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

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

Parent Directory | Revision Log | View Patch Patch

-revision 1.8 by wakaba,
Sun Apr 27 08:56:34 2008 UTC
+revision 1.16 by wakaba,
Tue Apr 29 04:07:18 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 47
+Line 54
        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) {
+     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 69
+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 96
+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 128
+Line 151
      i.s = i.s.replace (/^<\/([^>]+)(?:>|$)/, function (s, e) {
        if (p.insertionPoint < s.length ||
            (p.insertionPoint <= s.length &&
-            s.substring (s.length - 1, 1) != '>')) {
+            s.substring (s.length - 1, s.length) != '>')) {
          token = {type: 'abort'};
          return s;
        }
-Line 140
+Line 163
      i.s = i.s.replace (/^<([^>]+)(?:>|$)/, 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 150
+Line 173
          tagName = v.toLowerCase ();
          return '';
        });
-       e = e.replace (/^\s*([^\s=]+)\s*(?:=\s*(?:"([^"]*)"|'([^']*)'|([^"']+)))?/,
+       while (true) {
-       function (x, attrName, attrValue1, attrValue2, attrValue3) {
+         var m = false;
-         v = attrValue1 || attrValue2 || attrValue3;
+         e = e.replace (/^\s*([^\s=]+)\s*(?:=\s*(?:"([^"]*)"|'([^']*)'|([^"'\s]*)))?/,
-         v = v.replace (/&quot;/g, '"').replace (/&apos;/g, "'")
+         function (x, attrName, attrValue1, attrValue2, attrValue3) {
-             .replace (/&amp;/g, '&');
+           v = attrValue1 || attrValue2 || attrValue3;
-         attrs[attrName.toLowerCase ()] = v;
+           v = v.replace (/&quot;/g, '"').replace (/&apos;/g, "'")
-         return '';
+               .replace (/&amp;/g, '&');
-       });
+           attrs[attrName.toLowerCase ()] = v;
+           m = true;
+           return '';
+         });
+         if (!m) break;
+       }
        if (e.length) {
          log ('Broken start tag: "' + e + '"');
        }
-Line 198
+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 210
+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 224
+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 294
+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 324
+Line 423
      // "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 345
+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 401
+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 420
+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) {
            log ('Error: There is a script that will execute as soon as the parser resumes.');
          }
          p.scriptExecutedWhenParserResumes = e;
-         log ('Running a script: aborted (src)');
+         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.
        }
-Line 479
+Line 615
        var m;
        if (m = uri.match (/^javascript:\s*(?:'([^']*)'|"([^"]+)")\s*$/i)) {
          if (m[1]) {
-           return m[1];
+           return unescapeJSLiteral (m[1]);
          } else if (m[2]) {
-           return m[2];
+           return unescapeJSLiteral (m[2]);
          } else {
            return null;
          }
-Line 502
+Line 638
          matched = true;
          var args = [];
          t.replace (/('[^']*'|"[^"]*")/g, function (s, v) {
-           args.push (v.substring (1, v.length - 1));
+           args.push (unescapeJSLiteral (v.substring (1, v.length - 1)));
            return '';
          });
          doc.write.apply (doc, args);
          return '';
        });
+       var noDocumentElement = false;
+       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*/,
+       function (s, t, u) {
+         matched = true;
+         var args = [unescapeJSLiteral (t ? t : u)];
+         noDocumentElement = !doc._insertExternalScript.apply (doc, args);
+         return '';
+       });
+       if (noDocumentElement) {
+         log ('Script error: documentElement is null');
+         break;
+       }
+       s = s.replace (/^\s*w\s*\(\s*document\.documentElement\.innerHTML\s*\)\s*;\s*/,
+       function (s, t) {
+         matched = true;
+         log (dumpTree (doc, ''));
+         return '';
+       });
        if (s == '') break;
        if (!matched) {
          log ('Script parse error: "' + s + '"');
-Line 516
+Line 670
      }
    } // parseAndRunScript
+   function unescapeJSLiteral (s) {
+     return s.replace (/\\u([0-9A-Fa-f]{4})/g, function (t, v) {
+       return String.fromCharCode (parseInt ('0x' + v));
+     });
+   } // unescapeJSLiteral
    function JSText (data) {
      this.data = data;
    } // JSText
-Line 584
+Line 744
    }; // document.open
    JSDocument.prototype.write = function () {
+     log ('document.write: start');
      logIndentLevel++;
      var p = this._parser;
-Line 607
+Line 768
      if (p.scriptExecutedAfterParserResumes) {
        log ('document.write: processed later (there is an unprocessed <script src>)');
        logIndentLevel--;
+       log ('document.write: return');
        return;
      }
-Line 620
+Line 782
      // to do something here?
      // 5. Return
+     logIndentLevel--;
      log ('document.write: return');
-     logIndentLevel--;
      return;
    }; // document.write
+   JSDocument.prototype._insertExternalScript = function (uri) {
+     var s = new JSElement (this, 'script');
+     s.src = uri;
+     if (this.documentElement) {
+       this.documentElement.appendChild (s);
+       return true;
+     } else {
+       return false;
+     }
+   }; // _insertExternalScript
+   JSDocument.prototype.__defineGetter__ ('documentElement', function () {
+     var cn = this.childNodes;
+     for (var i = 0; i < cn.length; i++) {
+       if (cn[i] instanceof JSElement) {
+         return cn[i]
+       }
+     }
+     return null;
+   });
    JSElement.prototype.__defineGetter__ ('text', function () {
      var r = '';
      for (var i = 0; i < this.childNodes.length; i++) {
-Line 644
+Line 827
          r += '| ' + indent + node.localName + '\n';
          if (node.async) r += '| ' + indent + '  async=""\n';
          if (node.defer) r += '| ' + indent + '  defer=""\n';
-         if (node.src) r += '| ' + indent + '  src="' + node.src + '"\n';
+         if (node.src != null) {
+           r += '| ' + indent + '  src="' + node.src + '"\n';
+         }
          r += dumpTree (node, indent + '  ');
        } else if (node instanceof JSText) {
          r += '| ' + indent + '"' + node.data + '"\n';
-Line 693 
 document.write ('aaaaaaa&lt;/p>&lt;scrip
+Line 878 
 document.write ('aaaaaaa&lt;/p>&lt;scrip
  &lt;p>
  </textarea>
- <h2>Log</h2>
+ <h2 id=log>Log</h2>
  <p><output></output>
- <h2>Note</h2>
+ <h2 id=notes>Notes</h2>
  <p>This is a <em>simplified</em> implementation of
  <a href="http://www.whatwg.org/specs/web-apps/current-work/#parsing">HTML5
-Line 711 
 algorithm.  Especially, this parser:
+Line 896 
 algorithm.  Especially, this parser:
  algorithm, and so on.
  <li>Does not raise parse errors for invalid attribute specifications in start
  or end tags.
- <li>Does not support CDATA/PCDATA element other than <code>script</code>.
+ <li>Does not support PCDATA elements (<code>title</code> and
+ <code>textarea</code>).
+ <li>Does not strip the first newline in <code>pre</code> elements.
  <li>Does not support <code>&lt;!--</code>..<code>--></code> parsing rule
  in <code>script</code> element.
  <li>Does not support foreign (SVG or MathML) elements.
  <li>Only supports <code>script</code> <code>type</code>
  <code>text/javascript</code>.  <code>type</code> and <code>language</code>
  attributes are ignored.
- <li>Only supports <code>document.write</code>.
+ <li>Only supports limited statements.  It must consist of zero or more
- The script code must be match to the regular expression
+ of statements looking similar to the following statements, possibly
- <code>^\s*(?:document\.write\s*\(<var>v</var>\s*(?:,\s*<var>v</var>\s*)*\)\s*;\s*)*$</code>
+ introduced, followed, or separated by white space characters:
- where <var>v</var> is <code>"[^"]*"|'[^']*'</code>.
+   <ul>
+   <li><code>document.write ("<var>string</var>", ["<var>string</var>", ...]);</code>.
+   <li><code>var s = document.createElement ("script");
+             s.src = "<var>string</var>";
+             document.documentElement.appendChild (s);</code>
+   <li><code>w (document.documentElement.innerHTML);</code> (This statement
+   can be used to dump the document, even when the document has no
+   document element.  The output format is the tree dump format used
+   in html5lib test data, not <abbr>HTML</abbr>.)
+   </ul>
+ Note that strings may be delimited by <code>'</code>s instead of
+ <code>"</code>s.
  <li>Only supports <code>javascript:</code>
  <abbr title="Uniform Resourace Identifiers">URI</abbr> scheme in the
  <code>src</code> attribute of the <code>script</code> element.  In addition,
  the <abbr title="Uniform Resource Identifiers">URI</abbr> must be conform to
  the regular expression <code>^javascript:\s*(?:"[^"]*"|'[^']*')\s*$</code>.
+ <li>Only supports <code>\u<var>HHHH</var></code> escapes in JavaScript
+ string literals.
+ <li>Does not handle <i>stop parsing</i> phase correctly if the document is
+ replaced by <code>document.open ()</code> call.  In other word, delayed
+ (deferred or asynchronous) script executions and event firings might be
+ treated in a wrong way if a <code>document.open ()</code> invocation
+ is implicitly done by <code>document.write ()</code> in a delayed script.
  </ul>
  <p>For some reason, this parser does not work in browsers that do
  not support JavaScript 1.5.
- <!-- TODO: multiple attributes are not supported yet -->
+ <!-- TODO: |src| attribute value should refer the value at the time
+ when it is inserted into the document, not the value when the script is
+ executed.  Currently it does not matter, since we don't allow dynamic
+ modification to the |src| content/DOM attribute value yet. -->
  </body>
- </html>
+ </html>
+ <!-- $Date$ -->
+ <!--
+ Copyright 2008 Wakaba <w@suika.fam.cx>
+ This program is free software; you can redistribute it and/or
+ modify it under the terms of the GNU General Public License
+ as published by the Free Software Foundation; either version 2
+ of the License, or (at your option) any later version.
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ GNU General Public License for more details.
+ You should have received a copy of the GNU General Public License
+ along with this program; if not, write to the Free Software
+ Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ -->

 Legend:



Removed from v.1.8
 


changed lines


 
Added in v.1.16
 Legend:



Removed from v.1.8
 


changed lines


 
Added in v.1.16
-Removed from v.1.8
+Added in v.1.16

admin@suikawiki.org	ViewVC Help
Powered by ViewVC 1.1.24