/[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 1.9 by wakaba, Sun Apr 27 09:16:11 2008 UTC revision 1.13 by wakaba, Sun Apr 27 11:27:04 2008 UTC
# Line 2  Line 2 
2  <html lang=en>  <html lang=en>
3  <head>  <head>
4  <title>Live Scripting HTML Parser</title>  <title>Live Scripting HTML Parser</title>
5    <link rel=author href="http://suika.fam.cx/~wakaba/who?">
6    <link rel=license href="http://suika.fam.cx/c/gnu/gpl"
7        title="GNU GPL2 or later">
8  <style>  <style>
9    h1, h2 {    h1 {
10        margin: 0;
11        font-size: 150%;
12      }
13      h2 {
14      margin: 0;      margin: 0;
15      font-size: 100%;      font-size: 100%;
16    }    }
17    p, pre {    p {
18      margin: 0;      margin: 0 1em;
19    }    }
20    textarea {    textarea {
21      width: 100%;      width: 100%;
# Line 47  Line 54 
54        var p = new Parser (new InputStream (v));        var p = new Parser (new InputStream (v));
55        var doc = p.doc;        var doc = p.doc;
56        p.parse ();        p.parse ();
57          
58        log (dumpTree (doc, ''));        log (dumpTree (doc, ''));
59          
60          if (p.hasAsyncScript) {
61            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');
62          }
63      }      }
64    } // update2    } // update2
65    
# Line 73  Line 85 
85      this.openElements = [doc];      this.openElements = [doc];
86      this.input = i;      this.input = i;
87      this.scriptsExecutedAfterParsing = [];      this.scriptsExecutedAfterParsing = [];
88        this.scriptsExecutedSoon = [];
89        this.scriptsExecutedAsynchronously = [];
90    } // Parser    } // Parser
91    
92    Parser.prototype.getNextToken = function () {    Parser.prototype.getNextToken = function () {
# Line 329  Line 343 
343    
344      // "When a script completes loading" rules start applying.      // "When a script completes loading" rules start applying.
345    
346      // TODO: Handles "list of scripts that will execute as soon as possible"      while (this.scriptsExecutedSoon.length > 0 ||
347      // and "list of scripts that will execute asynchronously"             this.scriptsExecutedAsynchronously.length > 0) {
348          // Handle "list of scripts that will execute as soon as possible".
349          while (this.scriptsExecutedSoon.length > 0) {
350            var e = this.scriptsExecutedSoon.shift ();
351      
352            // If it has completed loading
353            log ('Execute an external script not inserted by parser...');
354            executeScript (this.doc, e);
355    
356            // NOTE: It MAY be executed before the end of the parsing, according
357            // to the spec.
358            this.hasAsyncScript = true;
359          }
360    
361          // Handle "list of scripts that will execute asynchronously".
362          while (this.scriptsExecutedAsynchronously.length > 0) {
363            var e = this.scriptsExecutedAsynchronously.shift ();
364    
365            // Step 1.
366            // We assume that all scripts have been loaded at this time.
367      
368            // Step 2.
369            log ('Execute an asynchronous script...');
370            executeScript (this.doc, e);
371    
372            // Step 3.
373            //
374    
375            // Step 4.
376            //
377    
378            this.hasAsyncScript = true;
379          }
380        }
381    
382      // Handle "list of scripts that will execute when the document has finished      // Handle "list of scripts that will execute when the document has finished
383      // parsing".      // parsing".
# Line 425  Line 472 
472          p.scriptsExecutedAfterParsing.push (e);          p.scriptsExecutedAfterParsing.push (e);
473          log ('Running a script: aborted (defer)');          log ('Running a script: aborted (defer)');
474        } else if (e.async && e.src != null) {        } else if (e.async && e.src != null) {
475          // TODO          p.scriptsExecutedAsynchronously.push (e);
476        } else if (e.async && e.src == null          log ('Running a script: aborted (async src)');
477                   /* && list of scripts that will execute asynchronously is not empty */) {        } else if (e.async && e.src == null &&
478          // TODO                   p.scriptsExecutedAsynchronously.length > 0) {
479            p.scriptsExecutedAsynchronously.push (e);
480            log ('Running a script: aborted (async)');
481            // ISSUE: What is the difference with the case above?
482        } else if (e.src != null && e.manakaiParserInserted) {        } else if (e.src != null && e.manakaiParserInserted) {
483          if (p.scriptExecutedWhenParserResumes) {          if (p.scriptExecutedWhenParserResumes) {
484            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.');
485          }          }
486          p.scriptExecutedWhenParserResumes = e;          p.scriptExecutedWhenParserResumes = e;
487          log ('Running a script: aborted (src)');          log ('Running a script: aborted (src parser-inserted)');
488        } else if (e.src != null) {        } else if (e.src != null) {
489          // TODO          p.scriptsExecutedSoon.push (e);
490            log ('Running a script: aborted (src)');
491        } else {        } else {
492          executeScript (doc, e); // even if other scripts are already executing.          executeScript (doc, e); // even if other scripts are already executing.
493        }        }
# Line 484  Line 535 
535        var m;        var m;
536        if (m = uri.match (/^javascript:\s*(?:'([^']*)'|"([^"]+)")\s*$/i)) {        if (m = uri.match (/^javascript:\s*(?:'([^']*)'|"([^"]+)")\s*$/i)) {
537          if (m[1]) {          if (m[1]) {
538            return m[1];            return unescapeJSLiteral (m[1]);
539          } else if (m[2]) {          } else if (m[2]) {
540            return m[2];            return unescapeJSLiteral (m[2]);
541          } else {          } else {
542            return null;            return null;
543          }          }
# Line 507  Line 558 
558          matched = true;          matched = true;
559          var args = [];          var args = [];
560          t.replace (/('[^']*'|"[^"]*")/g, function (s, v) {          t.replace (/('[^']*'|"[^"]*")/g, function (s, v) {
561            args.push (v.substring (1, v.length - 1));            args.push (unescapeJSLiteral (v.substring (1, v.length - 1)));
562            return '';            return '';
563          });          });
564          doc.write.apply (doc, args);          doc.write.apply (doc, args);
565          return '';          return '';
566        });        });
567          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*/,
568          function (s, t, u) {
569            matched = true;
570            var args = [unescapeJSLiteral (t ? t : u)];
571            doc._insertExternalScript.apply (doc, args);
572            return '';
573          });
574        if (s == '') break;        if (s == '') break;
575        if (!matched) {        if (!matched) {
576          log ('Script parse error: "' + s + '"');          log ('Script parse error: "' + s + '"');
# Line 521  Line 579 
579      }      }
580    } // parseAndRunScript    } // parseAndRunScript
581    
582      function unescapeJSLiteral (s) {
583        return s.replace (/\\u([0-9A-Fa-f]{4})/g, function (t, v) {
584          return String.fromCharCode (parseInt ('0x' + v));
585        });
586      } // unescapeJSLiteral
587    
588    function JSText (data) {    function JSText (data) {
589      this.data = data;      this.data = data;
590    } // JSText    } // JSText
# Line 631  Line 695 
695      return;      return;
696    }; // document.write    }; // document.write
697    
698      JSDocument.prototype._insertExternalScript = function (uri) {
699        var s = new JSElement (this, 'script');
700        s.src = uri;
701        this.documentElement.appendChild (s);
702      }; // _insertExternalScript
703    
704      JSDocument.prototype.__defineGetter__ ('documentElement', function () {
705        var cn = this.childNodes;
706        for (var i = 0; i < cn.length; i++) {
707          if (cn[i] instanceof JSElement) {
708            return cn[i]
709          }
710        }
711        return null;
712      });
713    
714    JSElement.prototype.__defineGetter__ ('text', function () {    JSElement.prototype.__defineGetter__ ('text', function () {
715      var r = '';      var r = '';
716      for (var i = 0; i < this.childNodes.length; i++) {      for (var i = 0; i < this.childNodes.length; i++) {
# Line 700  document.write ('aaaaaaa&lt;/p>&lt;scrip Line 780  document.write ('aaaaaaa&lt;/p>&lt;scrip
780  &lt;p>  &lt;p>
781  </textarea>  </textarea>
782    
783  <h2>Log</h2>  <h2 id=log>Log</h2>
784  <p><output></output>  <p><output></output>
785    
786  <h2>Note</h2>  <h2 id=notes>Notes</h2>
787    
788  <p>This is a <em>simplified</em> implementation of  <p>This is a <em>simplified</em> implementation of
789  <a href="http://www.whatwg.org/specs/web-apps/current-work/#parsing">HTML5  <a href="http://www.whatwg.org/specs/web-apps/current-work/#parsing">HTML5
# Line 725  in <code>script</code> element. Line 805  in <code>script</code> element.
805  <li>Only supports <code>script</code> <code>type</code>  <li>Only supports <code>script</code> <code>type</code>
806  <code>text/javascript</code>.  <code>type</code> and <code>language</code>  <code>text/javascript</code>.  <code>type</code> and <code>language</code>
807  attributes are ignored.  attributes are ignored.
808  <li>Only supports <code>document.write</code>.  <li>Only supports limited statements.  It must consist of zero or more
809  The script code must be match to the regular expression  of statements looking similar to the following statements, possibly
810  <code>^\s*(?:document\.write\s*\(<var>v</var>\s*(?:,\s*<var>v</var>\s*)*\)\s*;\s*)*$</code>  introduced, followed, or separated by white space characters:
811  where <var>v</var> is <code>"[^"]*"|'[^']*'</code>.    <ul>
812      <li><code>document.write ("<var>string</var>", ["<var>string</var>", ...]);</code>.
813      <li><code>var s = document.createElement ("script");
814                s.src = "<var>string</var>";
815                document.documentElement.appendChild (s);</code>
816      </ul>
817    Note that strings may be delimited by <code>'</code>s instead of
818    <code>"</code>s.
819  <li>Only supports <code>javascript:</code>  <li>Only supports <code>javascript:</code>
820  <abbr title="Uniform Resourace Identifiers">URI</abbr> scheme in the  <abbr title="Uniform Resourace Identifiers">URI</abbr> scheme in the
821  <code>src</code> attribute of the <code>script</code> element.  In addition,  <code>src</code> attribute of the <code>script</code> element.  In addition,
822  the <abbr title="Uniform Resource Identifiers">URI</abbr> must be conform to  the <abbr title="Uniform Resource Identifiers">URI</abbr> must be conform to
823  the regular expression <code>^javascript:\s*(?:"[^"]*"|'[^']*')\s*$</code>.  the regular expression <code>^javascript:\s*(?:"[^"]*"|'[^']*')\s*$</code>.
824    <li>Only supports <code>\u<var>HHHH</var></code> escapes in JavaScript
825    string literals.
826    <li>Does not handle <i>stop parsing</i> phase correctly if the document is
827    replaced by <code>document.open ()</code> call.  In other word, delayed
828    (deferred or asynchronous) script executions and event firings might be
829    treated in a wrong way if a <code>document.open ()</code> invocation
830    is implicitly done by <code>document.write ()</code> in a delayed script.
831  </ul>  </ul>
832    
833  <p>For some reason, this parser does not work in browsers that do  <p>For some reason, this parser does not work in browsers that do
834  not support JavaScript 1.5.  not support JavaScript 1.5.
835    
836    <!-- TODO: |src| attribute value should refer the value at the time
837    when it is inserted into the document, not the value when the script is
838    executed.  Currently it does not matter, since we don't allow dynamic
839    modification to the |src| content/DOM attribute value yet. -->
840    
841  </body>  </body>
 </html>  
842    </html>
843    <!-- $Date$ -->
844    <!--
845    
846    Copyright 2008 Wakaba <w@suika.fam.cx>
847    
848    This program is free software; you can redistribute it and/or
849    modify it under the terms of the GNU General Public License
850    as published by the Free Software Foundation; either version 2
851    of the License, or (at your option) any later version.
852    
853    This program is distributed in the hope that it will be useful,
854    but WITHOUT ANY WARRANTY; without even the implied warranty of
855    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
856    GNU General Public License for more details.
857    
858    You should have received a copy of the GNU General Public License
859    along with this program; if not, write to the Free Software
860    Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
861    
862    -->
Legend:
Removed from v.1.9  
changed lines
  Added in v.1.13
admin@suikawiki.org
 ViewVC Help
Powered by ViewVC 1.1.24