| 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303 | /** * html2Json 改造来自: https://github.com/Jxck/html2json *  *  * author: Di (微信小程序开发工程师) * organization: WeAppDev(微信小程序开发论坛)(http://weappdev.com) *               垂直微信小程序开发交流社区 *  * github地址: https://github.com/icindy/wxParse *  * for: 微信小程序富文本解析 * detail : http://weappdev.com/t/wxparse-alpha0-1-html-markdown/184 */var __placeImgeUrlHttps = "https";var __emojisReg = '';var __emojisBaseSrc = '';var __emojis = {};var wxDiscode = require('./wxDiscode.js');var HTMLParser = require('./htmlparser.js');// Empty Elements - HTML 5var empty = makeMap("area,base,basefont,br,col,frame,hr,img,input,link,meta,param,embed,command,keygen,source,track,wbr");// Block Elements - HTML 5var block = makeMap("br,a,code,address,article,applet,aside,audio,blockquote,button,canvas,center,dd,del,dir,div,dl,dt,fieldset,figcaption,figure,footer,form,frameset,h1,h2,h3,h4,h5,h6,header,hgroup,hr,iframe,ins,isindex,li,map,menu,noframes,noscript,object,ol,output,p,pre,section,script,table,tbody,td,tfoot,th,thead,tr,ul,video");// Inline Elements - HTML 5var inline = makeMap("abbr,acronym,applet,b,basefont,bdo,big,button,cite,del,dfn,em,font,i,iframe,img,input,ins,kbd,label,map,object,q,s,samp,script,select,small,span,strike,strong,sub,sup,textarea,tt,u,var");// Elements that you can, intentionally, leave open// (and which close themselves)var closeSelf = makeMap("colgroup,dd,dt,li,options,p,td,tfoot,th,thead,tr");// Attributes that have their values filled in disabled="disabled"var fillAttrs = makeMap("checked,compact,declare,defer,disabled,ismap,multiple,nohref,noresize,noshade,nowrap,readonly,selected");// Special Elements (can contain anything)var special = makeMap("wxxxcode-style,script,style,view,scroll-view,block");function makeMap(str) {    var obj = {}, items = str.split(",");    for (var i = 0; i < items.length; i++)        obj[items[i]] = true;    return obj;}function q(v) {    return '"' + v + '"';}function removeDOCTYPE(html) {    return html        .replace(/<\?xml.*\?>\n/, '')        .replace(/<.*!doctype.*\>\n/, '')        .replace(/<.*!DOCTYPE.*\>\n/, '');}function trimHtml(html) {  return html        .replace(/\r?\n+/g, '')        .replace(/<!--.*?-->/ig, '')        .replace(/\/\*.*?\*\//ig, '')        .replace(/[ ]+</ig, '<')}function html2json(html, bindName) {    //处理字符串    html = removeDOCTYPE(html);    html = trimHtml(html);    html = wxDiscode.strDiscode(html);    //生成node节点    var bufArray = [];    var results = {        node: bindName,        nodes: [],        images:[],        imageUrls:[]    };    var index = 0;    HTMLParser(html, {        start: function (tag, attrs, unary) {            //debug(tag, attrs, unary);            // node for this element            var node = {                node: 'element',                tag: tag,            };            if (bufArray.length === 0) {                node.index = index.toString()                index += 1            } else {                var parent = bufArray[0];                if (parent.nodes === undefined) {                    parent.nodes = [];                }                node.index = parent.index + '.' + parent.nodes.length            }            if (block[tag]) {                node.tagType = "block";            } else if (inline[tag]) {                node.tagType = "inline";            } else if (closeSelf[tag]) {                node.tagType = "closeSelf";            }            if (attrs.length !== 0) {                node.attr = attrs.reduce(function (pre, attr) {                    var name = attr.name;                    var value = attr.value;                    if (name == 'class') {                        console.dir(value);                        //  value = value.join("")                        node.classStr = value;                    }                    // has multi attibutes                    // make it array of attribute                    if (name == 'style') {                        console.dir(value);                        //  value = value.join("")                        node.styleStr = value;                    }                    if (value.match(/ /)) {                        value = value.split(' ');                    }                                        // if attr already exists                    // merge it                    if (pre[name]) {                        if (Array.isArray(pre[name])) {                            // already array, push to last                            pre[name].push(value);                        } else {                            // single value, make it array                            pre[name] = [pre[name], value];                        }                    } else {                        // not exist, put it                        pre[name] = value;                    }                    return pre;                }, {});            }            //对img添加额外数据            if (node.tag === 'img') {                node.imgIndex = results.images.length;                var imgUrl = node.attr.src;                if (imgUrl[0] == '') {                    imgUrl.splice(0, 1);                }                imgUrl = wxDiscode.urlToHttpUrl(imgUrl, __placeImgeUrlHttps);                node.attr.src = imgUrl;                node.from = bindName;                results.images.push(node);                results.imageUrls.push(imgUrl);            }                        // 处理font标签样式属性            if (node.tag === 'font') {                var fontSize = ['x-small', 'small', 'medium', 'large', 'x-large', 'xx-large', '-webkit-xxx-large'];                var styleAttrs = {                    'color': 'color',                    'face': 'font-family',                    'size': 'font-size'                };                if (!node.attr.style) node.attr.style = [];                if (!node.styleStr) node.styleStr = '';                for (var key in styleAttrs) {                    if (node.attr[key]) {                        var value = key === 'size' ? fontSize[node.attr[key]-1] : node.attr[key];                        node.attr.style.push(styleAttrs[key]);                        node.attr.style.push(value);                        node.styleStr += styleAttrs[key] + ': ' + value + ';';                    }                }            }            //临时记录source资源            if(node.tag === 'source'){                results.source = node.attr.src;            }                        if (unary) {                // if this tag doesn't have end tag                // like <img src="hoge.png"/>                // add to parents                var parent = bufArray[0] || results;                if (parent.nodes === undefined) {                    parent.nodes = [];                }                parent.nodes.push(node);            } else {                bufArray.unshift(node);            }        },        end: function (tag) {            //debug(tag);            // merge into parent tag            var node = bufArray.shift();            if (node.tag !== tag) console.error('invalid state: mismatch end tag');            //当有缓存source资源时于于video补上src资源            if(node.tag === 'video' && results.source){                node.attr.src = results.source;                delete results.source;            }                        if (bufArray.length === 0) {                results.nodes.push(node);            } else {                var parent = bufArray[0];                if (parent.nodes === undefined) {                    parent.nodes = [];                }                parent.nodes.push(node);            }        },        chars: function (text) {            //debug(text);            var node = {                node: 'text',                text: text,                textArray:transEmojiStr(text)            };                        if (bufArray.length === 0) {                node.index = index.toString()                index += 1                results.nodes.push(node);            } else {                var parent = bufArray[0];                if (parent.nodes === undefined) {                    parent.nodes = [];                }                node.index = parent.index + '.' + parent.nodes.length                parent.nodes.push(node);            }        },        comment: function (text) {            //debug(text);            // var node = {            //     node: 'comment',            //     text: text,            // };            // var parent = bufArray[0];            // if (parent.nodes === undefined) {            //     parent.nodes = [];            // }            // parent.nodes.push(node);        },    });    return results;};function transEmojiStr(str){  // var eReg = new RegExp("["+__reg+' '+"]");//   str = str.replace(/\[([^\[\]]+)\]/g,':$1:')    var emojiObjs = [];  //如果正则表达式为空  if(__emojisReg.length == 0 || !__emojis){      var emojiObj = {}      emojiObj.node = "text";      emojiObj.text = str;      array = [emojiObj];      return array;  }  //这个地方需要调整  str = str.replace(/\[([^\[\]]+)\]/g,':$1:')  var eReg = new RegExp("[:]");  var array = str.split(eReg);  for(var i = 0; i < array.length; i++){    var ele = array[i];    var emojiObj = {};    if(__emojis[ele]){      emojiObj.node = "element";      emojiObj.tag = "emoji";      emojiObj.text = __emojis[ele];      emojiObj.baseSrc= __emojisBaseSrc;    }else{      emojiObj.node = "text";      emojiObj.text = ele;    }    emojiObjs.push(emojiObj);  }    return emojiObjs;}function emojisInit(reg='',baseSrc="/wxParse/emojis/",emojis){    __emojisReg = reg;    __emojisBaseSrc=baseSrc;    __emojis=emojis;}module.exports = {    html2json: html2json,    emojisInit:emojisInit};
 |