"># emit engine</span>
<span id="L164" class="LineNr">164 </span> result <span class="Special">&lt;-</span> shift-left <span class="Constant">8</span>
<span id="L165" class="LineNr">165 </span> result <span class="Special">&lt;-</span> or c
<span id="L166" class="LineNr">166 </span> result <span class="Special">&lt;-</span> or first
<span id="L167" class="LineNr">167 </span> <span class="muComment">#</span>
<span id="L168" class="LineNr">168 </span> <span class="PreProc">return</span> result
<span id="L169" class="LineNr">169 </span><span class="Delimiter">}</span>
<span id="L170" class="LineNr">170 </span>
<span id="L171" class="LineNr">171 </span><span class="muComment"># single-byte code point have identical graphemes</span>
<span id="L172" class="LineNr">172 </span><span class="PreProc">fn</span> <span class="muTest"><a href='403unicode.mu.html#L172'>test-to-grapheme-single-byte</a></span> <span class="Delimiter">{</span>
<span id="L173" class="LineNr">173 </span> <span class="PreProc">var</span> in-int/<span class="muRegEcx">ecx</span>: int <span class="Special">&lt;-</span> copy <span class="Constant">0</span>
<span id="L174" class="LineNr">174 </span> <span class="Delimiter">{</span>
<span id="L175" class="LineNr">175 </span> compare in-int, <span class="Constant">0x7f</span>
<span id="L176" class="LineNr">176 </span> <span class="PreProc">break-if-&gt;</span>
<span id="L177" class="LineNr">177 </span> <span class="PreProc">var</span> in/eax: code-point <span class="Special">&lt;-</span> copy in-int
<span id="L178" class="LineNr">178 </span> <span class="PreProc">var</span> out/eax: grapheme <span class="Special">&lt;-</span> <a href='403unicode.mu.html#L104'>to-grapheme</a> in
<span id="L179" class="LineNr">179 </span> <span class="PreProc">var</span> out-int/eax: int <span class="Special">&lt;-</span> copy out
<span id="L180" class="LineNr">180 </span> <a href='502test.mu.html#L2'>check-ints-equal</a> out-int, in-int, <span class="Constant">&quot;F - test-to-grapheme-single-byte&quot;</span>
<span id="L181" class="LineNr">181 </span> in-int <span class="Special">&lt;-</span> increment
<span id="L182" class="LineNr">182 </span> <span class="PreProc">loop</span>
<span id="L183" class="LineNr">183 </span> <span class="Delimiter">}</span>
<span id="L184" class="LineNr">184 </span><span class="Delimiter">}</span>
<span id="L185" class="LineNr">185 </span>
<span id="L186" class="LineNr">186 </span> <span class="muComment"># byte | byte | byte | byte</span>
<span id="L187" class="LineNr">187 </span><span class="muComment"># smallest 2-byte utf-8</span>
<span id="L188" class="LineNr">188 </span><span class="PreProc">fn</span> <span class="muTest"><a href='403unicode.mu.html#L188'>test-to-grapheme-two-bytes-min</a></span> <span class="Delimiter">{</span>
<span id="L189" class="LineNr">189 </span> <span class="PreProc">var</span> in/eax: code-point <span class="Special">&lt;-</span> copy <span class="Constant">0x80</span> <span class="muComment"># 10 00-0000</span>
<span id="L190" class="LineNr">190 </span> <span class="PreProc">var</span> out/eax: grapheme <span class="Special">&lt;-</span> <a href='403unicode.mu.html#L104'>to-grapheme</a> in
<span id="L191" class="LineNr">191 </span> <span class="PreProc">var</span> out-int/eax: int <span class="Special">&lt;-</span> copy out
<span id="L192" class="LineNr">192 </span> <a href='502test.mu.html#L2'>check-ints-equal</a> out-int, <span class="Constant">0x80c2</span>, <span class="Constant">&quot;F - <a href='403unicode.mu.html#L104'>to-grapheme</a>/2a&quot;</span> <span class="muComment"># 110 0-0010 10 00-0000</span>
<span id="L193" class="LineNr">193 </span><span class="Delimiter">}</span>
<span id="L194" class="LineNr">194 </span>
<span id="L195" class="LineNr">195 </span><span class="muComment"># largest 2-byte utf-8</span>
<span id="L196" class="LineNr">196 </span><span class="PreProc">fn</span> <span class="muTest"><a href='403unicode.mu.html#L196'>test-to-grapheme-two-bytes-max</a></span> <span class="Delimiter">{</span>
<span id="L197" class="LineNr">197 </span> <span class="PreProc">var</span> in/eax: code-point <span class="Special">&lt;-</span> copy <span class="Constant">0x7ff</span> <span class="muComment"># 1-1111 11-1111</span>
<span id="L198" class="LineNr">198 </span> <span class="PreProc">var</span> out/eax: grapheme <span class="Special">&lt;-</span> <a href='403unicode.mu.html#L104'>to-grapheme</a> in
<span id="L199" class="LineNr">199 </span> <span class="PreProc">var</span> out-int/eax: int <span class="Special">&lt;-</span> copy out
<span id="L200" class="LineNr">200 </span> <a href='502test.mu.html#L2'>check-ints-equal</a> out-int, <span class="Constant">0xbfdf</span>, <span class="Constant">&quot;F - <a href='403unicode.mu.html#L104'>to-grapheme</a>/2b&quot;</span> <span class="muComment"># 110 1-1111 10 11-1111</span>
<span id="L201" class="LineNr">201 </span><span class="Delimiter">}</span>
<span id="L202" class="LineNr">202 </span>
<span id="L203" class="LineNr">203 </span><span class="muComment"># smallest 3-byte utf-8</span>
<span id="L204" class="LineNr">204 </span><span class="PreProc">fn</span> <span class="muTest"><a href='403unicode.mu.html#L204'>test-to-grapheme-three-bytes-min</a></span> <span class="Delimiter">{</span>
<span id="L205" class="LineNr">205 </span> <span class="PreProc">var</span> in/eax: code-point <span class="Special">&lt;-</span> copy <span class="Constant">0x800</span> <span class="muComment"># 10-0000 00-0000</span>
<span id="L206" class="LineNr">206 </span> <span class="PreProc">var</span> out/eax: grapheme <span class="Special">&lt;-</span> <a href='403unicode.mu.html#L104'>to-grapheme</a> in
<span id="L207" class="LineNr">207 </span> <span class="PreProc">var</span> out-int/eax: int <span class="Special">&lt;-</span> copy out
<span id="L208" class="LineNr">208 </span> <a href='502test.mu.html#L2'>check-ints-equal</a> out-int, <span class="Constant">0x80a0e0</span>, <span class="Constant">&quot;F - <a href='403unicode.mu.html#L104'>to-grapheme</a>/3a&quot;</span> <span class="muComment"># 1110 0000 10 10-0000 10 00-0000</span>
<span id="L209" class="LineNr">209 </span><span class="Delimiter">}</span>
<span id="L210" class="LineNr">210 </span>
<span id="L211" class="LineNr">211 </span><span class="muComment"># largest 3-byte utf-8</span>
<span id="L212" class="LineNr">212 </span><span class="PreProc">fn</span> <span class="muTest"><a href='403unicode.mu.html#L212'>test-to-grapheme-three-bytes-max</a></span> <span class="Delimiter">{</span>
<span id="L213" class="LineNr">213 </span> <span class="PreProc">var</span> in/eax: code-point <span class="Special">&lt;-</span> copy <span class="Constant">0xffff</span> <span class="muComment"># 1111 11-1111 11-1111</span>
<span id="L214" class="LineNr">214 </span> <span class="PreProc">var</span> out/eax: grapheme <span class="Special">&lt;-</span> <a href='403unicode.mu.html#L104'>to-grapheme</a> in
<span id="L215" class="LineNr">215 </span> <span class="PreProc">var</span> out-int/eax: int <span class="Special">&lt;-</span> copy out
<span id="L216" class="LineNr">216 </span> <a href='502test.mu.html#L2'>check-ints-equal</a> out-int, <span class="Constant">0xbfbfef</span>, <span class="Constant">&quot;F - <a href='403unicode.mu.html#L104'>to-grapheme</a>/3b&quot;</span> <span class="muComment"># 1110 1111 10 11-1111 10 11-1111</span>
<span id="L217" class="LineNr">217 </span><span class="Delimiter">}</span>
<span id="L218" class="LineNr">218 </span>
<span id="L219" class="LineNr">219 </span><span class="muComment"># smallest 4-byte utf-8</span>
<span id="L220" class="LineNr">220 </span><span class="PreProc">fn</span> <span class="muTest"><a href='403unicode.mu.html#L220'>test-to-grapheme-four-bytes-min</a></span> <span class="Delimiter">{</span>
<span id="L221" class="LineNr">221 </span> <span class="PreProc">var</span> in/eax: code-point <span class="Special">&lt;-</span> copy <span class="Constant">0x10000</span> <span class="muComment"># 1-0000 00-0000 00-0000</span>
<span id="L222" class="LineNr">222 </span> <span class="PreProc">var</span> out/eax: grapheme <span class="Special">&lt;-</span> <a href='403unicode.mu.html#L104'>to-grapheme</a> in
<span id="L223" class="LineNr">223 </span> <span class="PreProc">var</span> out-int/eax: int <span class="Special">&lt;-</span> copy out
<span id="L224" class="LineNr">224 </span> <a href='502test.mu.html#L2'>check-ints-equal</a> out-int, <span class="Constant">0x808090f0</span>, <span class="Constant">&quot;F - <a href='403unicode.mu.html#L104'>to-grapheme</a>/4a&quot;</span> <span class="muComment"># 1111-0 000 10 01-0000 10 00-0000 10 00-0000</span>
<span id="L225" class="LineNr">225 </span><span class="Delimiter">}</span>
<span id="L226" class="LineNr">226 </span>
<span id="L227" class="LineNr">227 </span><span class="muComment"># largest 4-byte utf-8</span>
<span id="L228" class="LineNr">228 </span><span class="PreProc">fn</span> <span class="muTest"><a href='403unicode.mu.html#L228'>test-to-grapheme-four-bytes-max</a></span> <span class="Delimiter">{</span>
<span id="L229" class="LineNr">229 </span> <span class="PreProc">var</span> in/eax: code-point <span class="Special">&lt;-</span> copy <span class="Constant">0x1fffff</span> <span class="muComment"># 111 11-1111 11-1111 11-1111</span>
<span id="L230" class="LineNr">230 </span> <span class="PreProc">var</span> out/eax: grapheme <span class="Special">&lt;-</span> <a href='403unicode.mu.html#L104'>to-grapheme</a> in
<span id="L231" class="LineNr">231 </span> <span class="PreProc">var</span> out-int/eax: int <span class="Special">&lt;-</span> copy out
<span id="L232" class="LineNr">232 </span> <a href='502test.mu.html#L2'>check-ints-equal</a> out-int, <span class="Constant">0xbfbfbff7</span>, <span class="Constant">&quot;F - <a href='403unicode.mu.html#L104'>to-grapheme</a>/4b&quot;</span> <span class="muComment"># 1111-0 111 10 11-1111 10 11-1111 10 11-1111</span>
<span id="L233" class="LineNr">233 </span><span class="Delimiter">}</span>
<span id="L234" class="LineNr">234 </span>
<span id="L235" class="LineNr">235 </span><span class="muComment"># read the next grapheme from a stream of bytes</span>
<span id="L236" class="LineNr">236 </span><span class="PreProc">fn</span> <span class="muFunction"><a href='403unicode.mu.html#L236'>read-grapheme</a></span> in: (addr stream byte)<span class="PreProc"> -&gt; </span>_/eax: grapheme <span class="Delimiter">{</span>
<span id="L237" class="LineNr">237 </span> <span class="muComment"># if at eof, return EOF</span>
<span id="L238" class="LineNr">238 </span> <span class="Delimiter">{</span>
<span id="L239" class="LineNr">239 </span> <span class="PreProc">var</span> eof?/eax: boolean <span class="Special">&lt;-</span> <a href='309stream.subx.html#L6'>stream-empty?</a> in
<span id="L240" class="LineNr">240 </span> compare eof?, <span class="Constant">0</span>/false
<span id="L241" class="LineNr">241 </span> <span class="PreProc">break-if-=</span>
<span id="L242" class="LineNr">242 </span> <span class="PreProc">return</span> <span class="Constant">0xffffffff</span>
<span id="L243" class="LineNr">243 </span> <span class="Delimiter">}</span>
<span id="L244" class="LineNr">244 </span> <span class="PreProc">var</span> c/eax: byte <span class="Special">&lt;-</span> <a href='112read-byte.subx.html#L13'>read-byte</a> in
<span id="L245" class="LineNr">245 </span> <span class="PreProc">var</span> num-trailers/<span class="muRegEcx">ecx</span>: int <span class="Special">&lt;-</span> copy <span class="Constant">0</span>
<span id="L246" class="LineNr">246 </span> $read-grapheme:compute-length: <span class="Delimiter">{</span>
<span id="L247" class="LineNr">247 </span> <span class="muComment"># single byte: just return it</span>
<span id="L248" class="LineNr">248 </span> compare c, <span class="Constant">0xc0</span>
<span id="L249" class="LineNr">249 </span> <span class="Delimiter">{</span>
<span id="L250" class="LineNr">250 </span> <span class="PreProc">break-if-&gt;=</span>
<span id="L251" class="LineNr">251 </span> <span class="PreProc">var</span> g/eax: grapheme <span class="Special">&lt;-</span> copy c
<span id="L252" class="LineNr">252 </span> <span class="PreProc">return</span> g
<span id="L253" class="LineNr">253 </span> <span class="Delimiter">}</span>
<span id="L254" class="LineNr">254 </span> compare c, <span class="Constant">0xfe</span>
<span id="L255" class="LineNr">255 </span> <span class="Delimiter">{</span>
<span id="L256" class="LineNr">256 </span> <span class="PreProc">break-if-&lt;</span>
<span id="L257" class="LineNr">257 </span> <span class="PreProc">var</span> g/eax: grapheme <span class="Special">&lt;-</span> copy c
<span id="L258" class="LineNr">258 </span> <span class="PreProc">return</span> g
<span id="L259" class="LineNr">259 </span> <span class="Delimiter">}</span>
<span id="L260" class="LineNr">260 </span> <span class="muComment"># 2 bytes</span>
<span id="L261" class="LineNr">261 </span> compare c, <span class="Constant">0xe0</span>
<span id="L262" class="LineNr">262 </span> <span class="Delimiter">{</span>
<span id="L263" class="LineNr">263 </span> <span class="PreProc">break-if-&gt;=</span>
<span id="L264" class="LineNr">264 </span> num-trailers <span class="Special">&lt;-</span> copy <span class="Constant">1</span>
<span id="L265" class="LineNr">265 </span> <span class="PreProc">break</span> $read-grapheme:compute-length
<span id="L266" class="LineNr">266 </span> <span class="Delimiter">}</span>
<span id="L267" class="LineNr">267 </span> <span class="muComment"># 3 bytes</span>
<span id="L268" class="LineNr">268 </span> compare c, <span class="Constant">0xf0</span>
<span id="L269" class="LineNr">269 </span> <span class="Delimiter">{</span>
<span id="L270" class="LineNr">270 </span> <span class="PreProc">break-if-&gt;=</span>
<span id="L271" class="LineNr">271 </span> num-trailers <span class="Special">&lt;-</span> copy <span class="Constant">2</span>
<span id="L272" class="LineNr">272 </span> <span class="PreProc">break</span> $read-grapheme:compute-length
<span id="L273" class="LineNr">273 </span> <span class="Delimiter">}</span>
<span id="L274" class="LineNr">274 </span> <span class="muComment"># 4 bytes</span>
<span id="L275" class="LineNr">275 </span> compare c, <span class="Constant">0xf8</span>
<span id="L276" class="LineNr">276 </span> <span class="Delimiter">{</span>
<span id="L277" class="LineNr">277 </span> <span class="PreProc">break-if-&gt;=</span>
<span id="L278" class="LineNr">278 </span> num-trailers <span class="Special">&lt;-</span> copy <span class="Constant">3</span>
<span id="L279" class="LineNr">279 </span> <span class="PreProc">break</span> $read-grapheme:compute-length
<span id="L280" class="LineNr">280 </span> <span class="Delimiter">}</span>
<span id="L281" class="LineNr">281 </span> <span class="muComment"># TODO: print error message</span>
<span id="L282" class="LineNr">282 </span> <span class="PreProc">return</span> <span class="Constant">0</span>
<span id="L283" class="LineNr">283 </span> <span class="Delimiter">}</span>
<span id="L284" class="LineNr">284 </span> <span class="muComment"># prepend trailer bytes</span>
<span id="L285" class="LineNr">285 </span> <span class="PreProc">var</span> result/<span class="muRegEdi">edi</span>: grapheme <span class="Special">&lt;-</span> copy c
<span id="L286" class="LineNr">286 </span> <span class="PreProc">var</span> num-byte-shifts/<span class="muRegEdx">edx</span>: int <span class="Special">&lt;-</span> copy <span class="Constant">1</span>
<span id="L287" class="LineNr">287 </span> <span class="Delimiter">{</span>
<span id="L288" class="LineNr">288 </span> compare num-trailers, <span class="Constant">0</span>
<span id="L289" class="LineNr">289 </span> <span class="PreProc">break-if-&lt;=</span>
<span id="L290" class="LineNr">290 </span> <span class="PreProc">var</span> tmp/eax: byte <span class="Special">&lt;-</span> <a href='112read-byte.subx.html#L13'>read-byte</a> in
<span id="L291" class="LineNr">291 </span> <span class="PreProc">var</span> tmp2/eax: int <span class="Special">&lt;-</span> copy tmp
<span id="L292" class="LineNr">292 </span> tmp2 <span class="Special">&lt;-</span> <a href='403unicode.mu.html#L322'>shift-left-bytes</a> tmp2, num-byte-shifts
<span id="L293" class="LineNr">293 </span> result <span class="Special">&lt;-</span> or tmp2
<span id="L294" class="LineNr">294 </span> <span class="muComment"># update loop state</span>
<span id="L295" class="LineNr">295 </span> num-byte-shifts <span class="Special">&lt;-</span> increment
<span id="L296" class="LineNr">296 </span> num-trailers <span class="Special">&lt;-</span> decrement
<span id="L297" class="LineNr">297 </span> <span class="PreProc">loop</span>
<span id="L298" class="LineNr">298 </span> <span class="Delimiter">}</span>
<span id="L299" class="LineNr">299 </span> <span class="PreProc">return</span> result
<span id="L300" class="LineNr">300 </span><span class="Delimiter">}</span>
<span id="L301" class="LineNr">301 </span>
<span id="L302" class="LineNr">302 </span><span class="PreProc">fn</span> <span class="muFunction"><a href='403unicode.mu.html#L302'>grapheme-length</a></span> g: grapheme<span class="PreProc"> -&gt; </span>_/<span class="muRegEdx">edx</span>: int <span class="Delimiter">{</span>
<span id="L303" class="LineNr">303 </span> <span class="Delimiter">{</span>
<span id="L304" class="LineNr">304 </span> compare g, <span class="Constant">0xff</span>
<span id="L305" class="LineNr">305 </span> <span class="PreProc">break-if-&gt;</span>
<span id="L306" class="LineNr">306 </span> <span class="PreProc">return</span> <span class="Constant">1</span>
<span id="L307" class="LineNr">307 </span> <span class="Delimiter">}</span>
<span id="L308" class="LineNr">308 </span> <span class="Delimiter">{</span>
<span id="L309" class="LineNr">309 </span> compare g, <span class="Constant">0xffff</span>
<span id="L310" class="LineNr">310 </span> <span class="PreProc">break-if-&gt;</span>
<span id="L311" class="LineNr">311 </span> <span class="PreProc">return</span> <span class="Constant">2</span>
<span id="L312" class="LineNr">312 </span> <span class="Delimiter">}</span>
<span id="L313" class="LineNr">313 </span> <span class="Delimiter">{</span>
<span id="L314" class="LineNr">314 </span> compare g, <span class="Constant">0xffffff</span>
<span id="L315" class="LineNr">315 </span> <span class="PreProc">break-if-&gt;</span>
<span id="L316" class="LineNr">316 </span> <span class="PreProc">return</span> <span class="Constant">3</span>
<span id="L317" class="LineNr">317 </span> <span class="Delimiter">}</span>
<span id="L318" class="LineNr">318 </span> <span class="PreProc">return</span> <span class="Constant">4</span>
<span id="L319" class="LineNr">319 </span><span class="Delimiter">}</span>
<span id="L320" class="LineNr">320 </span>
<span id="L321" class="LineNr">321 </span><span class="muComment"># needed because available primitives only shift by a literal/constant number of bits</span>
<span id="L322" class="LineNr">322 </span><span class="PreProc">fn</span> <span class="muFunction"><a href='403unicode.mu.html#L322'>shift-left-bytes</a></span> n: int, k: int<span class="PreProc"> -&gt; </span>_/eax: int <span class="Delimiter">{</span>
<span id="L323" class="LineNr">323 </span> <span class="PreProc">var</span> i/<span class="muRegEcx">ecx</span>: int <span class="Special">&lt;-</span> copy <span class="Constant">0</span>
<span id="L324" class="LineNr">324 </span> <span class="PreProc">var</span> result/eax: int <span class="Special">&lt;-</span> copy n
<span id="L325" class="LineNr">325 </span> <span class="Delimiter">{</span>
<span id="L326" class="LineNr">326 </span> compare i, k
<span id="L327" class="LineNr">327 </span> <span class="PreProc">break-if-&gt;=</span>
<span id="L328" class="LineNr">328 </span> compare i, <span class="Constant">4</span> <span class="muComment"># only 4 bytes in 32 bits</span>
<span id="L329" class="LineNr">329 </span> <span class="PreProc">break-if-&gt;=</span>
<span id="L330" class="LineNr">330 </span> result <span class="Special">&lt;-</span> shift-left <span class="Constant">8</span>
<span id="L331" class="LineNr">331 </span> i <span class="Special">&lt;-</span> increment
<span id="L332" class="LineNr">332 </span> <span class="PreProc">loop</span>
<span id="L333" class="LineNr">333 </span> <span class="Delimiter">}</span>
<span id="L334" class="LineNr">334 </span> <span class="PreProc">return</span> result
<span id="L335" class="LineNr">335 </span><span class="Delimiter">}</span>
<span id="L336" class="LineNr">336 </span>
<span id="L337" class="LineNr">337 </span><span class="muComment"># write a grapheme to a stream of bytes</span>
<span id="L338" class="LineNr">338 </span><span class="muComment"># this is like write-to-stream, except we skip leading 0 bytes</span>
<span id="L339" class="LineNr">339 </span><span class="PreProc">fn</span> <span class="muFunction"><a href='403unicode.mu.html#L339'>write-grapheme</a></span> out: (addr stream byte), g: grapheme <span class="Delimiter">{</span>
<span id="L340" class="LineNr">340 </span>$write-grapheme:body: <span class="Delimiter">{</span>
<span id="L341" class="LineNr">341 </span> <span class="PreProc">var</span> c/eax: int <span class="Special">&lt;-</span> copy g
<span id="L342" class="LineNr">342 </span> <a href='115write-byte.subx.html#L12'>append-byte</a> out, c <span class="muComment"># first byte is always written</span>
<span id="L343" class="LineNr">343 </span> c <span class="Special">&lt;-</span> shift-right <span class="Constant">8</span>
<span id="L344" class="LineNr">344 </span> compare c, <span class="Constant">0</span>
<span id="L345" class="LineNr">345 </span> <span class="PreProc">break-if-=</span> $write-grapheme:body
<span id="L346" class="LineNr">346 </span> <a href='115write-byte.subx.html#L12'>append-byte</a> out, c
<span id="L347" class="LineNr">347 </span> c <span class="Special">&lt;-</span> shift-right <span class="Constant">8</span>
<span id="L348" class="LineNr">348 </span> compare c, <span class="Constant">0</span>
<span id="L349" class="LineNr">349 </span> <span class="PreProc">break-if-=</span> $write-grapheme:body
<span id="L350" class="LineNr">350 </span> <a href='115write-byte.subx.html#L12'>append-byte</a> out, c
<span id="L351" class="LineNr">351 </span> c <span class="Special">&lt;-</span> shift-right <span class="Constant">8</span>
<span id="L352" class="LineNr">352 </span> compare c, <span class="Constant">0</span>
<span id="L353" class="LineNr">353 </span> <span class="PreProc">break-if-=</span> $write-grapheme:body
<span id="L354" class="LineNr">354 </span> <a href='115write-byte.subx.html#L12'>append-byte</a> out, c
<span id="L355" class="LineNr">355 </span><span class="Delimiter">}</span>
<span id="L356" class="LineNr">356 </span><span class="Delimiter">}</span>
</pre>
</body>
</html>
<!-- vim: set foldmethod=manual : -->