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

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

Parent Directory | Revision Log | View Patch Patch

-revision 1.11 by wakaba,
Sun Apr 27 10:44:36 2008 UTC
+revision 1.18 by wakaba,
Sun Aug 31 09:46:14 2008 UTC
 Line 2
  <html lang=en>
  <head>
  <title>Live Scripting HTML Parser</title>
+ <link rel=author href="http://suika.fam.cx/~wakaba/who?">
+ <link rel=license href="http://suika.fam.cx/c/gnu/gpl"
+     title="GNU GPL2 or later">
  <style>
-   h1, h2 {
+   h1 {
+     margin: 0;
+     font-size: 150%;
+   }
+   h2 {
      margin: 0;
      font-size: 100%;
    }
-   p, pre {
+   p {
-     margin: 0;
+     margin: 0 1em;
    }
    textarea {
      width: 100%;
-Line 58
+Line 65
    var logIndentLevel = 0;
    function log (s) {
+     var indent = '';
      for (var i = 0; i < logIndentLevel; i++) {
-       s = '  ' + s;
+       indent += '  ';
      }
+     s = indent + s.replace (/\n/g, "\n" + indent);
      document.logElement.appendChild (document.createTextNode (s + "\n"));
    } // log
-Line 74
+Line 83
        doc = new JSDocument (this);
        doc.manakaiIsHTML = true;
      }
+     this.nextToken = [];
      this.doc = doc;
      this.openElements = [doc];
      this.input = i;
      this.scriptsExecutedAfterParsing = [];
      this.scriptsExecutedSoon = [];
+     this.scriptsExecutedAsynchronously = [];
    } // Parser
    Parser.prototype.getNextToken = function () {
+     if (this.nextToken.length) {
+       return this.nextToken.shift ();
+     }
      var p = this;
      var i = this.input;
-     if (this.parseMode == 'script') {
+     if (this.parseMode == 'cdata') {
+       var tagName = this.endTagName;
        var token;
        if (p.insertionPoint <= 0) {
          return {type: 'abort'};
-Line 102
+Line 118
          return '';
        });
        if (token) return token;
-       i.s = i.s.replace (/^<\/[Ss][Cc][Rr][Ii][Pp][Tt]>/, function (s) {
+       var pattern = new RegExp ('^</' + tagName + '>', 'i');
+       i.s = i.s.replace (pattern, function (s) {
          if (p.insertionPoint < s.length) {
            token = {type: 'abort'};
            return s;
          }
-         token = {type: 'end-tag', value: 'script'};
+         token = {type: 'end-tag', value: tagName};
          p.insertionPoint -= s.length;
          return '';
        });
        if (token) return token;
        var m;
-       if ((p.insertionPoint < '</script'.length) &&
+       if ((p.insertionPoint < ('</' + tagName).length) &&
-           (m = i.s.match (/^<\/([SCRIPTscript]+)/))) {
+           (m = i.s.match (/^<\/([A-Za-z]+)/))) {
          var v = m[1].substring (0, p.insertionPoint).toLowerCase ();
-         if (v == 'script'.substring (0, p.insertionPoint - '</'.length)) {
+         if (v == tagName.substring (0, p.insertionPoint - '</'.length)) {
            return {type: 'abort'};
          }
        }
-Line 134
+Line 151
      i.s = i.s.replace (/^<\/([^>]+)(?:>|$)/, function (s, e) {
        if (p.insertionPoint < s.length ||
            (p.insertionPoint <= s.length &&
-            s.substring (s.length - 1, 1) != '>')) {
+            s.substring (s.length - 1, s.length) != '>')) {
          token = {type: 'abort'};
          return s;
        }
-Line 146
+Line 163
      i.s = i.s.replace (/^<([^>]+)(?:>|$)/, function (s, e) {
        if (p.insertionPoint < s.length ||
            (p.insertionPoint <= s.length &&
-            s.substring (s.length - 1, 1) != '>')) {
+            s.substring (s.length - 1, s.length) != '>')) {
          token = {type: 'abort'};
          return s;
        }
-Line 209
+Line 226
        var token = this.getNextToken ();
        log ('token: ' + token.type + ' "' + token.value + '"');
+       if (this.cdataEndTagRequired) {
+         // Generic CDATA parsing algorithm
+         if (token.type != 'abort') {
+           // 7.
+           if (token.type == 'end-tag' && token.value == this.endTagName) {
+             // 7.1. Ignores it.
+             //
+           } else {
+             // 7.2. Parse error.
+             log ('Parse error: no </' + this.endTagName + '>');
+             this.nextToken.unshift (token);
+           }
+           this.cdataEndTagRequired = false;
+           continue;
+         }
+       }
        if (token.type == 'start-tag') {
          if (token.value == 'script') {
            // 1. Create an element for the token in the HTML namespace.
-Line 221
+Line 256
            el.manakaiParserInserted = true;
            // 3. Switch the tokeniser's content model flag to the CDATA state.
-           this.parseMode = 'script';
+           this.parseMode = 'cdata';
+           this.endTagName = 'script';
            // 4.1. Collect all the character tokens.
            while (true) {
-Line 235
+Line 271
              // 4.2. Until it returns a token that is not a character token, or
              // until it stops tokenising.
              } else if (token.type == 'eof' ||
-                        (token.type == 'end-tag' && token.value == 'script') ||
+                        token.type == 'end-tag' ||
                         token.type == 'abort') {
                // 6. Switched back to the PCDATA state.
                this.parseMode = 'pcdata';
                // 7.1. If the next token is not an end tag token with ...
-               if (token.type != 'end-tag') {
+               if (!(token.type == 'end-tag' && token.value == 'script')) {
                  // 7.2. This is a parse error.
                  log ('Parse error: no </' + 'script>');
+                 this.nextToken.unshift (token);
                  // 7.3. Mark the script element as "already executed".
                  el.manakaiAlreadyExecuted = true;
-Line 275
+Line 312
            oldInsertionPoint += this.insertionPoint;
            this.setInsertionPoint (oldInsertionPoint);
-           // 12. If there is a script that will execute as soon as ...
+           // 12. If there is a pending external script
-           while (this.scriptExecutedWhenParserResumes) {
+           while (this.pendingExternalScript) {
              // 12.1. If the tree construction stage is being called reentrantly
              if (this.reentrant) {
                log ('parse: abort (reentrance)');
-Line 286
+Line 323
              // 12.2. Otherwise
              } else {
                // 1.
-               var script = this.scriptExecutedWhenParserResumes;
+               var script = this.pendingExternalScript;
-               this.scriptExecutedWhenParserResumes = null;
+               this.pendingExternalScript = null;
                // 2. Pause until the script has completed loading.
                //
-Line 305
+Line 342
                //
              }
            }
+         } else if (token.value == 'style' ||
+                    token.value == 'noscript' ||
+                    token.value == 'xmp') {
+           // 1. Create an element for the token in the HTML namespace.
+           var el = new JSElement (this.doc, token.value);
+           // 2. Append the new element to the current node.
+           this.openElements[this.openElements.length - 1].appendChild (el);
+           // 3. Switch the tokeniser's content model flag to the CDATA state.
+           this.parseMode = 'cdata';
+           this.endTagName = token.value;
+           // 4.1. Collect all the character tokens.
+           while (true) {
+             var token = this.getNextToken ();
+             log ('token: ' + token.type + ' "' + token.value + '"');
+             if (token.type == 'char') {
+               // 5. Append a single Text node to the script element node.
+               el.manakaiAppendText (token.value);
+             // 4.2. Until it returns a token that is not a character token, or
+             // until it stops tokenising.
+             } else if (token.type == 'eof' ||
+                        token.type == 'end-tag' ||
+                        token.type == 'abort') {
+               // 6. Switched back to the PCDATA state.
+               this.parseMode = 'pcdata';
+               if (token.type == 'abort') {
+                 this.cdataEndTagRequired = true;
+                 break;
+               }
+               // 7.1. If the next token is not an end tag token with ...
+               if (!(token.type == 'end-tag' &&
+                     token.value == this.endTagName)) {
+                 // 7.2. This is a parse error.
+                 log ('Parse error: no </' + this.endTagName + '>');
+                 this.nextToken.unshift (token);
+                 // 7.3. Mark the script element as "already executed".
+                 el.manakaiAlreadyExecuted = true;
+               } else {
+                 // 7.4. Ignore it.
+                 //
+               }
+               break;
+             }
+           }
          } else {
            var el = new JSElement (this.doc, token.value);
            this.openElements[this.openElements.length - 1].appendChild (el);
-Line 335
+Line 423
      // "When a script completes loading" rules start applying.
-     // List of scripts that will execute as soon as possible
+     while (this.scriptsExecutedSoon.length > 0 ||
-     for (var i = 0; i < this.scriptsExecutedSoon.length; i++) {
+            this.scriptsExecutedAsynchronously.length > 0) {
-       var e = this.scriptsExecutedSoon[i];
+       // Handle "list of scripts that will execute as soon as possible".
+       while (this.scriptsExecutedSoon.length > 0) {
+         var e = this.scriptsExecutedSoon.shift ();
+         // If it has completed loading
+         log ('Execute an external script not inserted by parser...');
+         executeScript (this.doc, e);
+         // NOTE: It MAY be executed before the end of the parsing, according
+         // to the spec.
+         this.hasAsyncScript = true;
+       }
+       // Handle "list of scripts that will execute asynchronously".
+       while (this.scriptsExecutedAsynchronously.length > 0) {
+         var e = this.scriptsExecutedAsynchronously.shift ();
+         // Step 1.
+         // We assume that all scripts have been loaded at this time.
+         // Step 2.
+         log ('Execute an asynchronous script...');
+         executeScript (this.doc, e);
+         // Step 3.
+         //
+         // Step 4.
+         //
-       // If it has completed loading
+         this.hasAsyncScript = true;
-       log ('Execute an external script not inserted by parser...');
+       }
-       executeScript (this.doc, e);
-       // NOTE: It MAY be executed before the end of the parsing, according
-       // to the spec.
-       this.hasAsyncScript = true;
      }
-     // TODO: Handles
-     // "list of scripts that will execute asynchronously"
      // Handle "list of scripts that will execute when the document has finished
      // parsing".
      var list = this.scriptsExecutedAfterParsing;
-Line 369
+Line 477
      log ('DOMContentLoaded event fired');
-     // "delays tha load event" things has completed:
+     // "delays the load event" things has completed:
      // readyState = 'complete'
      log ('load event fired');
-Line 425
+Line 533
        // 2.4. If the script element has its "already executed" flag set
        if (e.manakaiAlreadyExecuted) {
          // 2.5. Abort these steps at this point.
-         log ('Running a script: aborted');
+         log ('Running a script: aborted (already executed)');
          logIndentLevel--;
          return e;
        }
-Line 444
+Line 552
          p.scriptsExecutedAfterParsing.push (e);
          log ('Running a script: aborted (defer)');
        } else if (e.async && e.src != null) {
-         // TODO
+         p.scriptsExecutedAsynchronously.push (e);
-       } else if (e.async && e.src == null
+         log ('Running a script: aborted (async src)');
-                  /* && list of scripts that will execute asynchronously is not empty */) {
+       } else if (e.async && e.src == null &&
-         // TODO
+                  p.scriptsExecutedAsynchronously.length > 0) {
+         p.scriptsExecutedAsynchronously.push (e);
+         log ('Running a script: aborted (async)');
+         // ISSUE: What is the difference with the case above?
        } else if (e.src != null && e.manakaiParserInserted) {
-         if (p.scriptExecutedWhenParserResumes) {
+         if (p.pendingExternalScript) {
-           log ('Error: There is a script that will execute as soon as the parser resumes.');
+           log ('Error: There is a pending external script.');
          }
-         p.scriptExecutedWhenParserResumes = e;
+         p.pendingExternalScript = e;
          log ('Running a script: aborted (src parser-inserted)');
        } else if (e.src != null) {
          p.scriptsExecutedSoon.push (e);
-Line 487
+Line 598
      }
      // If the load was successful
-     log ('load event fired at the script element');
      if (true) {
      // Scripting is enabled, Document.designMode is disabled,
-Line 496
+Line 606
        parseAndRunScript (doc, s);
      }
+     log ('load event fired at the script element');
      log ('executing a script block: end');
    } // executeScript
-Line 533
+Line 645
          doc.write.apply (doc, args);