--- /dev/null
+\r
+/*\r
+ * Original code by Erik John Resig (ejohn.org)\r
+ * http://ejohn.org/blog/pure-javascript-html-parser/\r
+ *\r
+ */\r
+\r
+X_Dom_Parser = {\r
+ CHARS : {\r
+ A:1,B:1,C:1,D:1,E:1,F:1,G:1,H:1,I:1,J:1,K:1,L:1,M:1,N:1,O:1,P:1,Q:1,R:1,S:1,T:1,U:1,V:1,W:1,X:1,Y:1,Z:1,\r
+ a:2,b:2,c:2,d:2,e:2,f:2,g:2,h:2,i:2,j:2,k:2,l:2,m:2,n:2,o:2,p:2,q:2,r:2,s:2,t:2,u:2,v:2,w:2,x:2,y:2,z:2,\r
+ '0' : 4, '1' : 4, '2' : 4, '3' : 4, '4' : 4, '5' : 4, '6' : 4, '7' : 4, '8' : 4, '9' : 4,\r
+ \r
+ '\t' : 16, '\r\n' : 16, '\r' : 16, '\n' : 16, '\f' : 16, '\b' : 16, ' ' : 16\r
+ },\r
+ alphabets : 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz',\r
+ whiteSpace : '\t\r\n\f\b ',\r
+\r
+ // Empty Elements - HTML 4.01\r
+ empty : X.Dom.DTD.EMPTY,\r
+\r
+ // Block Elements - HTML 4.01\r
+ block : {ADDRESS:1,APPLET:1,BLOCKQUOTE:1,BUTTON:1,CENTER:1,DD:1,DEL:1,DIR:1,DIV:1,DL:1,DT:1,FIELDSET:1,FORM:1,FRAMESET:1,HR:1,IFRAME:1,INS:1,\r
+ ISINDEX:1,LI:1,MAP:1,MENU:1,NOFRAMES:1,NOSCRIPT:1,OBJECT:1,OL:1,P:1,PRE:1,SCRIPT:1,TABLE:1,TBODY:1,TD:1,TFOOT:1,TH:1,THEAD:1,TR:1,UL:1 },\r
+ // Inline Elements - HTML 4.01\r
+ inline : {A:1,ABBR:1,ACRONYM:1,APPLET:1,B:1,BASEFONT:1,BDO:1,BIG:1,BR:1,BUTTON:1,CITE:1,CODE:1,DEL:1,DFN:1,EM:1,FONT:1,I:1,IFRAME:1,IMG:1,\r
+ INPUT:1,INS:1,KBD:1,LABEL:1,MAP:1,OBJECT:1,Q:1,S:1,SAMP:1,SCRIPT:1,SELECT:1,SMALL:1,SPAN:1,STRIKE:1,STRONG:1,SUB:1,SUP:1,TEXTAREA:1,TT:1,U:1,VAR:1},\r
+ // Elements that you can, intentionally, leave open\r
+ // (and which close themselves)\r
+ closeSelf : {OLGROUP:1,DD:1,DT:1,LI:1,OPTIONS:1,P:1,TBODY:1,TD:1,TFOOT:1,TH:1,THEAD:1,TR:1}, // add tbody\r
+\r
+ sisters : {\r
+ TH : { TD : 1 },\r
+ TD : { TH : 1 },\r
+ DT : { DD : 1 },\r
+ DD : { DT : 1 },\r
+ COLGROUP : { CAPTION : 1 },\r
+ THEAD : { CAPTION : 1, COLGROUP : 1 },\r
+ TFOOT : { CAPTION : 1, COLGROUP : 1, THEAD : 1, TBODY : 1 },\r
+ TBODY : { CAPTION : 1, COLGROUP : 1, THEAD : 1, TFOOT : 1 }\r
+ },\r
+ /*\r
+ * http://www.tohoho-web.com/html/tbody.htm\r
+ * HTML4.01では、ヘッダとフッタを先読みして表示するために、<tbody> よりも <tfoot> の方を先に記述しなくてはならないと定義されています。\r
+ * IE5.0 などでは HEAD → BODY → FOOT の順に表示するのですが、\r
+ * <tfoot> に未対応の古いブラウザでは、HEAD → FOOT → BODY の順に表示されてしまいます。\r
+ * また、HTML5 では、<tfoot> と <tbody> の順番はどちらでもよいことになりました。\r
+ */\r
+\r
+ // Attributes that have their values filled in disabled="disabled"\r
+ fillAttrs : X.Dom.Attr.noValue, //{checked:1,compact:1,declare:1,defer:1,disabled:1,ismap:1,multiple:1,nohref:1,noresize:1,noshade:1,nowrap:1,readonly:1,selected:1};\r
+\r
+ // Special Elements (can contain anything)\r
+ special : { SCRIPT : 1, STYLE : 1, PLAINTEXT : 1, XMP : 1, TEXTAREA : 1 },\r
+ \r
+ exec : function( html, handler, async ){\r
+ var special = X_Dom_Parser.special,\r
+ //plainText = X_Dom_Parser.plainText,\r
+ startTime = async && X.getTime(),\r
+ _parseStartTag = X_Dom_Parser._parseStartTag,\r
+ _parseEndTag = X_Dom_Parser._parseEndTag,\r
+ stack = async ? async[ 1 ] : [],\r
+ lastHtml = html,\r
+ chars, last, text, index;\r
+\r
+ while ( html ) {\r
+ chars = true;\r
+ last = stack[ stack.length - 1 ];\r
+ \r
+ // Make sure we're not in a script or style element\r
+ if ( last && special[ last ] === 1 ) {\r
+ if( 0 <= ( index = html.toUpperCase().indexOf( '</' + last ) ) ){\r
+ handler.chars( html.substring( 0, index ) );\r
+ if( index = _parseEndTag( stack, handler, html ) ){\r
+ html = html.substring( index );\r
+ } else {\r
+ handler.chars( html );\r
+ html = '';\r
+ };\r
+ } else {\r
+ handler.chars( html );\r
+ html = '';\r
+ };\r
+ } else {\r
+ // Comment\r
+ if ( html.indexOf("<!--") === 0 ) {\r
+ if ( 0 < ( index = html.indexOf("-->") ) ) {\r
+ handler.comment( html.substring( 4, index ) );\r
+ html = html.substring( index + 3 );\r
+ chars = false;\r
+ };\r
+ \r
+ // end tag\r
+ } else if ( html.indexOf("</") === 0 ) {\r
+ if ( 2 < ( index = _parseEndTag( stack, handler, html ) ) ) {\r
+ html = html.substring( index );\r
+ chars = false;\r
+ };\r
+ \r
+ // start tag\r
+ } else if ( html.indexOf("<") === 0 ) {\r
+ if( index = _parseStartTag( stack, last, handler, html ) ){\r
+ html = html.substring( index );\r
+ chars = false;\r
+ } else\r
+ if( index === false ){\r
+ return;\r
+ };\r
+ };\r
+\r
+ if ( chars ) {\r
+ index = html.indexOf("<");\r
+ \r
+ text = index < 0 ? html : html.substring( 0, index );\r
+ html = index < 0 ? '' : html.substring( index );\r
+ \r
+ handler.chars( text );\r
+ };\r
+\r
+ };\r
+\r
+ if( html === lastHtml ){\r
+ handler.err( html );\r
+ return;\r
+ };\r
+ \r
+ if( async && startTime + 15 <= X.getTime() && html ){\r
+ handler.progress( 1 - html.length / async[ 0 ] );\r
+ X.Timer.once( 0, X_Dom_Parser.exec, [ html, handler, async ] );\r
+ return;\r
+ };\r
+ \r
+ lastHtml = html;\r
+ };\r
+ \r
+ // Clean up any remaining tags\r
+ X_Dom_Parser.parseEndTag( stack, handler );\r
+ \r
+ async && handler.complete();\r
+ },\r
+\r
+ _parseStartTag : function( stack, last, handler, html ){\r
+ var alphabets = X_Dom_Parser.CHARS,\r
+ whiteSpace = X_Dom_Parser.CHARS,\r
+ saveAttr = X_Dom_Parser.saveAttr,\r
+ uri = X.Dom.DTD.ATTR_VAL_IS_URI,\r
+ phase = 0,\r
+ l = html.length,\r
+ i = 0,\r
+ attrs = [],\r
+ tagName, empty = false,\r
+ chr, start, attrName, quot, escape;\r
+ \r
+ while( i < l && phase < 9 ){\r
+ chr = html.charAt( i );\r
+ switch( phase ){\r
+ case 0 :\r
+ chr === '<' && ( ++phase );\r
+ break;\r
+ case 1 : // タグ名の開始を待つ\r
+ ( alphabets[ chr ] & 3 ) && ( ++phase && ( start = i ) );\r
+ break;\r
+ case 2 : // タグ名の終わりの空白文字を待つ\r
+ ( whiteSpace[ chr ] & 16 ) ?\r
+ ( ++phase && ( tagName = html.substring( start, i ) ) ) :\r
+ ( chr === '>' || ( empty = html.substr( i, 2 ) === '/>' ) ) &&\r
+ ( ( tagName = html.substring( start, i ) ) && ( phase = 9 ) );\r
+ break;\r
+ case 3 : // 属性名の開始を待つ\r
+ ( alphabets[ chr ] & 3 ) ?\r
+ ( ++phase && ( start = i ) ) :\r
+ ( chr === '>' || ( empty = html.substr( i, 2 ) === '/>' ) ) &&\r
+ ( phase = 9 );\r
+ break;\r
+ case 4 : // 属性名の終わりを待つ\r
+ chr === '=' ?\r
+ ( ( phase = 6 ) && ( attrName = html.substring( start, i ) ) ) :\r
+ ( whiteSpace[ chr ] & 16 ) &&\r
+ ( ( phase = 5 ) && ( attrName = html.substring( start, i ) ) );\r
+ break;\r
+ case 5 : // 属性の = または次の属性または htmlタグの閉じ\r
+ ( whiteSpace[ chr ] & 16 ) ?// ie4 未対応の属性には cite = http:// となる\r
+ 1 :\r
+ ( alphabets[ chr ] & 3 ) ?\r
+ ( ( phase = 4 ) && ( attrs[ attrs.length ] = attrName ) && ( start = i ) ) :\r
+ chr === '=' ?\r
+ ( phase = 6 ) :\r
+ ( chr === '>' || ( empty = html.substr( i, 2 ) === '/>' ) ) &&\r
+ ( ( phase = 9 ) && ( attrs[ attrs.length ] = attrName ) );\r
+ break;\r
+ case 6 : // 属性値の開始 quot を待つ\r
+ ( chr === '"' || chr === "'" ) ?\r
+ ( ( phase = 7 ) && ( quot = chr ) && ( start = i + 1 ) ):\r
+ !( whiteSpace[ chr ] & 16 ) &&\r
+ ( ( phase = 8 ) && ( start = i ) ); // no quot\r
+ break;\r
+ case 7 : //属性値の閉じ quot を待つ\r
+ !escape && ( chr === quot ) && ( phase = 3 ) && saveAttr( attrs, attrName, html.substring( start, i ) );\r
+ break;\r
+ case 8 : //閉じ quot のない属性の値\r
+ ( whiteSpace[ chr ] & 16 ) ?\r
+ ( ( phase = 3 ) && saveAttr( attrs, attrName, html.substring( start, i ) ) ) :\r
+ ( chr === '>' ) ?\r
+ ( ( phase = 9 ) && saveAttr( attrs, attrName, html.substring( start, i ) ) ) :\r
+ !escape && !uri[ attrName ] && ( empty = html.substr( i, 2 ) === '/>' ) && // attr の val が uri で / で終わりかつ、未対応属性の場合\r
+ ( phase = 9 );\r
+ break;\r
+ };\r
+ escape = chr === '\\' && !escape; // \\\\ is not escape for "\r
+ ++i;\r
+ };\r
+ if( phase === 9 ){\r
+ if( empty ) ++i;\r
+ if( X_Dom_Parser.parseStartTag( stack, last, handler, tagName.toUpperCase(), attrs, empty, i ) === false ) return false;\r
+ return i;\r
+ };\r
+ return 0; // error\r
+ },\r
+\r
+ _parseEndTag : function( stack, handler, html ){\r
+ var alphabets = X_Dom_Parser.CHARS,\r
+ whiteSpace = X_Dom_Parser.CHARS,\r
+ phase = 0,\r
+ l = html.length,\r
+ i = 0,\r
+ tagName,\r
+ chr, start;\r
+ \r
+ while( i < l && phase < 9 ){\r
+ chr = html.charAt( i );\r
+ switch( phase ){\r
+ case 0 :\r
+ html.substr( i, 2 ) === '</' && ( ++phase && ++i );\r
+ break;\r
+ case 1 : // タグ名の開始を待つ\r
+ ( alphabets[ chr ] & 3 ) && ( ++phase && ( start = i ) );\r
+ break;\r
+ case 2 : // タグ名の終わりの空白文字を待つ\r
+ ( whiteSpace[ chr ] & 16 ) && ( ++phase );\r
+ ( chr === '>' ) && ( phase = 9 );\r
+ ( phase !== 2 ) && ( tagName = html.substring( start, i ) );\r
+ break;\r
+ case 3 : // タグの終了を待つ\r
+ chr === '>' && ( phase = 9 );\r
+ break;\r
+ };\r
+ ++i;\r
+ };\r
+ if( phase === 9 ){\r
+ X_Dom_Parser.parseEndTag( stack, handler, tagName.toUpperCase() );\r
+ return i;\r
+ };\r
+ return 0; // error\r
+ },\r
+\r
+ saveAttr : function( attrs, name, value ){\r
+ name = name.toLowerCase();\r
+ value = X_Dom_Parser.fillAttrs[ name ] === 1 ? name : value;\r
+ attrs[ attrs.length ] = {\r
+ name : name,\r
+ value : value,\r
+ escaped :\r
+ value.indexOf( '"' ) !== -1 ?\r
+ value.split( '"' ).join( '\\"' ).split( '\\\\"' ).join( '\\"' ) :\r
+ value\r
+ };\r
+ },\r
+\r
+ parseStartTag : function( stack, last, handler, tagName, attrs, unary, index ) {\r
+ var inline = X_Dom_Parser.inline,\r
+ parseEndTag = X_Dom_Parser.parseEndTag,\r
+ sisters = X_Dom_Parser.sisters;\r
+ if ( X_Dom_Parser.block[ tagName ] === 1 ) {\r
+ while ( last && inline[ last ] === 1 ) {\r
+ parseEndTag( stack, handler, last );\r
+ last = stack[ stack.length - 1 ];\r
+ };\r
+ };\r
+ last && X_Dom_Parser.closeSelf[ tagName ] === 1 && ( last === tagName || ( sisters[ tagName ] && sisters[ tagName ][ last ] === 1 ) ) && parseEndTag( stack, handler, last );\r
+ unary = unary || X_Dom_Parser.empty[ tagName ];\r
+ !unary && ( stack[ stack.length ] = tagName );\r
+ \r
+ return handler.start( tagName, attrs, unary, index );\r
+ },\r
+\r
+ parseEndTag : function( stack, handler, tagName ) {\r
+ var pos = 0, i = stack.length;\r
+ // If no tag name is provided, clean shop\r
+ \r
+ // Find the closest opened tag of the same type\r
+ if ( tagName )\r
+ for ( pos = i; 0 <= pos; )\r
+ if ( stack[ --pos ] === tagName )\r
+ break;\r
+ \r
+ if ( 0 <= pos ) {\r
+ // Close all the open elements, up the stack\r
+ for ( ; pos < i; )\r
+ handler.end( stack[ --i ] );\r
+ \r
+ // Remove the open elements from the stack\r
+ stack.length = pos;\r
+ };\r
+ }\r
+ \r
+};\r
+\r
+X.Dom._htmlStringToXNode = {\r
+ flat : null,\r
+ nest : [],\r
+ err : function( html ){\r
+ X.Dom._htmlStringToXNode.flat.length = 0;\r
+ X.Dom._htmlStringToXNode.ignoreError !== true && X.Logger.warn( 'X_Dom_Parser() error ' + html );\r
+ },\r
+ start : function( tagName, attrs, noChild, length ){\r
+ var xnode,\r
+ nest = X.Dom._htmlStringToXNode.nest,\r
+ flat = X.Dom._htmlStringToXNode.flat,\r
+ l = nest.length,\r
+ attr, name, i, _attrs; //, toIndex;\r
+ if( l ){\r
+ xnode = nest[ l - 1 ].create( tagName );\r
+ } else {\r
+ xnode = flat[ flat.length ] = X.Dom.Node.create( tagName );\r
+ };\r
+ if( !noChild ) nest[ l ] = xnode;\r
+ if( i = attrs.length ){\r
+ _attrs = {};\r
+ for( ; i; ){\r
+ if( attr = attrs[ --i ] ){\r
+ if( typeof attr === 'string' ){\r
+ name = attr;\r
+ _attrs[ name ] = true;\r
+ } else {\r
+ name = attr.name;\r
+ _attrs[ name ] = attr.escaped;\r
+ };\r
+ };\r
+ };\r
+ xnode.attr( _attrs );\r
+ };\r
+ },\r
+ end : function(){\r
+ 0 < X.Dom._htmlStringToXNode.nest.length && ( --X.Dom._htmlStringToXNode.nest.length );\r
+ },\r
+ chars : function( text ){\r
+ if( X.Dom._htmlStringToXNode.nest.length ){\r
+ X.Dom._htmlStringToXNode.nest[ X.Dom._htmlStringToXNode.nest.length - 1 ].createText( text );\r
+ } else {\r
+ X.Dom._htmlStringToXNode.flat[ X.Dom._htmlStringToXNode.flat.length ] = X.Dom.Node.createText( text );\r
+ };\r
+ },\r
+ comment : X.emptyFunction\r
+};\r
+\r
+X.Dom.parse = function( html, ignoreError ){\r
+ var worker = X.Dom._htmlStringToXNode, ret;\r
+ worker.flat = [];\r
+ worker.nest.length = 0;\r
+ worker.ignoreError = ignoreError;\r
+ X_Dom_Parser.exec( html, worker );\r
+ ret = worker.flat;\r
+ delete worker.flat;\r
+ return ret;\r
+};\r
+\r
+X.Dom._asyncHtmlStringToXNode = {\r
+ err : function( html ){\r
+ X.Dom._htmlStringToXNode.err( html );\r
+ this.asyncDispatch( 0, { type : X.Event.ERROR } );\r
+ },\r
+ start : X.Dom._htmlStringToXNode.start,\r
+ end : X.Dom._htmlStringToXNode.end,\r
+ chars : X.Dom._htmlStringToXNode.chars,\r
+ comment : X.emptyFunction,\r
+ \r
+ progress : function( pct ){\r
+ this.asyncDispatch( 0, { type : X.Event.PROGRESS, percent : pct } );\r
+ },\r
+ complete : function(){\r
+ var ret = X.Dom._htmlStringToXNode.flat;\r
+ delete X.Dom._htmlStringToXNode.flat;\r
+ this.asyncDispatch( 0, { type : X.Event.SUCCESS, xnodes : ret } );\r
+ }\r
+};\r
+\r
+X.Dom.asyncParse = function( html, ignoreError ){\r
+ var dispatcher = X.Class._override( new X.EventDispatcher(), X.Dom._asyncHtmlStringToXNode ),\r
+ worker = X.Dom._htmlStringToXNode;\r
+ dispatcher.listenOnce( X.Event.SUCCESS, dispatcher, dispatcher.kill );\r
+ worker.flat = [];\r
+ worker.nest.length = 0;\r
+ worker.ignoreError = ignoreError;\r
+ X_Dom_Parser.exec( html, dispatcher, [ html.length, [] ] );\r
+ return dispatcher;\r
+};\r
OSDN Git Service
