/[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 1.13 by wakaba, Sun Apr 27 11:27:04 2008 UTC revision 1.16 by wakaba, Tue Apr 29 04:07:18 2008 UTC
# Line 65  Line 65 
65    
66    var logIndentLevel = 0;    var logIndentLevel = 0;
67    function log (s) {    function log (s) {
68        var indent = '';
69      for (var i = 0; i < logIndentLevel; i++) {      for (var i = 0; i < logIndentLevel; i++) {
70        s = '  ' + s;        indent += '  ';
71      }      }
72        s = indent + s.replace (/\n/g, "\n" + indent);
73      document.logElement.appendChild (document.createTextNode (s + "\n"));      document.logElement.appendChild (document.createTextNode (s + "\n"));
74    } // log    } // log
75    
# Line 81  Line 83 
83        doc = new JSDocument (this);        doc = new JSDocument (this);
84        doc.manakaiIsHTML = true;        doc.manakaiIsHTML = true;
85      }      }
86        this.nextToken = [];
87      this.doc = doc;      this.doc = doc;
88      this.openElements = [doc];      this.openElements = [doc];
89      this.input = i;      this.input = i;
# Line 90  Line 93 
93    } // Parser    } // Parser
94    
95    Parser.prototype.getNextToken = function () {    Parser.prototype.getNextToken = function () {
96        if (this.nextToken.length) {
97          return this.nextToken.shift ();
98        }
99    
100      var p = this;      var p = this;
101      var i = this.input;      var i = this.input;
102      if (this.parseMode == 'script') {      if (this.parseMode == 'cdata') {
103          var tagName = this.endTagName;
104        var token;        var token;
105        if (p.insertionPoint <= 0) {        if (p.insertionPoint <= 0) {
106          return {type: 'abort'};          return {type: 'abort'};
# Line 110  Line 118 
118          return '';          return '';
119        });        });
120        if (token) return token;        if (token) return token;
121        i.s = i.s.replace (/^<\/[Ss][Cc][Rr][Ii][Pp][Tt]>/, function (s) {        var pattern = new RegExp ('^</' + tagName + '>', 'i');
122          i.s = i.s.replace (pattern, function (s) {
123          if (p.insertionPoint < s.length) {          if (p.insertionPoint < s.length) {
124            token = {type: 'abort'};            token = {type: 'abort'};
125            return s;            return s;
126          }          }
127          token = {type: 'end-tag', value: 'script'};          token = {type: 'end-tag', value: tagName};
128          p.insertionPoint -= s.length;          p.insertionPoint -= s.length;
129          return '';          return '';
130        });        });
131        if (token) return token;        if (token) return token;
132        var m;        var m;
133        if ((p.insertionPoint < '</script'.length) &&        if ((p.insertionPoint < ('</' + tagName).length) &&
134            (m = i.s.match (/^<\/([SCRIPTscript]+)/))) {            (m = i.s.match (/^<\/([A-Za-z]+)/))) {
135          var v = m[1].substring (0, p.insertionPoint).toLowerCase ();          var v = m[1].substring (0, p.insertionPoint).toLowerCase ();
136          if (v == 'script'.substring (0, p.insertionPoint - '</'.length)) {          if (v == tagName.substring (0, p.insertionPoint - '</'.length)) {
137            return {type: 'abort'};            return {type: 'abort'};
138          }          }
139        }        }
# Line 142  Line 151 
151      i.s = i.s.replace (/^<\/([^>]+)(?:>|$)/, function (s, e) {      i.s = i.s.replace (/^<\/([^>]+)(?:>|$)/, function (s, e) {
152        if (p.insertionPoint < s.length ||        if (p.insertionPoint < s.length ||
153            (p.insertionPoint <= s.length &&            (p.insertionPoint <= s.length &&
154             s.substring (s.length - 1, 1) != '>')) {             s.substring (s.length - 1, s.length) != '>')) {
155          token = {type: 'abort'};          token = {type: 'abort'};
156          return s;          return s;
157        }        }
# Line 154  Line 163 
163      i.s = i.s.replace (/^<([^>]+)(?:>|$)/, function (s, e) {      i.s = i.s.replace (/^<([^>]+)(?:>|$)/, function (s, e) {
164        if (p.insertionPoint < s.length ||        if (p.insertionPoint < s.length ||
165            (p.insertionPoint <= s.length &&            (p.insertionPoint <= s.length &&
166             s.substring (s.length - 1, 1) != '>')) {             s.substring (s.length - 1, s.length) != '>')) {
167          token = {type: 'abort'};          token = {type: 'abort'};
168          return s;          return s;
169        }        }
# Line 217  Line 226 
226        var token = this.getNextToken ();        var token = this.getNextToken ();
227        log ('token: ' + token.type + ' "' + token.value + '"');        log ('token: ' + token.type + ' "' + token.value + '"');
228    
229          if (this.cdataEndTagRequired) {
230            // Generic CDATA parsing algorithm
231    
232            if (token.type != 'abort') {
233              // 7.
234              if (token.type == 'end-tag' && token.value == this.endTagName) {
235                // 7.1. Ignores it.
236                //
237              } else {
238                // 7.2. Parse error.
239                log ('Parse error: no </' + this.endTagName + '>');
240                this.nextToken.unshift (token);
241              }
242              this.cdataEndTagRequired = false;
243              continue;
244            }
245          }
246    
247        if (token.type == 'start-tag') {        if (token.type == 'start-tag') {
248          if (token.value == 'script') {          if (token.value == 'script') {
249            // 1. Create an element for the token in the HTML namespace.            // 1. Create an element for the token in the HTML namespace.
# Line 229  Line 256 
256            el.manakaiParserInserted = true;            el.manakaiParserInserted = true;
257    
258            // 3. Switch the tokeniser's content model flag to the CDATA state.            // 3. Switch the tokeniser's content model flag to the CDATA state.
259            this.parseMode = 'script';            this.parseMode = 'cdata';
260              this.endTagName = 'script';
261    
262            // 4.1. Collect all the character tokens.            // 4.1. Collect all the character tokens.
263            while (true) {            while (true) {
# Line 243  Line 271 
271              // 4.2. Until it returns a token that is not a character token, or              // 4.2. Until it returns a token that is not a character token, or
272              // until it stops tokenising.              // until it stops tokenising.
273              } else if (token.type == 'eof' ||              } else if (token.type == 'eof' ||
274                         (token.type == 'end-tag' && token.value == 'script') ||                         token.type == 'end-tag' ||
275                         token.type == 'abort') {                         token.type == 'abort') {
276                // 6. Switched back to the PCDATA state.                // 6. Switched back to the PCDATA state.
277                this.parseMode = 'pcdata';                this.parseMode = 'pcdata';
278    
279                // 7.1. If the next token is not an end tag token with ...                // 7.1. If the next token is not an end tag token with ...
280                if (token.type != 'end-tag') {                if (!(token.type == 'end-tag' && token.value == 'script')) {
281                  // 7.2. This is a parse error.                  // 7.2. This is a parse error.
282                  log ('Parse error: no </' + 'script>');                  log ('Parse error: no </' + 'script>');
283                    this.nextToken.unshift (token);
284    
285                  // 7.3. Mark the script element as "already executed".                  // 7.3. Mark the script element as "already executed".
286                  el.manakaiAlreadyExecuted = true;                  el.manakaiAlreadyExecuted = true;
# Line 313  Line 342 
342                //                //
343              }              }
344            }            }
345            } else if (token.value == 'style' ||
346                       token.value == 'noscript' ||
347                       token.value == 'xmp') {
348              // 1. Create an element for the token in the HTML namespace.
349              var el = new JSElement (this.doc, token.value);
350    
351              // 2. Append the new element to the current node.
352              this.openElements[this.openElements.length - 1].appendChild (el);
353    
354              // 3. Switch the tokeniser's content model flag to the CDATA state.
355              this.parseMode = 'cdata';
356              this.endTagName = token.value;
357    
358              // 4.1. Collect all the character tokens.
359              while (true) {
360                var token = this.getNextToken ();
361                log ('token: ' + token.type + ' "' + token.value + '"');
362    
363                if (token.type == 'char') {
364                  // 5. Append a single Text node to the script element node.
365                  el.manakaiAppendText (token.value);
366    
367                // 4.2. Until it returns a token that is not a character token, or
368                // until it stops tokenising.
369                } else if (token.type == 'eof' ||
370                           token.type == 'end-tag' ||
371                           token.type == 'abort') {
372                  // 6. Switched back to the PCDATA state.
373                  this.parseMode = 'pcdata';
374    
375                  if (token.type == 'abort') {
376                    this.cdataEndTagRequired = true;
377                    break;
378                  }
379    
380                  // 7.1. If the next token is not an end tag token with ...
381                  if (!(token.type == 'end-tag' &&
382                        token.value == this.endTagName)) {
383                    // 7.2. This is a parse error.
384                    log ('Parse error: no </' + this.endTagName + '>');
385                    this.nextToken.unshift (token);
386    
387                    // 7.3. Mark the script element as "already executed".
388                    el.manakaiAlreadyExecuted = true;
389                  } else {
390                    // 7.4. Ignore it.
391                    //
392                  }
393                  break;
394                }
395              }
396          } else {          } else {
397            var el = new JSElement (this.doc, token.value);            var el = new JSElement (this.doc, token.value);
398            this.openElements[this.openElements.length - 1].appendChild (el);            this.openElements[this.openElements.length - 1].appendChild (el);
# Line 397  Line 477 
477    
478      log ('DOMContentLoaded event fired');      log ('DOMContentLoaded event fired');
479    
480      // "delays tha load event" things has completed:      // "delays the load event" things has completed:
481      // readyState = 'complete'      // readyState = 'complete'
482      log ('load event fired');      log ('load event fired');
483    
# Line 453  Line 533 
533        // 2.4. If the script element has its "already executed" flag set        // 2.4. If the script element has its "already executed" flag set
534        if (e.manakaiAlreadyExecuted) {        if (e.manakaiAlreadyExecuted) {
535          // 2.5. Abort these steps at this point.          // 2.5. Abort these steps at this point.
536          log ('Running a script: aborted');          log ('Running a script: aborted (already executed)');
537          logIndentLevel--;          logIndentLevel--;
538          return e;          return e;
539        }        }
# Line 564  Line 644 
644          doc.write.apply (doc, args);          doc.write.apply (doc, args);
645          return '';          return '';
646        });        });
647        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*/,        var noDocumentElement = false;
648          s = s.replace (/^\s*var\s+s\s*=\s*document\.createElement\s*\(\s*['"]script['"]\s*\)\s*;\s*s\.src\s*=\s*(?:'([^']*)'|"([^"]*)")\s*;\s*document\.documentElement\.appendChild\s*\(\s*s\s*\)\s*;\s*/,
649        function (s, t, u) {        function (s, t, u) {
650          matched = true;          matched = true;
651          var args = [unescapeJSLiteral (t ? t : u)];          var args = [unescapeJSLiteral (t ? t : u)];
652          doc._insertExternalScript.apply (doc, args);          noDocumentElement = !doc._insertExternalScript.apply (doc, args);
653            return '';
654          });
655          if (noDocumentElement) {
656            log ('Script error: documentElement is null');
657            break;
658          }
659          s = s.replace (/^\s*w\s*\(\s*document\.documentElement\.innerHTML\s*\)\s*;\s*/,
660          function (s, t) {
661            matched = true;
662            log (dumpTree (doc, ''));
663          return '';          return '';
664        });        });
665        if (s == '') break;        if (s == '') break;
# Line 653  Line 744 
744    }; // document.open    }; // document.open
745    
746    JSDocument.prototype.write = function () {    JSDocument.prototype.write = function () {
747        log ('document.write: start');
748      logIndentLevel++;      logIndentLevel++;
749    
750      var p = this._parser;      var p = this._parser;
# Line 676  Line 768 
768      if (p.scriptExecutedAfterParserResumes) {      if (p.scriptExecutedAfterParserResumes) {
769        log ('document.write: processed later (there is an unprocessed <script src>)');        log ('document.write: processed later (there is an unprocessed <script src>)');
770        logIndentLevel--;        logIndentLevel--;
771          log ('document.write: return');
772        return;        return;
773      }      }
774    
# Line 689  Line 782 
782      // to do something here?      // to do something here?
783    
784      // 5. Return      // 5. Return
785        logIndentLevel--;
786      log ('document.write: return');      log ('document.write: return');
787    
     logIndentLevel--;  
788      return;      return;
789    }; // document.write    }; // document.write
790    
791    JSDocument.prototype._insertExternalScript = function (uri) {    JSDocument.prototype._insertExternalScript = function (uri) {
792      var s = new JSElement (this, 'script');      var s = new JSElement (this, 'script');
793      s.src = uri;      s.src = uri;
794      this.documentElement.appendChild (s);      if (this.documentElement) {
795          this.documentElement.appendChild (s);
796          return true;
797        } else {
798          return false;
799        }
800    }; // _insertExternalScript    }; // _insertExternalScript
801    
802    JSDocument.prototype.__defineGetter__ ('documentElement', function () {    JSDocument.prototype.__defineGetter__ ('documentElement', function () {
# Line 798  algorithm.  Especially, this parser: Line 896  algorithm.  Especially, this parser:
896  algorithm, and so on.  algorithm, and so on.
897  <li>Does not raise parse errors for invalid attribute specifications in start  <li>Does not raise parse errors for invalid attribute specifications in start
898  or end tags.  or end tags.
899  <li>Does not support CDATA/PCDATA element other than <code>script</code>.  <li>Does not support PCDATA elements (<code>title</code> and
900    <code>textarea</code>).
901    <li>Does not strip the first newline in <code>pre</code> elements.
902  <li>Does not support <code>&lt;!--</code>..<code>--></code> parsing rule  <li>Does not support <code>&lt;!--</code>..<code>--></code> parsing rule
903  in <code>script</code> element.  in <code>script</code> element.
904  <li>Does not support foreign (SVG or MathML) elements.  <li>Does not support foreign (SVG or MathML) elements.
# Line 813  introduced, followed, or separated by wh Line 913  introduced, followed, or separated by wh
913    <li><code>var s = document.createElement ("script");    <li><code>var s = document.createElement ("script");
914              s.src = "<var>string</var>";              s.src = "<var>string</var>";
915              document.documentElement.appendChild (s);</code>              document.documentElement.appendChild (s);</code>
916      <li><code>w (document.documentElement.innerHTML);</code> (This statement
917      can be used to dump the document, even when the document has no
918      document element.  The output format is the tree dump format used
919      in html5lib test data, not <abbr>HTML</abbr>.)
920    </ul>    </ul>
921  Note that strings may be delimited by <code>'</code>s instead of  Note that strings may be delimited by <code>'</code>s instead of
922  <code>"</code>s.  <code>"</code>s.
Legend:
Removed from v.1.13  
changed lines
  Added in v.1.16
admin@suikawiki.org
 ViewVC Help
Powered by ViewVC 1.1.24