Fix the Khmer font issue.

Update the font subset script. List the characters needed in the document. TODO: generate the non-Khmer subset font
w3c · Jan 23, 2019 · 9cbaf56 · 9cbaf56
1 parent 8c1b3e8
commit 9cbaf56
Show file tree

Hide file tree

Showing 4 changed files with 32 additions and 14 deletions.
diff --git a/NotoSansKhmer-Regular.woff b/NotoSansKhmer-Regular.woff
diff --git a/SubsetFont.user.js b/SubsetFont.user.js
@@ -7,14 +7,14 @@
 // @grant       none
 // ==/UserScript==
 
-// author = addison@lab126.com
+// author = addison@amazon.com
 
 var text = document.body.innerHTML;
 
 var charsToGet = [];
 var request = '';
 
-var re = new RegExp("[\u0100-\uffff]", 'g');
+var re = new RegExp("([\u0100-\ud7ff\ue000-\uffff]|[\ud800-\udbff][\udc00-\udfff])", 'g');
 
 var unique = 0;
 
@@ -23,13 +23,14 @@ while ((m = re.exec(text)) !== null) {
   var item = m[0];
   if (charsToGet.length === 0 || charsToGet.indexOf(item) < 0) {
     charsToGet.push(item);
-    request = request + item;
     unique++;
   }
 }
+charsToGet.sort();
 
-
-//alert(re.test(text) + ' ' + re + ' ' + text.substring(0,100));
+for (var i = 0; i < charsToGet.length; i++) {
+	request = request + charsToGet[i];
+}
 
 var mypanel = document.getElementsByClassName('subsetFont')[0];
 
@@ -40,12 +41,14 @@ mypanel.innerHTML = '';
 
 mypanel.setAttribute('id', 'subsetFont');
 mypanel.setAttribute('class', "subsetFont");
-mypanel.style.position = 'fixed';
-mypanel.style.bottom   = 0;
-mypanel.style.right    = 0;
-mypanel.style.overflow = 'auto';
-mypanel.style.inset    = '0 0 0 10px';
-mypanel.style.zIndex   = '1001'; // make it appear in front of most things
+mypanel.style.position   = 'fixed';
+mypanel.style.bottom     = 0;
+mypanel.style.right      = 0;
+mypanel.style.background = 'white';
+mypanel.style.border     = '1px solid black';
+mypanel.style.overflow   = 'auto';
+mypanel.style.inset      = '0 0 0 10px';
+mypanel.style.zIndex     = '1001'; // make it appear in front of most things
 
 var pre = document.createElement('pre');
 pre.appendChild(document.createTextNode('count: ' + unique));
@@ -54,4 +57,4 @@ pre = document.createElement('pre');
 pre.appendChild(document.createTextNode(request));
 mypanel.appendChild(pre);
 
-document.body.appendChild(mypanel);
+document.body.appendChild(mypanel);
diff --git a/index.html b/index.html
@@ -2,6 +2,14 @@
 <html dir="ltr" lang="en">
   <head>
     <meta charset="UTF-8">
+
+    <!--
+    Editor's note: the current list of Unicode characters used in this document is:
+    
+    U+0130 U+0131 U+0133 U+0141 U+017E U+01C4 U+01C5 U+01C6 U+01FA U+0300 U+0301 U+0307 U+030A U+0323 U+0327 U+0342 U+0398 U+03A1 U+03A9 U+03B1 U+03B8 U+03B9 U+03C9 U+0414 U+0420 U+0434 U+0627 U+0628 U+062A U+0646 U+0647 U+0654 U+08A1 U+0915 U+0921 U+0928 U+092F U+093F U+0942 U+094B U+1100 U+1161 U+1780 U+178A U+178F U+17D2 U+1E9E U+1F23 U+1F7C U+1F93 U+1F9B U+1FB6 U+1FB7 U+1FF2 U+200B U+200C U+200D U+2014 U+2019 U+2044 U+2079 U+2089 U+208A U+20AC U+210C U+210D U+2126 U+212B U+215F U+2189 U+21D2 U+2260 U+2460 U+2469 U+2474 U+2488 U+24B6 U+25CC U+2EF3 U+3002 U+30A2 U+30AB U+30C8 U+30D1 U+30E6 U+30F3 U+30FC U+3250 U+329E U+3300 U+3350 U+3389 U+3392 U+5370 U+9F9F U+AC00 U+1F1F2 U+1F1FF U+1F3F4 U+1F3FB U+1F3FC U+1F3FD U+1F3FE U+1F3FF U+1F467 U+1F468 U+1F469 U+1F46A U+E0062 U+E0063 U+E0067 U+E0073 U+E0074 U+E007F U+FE37 U+FEE9 U+FEEA U+FEEB U+FEEC U+FF21 U+FF5B U+FF76
+    -->
+
+
     <title>Character Model for the World Wide Web: String Matching</title>
     <link rel="canonical" href="http://www.w3.org/TR/2015/WD-charmod-norm-20151119/"/>
     <!-- local styles. Includes the styles from http://www.w3.org/International/docs/styleguide -->
@@ -11,7 +19,7 @@
       var respecConfig = {
           // specification status (e.g. WD, LCWD, NOTE, etc.). If in doubt use ED.
           specStatus:				"ED",
-          publishDate:  			"2019-01-19",
+          publishDate:  			"2019-01-23",
           previousPublishDate:  	"2015-11-19",
           previousMaturity:  		"WD",
 

diff --git a/local.css b/local.css
@@ -6,6 +6,13 @@
   src: url('W3CCharmodNormSubset-Regular.woff2') format('woff2');
 }
 
+@font-face {
+   font-family: Khmer;
+   font-style: normal;
+   font-weight: 400;
+   src: url('NotoSansKhmer-Regular.woff') format('woff');
+}
+
 body {
 	counter-reset: requirement;
 }
@@ -207,7 +214,7 @@ div.exampleBox {
 
 .exampleChar {
     font-size: 36pt;
-    font-family: "NoToFu", "Palatino", "Times", "Times New Roman", "Verdana", serif;
+    font-family: "Khmer", "NoToFu", "Palatino", "Times", "Times New Roman", "Verdana", serif;
 }
 
 .bidiExample {