/[suikacvs]/markup/html/scripting-parser/parser.html
Suika

Diff of /markup/html/scripting-parser/parser.html

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1.7 by wakaba, Fri Apr 25 23:03:35 2008 UTC revision 1.8 by wakaba, Sun Apr 27 08:56:34 2008 UTC
# Line 1  Line 1 
1  <!DOCTYPE HTML>  <!DOCTYPE HTML>
2  <html lang=en>  <html lang=en>
3  <head>  <head>
4  <title>Demo of HTML5 Parsing Algorithm with Scripting Enabled</title>  <title>Live Scripting HTML Parser</title>
5  <style>  <style>
6    h1, h2 {    h1, h2 {
7      margin: 0;      margin: 0;
# Line 34  Line 34 
34    } // update    } // update
35    
36    function update2 () {    function update2 () {
     document.logElement.textContent = '';  
37      var v = document.sourceElement.value;      var v = document.sourceElement.value;
38      var p = new Parser (new InputStream (v));      if (v != document.previousSourceText) {
39      var doc = p.doc;        document.previousSourceText = v;
40      p.parse ();        document.links['permalink'].href
41      log (dumpTree (doc, ''));            = location.pathname + '?s=' + encodeURIComponent (v);
42          document.links['ldvlink'].href
43      document.links['permalink'].href            = 'http://software.hixie.ch/utilities/js/live-dom-viewer/?'
44          = location.href + '?s=' + encodeURIComponent (v);            + encodeURIComponent (v);
45    
46          document.logElement.textContent = '';
47          var p = new Parser (new InputStream (v));
48          var doc = p.doc;
49          p.parse ();
50          log (dumpTree (doc, ''));
51        }
52    } // update2    } // update2
53    
54    var logIndentLevel = 0;    var logIndentLevel = 0;
# Line 65  Line 71 
71      }      }
72      this.doc = doc;      this.doc = doc;
73      this.openElements = [doc];      this.openElements = [doc];
74      this.in = i;      this.input = i;
75      this.scriptsExecutedAfterParsing = [];      this.scriptsExecutedAfterParsing = [];
76    } // Parser    } // Parser
77    
78    Parser.prototype.getNextToken = function () {    Parser.prototype.getNextToken = function () {
79      var p = this;      var p = this;
80      var i = this.in;      var i = this.input;
81      if (this.parseMode == 'script') {      if (this.parseMode == 'script') {
82        var token;        var token;
83        if (p.insertionPoint <= 0) {        if (p.insertionPoint <= 0) {
# Line 350  Line 356 
356      if (ip == undefined || ip == null || isNaN (ip)) {      if (ip == undefined || ip == null || isNaN (ip)) {
357        log ('insertion point: set to undefined');        log ('insertion point: set to undefined');
358        this.insertionPoint = undefined;        this.insertionPoint = undefined;
359      } else if (ip == this.in.s.length) {      } else if (ip == this.input.s.length) {
360        log ('insertion point: end of file');        log ('insertion point: end of file');
361        this.insertionPoint = ip;        this.insertionPoint = ip;
362      } else {      } else {
363        log ('insertion point: set to ' + ip +        log ('insertion point: set to ' + ip +
364             ' (before "' + this.in.s.substring (0, 10) + '")');             ' (before "' + this.input.s.substring (0, 10) + '")');
365        this.insertionPoint = ip;        this.insertionPoint = ip;
366      }      }
367    }; // setInsertionPoint    }; // setInsertionPoint
# Line 537  Line 543 
543      // Step 3.      // Step 3.
544      if (this._parser &&      if (this._parser &&
545          !this._parser.scriptCreated &&          !this._parser.scriptCreated &&
546          this._parser.in.insertionPoint != undefined) {          this._parser.input.insertionPoint != undefined) {
547        log ('document.open () in parsing mode is ignored');        log ('document.open () in parsing mode is ignored');
548        return this;        return this;
549      }      }
# Line 571  Line 577 
577      }      }
578    
579      // Step 11.      // Step 11.
580      this._parser.setInsertionPoint (this._parser.in.s.length);      this._parser.setInsertionPoint (this._parser.input.s.length);
581    
582      // Step 12.      // Step 12.
583      return this;      return this;
# Line 591  Line 597 
597      // 2. ... inserted into the input stream just before the insertion point.      // 2. ... inserted into the input stream just before the insertion point.
598      var s = Array.join (arguments, '');      var s = Array.join (arguments, '');
599      log ('document.write: insert "' + s + '"' +      log ('document.write: insert "' + s + '"' +
600           ' before "' + p.in.s.substring (p.insertionPoint, p.insertionPoint + 10) + '"');           ' before "' +
601      p.in.s = p.in.s.substring (0, p.insertionPoint) + s           p.input.s.substring (p.insertionPoint, p.insertionPoint + 10) + '"');
602          + p.in.s.substring (p.insertionPoint, p.in.s.length);      p.input.s = p.input.s.substring (0, p.insertionPoint) + s
603            + p.input.s.substring (p.insertionPoint, p.input.s.length);
604      p.insertionPoint += s.length;      p.insertionPoint += s.length;
605    
606      // 3. If there is a script that will execute as soon as the parser resumes      // 3. If there is a script that will execute as soon as the parser resumes
# Line 651  Line 658 
658  </head>  </head>
659  <body onload="  <body onload="
660    document.sourceElement = document.getElementsByTagName ('textarea')[0];    document.sourceElement = document.getElementsByTagName ('textarea')[0];
661    
662      var q = location.search;
663      if (q != null) {
664        q = q.substring (1).split (/;/);
665        for (var i = 0; i < q.length; i++) {
666          var v = q[i].split (/=/, 2);
667          v[0] = decodeURIComponent (v[0]);
668          v[1] = decodeURIComponent (v[1] || '');
669          if (v[0] == 's') {
670            document.sourceElement.value = v[1];
671          }
672        }
673      }
674    
675    document.logElement = document.getElementsByTagName ('output')[0];    document.logElement = document.getElementsByTagName ('output')[0];
676    update ();    update ();
677  ">  ">
678  <h1>Live Scripting Parser</h1>  <h1>Live Scripting <abbr title="Hypertext Markup Language">HTML</abbr>
679    Parser</h1>
680    
681  <h2>Markup to test  <h2>Markup to test
682  (<a href=data:, id=permalink rel=bookmark>permalink</a>)</h2>  (<a href=data:, id=permalink rel=bookmark>permalink</a>,
683    <a href="http://software.hixie.ch/utilities/js/live-dom-viewer/"
684        id=ldvlink>Live <abbr title="Document Object Model">DOM</abbr>
685        Viewer</a>)</h2>
686  <p>  <p>
687  <textarea onkeydown=" update () " onchange=" update () " oninput=" update () ">&lt;html>  <textarea onkeydown=" update () " onchange=" update () " oninput=" update () ">&lt;html>
688  &lt;head>&lt;/head>&lt;body>  &lt;head>&lt;/head>&lt;body>
# Line 671  document.write ('aaaaaaa&lt;/p>&lt;scrip Line 696  document.write ('aaaaaaa&lt;/p>&lt;scrip
696  <h2>Log</h2>  <h2>Log</h2>
697  <p><output></output>  <p><output></output>
698    
699  <!-- TODO: short description -->  <h2>Note</h2>
700    
701    <p>This is a <em>simplified</em> implementation of
702    <a href="http://www.whatwg.org/specs/web-apps/current-work/#parsing">HTML5
703    Parsing Algorithm</a>.  It only implements script-related part of the
704    algorithm.  Especially, this parser:
705    <ul>
706    <li>Does not support <code>DOCTYPE</code> and comment tokens.
707    <li>Does not support entities except for <code>&amp;quot;</code>,
708    <code>&amp;apos;</code>, and <code>&amp;amp;</code> in <code>script</code>
709    <code>src</code> attribute value.
710    <li>Does not support omissions of start or end tags, the <abbr>AAA</abbr>
711    algorithm, and so on.
712    <li>Does not raise parse errors for invalid attribute specifications in start
713    or end tags.
714    <li>Does not support CDATA/PCDATA element other than <code>script</code>.
715    <li>Does not support <code>&lt;!--</code>..<code>--></code> parsing rule
716    in <code>script</code> element.
717    <li>Does not support foreign (SVG or MathML) elements.
718    <li>Only supports <code>script</code> <code>type</code>
719    <code>text/javascript</code>.  <code>type</code> and <code>language</code>
720    attributes are ignored.
721    <li>Only supports <code>document.write</code>.
722    The script code must be match to the regular expression
723    <code>^\s*(?:document\.write\s*\(<var>v</var>\s*(?:,\s*<var>v</var>\s*)*\)\s*;\s*)*$</code>
724    where <var>v</var> is <code>"[^"]*"|'[^']*'</code>.
725    <li>Only supports <code>javascript:</code>
726    <abbr title="Uniform Resourace Identifiers">URI</abbr> scheme in the
727    <code>src</code> attribute of the <code>script</code> element.  In addition,
728    the <abbr title="Uniform Resource Identifiers">URI</abbr> must be conform to
729    the regular expression <code>^javascript:\s*(?:"[^"]*"|'[^']*')\s*$</code>.
730    </ul>
731    
732  <!-- TODO: permalink query -> textarea -->  <p>For some reason, this parser does not work in browsers that do
733    not support JavaScript 1.5.
734    
735  <!-- TODO: multiple attributes are not supported yet -->  <!-- TODO: multiple attributes are not supported yet -->
736    
Legend:
Removed from v.1.7  
changed lines
  Added in v.1.8
admin@suikawiki.org
 ViewVC Help
Powered by ViewVC 1.1.24