/[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 1.4 by wakaba, Sun Apr 20 12:19:13 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 {
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          log (dumpTree (doc, ''));
51        }
52      } // update2
53    
54      var logIndentLevel = 0;
55    function log (s) {    function log (s) {
56        for (var i = 0; i < logIndentLevel; i++) {
57          s = '  ' + s;
58        }
59      document.logElement.appendChild (document.createTextNode (s + "\n"));      document.logElement.appendChild (document.createTextNode (s + "\n"));
60    } // log    } // log
61    
# Line 42  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 77  Line 106 
106          return '';          return '';
107        });        });
108        if (token) return token;        if (token) return token;
109          var m;
110          if ((p.insertionPoint < '</script'.length) &&
111              (m = i.s.match (/^<\/([SCRIPTscript]+)/))) {
112            var v = m[1].substring (0, p.insertionPoint).toLowerCase ();
113            if (v == 'script'.substring (0, p.insertionPoint - '</'.length)) {
114              return {type: 'abort'};
115            }
116          }
117        i.s = i.s.replace (/^</,        i.s = i.s.replace (/^</,
118        function (s) {        function (s) {
119          token = {type: 'char', value: s};          token = {type: 'char', value: s};
# Line 88  Line 125 
125      }      }
126    
127      var token;      var token;
128      i.s = i.s.replace (/^<\/([^>]+)>/, function (s, e) {      i.s = i.s.replace (/^<\/([^>]+)(?:>|$)/, function (s, e) {
129        if (p.insertionPoint < s.length) {        if (p.insertionPoint < s.length ||
130              (p.insertionPoint <= s.length &&
131               s.substring (s.length - 1, 1) != '>')) {
132          token = {type: 'abort'};          token = {type: 'abort'};
133          return s;          return s;
134        }        }
# Line 98  Line 137 
137        return '';        return '';
138      });      });
139      if (token) return token;      if (token) return token;
140      i.s = i.s.replace (/^<([^>]+)>/, function (s, e) {      i.s = i.s.replace (/^<([^>]+)(?:>|$)/, function (s, e) {
141        if (p.insertionPoint < s.length) {        if (p.insertionPoint < s.length ||
142              (p.insertionPoint <= s.length &&
143               s.substring (s.length - 1, 1) != '>')) {
144          token = {type: 'abort'};          token = {type: 'abort'};
145          return s;          return s;
146        }        }
# Line 109  Line 150 
150          tagName = v.toLowerCase ();          tagName = v.toLowerCase ();
151          return '';          return '';
152        });        });
153        e = e.replace (/^\s*(\S+)\s*(?:=\s*"([^"]*)"|'([^']*)'|([^"']+))?/,        e = e.replace (/^\s*([^\s=]+)\s*(?:=\s*(?:"([^"]*)"|'([^']*)'|([^"']+)))?/,
154        function (x, attrName, attrValue1, attrValue2, attrValue3) {        function (x, attrName, attrValue1, attrValue2, attrValue3) {
155          attrs[attrName] = attrValue1 || attrValue2 || attrValue3;          v = attrValue1 || attrValue2 || attrValue3;
156            v = v.replace (/&quot;/g, '"').replace (/&apos;/g, "'")
157                .replace (/&amp;/g, '&');
158            attrs[attrName.toLowerCase ()] = v;
159          return '';          return '';
160        });        });
161          if (e.length) {
162            log ('Broken start tag: "' + e + '"');
163          }
164        token = {type: 'start-tag', value: tagName, attrs: attrs};        token = {type: 'start-tag', value: tagName, attrs: attrs};
165        p.insertionPoint -= s.length;        p.insertionPoint -= s.length;
166        return '';        return '';
# Line 144  Line 191 
191    } // getNextToken    } // getNextToken
192    
193    Parser.prototype.parse = function () {    Parser.prototype.parse = function () {
194      log ('start parsing');      logIndentLevel++;
195        log ('parse: start');
196    
197      while (true) {      while (true) {
198        var token = this.getNextToken ();        var token = this.getNextToken ();
# Line 212  Line 260 
260            this.openElements[this.openElements.length - 1].appendChild (el);            this.openElements[this.openElements.length - 1].appendChild (el);
261    
262            // 11. Let the insertion point have the value of the old ...            // 11. Let the insertion point have the value of the old ...
263    
264              oldInsertionPoint += this.insertionPoint;
265            this.setInsertionPoint (oldInsertionPoint);            this.setInsertionPoint (oldInsertionPoint);
266    
267            // 12. If there is a script that will execute as soon as ...            // 12. If there is a script that will execute as soon as ...
268                        while (this.scriptExecutedWhenParserResumes) {
269                // 12.1. If the tree construction stage is being called reentrantly
270                if (this.reentrant) {
271                  log ('parse: abort (reentrance)');
272                  logIndentLevel--;
273                  return;
274    
275                // 12.2. Otherwise
276                } else {
277                  // 1.
278                  var script = this.scriptExecutedWhenParserResumes;
279                  this.scriptExecutedWhenParserResumes = null;
280    
281                  // 2. Pause until the script has completed loading.
282                  //
283    
284                  // 3. Let the insertion point to just before the next input char.
285                  this.setInsertionPoint (0);
286    
287                  // 4. Execute the script.
288                  executeScript (this.doc, script);
289    
290                  // 5. Let the insertion point be undefined again.
291                  this.setInsertionPoint (undefined);
292    
293                  // 6. If there is once again a script that will execute ...
294                  //
295                }
296              }
297          } else {          } else {
298            var el = new JSElement (this.doc, token.value);            var el = new JSElement (this.doc, token.value);
299            this.openElements[this.openElements.length - 1].appendChild (el);            this.openElements[this.openElements.length - 1].appendChild (el);
# Line 236  Line 313 
313          break;          break;
314        } else if (token.type == 'abort') {        } else if (token.type == 'abort') {
315          log ('parse: abort');          log ('parse: abort');
316            logIndentLevel--;
317          return;          return;
318        }        }
319      }      }
# Line 270  Line 348 
348      // "delays tha load event" things has completed:      // "delays tha load event" things has completed:
349      // readyState = 'complete'      // readyState = 'complete'
350      log ('load event fired');      log ('load event fired');
351    
352        logIndentLevel--;
353    } // parse    } // parse
354    
355    Parser.prototype.setInsertionPoint = function (ip) {    Parser.prototype.setInsertionPoint = function (ip) {
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 303  Line 383 
383      e.parentNode = this;      e.parentNode = this;
384    
385      if (e.localName == 'script') {      if (e.localName == 'script') {
386          logIndentLevel++;
387        log ('Running a script: start');        log ('Running a script: start');
388    
389        var doc = this.ownerDocument || this;        var doc = this.ownerDocument || this;
# Line 321  Line 402 
402        if (e.manakaiAlreadyExecuted) {        if (e.manakaiAlreadyExecuted) {
403          // 2.5. Abort these steps at this point.          // 2.5. Abort these steps at this point.
404          log ('Running a script: aborted');          log ('Running a script: aborted');
405            logIndentLevel--;
406          return e;          return e;
407        }        }
408    
# Line 343  Line 425 
425                   /* && list of scripts that will execute asynchronously is not empty */) {                   /* && list of scripts that will execute asynchronously is not empty */) {
426          // TODO          // TODO
427        } else if (e.src != null && e.manakaiParserInserted) {        } else if (e.src != null && e.manakaiParserInserted) {
428          // TODO          if (p.scriptExecutedWhenParserResumes) {
429              log ('Error: There is a script that will execute as soon as the parser resumes.');
430            }
431            p.scriptExecutedWhenParserResumes = e;
432            log ('Running a script: aborted (src)');
433        } else if (e.src != null) {        } else if (e.src != null) {
434          // TODO          // TODO
435        } else {        } else {
# Line 351  Line 437 
437        }        }
438    
439        log ('Running a script: end');        log ('Running a script: end');
440          logIndentLevel--;
441      }      }
442    
443      return e;      return e;
# Line 359  Line 446 
446    function executeScript (doc, e) {    function executeScript (doc, e) {
447      log ('executing a script block: start');      log ('executing a script block: start');
448    
449      // If the load resulted in an error, then ... firing an error event ...      var s;
450        if (e.src != null) {
451          s = getExternalScript (e.src);
452    
453          // If the load resulted in an error, then ... firing an error event ...
454          if (s == null) {
455            log ('error event fired at the script element');
456            return;
457          }
458    
459          log ('External script loaded: "' + s + '"');
460        } else {
461          s = e.text;
462        }
463    
464      // If the load was successful      // If the load was successful
465      log ('load event fired at the script element');      log ('load event fired at the script element');
# Line 368  Line 468 
468      // Scripting is enabled, Document.designMode is disabled,      // Scripting is enabled, Document.designMode is disabled,
469      // Document is the active document in its browsing context      // Document is the active document in its browsing context
470    
       var s;  
       if (e.src != null) {  
         // TODO: from external file  
       } else {  
         s = e.text;  
       }  
   
471        parseAndRunScript (doc, s);        parseAndRunScript (doc, s);
472      }      }
473    
474      log ('executing a script block: end');      log ('executing a script block: end');
475    } // executeScript    } // executeScript
476    
477      function getExternalScript (uri) {
478        if (uri.match (/^javascript:/i)) {
479          var m;
480          if (m = uri.match (/^javascript:\s*(?:'([^']*)'|"([^"]+)")\s*$/i)) {
481            if (m[1]) {
482              return m[1];
483            } else if (m[2]) {
484              return m[2];
485            } else {
486              return null;
487            }
488          } else {
489            log ('Complex javascript: URI is not supported: <' + uri + '>');
490            return null;
491          }
492        } else {
493          log ('URI scheme not supported: <' + uri + '>');
494          return null;
495        }
496      } // getExternalScript
497    
498    function parseAndRunScript (doc, s) {    function parseAndRunScript (doc, s) {
499      while (true) {      while (true) {
500        var matched = false;        var matched = false;
# Line 429  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 463  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;
584    }; // document.open    }; // document.open
585    
586    JSDocument.prototype.write = function () {    JSDocument.prototype.write = function () {
587        logIndentLevel++;
588    
589      var p = this._parser;      var p = this._parser;
590    
591      // 1. If the insertion point is undefined, the open() method must be ...      // 1. If the insertion point is undefined, the open() method must be ...
# Line 481  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
607      // TODO      if (p.scriptExecutedAfterParserResumes) {
608          log ('document.write: processed later (there is an unprocessed <script src>)');
609          logIndentLevel--;
610          return;
611        }
612    
613      // 4. Process the characters that were inserted, ...      // 4. Process the characters that were inserted, ...
614        var originalReentrant = p.reentrant;
615        p.reentrant = true;
616      p.parse ();      p.parse ();
617        p.reentrant = originalReentrant;
618        // TODO: "Abort the processing of any nested invokations of the tokeniser,
619        // yielding control back to the caller." (<script> parsing).  Do we need
620        // to do something here?
621    
622      // 5. Return      // 5. Return
623      log ('document.write: return');      log ('document.write: return');
624    
625        logIndentLevel--;
626      return;      return;
627    }; // document.write    }; // document.write
628    
# Line 529  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 <abbr title="Hypertext Markup Language">HTML</abbr>
679    Parser</h1>
680    
681  <textarea onchange=" update () ">&lt;html>  <h2>Markup to test
682    (<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>
687    <textarea onkeydown=" update () " onchange=" update () " oninput=" update () ">&lt;html>
688  &lt;head>&lt;/head>&lt;body>  &lt;head>&lt;/head>&lt;body>
689  &lt;p>  &lt;p>
690  &lt;script>  &lt;script>
# Line 542  document.write ('aaaaaaa&lt;/p>&lt;scrip Line 693  document.write ('aaaaaaa&lt;/p>&lt;scrip
693  &lt;p>  &lt;p>
694  </textarea>  </textarea>
695    
696  <output></output>  <h2>Log</h2>
697    <p><output></output>
698    
699    <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    <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 -->
736    
737  </body>  </body>
738  </html>  </html>
Legend:
Removed from v.1.4  
changed lines
  Added in v.1.8
admin@suikawiki.org
 ViewVC Help
Powered by ViewVC 1.1.24