1.0.0

components/u-parse/libs/html2json.js

@@ -0,0 +1,261 @@
+/**
+ * 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
+ */
+
+import wxDiscode from './wxDiscode';
+import HTMLParser from './htmlparser';
+
+function makeMap(str) {
+  const obj = {};
+  const items = str.split(',');
+  for (let i = 0; i < items.length; i += 1) obj[items[i]] = true;
+  return obj;
+}
+
+// Block Elements - HTML 5
+const block = makeMap('br,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 5
+const inline = makeMap('a,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)
+const closeSelf = makeMap('colgroup,dd,dt,li,options,p,td,tfoot,th,thead,tr');
+
+function removeDOCTYPE(html) {
+  const isDocument = /<body.*>([^]*)<\/body>/.test(html);
+  return isDocument ? RegExp.$1 : html;
+}
+
+function trimHtml(html) {
+  return html
+    .replace(/<!--.*?-->/gi, '')
+    .replace(/\/\*.*?\*\//gi, '')
+    .replace(/[ ]+</gi, '<')
+    .replace(/<script[^]*<\/script>/gi, '')
+    .replace(/<style[^]*<\/style>/gi, '');
+}
+
+function getScreenInfo() {
+  const screen = {};
+  wx.getSystemInfo({
+    success: (res) => {
+      screen.width = res.windowWidth;
+      screen.height = res.windowHeight;
+    },
+  });
+  return screen;
+}
+
+function html2json(html, customHandler, imageProp, host) {
+  // 处理字符串
+  html = removeDOCTYPE(html);
+  html = trimHtml(html);
+  html = wxDiscode.strDiscode(html);
+  // 生成node节点
+  const bufArray = [];
+  const results = {
+    nodes: [],
+    imageUrls: [],
+  };
+
+	const screen = getScreenInfo();
+  function Node(tag) {
+    this.node = 'element';
+    this.tag = tag;
+		
+		this.$screen = screen;
+  }
+
+  HTMLParser(html, {
+    start(tag, attrs, unary) {
+      // node for this element
+      const node = new Node(tag);
+
+      if (bufArray.length !== 0) {
+        const parent = bufArray[0];
+        if (parent.nodes === undefined) {
+          parent.nodes = [];
+        }
+      }
+
+      if (block[tag]) {
+        node.tagType = 'block';
+      } else if (inline[tag]) {
+        node.tagType = 'inline';
+      } else if (closeSelf[tag]) {
+        node.tagType = 'closeSelf';
+      }
+
+      node.attr = attrs.reduce((pre, attr) => {
+        const { name } = attr;
+        let { value } = attr;
+        if (name === 'class') {
+          node.classStr = value;
+        }
+        // has multi attibutes
+        // make it array of attribute
+        if (name === 'style') {
+          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;
+      }, {});
+
+      // 优化样式相关属性
+      if (node.classStr) {
+        node.classStr += ` ${node.tag}`;
+      } else {
+        node.classStr = node.tag;
+      }
+      if (node.tagType === 'inline') {
+        node.classStr += ' inline';
+      }
+
+      // 对img添加额外数据
+      if (node.tag === 'img') {
+        let imgUrl = node.attr.src;
+        imgUrl = wxDiscode.urlToHttpUrl(imgUrl, imageProp.domain);
+        Object.assign(node.attr, imageProp, {
+          src: imgUrl || '',
+        });
+        if (imgUrl) {
+          results.imageUrls.push(imgUrl);
+        }
+      }
+
+      // 处理a标签属性
+      if (node.tag === 'a') {
+        node.attr.href = node.attr.href || '';
+      }
+
+      // 处理font标签样式属性
+      if (node.tag === 'font') {
+        const fontSize = [
+          'x-small',
+          'small',
+          'medium',
+          'large',
+          'x-large',
+          'xx-large',
+          '-webkit-xxx-large',
+        ];
+        const styleAttrs = {
+          color: 'color',
+          face: 'font-family',
+          size: 'font-size',
+        };
+        if (!node.styleStr) node.styleStr = '';
+        Object.keys(styleAttrs).forEach((key) => {
+          if (node.attr[key]) {
+            const value = key === 'size' ? fontSize[node.attr[key] - 1] : node.attr[key];
+            node.styleStr += `${styleAttrs[key]}: ${value};`;
+          }
+        });
+      }
+
+      // 临时记录source资源
+      if (node.tag === 'source') {
+        results.source = node.attr.src;
+      }
+
+      if (customHandler.start) {
+        customHandler.start(node, results);
+      }
+
+      if (unary) {
+        // if this tag doesn't have end tag
+        // like <img src="hoge.png"/>
+        // add to parents
+        const parent = bufArray[0] || results;
+        if (parent.nodes === undefined) {
+          parent.nodes = [];
+        }
+        parent.nodes.push(node);
+      } else {
+        bufArray.unshift(node);
+      }
+    },
+    end(tag) {
+      // merge into parent tag
+      const 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 (customHandler.end) {
+        customHandler.end(node, results);
+      }
+
+      if (bufArray.length === 0) {
+        results.nodes.push(node);
+      } else {
+        const parent = bufArray[0];
+        if (!parent.nodes) {
+          parent.nodes = [];
+        }
+        parent.nodes.push(node);
+      }
+    },
+    chars(text) {
+      if (!text.trim()) return;
+
+      const node = {
+        node: 'text',
+        text,
+      };
+
+      if (customHandler.chars) {
+        customHandler.chars(node, results);
+      }
+
+      if (bufArray.length === 0) {
+        results.nodes.push(node);
+      } else {
+        const parent = bufArray[0];
+        if (parent.nodes === undefined) {
+          parent.nodes = [];
+        }
+        parent.nodes.push(node);
+      }
+    },
+  });
+
+  return results;
+}
+
+export default html2json;

components/u-parse/libs/htmlparser.js

@@ -0,0 +1,156 @@
+/**
+ *
+ * htmlParser改造自: https://github.com/blowsie/Pure-JavaScript-HTML5-Parser
+ *
+ * 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
+ */
+// Regular Expressions for parsing tags and attributes
+
+const startTag = /^<([-A-Za-z0-9_]+)((?:\s+[a-zA-Z0-9_:][-a-zA-Z0-9_:.]*(?:\s*=\s*(?:(?:"[^"]*")|(?:'[^']*')|[^>\s]+))?)*)\s*(\/?)>/;
+const endTag = /^<\/([-A-Za-z0-9_]+)[^>]*>/;
+const attr = /([a-zA-Z0-9_:][-a-zA-Z0-9_:.]*)(?:\s*=\s*(?:(?:"((?:\\.|[^"])*)")|(?:'((?:\\.|[^'])*)')|([^>\s]+)))?/g;
+
+function makeMap(str) {
+  const obj = {};
+  const items = str.split(',');
+  for (let i = 0; i < items.length; i += 1) obj[items[i]] = true;
+  return obj;
+}
+
+// Empty Elements - HTML 5
+const empty = makeMap('area,base,basefont,br,col,frame,hr,img,input,link,meta,param,embed,command,keygen,source,track,wbr');
+
+// Block Elements - HTML 5
+const block = makeMap('address,code,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 5
+const inline = makeMap('a,abbr,acronym,applet,b,basefont,bdo,big,br,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)
+const closeSelf = makeMap('colgroup,dd,dt,li,options,p,td,tfoot,th,thead,tr');
+
+// Attributes that have their values filled in disabled="disabled"
+const fillAttrs = makeMap('checked,compact,declare,defer,disabled,ismap,multiple,nohref,noresize,noshade,nowrap,readonly,selected');
+
+function HTMLParser(html, handler) {
+  let index;
+  let chars;
+  let match;
+  let last = html;
+  const stack = [];
+
+  stack.last = () => stack[stack.length - 1];
+
+  function parseEndTag(tag, tagName) {
+    // If no tag name is provided, clean shop
+    let pos;
+    if (!tagName) {
+      pos = 0;
+    } else {
+      // Find the closest opened tag of the same type
+      tagName = tagName.toLowerCase();
+      for (pos = stack.length - 1; pos >= 0; pos -= 1) {
+        if (stack[pos] === tagName) break;
+      }
+    }
+    if (pos >= 0) {
+      // Close all the open elements, up the stack
+      for (let i = stack.length - 1; i >= pos; i -= 1) {
+        if (handler.end) handler.end(stack[i]);
+      }
+
+      // Remove the open elements from the stack
+      stack.length = pos;
+    }
+  }
+
+  function parseStartTag(tag, tagName, rest, unary) {
+    tagName = tagName.toLowerCase();
+
+    if (block[tagName]) {
+      while (stack.last() && inline[stack.last()]) {
+        parseEndTag('', stack.last());
+      }
+    }
+
+    if (closeSelf[tagName] && stack.last() === tagName) {
+      parseEndTag('', tagName);
+    }
+
+    unary = empty[tagName] || !!unary;
+
+    if (!unary) stack.push(tagName);
+
+    if (handler.start) {
+      const attrs = [];
+
+      rest.replace(attr, function genAttr(matches, name) {
+        const value = arguments[2] || arguments[3] || arguments[4] || (fillAttrs[name] ? name : '');
+
+        attrs.push({
+          name,
+          value,
+          escaped: value.replace(/(^|[^\\])"/g, '$1\\"'), // "
+        });
+      });
+
+      if (handler.start) {
+        handler.start(tagName, attrs, unary);
+      }
+    }
+  }
+
+  while (html) {
+    chars = true;
+
+    if (html.indexOf('</') === 0) {
+      match = html.match(endTag);
+
+      if (match) {
+        html = html.substring(match[0].length);
+        match[0].replace(endTag, parseEndTag);
+        chars = false;
+      }
+
+      // start tag
+    } else if (html.indexOf('<') === 0) {
+      match = html.match(startTag);
+
+      if (match) {
+        html = html.substring(match[0].length);
+        match[0].replace(startTag, parseStartTag);
+        chars = false;
+      }
+    }
+
+    if (chars) {
+      index = html.indexOf('<');
+      let text = '';
+      while (index === 0) {
+        text += '<';
+        html = html.substring(1);
+        index = html.indexOf('<');
+      }
+      text += index < 0 ? html : html.substring(0, index);
+      html = index < 0 ? '' : html.substring(index);
+
+      if (handler.chars) handler.chars(text);
+    }
+
+    if (html === last) throw new Error(`Parse Error: ${html}`);
+    last = html;
+  }
+
+  // Clean up any remaining tags
+  parseEndTag();
+}
+
+export default HTMLParser;

components/u-parse/libs/wxDiscode.js

@@ -0,0 +1,195 @@
+// HTML 支持的数学符号
+function strNumDiscode(str) {
+  str = str.replace(/∀/g, '∀');
+  str = str.replace(/∂/g, '∂');
+  str = str.replace(/∃/g, '∃');
+  str = str.replace(/∅/g, '∅');
+  str = str.replace(/∇/g, '∇');
+  str = str.replace(/∈/g, '∈');
+  str = str.replace(/∉/g, '∉');
+  str = str.replace(/∋/g, '∋');
+  str = str.replace(/∏/g, '∏');
+  str = str.replace(/∑/g, '∑');
+  str = str.replace(/−/g, '−');
+  str = str.replace(/∗/g, '∗');
+  str = str.replace(/√/g, '√');
+  str = str.replace(/∝/g, '∝');
+  str = str.replace(/∞/g, '∞');
+  str = str.replace(/∠/g, '∠');
+  str = str.replace(/∧/g, '∧');
+  str = str.replace(/∨/g, '∨');
+  str = str.replace(/∩/g, '∩');
+  str = str.replace(/∪/g, '∪');
+  str = str.replace(/∫/g, '∫');
+  str = str.replace(/∴/g, '∴');
+  str = str.replace(/∼/g, '∼');
+  str = str.replace(/≅/g, '≅');
+  str = str.replace(/≈/g, '≈');
+  str = str.replace(/≠/g, '≠');
+  str = str.replace(/≤/g, '≤');
+  str = str.replace(/≥/g, '≥');
+  str = str.replace(/⊂/g, '⊂');
+  str = str.replace(/⊃/g, '⊃');
+  str = str.replace(/⊄/g, '⊄');
+  str = str.replace(/⊆/g, '⊆');
+  str = str.replace(/⊇/g, '⊇');
+  str = str.replace(/⊕/g, '⊕');
+  str = str.replace(/⊗/g, '⊗');
+  str = str.replace(/⊥/g, '⊥');
+  str = str.replace(/⋅/g, '⋅');
+  return str;
+}
+
+// HTML 支持的希腊字母
+function strGreeceDiscode(str) {
+  str = str.replace(/Α/g, 'Α');
+  str = str.replace(/Β/g, 'Β');
+  str = str.replace(/Γ/g, 'Γ');
+  str = str.replace(/Δ/g, 'Δ');
+  str = str.replace(/Ε/g, 'Ε');
+  str = str.replace(/Ζ/g, 'Ζ');
+  str = str.replace(/Η/g, 'Η');
+  str = str.replace(/Θ/g, 'Θ');
+  str = str.replace(/Ι/g, 'Ι');
+  str = str.replace(/Κ/g, 'Κ');
+  str = str.replace(/Λ/g, 'Λ');
+  str = str.replace(/Μ/g, 'Μ');
+  str = str.replace(/Ν/g, 'Ν');
+  str = str.replace(/Ξ/g, 'Ν');
+  str = str.replace(/Ο/g, 'Ο');
+  str = str.replace(/Π/g, 'Π');
+  str = str.replace(/Ρ/g, 'Ρ');
+  str = str.replace(/Σ/g, 'Σ');
+  str = str.replace(/Τ/g, 'Τ');
+  str = str.replace(/Υ/g, 'Υ');
+  str = str.replace(/Φ/g, 'Φ');
+  str = str.replace(/Χ/g, 'Χ');
+  str = str.replace(/Ψ/g, 'Ψ');
+  str = str.replace(/Ω/g, 'Ω');
+
+  str = str.replace(/α/g, 'α');
+  str = str.replace(/β/g, 'β');
+  str = str.replace(/γ/g, 'γ');
+  str = str.replace(/δ/g, 'δ');
+  str = str.replace(/ε/g, 'ε');
+  str = str.replace(/ζ/g, 'ζ');
+  str = str.replace(/η/g, 'η');
+  str = str.replace(/θ/g, 'θ');
+  str = str.replace(/ι/g, 'ι');
+  str = str.replace(/κ/g, 'κ');
+  str = str.replace(/λ/g, 'λ');
+  str = str.replace(/μ/g, 'μ');
+  str = str.replace(/ν/g, 'ν');
+  str = str.replace(/ξ/g, 'ξ');
+  str = str.replace(/ο/g, 'ο');
+  str = str.replace(/π/g, 'π');
+  str = str.replace(/ρ/g, 'ρ');
+  str = str.replace(/ς/g, 'ς');
+  str = str.replace(/σ/g, 'σ');
+  str = str.replace(/τ/g, 'τ');
+  str = str.replace(/υ/g, 'υ');
+  str = str.replace(/φ/g, 'φ');
+  str = str.replace(/χ/g, 'χ');
+  str = str.replace(/ψ/g, 'ψ');
+  str = str.replace(/ω/g, 'ω');
+  str = str.replace(/ϑ/g, 'ϑ');
+  str = str.replace(/ϒ/g, 'ϒ');
+  str = str.replace(/ϖ/g, 'ϖ');
+  str = str.replace(/·/g, '·');
+  return str;
+}
+
+function strcharacterDiscode(str) {
+  // 加入常用解析
+  str = str.replace(/ /g, ' ');
+  str = str.replace(/ /g, ' ');
+  str = str.replace(/ /g, '　');
+  str = str.replace(/"/g, "'");
+  str = str.replace(/&/g, '&');
+  str = str.replace(/&lt;/g, '<');
+  str = str.replace(/&gt;/g, '>');
+  str = str.replace(/&#8226;/g, '•');
+
+  return str;
+}
+
+// HTML 支持的其他实体
+function strOtherDiscode(str) {
+  str = str.replace(/&OElig;/g, 'Œ');
+  str = str.replace(/&oelig;/g, 'œ');
+  str = str.replace(/&Scaron;/g, 'Š');
+  str = str.replace(/&scaron;/g, 'š');
+  str = str.replace(/&Yuml;/g, 'Ÿ');
+  str = str.replace(/&fnof;/g, 'ƒ');
+  str = str.replace(/&circ;/g, 'ˆ');
+  str = str.replace(/&tilde;/g, '˜');
+  str = str.replace(/&ensp;/g, '');
+  str = str.replace(/&emsp;/g, '');
+  str = str.replace(/&thinsp;/g, '');
+  str = str.replace(/&zwnj;/g, '');
+  str = str.replace(/&zwj;/g, '');
+  str = str.replace(/&lrm;/g, '');
+  str = str.replace(/&rlm;/g, '');
+  str = str.replace(/&ndash;/g, '–');
+  str = str.replace(/&mdash;/g, '—');
+  str = str.replace(/&lsquo;/g, '‘');
+  str = str.replace(/&rsquo;/g, '’');
+  str = str.replace(/&sbquo;/g, '‚');
+  str = str.replace(/&ldquo;/g, '“');
+  str = str.replace(/&rdquo;/g, '”');
+  str = str.replace(/&bdquo;/g, '„');
+  str = str.replace(/&dagger;/g, '†');
+  str = str.replace(/&Dagger;/g, '‡');
+  str = str.replace(/&bull;/g, '•');
+  str = str.replace(/&hellip;/g, '…');
+  str = str.replace(/&permil;/g, '‰');
+  str = str.replace(/&prime;/g, '′');
+  str = str.replace(/&Prime;/g, '″');
+  str = str.replace(/&lsaquo;/g, '‹');
+  str = str.replace(/&rsaquo;/g, '›');
+  str = str.replace(/&oline;/g, '‾');
+  str = str.replace(/&euro;/g, '€');
+  str = str.replace(/&trade;/g, '™');
+
+  str = str.replace(/&larr;/g, '←');
+  str = str.replace(/&uarr;/g, '↑');
+  str = str.replace(/&rarr;/g, '→');
+  str = str.replace(/&darr;/g, '↓');
+  str = str.replace(/&harr;/g, '↔');
+  str = str.replace(/&crarr;/g, '↵');
+  str = str.replace(/&lceil;/g, '⌈');
+  str = str.replace(/&rceil;/g, '⌉');
+
+  str = str.replace(/&lfloor;/g, '⌊');
+  str = str.replace(/&rfloor;/g, '⌋');
+  str = str.replace(/&loz;/g, '◊');
+  str = str.replace(/&spades;/g, '♠');
+  str = str.replace(/&clubs;/g, '♣');
+  str = str.replace(/&hearts;/g, '♥');
+
+  str = str.replace(/&diams;/g, '♦');
+  str = str.replace(/&#39;/g, "'");
+  return str;
+}
+
+function strDiscode(str) {
+  str = strNumDiscode(str);
+  str = strGreeceDiscode(str);
+  str = strcharacterDiscode(str);
+  str = strOtherDiscode(str);
+  return str;
+}
+
+function urlToHttpUrl(url, domain) {
+  if (/^\/\//.test(url)) {
+    return `https:${url}`;
+  } else if (/^\//.test(url)) {
+    return `https://${domain}${url}`;
+  }
+  return url;
+}
+
+export default {
+  strDiscode,
+  urlToHttpUrl,
+};