/[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 1.6 by wakaba, Fri Apr 25 13:42:51 2008 UTC revision 1.10 by wakaba, Sun Apr 27 10:34:18 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 {
7        margin: 0;
8        font-size: 100%;
9      }
10      p, pre {
11        margin: 0;
12      }
13    textarea {    textarea {
14       display: block;      width: 100%;
15       width: 80%;      -width: 99%;
16       margin-left: auto;      height: 10em;
      margin-right: auto;  
      min-height: 20em;  
17    }    }
18    output {    output {
19      display: block;      display: block;
# Line 18  Line 23 
23    }    }
24  </style>  </style>
25  <script>  <script>
26      var delayedUpdater = 0;
27    
28    function update () {    function update () {
29      document.logElement.textContent = '';      if (delayedUpdater) {
30      var p = new Parser (new InputStream (document.sourceElement.value));        clearTimeout (delayedUpdater);
31      var doc = p.doc;        delayedUpdater = 0;
32      p.parse ();      }
33      log (dumpTree (doc, ''));      delayedUpdater = setTimeout (update2, 100);
34    } // update    } // update
35    
36      function update2 () {
37        var v = document.sourceElement.value;
38        if (v != document.previousSourceText) {
39          document.previousSourceText = v;
40          document.links['permalink'].href
41              = location.pathname + '?s=' + encodeURIComponent (v);
42          document.links['ldvlink'].href
43              = 'http://software.hixie.ch/utilities/js/live-dom-viewer/?'
44              + encodeURIComponent (v);
45    
46          document.logElement.textContent = '';
47          var p = new Parser (new InputStream (v));
48          var doc = p.doc;
49          p.parse ();
50          
51          log (dumpTree (doc, ''));
52          
53          if (p.hasAsyncScript) {
54            log ('Some script codes are executed asynchronously; it means that the document might be rendered in different ways depending on the network condition and other factors');
55          }
56        }
57      } // update2
58    
59    var logIndentLevel = 0;    var logIndentLevel = 0;
60    function log (s) {    function log (s) {
61      for (var i = 0; i < logIndentLevel; i++) {      for (var i = 0; i < logIndentLevel; i++) {
# Line 46  Line 76 
76      }      }
77      this.doc = doc;      this.doc = doc;
78      this.openElements = [doc];      this.openElements = [doc];
79      this.in = i;      this.input = i;
80      this.scriptsExecutedAfterParsing = [];      this.scriptsExecutedAfterParsing = [];
81        this.scriptsExecutedSoon = [];
82    } // Parser    } // Parser
83    
84    Parser.prototype.getNextToken = function () {    Parser.prototype.getNextToken = function () {
85      var p = this;      var p = this;
86      var i = this.in;      var i = this.input;
87      if (this.parseMode == 'script') {      if (this.parseMode == 'script') {
88        var token;        var token;
89        if (p.insertionPoint <= 0) {        if (p.insertionPoint <= 0) {
# Line 125  Line 156 
156          tagName = v.toLowerCase ();          tagName = v.toLowerCase ();
157          return '';          return '';
158        });        });
159        e = e.replace (/^\s*([^\s=]+)\s*(?:=\s*(?:"([^"]*)"|'([^']*)'|([^"']+)))?/,        while (true) {
160        function (x, attrName, attrValue1, attrValue2, attrValue3) {          var m = false;
161          v = attrValue1 || attrValue2 || attrValue3;          e = e.replace (/^\s*([^\s=]+)\s*(?:=\s*(?:"([^"]*)"|'([^']*)'|([^"'\s]*)))?/,
162          v = v.replace (/&quot;/g, '"').replace (/&apos;/g, "'")          function (x, attrName, attrValue1, attrValue2, attrValue3) {
163              .replace (/&amp;/g, '&');            v = attrValue1 || attrValue2 || attrValue3;
164          attrs[attrName.toLowerCase ()] = v;            v = v.replace (/&quot;/g, '"').replace (/&apos;/g, "'")
165          return '';                .replace (/&amp;/g, '&');
166        });            attrs[attrName.toLowerCase ()] = v;
167              m = true;
168              return '';
169            });
170            if (!m) break;
171          }
172        if (e.length) {        if (e.length) {
173          log ('Broken start tag: "' + e + '"');          log ('Broken start tag: "' + e + '"');
174        }        }
# Line 235  Line 271 
271            this.openElements[this.openElements.length - 1].appendChild (el);            this.openElements[this.openElements.length - 1].appendChild (el);
272    
273            // 11. Let the insertion point have the value of the old ...            // 11. Let the insertion point have the value of the old ...
274    
275            oldInsertionPoint += this.insertionPoint;            oldInsertionPoint += this.insertionPoint;
276            this.setInsertionPoint (oldInsertionPoint);            this.setInsertionPoint (oldInsertionPoint);
277    
# Line 298  Line 335 
335    
336      // "When a script completes loading" rules start applying.      // "When a script completes loading" rules start applying.
337    
338      // TODO: Handles "list of scripts that will execute as soon as possible"      // List of scripts that will execute as soon as possible
339      // and "list of scripts that will execute asynchronously"      for (var i = 0; i < this.scriptsExecutedSoon.length; i++) {
340          var e = this.scriptsExecutedSoon[i];
341    
342          // If it has completed loading
343          log ('Execute an external script not inserted by parser...');
344          executeScript (this.doc, e);
345    
346          // NOTE: It MAY be executed before the end of the parsing, according
347          // to the spec.
348          this.hasAsyncScript = true;
349        }
350    
351        // TODO: Handles
352        // "list of scripts that will execute asynchronously"
353    
354      // Handle "list of scripts that will execute when the document has finished      // Handle "list of scripts that will execute when the document has finished
355      // parsing".      // parsing".
# Line 330  Line 380 
380      if (ip == undefined || ip == null || isNaN (ip)) {      if (ip == undefined || ip == null || isNaN (ip)) {
381        log ('insertion point: set to undefined');        log ('insertion point: set to undefined');
382        this.insertionPoint = undefined;        this.insertionPoint = undefined;
383      } else if (ip == this.in.s.length) {      } else if (ip == this.input.s.length) {
384        log ('insertion point: end of file');        log ('insertion point: end of file');
385        this.insertionPoint = ip;        this.insertionPoint = ip;
386      } else {      } else {
387        log ('insertion point: set to ' + ip +        log ('insertion point: set to ' + ip +
388             ' (before "' + this.in.s.substring (0, 10) + '")');             ' (before "' + this.input.s.substring (0, 10) + '")');
389        this.insertionPoint = ip;        this.insertionPoint = ip;
390      }      }
391    }; // setInsertionPoint    }; // setInsertionPoint
# Line 403  Line 453 
453            log ('Error: There is a script that will execute as soon as the parser resumes.');            log ('Error: There is a script that will execute as soon as the parser resumes.');
454          }          }
455          p.scriptExecutedWhenParserResumes = e;          p.scriptExecutedWhenParserResumes = e;
456          log ('Running a script: aborted (src)');          log ('Running a script: aborted (src parser-inserted)');
457        } else if (e.src != null) {        } else if (e.src != null) {
458          // TODO          p.scriptsExecutedSoon.push (e);
459            log ('Running a script: aborted (src)');
460        } else {        } else {
461          executeScript (doc, e); // even if other scripts are already executing.          executeScript (doc, e); // even if other scripts are already executing.
462        }        }
# Line 453  Line 504 
504        var m;        var m;
505        if (m = uri.match (/^javascript:\s*(?:'([^']*)'|"([^"]+)")\s*$/i)) {        if (m = uri.match (/^javascript:\s*(?:'([^']*)'|"([^"]+)")\s*$/i)) {
506          if (m[1]) {          if (m[1]) {
507            return m[1];            return m[1].replace (/\\u([0-9A-F]{4})/g, function (s, v) {
508                return String.fromCharCode (parseInt ('0x' + v));
509              });
510          } else if (m[2]) {          } else if (m[2]) {
511            return m[2];            return m[2].replace (/\\u([0-9A-F]{4})/g, function (s, v) {
512                return String.fromCharCode (parseInt ('0x' + v));
513              });
514          } else {          } else {
515            return null;            return null;
516          }          }
# Line 482  Line 537 
537          doc.write.apply (doc, args);          doc.write.apply (doc, args);
538          return '';          return '';
539        });        });
540          s = s.replace (/^\s*var\s+s\s*=\s*document\.createElement\s*\(\s*['"]script['"]\s*\)\s*;\s*s\.src\s*=\s*(?:'(javascript:[^']*)'|"(javascript:[^"]*)")\s*;\s*document\.documentElement\.appendChild\s*\(\s*s\s*\)\s*;\s*/,
541          function (s, t, u) {
542            matched = true;
543            var args = [t ? t : u];
544            doc._insertExternalScript.apply (doc, args);
545            return '';
546          });
547        if (s == '') break;        if (s == '') break;
548        if (!matched) {        if (!matched) {
549          log ('Script parse error: "' + s + '"');          log ('Script parse error: "' + s + '"');
# Line 517  Line 579 
579      // Step 3.      // Step 3.
580      if (this._parser &&      if (this._parser &&
581          !this._parser.scriptCreated &&          !this._parser.scriptCreated &&
582          this._parser.in.insertionPoint != undefined) {          this._parser.input.insertionPoint != undefined) {
583        log ('document.open () in parsing mode is ignored');        log ('document.open () in parsing mode is ignored');
584        return this;        return this;
585      }      }
# Line 551  Line 613 
613      }      }
614    
615      // Step 11.      // Step 11.
616      this._parser.setInsertionPoint (this._parser.in.s.length);      this._parser.setInsertionPoint (this._parser.input.s.length);
617    
618      // Step 12.      // Step 12.
619      return this;      return this;
# Line 571  Line 633 
633      // 2. ... inserted into the input stream just before the insertion point.      // 2. ... inserted into the input stream just before the insertion point.
634      var s = Array.join (arguments, '');      var s = Array.join (arguments, '');
635      log ('document.write: insert "' + s + '"' +      log ('document.write: insert "' + s + '"' +
636           ' before "' + p.in.s.substring (p.insertionPoint, p.insertionPoint + 10) + '"');           ' before "' +
637      p.in.s = p.in.s.substring (0, p.insertionPoint) + s           p.input.s.substring (p.insertionPoint, p.insertionPoint + 10) + '"');
638          + p.in.s.substring (p.insertionPoint, p.in.s.length);      p.input.s = p.input.s.substring (0, p.insertionPoint) + s
639            + p.input.s.substring (p.insertionPoint, p.input.s.length);
640      p.insertionPoint += s.length;      p.insertionPoint += s.length;
641    
642      // 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 599  Line 662 
662      return;      return;
663    }; // document.write    }; // document.write
664    
665      JSDocument.prototype._insertExternalScript = function (uri) {
666        var s = new JSElement (this, 'script');
667        s.src = uri;
668        this.documentElement.appendChild (s);
669      }; // _insertExternalScript
670    
671      JSDocument.prototype.__defineGetter__ ('documentElement', function () {
672        var cn = this.childNodes;
673        for (var i = 0; i < cn.length; i++) {
674          if (cn[i] instanceof JSElement) {
675            return cn[i]
676          }
677        }
678        return null;
679      });
680    
681    JSElement.prototype.__defineGetter__ ('text', function () {    JSElement.prototype.__defineGetter__ ('text', function () {
682      var r = '';      var r = '';
683      for (var i = 0; i < this.childNodes.length; i++) {      for (var i = 0; i < this.childNodes.length; i++) {
# Line 617  Line 696 
696          r += '| ' + indent + node.localName + '\n';          r += '| ' + indent + node.localName + '\n';
697          if (node.async) r += '| ' + indent + '  async=""\n';          if (node.async) r += '| ' + indent + '  async=""\n';
698          if (node.defer) r += '| ' + indent + '  defer=""\n';          if (node.defer) r += '| ' + indent + '  defer=""\n';
699          if (node.src) r += '| ' + indent + '  src="' + node.src + '"\n';          if (node.src != null) {
700              r += '| ' + indent + '  src="' + node.src + '"\n';
701            }
702          r += dumpTree (node, indent + '  ');          r += dumpTree (node, indent + '  ');
703        } else if (node instanceof JSText) {        } else if (node instanceof JSText) {
704          r += '| ' + indent + '"' + node.data + '"\n';          r += '| ' + indent + '"' + node.data + '"\n';
# Line 631  Line 712 
712  </head>  </head>
713  <body onload="  <body onload="
714    document.sourceElement = document.getElementsByTagName ('textarea')[0];    document.sourceElement = document.getElementsByTagName ('textarea')[0];
715    
716      var q = location.search;
717      if (q != null) {
718        q = q.substring (1).split (/;/);
719        for (var i = 0; i < q.length; i++) {
720          var v = q[i].split (/=/, 2);
721          v[0] = decodeURIComponent (v[0]);
722          v[1] = decodeURIComponent (v[1] || '');
723          if (v[0] == 's') {
724            document.sourceElement.value = v[1];
725          }
726        }
727      }
728    
729    document.logElement = document.getElementsByTagName ('output')[0];    document.logElement = document.getElementsByTagName ('output')[0];
730    update ();    update ();
731  ">  ">
732    <h1>Live Scripting <abbr title="Hypertext Markup Language">HTML</abbr>
733    Parser</h1>
734    
735  <textarea onchange=" update () ">&lt;html>  <h2>Markup to test
736    (<a href=data:, id=permalink rel=bookmark>permalink</a>,
737    <a href="http://software.hixie.ch/utilities/js/live-dom-viewer/"
738        id=ldvlink>Live <abbr title="Document Object Model">DOM</abbr>
739        Viewer</a>)</h2>
740    <p>
741    <textarea onkeydown=" update () " onchange=" update () " oninput=" update () ">&lt;html>
742  &lt;head>&lt;/head>&lt;body>  &lt;head>&lt;/head>&lt;body>
743  &lt;p>  &lt;p>
744  &lt;script>  &lt;script>
# Line 644  document.write ('aaaaaaa&lt;/p>&lt;scrip Line 747  document.write ('aaaaaaa&lt;/p>&lt;scrip
747  &lt;p>  &lt;p>
748  </textarea>  </textarea>
749    
750  <output></output>  <h2 id=log>Log</h2>
751    <p><output></output>
752    
753    <h2 id=notes>Notes</h2>
754    
755    <p>This is a <em>simplified</em> implementation of
756    <a href="http://www.whatwg.org/specs/web-apps/current-work/#parsing">HTML5
757    Parsing Algorithm</a>.  It only implements script-related part of the
758    algorithm.  Especially, this parser:
759    <ul>
760    <li>Does not support <code>DOCTYPE</code> and comment tokens.
761    <li>Does not support entities except for <code>&amp;quot;</code>,
762    <code>&amp;apos;</code>, and <code>&amp;amp;</code> in <code>script</code>
763    <code>src</code> attribute value.
764    <li>Does not support omissions of start or end tags, the <abbr>AAA</abbr>
765    algorithm, and so on.
766    <li>Does not raise parse errors for invalid attribute specifications in start
767    or end tags.
768    <li>Does not support CDATA/PCDATA element other than <code>script</code>.
769    <li>Does not support <code>&lt;!--</code>..<code>--></code> parsing rule
770    in <code>script</code> element.
771    <li>Does not support foreign (SVG or MathML) elements.
772    <li>Only supports <code>script</code> <code>type</code>
773    <code>text/javascript</code>.  <code>type</code> and <code>language</code>
774    attributes are ignored.
775    <li>Only supports limited statements.  It must consist of zero or more
776    of statements looking similar to the following statements, possibly
777    introduced, followed, or separated by white space characters:
778      <ul>
779      <li><code>document.write ("<var>string</var>", ["<var>string</var>", ...]);</code>.
780      <li><code>var s = document.createElement ("script");
781                s.src = "<var>string</var>";
782                document.documentElement.appendChild (s);</code>
783      </ul>
784    Note that strings may be delimited by <code>'</code>s instead of
785    <code>"</code>s.
786    <li>Only supports <code>javascript:</code>
787    <abbr title="Uniform Resourace Identifiers">URI</abbr> scheme in the
788    <code>src</code> attribute of the <code>script</code> element.  In addition,
789    the <abbr title="Uniform Resource Identifiers">URI</abbr> must be conform to
790    the regular expression <code>^javascript:\s*(?:"[^"]*"|'[^']*')\s*$</code>.
791    <li>Only supports <code>\u<var>HHHH</var></code> escapes only in
792    <code>javascript:</code> URI.
793    </ul>
794    
795    <p>For some reason, this parser does not work in browsers that do
796    not support JavaScript 1.5.
797    
798    <!-- TODO: license -->
799    
800  </body>  </body>
801  </html>  </html>
Legend:
Removed from v.1.6  
changed lines
  Added in v.1.10
admin@suikawiki.org
 ViewVC Help
Powered by ViewVC 1.1.24