Parent Directory
| 
Revision Log
Scripting support, first version (only placeholder for document.write is implemented)
| 1 | <!DOCTYPE HTML> |
| 2 | <html lang=en> |
| 3 | <head> |
| 4 | <title>Demo of HTML5 Parsing Algorithm with Scripting Enabled</title> |
| 5 | <style> |
| 6 | textarea { |
| 7 | display: block; |
| 8 | width: 80%; |
| 9 | margin-left: auto; |
| 10 | margin-right: auto; |
| 11 | min-height: 20em; |
| 12 | } |
| 13 | output { |
| 14 | display: block; |
| 15 | font-family: monospace; |
| 16 | white-space: pre; |
| 17 | } |
| 18 | </style> |
| 19 | <script> |
| 20 | function update () { |
| 21 | document.logElement.textContent = ''; |
| 22 | var p = new Parser (new InputStream (document.sourceElement.value)); |
| 23 | p.parse (); |
| 24 | log (dumpTree (p.doc, '')); |
| 25 | } // update |
| 26 | |
| 27 | function log (s) { |
| 28 | document.logElement.appendChild (document.createTextNode (s + "\n")); |
| 29 | } // log |
| 30 | |
| 31 | function InputStream (s) { |
| 32 | this.s = s; |
| 33 | } // InputStream |
| 34 | |
| 35 | function Parser (i) { |
| 36 | this.parseMode = 'pcdata'; |
| 37 | this.doc = new JSDocument (this); |
| 38 | this.openElements = [this.doc]; |
| 39 | this.in = i; |
| 40 | } // Parser |
| 41 | |
| 42 | Parser.prototype.getNextToken = function () { |
| 43 | var i = this.in; |
| 44 | if (this.parseMode == 'script') { |
| 45 | var token; |
| 46 | i.s = i.s.replace (/^([\s\S]+?)<\/[Ss][Cc][Rr][Ii][Pp][Tt]>/, |
| 47 | function (s, t) { |
| 48 | token = {type: 'char', value: t}; |
| 49 | return '<' + '/script>'; |
| 50 | }); |
| 51 | if (token) return token; |
| 52 | i.s = i.s.replace (/^<\/[Ss][Cc][Rr][Ii][Pp][Tt]>/, function () { |
| 53 | token = {type: 'end-tag', value: 'script'}; |
| 54 | return ''; |
| 55 | }); |
| 56 | if (token) return token; |
| 57 | return {type: 'eof'}; |
| 58 | } |
| 59 | |
| 60 | var token; |
| 61 | i.s = i.s.replace (/^<\/([^>]+)>/, function (s, e) { |
| 62 | token = {type: 'end-tag', value: e.toLowerCase ()}; |
| 63 | return ''; |
| 64 | }); |
| 65 | if (token) return token; |
| 66 | i.s = i.s.replace (/^<([^>]+)>/, function (s, e) { |
| 67 | token = {type: 'start-tag', value: e.toLowerCase ()}; |
| 68 | return ''; |
| 69 | }); |
| 70 | if (token) return token; |
| 71 | i.s = i.s.replace (/^[^<]+/, function (s) { |
| 72 | token = {type: 'char', value: s}; |
| 73 | return ''; |
| 74 | }); |
| 75 | if (token) return token; |
| 76 | i.s = i.s.replace (/^[\s\S]/, function (s) { |
| 77 | token = {type: 'char', value: s}; |
| 78 | return ''; |
| 79 | }); |
| 80 | if (token) return token; |
| 81 | return {type: 'eof'}; |
| 82 | } // getNextToken |
| 83 | |
| 84 | Parser.prototype.parse = function () { |
| 85 | log ('start parsing'); |
| 86 | |
| 87 | while (true) { |
| 88 | var token = this.getNextToken (); |
| 89 | log ('token: ' + token.type + ' "' + token.value + '"'); |
| 90 | |
| 91 | if (token.type == 'start-tag') { |
| 92 | if (token.value == 'script') { |
| 93 | // 1. Create an element for the token in the HTML namespace. |
| 94 | var el = new JSElement (this.doc, token.value); |
| 95 | |
| 96 | // 2. Mark the element as being "parser-inserted". |
| 97 | el.manakaiParserInserted = true; |
| 98 | |
| 99 | // 3. Switch the tokeniser's content model flag to the CDATA state. |
| 100 | this.parseMode = 'script'; |
| 101 | |
| 102 | // 4.1. Collect all the character tokens. |
| 103 | while (true) { |
| 104 | var token = this.getNextToken (); |
| 105 | log ('token: ' + token.type + ' "' + token.value + '"'); |
| 106 | |
| 107 | if (token.type == 'char') { |
| 108 | // 5. Append a single Text node to the script element node. |
| 109 | el.manakaiAppendText (token.value); |
| 110 | |
| 111 | // 4.2. Until it returns a token that is not a character token, or |
| 112 | // TODO: 4.3. Until it stops tokenising. |
| 113 | } else if (token.type == 'eof' || |
| 114 | (token.type == 'end-tag' && token.value == 'script')) { |
| 115 | // 6. Switched back to the PCDATA state. |
| 116 | this.parseMode = 'pcdata'; |
| 117 | |
| 118 | // 7.1. If the next token is not an end tag token with ... |
| 119 | if (token.type != 'end-tag') { |
| 120 | // 7.2. This is a parse error. |
| 121 | log ('Parse error: no </' + 'script>'); |
| 122 | |
| 123 | // 7.3. Mark the script element as "already executed". |
| 124 | el.manakaiAlreadyExecuted = true; |
| 125 | } else { |
| 126 | // 7.4. Ignore it. |
| 127 | // |
| 128 | } |
| 129 | break; |
| 130 | } |
| 131 | } |
| 132 | |
| 133 | // 8.1. If the parser were originally created for the ... |
| 134 | if (this.fragmentParsingMode) { |
| 135 | // 8.2. Mark the script element as "already executed" and ... |
| 136 | el.alreadyExecuted = true; |
| 137 | continue; |
| 138 | } |
| 139 | |
| 140 | // 9.1. Let the old insertion point have the same value as the ... |
| 141 | |
| 142 | // 9.2. Let the insertion point be just before the next input ... |
| 143 | |
| 144 | // 10. Append the new element to the current node. |
| 145 |