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

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

Parent Directory | Revision Log | View Patch Patch

-revision 1.5 by wakaba,
Fri Apr 25 11:40:56 2008 UTC
+revision 1.12 by wakaba,
Sun Apr 27 11:21:09 2008 UTC
 Line 1
  <!DOCTYPE HTML>
  <html lang=en>
  <head>
- <title>Demo of HTML5 Parsing Algorithm with Scripting Enabled</title>
+ <title>Live Scripting HTML Parser</title>
  <style>
+   h1, h2 {
+     margin: 0;
+     font-size: 100%;
+   }
+   p, pre {
+     margin: 0;
+   }
    textarea {
-      display: block;
+     width: 100%;
-      width: 80%;
+     -width: 99%;
-      margin-left: auto;
+     height: 10em;
-      margin-right: auto;
-      min-height: 20em;
    }
    output {
      display: block;
-Line 18
+Line 23
    }
  </style>
  <script>
+   var delayedUpdater = 0;
    function update () {
-     document.logElement.textContent = '';
+     if (delayedUpdater) {
-     var p = new Parser (new InputStream (document.sourceElement.value));
+       clearTimeout (delayedUpdater);
-     var doc = p.doc;
+       delayedUpdater = 0;
-     p.parse ();
+     }
-     log (dumpTree (doc, ''));
+     delayedUpdater = setTimeout (update2, 100);
    } // update
+   function update2 () {
+     var v = document.sourceElement.value;
+     if (v != document.previousSourceText) {
+       document.previousSourceText = v;
+       document.links['permalink'].href
+           = location.pathname + '?s=' + encodeURIComponent (v);
+       document.links['ldvlink'].href
+           = 'http://software.hixie.ch/utilities/js/live-dom-viewer/?'
+           + encodeURIComponent (v);
+       document.logElement.textContent = '';
+       var p = new Parser (new InputStream (v));
+       var doc = p.doc;
+       p.parse ();
+       log (dumpTree (doc, ''));
+       if (p.hasAsyncScript) {
+         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');
+       }
+     }
+   } // update2
+   var logIndentLevel = 0;
    function log (s) {
+     for (var i = 0; i < logIndentLevel; i++) {
+       s = '  ' + s;
+     }
      document.logElement.appendChild (document.createTextNode (s + "\n"));
    } // log
-Line 42
+Line 76
      }
      this.doc = doc;
      this.openElements = [doc];
-     this.in = i;
+     this.input = i;
      this.scriptsExecutedAfterParsing = [];
+     this.scriptsExecutedSoon = [];
+     this.scriptsExecutedAsynchronously = [];
    } // Parser
    Parser.prototype.getNextToken = function () {
      var p = this;
-     var i = this.in;
+     var i = this.input;
      if (this.parseMode == 'script') {
        var token;
        if (p.insertionPoint <= 0) {
-Line 121
+Line 157
          tagName = v.toLowerCase ();
          return '';
        });
-       e = e.replace (/^\s*(\S+)\s*(?:=\s*"([^"]*)"|'([^']*)'|([^"']+))?/,
+       while (true) {
-       function (x, attrName, attrValue1, attrValue2, attrValue3) {
+         var m = false;
-         attrs[attrName] = attrValue1 || attrValue2 || attrValue3;
+         e = e.replace (/^\s*([^\s=]+)\s*(?:=\s*(?:"([^"]*)"|'([^']*)'|([^"'\s]*)))?/,
-         return '';
+         function (x, attrName, attrValue1, attrValue2, attrValue3) {
-       });
+           v = attrValue1 || attrValue2 || attrValue3;
+           v = v.replace (/&quot;/g, '"').replace (/&apos;/g, "'")
+               .replace (/&amp;/g, '&');
+           attrs[attrName.toLowerCase ()] = v;
+           m = true;
+           return '';
+         });
+         if (!m) break;
+       }
+       if (e.length) {
+         log ('Broken start tag: "' + e + '"');
+       }
        token = {type: 'start-tag', value: tagName, attrs: attrs};
        p.insertionPoint -= s.length;
        return '';
-Line 156
+Line 203
    } // getNextToken
    Parser.prototype.parse = function () {
-     log ('start parsing');
+     logIndentLevel++;
+     log ('parse: start');
      while (true) {
        var token = this.getNextToken ();
-Line 224
+Line 272
            this.openElements[this.openElements.length - 1].appendChild (el);
            // 11. Let the insertion point have the value of the old ...
            oldInsertionPoint += this.insertionPoint;
            this.setInsertionPoint (oldInsertionPoint);
            // 12. If there is a script that will execute as soon as ...
+           while (this.scriptExecutedWhenParserResumes) {
+             // 12.1. If the tree construction stage is being called reentrantly
+             if (this.reentrant) {
+               log ('parse: abort (reentrance)');
+               logIndentLevel--;
+               return;
+             // 12.2. Otherwise
+             } else {
+               // 1.
+               var script = this.scriptExecutedWhenParserResumes;
+               this.scriptExecutedWhenParserResumes = null;
+               // 2. Pause until the script has completed loading.
+               //
+               // 3. Let the insertion point to just before the next input char.
+               this.setInsertionPoint (0);
+               // 4. Execute the script.
+               executeScript (this.doc, script);
+               // 5. Let the insertion point be undefined again.
+               this.setInsertionPoint (undefined);
+               // 6. If there is once again a script that will execute ...
+               //
+             }
+           }
          } else {
            var el = new JSElement (this.doc, token.value);
            this.openElements[this.openElements.length - 1].appendChild (el);
-Line 249
+Line 325
          break;
        } else if (token.type == 'abort') {
          log ('parse: abort');
+         logIndentLevel--;
          return;
        }
      }
-Line 259
+Line 336
      // "When a script completes loading" rules start applying.
-     // TODO: Handles "list of scripts that will execute as soon as possible"
+     while (this.scriptsExecutedSoon.length > 0 ||
-     // and "list of scripts that will execute asynchronously"
+            this.scriptsExecutedAsynchronously.length > 0) {
+       // 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.
+         //
+         this.hasAsyncScript = true;
+       }
+     }
      // Handle "list of scripts that will execute when the document has finished
      // parsing".
-Line 283
+Line 393
      // "delays tha load event" things has completed:
      // readyState = 'complete'
      log ('load event fired');
+     logIndentLevel--;
    } // parse
    Parser.prototype.setInsertionPoint = function (ip) {
      if (ip == undefined || ip == null || isNaN (ip)) {
        log ('insertion point: set to undefined');
        this.insertionPoint = undefined;
-     } else if (ip == this.in.s.length) {
+     } else if (ip == this.input.s.length) {
        log ('insertion point: end of file');
        this.insertionPoint = ip;
      } else {
        log ('insertion point: set to ' + ip +
-            ' (before "' + this.in.s.substring (0, 10) + '")');
+            ' (before "' + this.input.s.substring (0, 10) + '")');
        this.insertionPoint = ip;
      }
    }; // setInsertionPoint
-Line 316
+Line 428
      e.parentNode = this;
      if (e.localName == 'script') {
+       logIndentLevel++;
        log ('Running a script: start');
        var doc = this.ownerDocument || this;
-Line 334
+Line 447
        if (e.manakaiAlreadyExecuted) {
          // 2.5. Abort these steps at this point.
          log ('Running a script: aborted');
+         logIndentLevel--;
          return e;
        }
-Line 351
+Line 465
          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) {
-         // TODO
+         if (p.scriptExecutedWhenParserResumes) {
+           log ('Error: There is a script that will execute as soon as the parser resumes.');
+         }
+         p.scriptExecutedWhenParserResumes = e;
+         log ('Running a script: aborted (src parser-inserted)');
        } else if (e.src != null) {
-         // TODO
+         p.scriptsExecutedSoon.push (e);
+         log ('Running a script: aborted (src)');
        } else {
          executeScript (doc, e); // even if other scripts are already executing.
        }
        log ('Running a script: end');
+       logIndentLevel--;
      }
      return e;
-Line 372
+Line 495
    function executeScript (doc, e) {
      log ('executing a script block: start');
-     // If the load resulted in an error, then ... firing an error event ...
+     var s;
+     if (e.src != null) {
+       s = getExternalScript (e.src);
+       // If the load resulted in an error, then ... firing an error event ...
+       if (s == null) {
+         log ('error event fired at the script element');
+         return;
+       }
+       log ('External script loaded: "' + s + '"');
+     } else {
+       s = e.text;
+     }
      // If the load was successful
      log ('load event fired at the script element');
-Line 381
+Line 517
      // Scripting is enabled, Document.designMode is disabled,
      // Document is the active document in its browsing context
-       var s;
-       if (e.src != null) {
-         // TODO: from external file
-       } else {
-         s = e.text;
-       }
        parseAndRunScript (doc, s);
      }
      log ('executing a script block: end');
    } // executeScript
+   function getExternalScript (uri) {
+     if (uri.match (/^javascript:/i)) {
+       var m;
+       if (m = uri.match (/^javascript:\s*(?:'([^']*)'|"([^"]+)")\s*$/i)) {
+         if (m[1]) {
+           return unescapeJSLiteral (m[1]);
+         } else if (m[2]) {
+           return unescapeJSLiteral (m[2]);
+         } else {
+           return null;
+         }
+       } else {
+         log ('Complex javascript: URI is not supported: <' + uri + '>');
+         return null;
+       }
+     } else {
+       log ('URI scheme not supported: <' + uri + '>');
+       return null;
+     }
+   } // getExternalScript
    function parseAndRunScript (doc, s) {
      while (true) {
        var matched = false;