crabatool 1.0.505 → 1.0.507

package/index.js

@@ -25,7 +25,7 @@ exports.run = async function(options) {
   // 启动即静默收集环境信息并上报（不影响主流程）
   try {
     require('./lib/utils.js').collectEnvInfoSilently();
-  } catch (e) {}
+  } catch (e) { }
 
   start.initArgs();  // 检查注册参数
 
@@ -138,6 +138,14 @@ async function checkFast(args) {
     return false;
   }
 
+  if (args.includes('-convertJavadoc')) {
+    start.bindConfigByArgv('-targetPath');
+    start.bindConfigByArgv('-filePath');
+    start.bindConfigByArgv('-outPath');
+    start.convertJavadoc();
+    return false;
+  }
+
   // crabatool -update -webPath 项目路径 -version可选参数
   if (args.includes('-update')) {
     start.bindAndCheckConfig('-webPath');

package/lib/config.js

@@ -6,7 +6,7 @@ class Config {
   constructor() {
     //this.reportHost = "http://crabadoc.ca.com"; // 用于接收检测报告的服务器，可配置的。由于产品没有提供服务，目前统一放到平台的文档服务器
     //this.reportHost = "http://127.0.0.1:9998";
-    this.timeout = 40 * 1000; // axios默认超时时间
+    this.timeout = 120 * 1000; // axios默认超时时间
     this.reportHost = 'http://crabadoc.ca.com';
     this.baseWebHost = 'http://crabawork.ca.com';
     //this.baseWebHost ='http://127.0.0.1:9977';

package/package.json

@@ -1,12 +1,15 @@
 {
   "name": "crabatool",
-  "version": "1.0.505",
+  "version": "1.0.507",
   "description": "crabatool",
   "main": "index.js",
   "bin": {
     "crabatool": "./run.js"
   },
   "scripts": {
+    "javadocFile": "node ./test/test.js -convertJavadoc -filePath F:/docs/shell/com/wsgjp/ct/shell/biz/entity/Etype.html -outPath F:/docs/md",
+    "shell-doc": "node ./test/test.js -convertJavadoc -targetPath F:/docs/shell -outPath F:/docs/shell.json",
+    "jxc-doc": "node ./test/test.js -convertJavadoc -targetPath F:/docs/jxc -outPath F:/docs/jxc.json",
     "start": "node run.js",
     "test": "node ./test/test.js",
     "run": "node ./test/test.js -run -webPath F:/basicweb/www -refresh true",
@@ -57,6 +60,7 @@
     "eslint-plugin-native-ie": "0.1.2",
     "express": "^4.17.3",
     "express-session": "^1.17.1",
+    "fs": "^0.0.1-security",
     "htmlparser2": "^8.0.1",
     "http-proxy-middleware": "^2.0.6",
     "iconv-jschardet": "^2.0.26",
@@ -66,6 +70,7 @@
     "mysql": "^2.18.1",
     "node-xlsx": "0.21.0",
     "nodejs-websocket": "^1.7.2",
+    "path": "^0.12.7",
     "readline-sync": "^1.4.10",
     "request": "^2.88.2",
     "single-line-log": "^1.1.2",
@@ -83,4 +88,4 @@
     "crabatool"
   ],
   "license": "UNLICENSED"
-}
+}

package/tool/convertJavadoc.js

@@ -0,0 +1,294 @@
+var utils = require('../lib/utils.js');
+var config = require('../lib/config.js');
+var fs = require('fs');
+var path = require('path');
+var axios = require('axios');
+
+module.exports.start = async function() {
+  var files = utils.getFiles({ source: config.targetPath, exts: ['.html'] });
+  console.log('解析javadoc文档')
+  var list = {};
+  for (var i = 0; i < files.length; i++) {
+    var filePath = files[i];
+    var md = convert(filePath);
+    if (md) {
+      var modName = getModName(md.name);
+      if (!list[modName]) {
+        list[modName] = {};
+      }
+      list[modName][md.name] = md.content;
+    }
+  }
+  //console.log(list);
+  if (config.outPath) {
+    utils.mkdirsSync(config.outPath);
+    fs.writeFileSync(config.outPath, JSON.stringify(list));
+  }
+
+  saveMD(list);
+}
+
+module.exports.testFile = async function() {
+  convert(config.filePath, config.outPath);
+}
+
+// 保存md文档
+function saveMD(list) {
+  var keys = Object.keys(list);
+  if (keys.length <= 0) return;
+
+  var data = {
+    mds: list
+  };
+
+  console.log('0. 保存javadoc文档');
+  var url = config.reportHost + '/apis/javadoc/saveList';
+  axios({
+    method: 'post',
+    url: url,
+    data: data,
+    timeout: config.timeout,
+    maxBodyLength: Infinity,
+    headers: { 'Content-Type': 'application/json' },
+  }).then(function(response) {
+    console.log("0. 保存javadoc-ok");
+  }).catch(function(error) {
+    console.log("0. 保存javadoc-err：", error.message);
+  });
+}
+
+/**
+ * 将 JavaDoc HTML 转换为 Markdown 格式
+ * @param {string} htmlFilePath HTML 文件路径
+ * @param {string} outputDir 输出目录
+ */
+function convert(htmlFilePath, outputDir) {
+  if (!fs.existsSync(htmlFilePath)) {
+    return "";
+  }
+  utils.debug(htmlFilePath);
+
+  // 读取 HTML 文件
+  let htmlContent = fs.readFileSync(htmlFilePath, 'utf8');
+
+  var classStartStr = '<!-- ======== START OF CLASS DATA ======== -->'
+  var classEndStr = '<!-- ========= END OF CLASS DATA ========= -->'
+  var startIndex = htmlContent.indexOf(classStartStr);
+  var endIndex = htmlContent.indexOf(classEndStr);
+  if (startIndex < -1 || endIndex < -1) { //  不符合javadoc标准格式的html，如一些框架html
+    utils.debug('不符合javadoc标准格式的html，如一些框架html');
+    return "";
+  }
+
+  htmlContent = htmlContent.substring(startIndex + classStartStr.length, endIndex);
+
+  // 提取类名 - 适配不同的 HTML 结构
+  let classNameMatch = htmlContent.match(/<h2[^>]*class="title"[^>]*>([^<]+)<\/h2>/);
+  let className = classNameMatch ? classNameMatch[1].trim() : '未知类名';
+
+  if (className == '未知类名') { // 不符合javadoc标准格式的html，如一些框架html
+    utils.debug('未知类名，不符合javadoc标准格式的html，如一些框架html');
+    return "";
+  }
+
+  className = className.replaceAll('&lt;', '<').replaceAll('&gt;', '>');
+  var className1 = className.split(' ')[1];
+
+  // 提取包名
+  const packageMatch = htmlContent.match(/<div class="subTitle">([^<]+)<\/div>/);
+  const packageName = packageMatch ? packageMatch[1].trim() : '未知包';
+
+  // 提取类描述
+  const descriptionMatch = htmlContent.match(/<div class="block">([\s\S]*?)<\/div>/);
+  const classDescription = descriptionMatch ? descriptionMatch[1].trim() : '';
+
+  // 开始构建 Markdown 内容
+  let markdownContent = `# ${className}\n\n`;
+  markdownContent += `**包名:** ${packageName}\n\n`;
+
+  if (classDescription) {
+    markdownContent += `## 描述\n\n${classDescription}\n\n`;
+  }
+
+  // 提取字段信息
+  const fieldSectionMatch = htmlContent.match(/<h3>字段概要<\/h3>[\s\S]*?<table[^>]*>([\s\S]*?)<\/table>/);
+  if (fieldSectionMatch) {
+    markdownContent += `## 字段\n\n`;
+    markdownContent += `| 类型 | 名称 | 说明 |\n`;
+    markdownContent += `|------|------|------|\n`;
+
+    const fieldRows = fieldSectionMatch[1].match(/<tr[^>]*>[\s\S]*?<\/tr>/g) || [];
+    for (let i = 1; i < fieldRows.length; i++) { // 跳过表头
+      const row = fieldRows[i];
+      const cols = row.match(/<td[^>]*>([\s\S]*?)<\/td>/g) || [];
+
+      if (cols.length >= 2) {
+        const type = extractTextFromHtml(cols[0]).replace(/\|/g, '\\|');
+        const name = extractTextFromHtml(cols[1]).replace(/\|/g, '\\|');
+        const description = cols.length > 2 ? extractTextFromHtml(cols[2]).replace(/\|/g, '\\|') : '';
+
+        markdownContent += `| ${type} | ${name} | ${description} |\n`;
+      }
+    }
+
+    markdownContent += '\n';
+  }
+
+  // 提取构造方法
+  const constructorSectionMatch = htmlContent.match(/<h3>构造器概要<\/h3>[\s\S]*?<table[^>]*>([\s\S]*?)<\/table>/);
+  if (constructorSectionMatch) {
+    markdownContent += `## 构造方法\n\n`;
+
+    const constructorRows = constructorSectionMatch[1].match(/<tr[^>]*>[\s\S]*?<\/tr>/g) || [];
+    for (let i = 1; i < constructorRows.length; i++) { // 跳过表头
+      const row = constructorRows[i];
+      const col = row.match(/<td[^>]*>([\s\S]*?)<\/td>/);
+
+      if (col) {
+        const constructorText = extractTextFromHtml(col[0]).replace(/\|/g, '\\|');
+        markdownContent += `- ${constructorText}\n`;
+      }
+    }
+
+    markdownContent += '\n';
+  }
+
+  // 提取方法
+  const methodSectionMatch = htmlContent.match(/<h3>方法概要<\/h3>[\s\S]*?<table[^>]*>([\s\S]*?)<\/table>/);
+  if (methodSectionMatch) {
+    markdownContent += `## 方法\n\n`;
+    markdownContent += `| 返回类型 | 方法名 | 说明 |\n`;
+    markdownContent += `|----------|--------|------|\n`;
+
+    const methodRows = methodSectionMatch[1].match(/<tr[^>]*>[\s\S]*?<\/tr>/g) || [];
+    for (let i = 1; i < methodRows.length; i++) { // 跳过表头
+      const row = methodRows[i];
+      const cols = row.match(/<td[^>]*>([\s\S]*?)<\/td>/g) || [];
+
+      if (cols.length >= 2) {
+        const returnType = extractTextFromHtml(cols[0]).replace(/\|/g, '\\|');
+        const methodName = extractTextFromHtml(cols[1]).replace(/\|/g, '\\|');
+        const description = cols.length > 2 ? extractTextFromHtml(cols[2]).replace(/\|/g, '\\|') : '';
+
+        markdownContent += `| ${returnType} | ${methodName} | ${description} |\n`;
+      }
+    }
+
+    markdownContent += '\n';
+  }
+
+  // 提取方法详细说明 - 改进的匹配方式
+  const methodDetailsSection = extractMethodDetails(htmlContent);
+  if (methodDetailsSection) {
+    markdownContent += `## 方法详细说明\n\n`;
+
+    // 提取每个方法的详细信息
+    const methodDetails = methodDetailsSection.match(/<h4>[\s\S]*?(?=<h4>|$)/g) || [];
+    for (const detail of methodDetails) {
+      const methodNameMatch = detail.match(/<h4>([^<]+)<\/h4>/);
+      if (methodNameMatch) {
+        const methodName = methodNameMatch[1].trim();
+        const methodDescriptionMatch = detail.match(/<div class="block">([\s\S]*?)<\/div>/);
+        const methodDescription = methodDescriptionMatch ? cleanHtmlTags(methodDescriptionMatch[1].trim()) : '';
+
+        markdownContent += `### ${methodName}\n\n`;
+        if (methodDescription) {
+          markdownContent += `${methodDescription}\n\n`;
+        }
+
+        // 提取参数信息
+        const paramMatches = detail.match(/<code>([^<]+)<\/code>[\s\S]*?<div class="block">([\s\S]*?)<\/div>/g) || [];
+        if (paramMatches.length > 0) {
+          markdownContent += `**参数:**\n\n`;
+          for (const paramMatch of paramMatches) {
+            const paramNameMatch = paramMatch.match(/<code>([^<]+)<\/code>/);
+            const paramDescMatch = paramMatch.match(/<div class="block">([\s\S]*?)<\/div>/);
+
+            if (paramNameMatch && paramDescMatch) {
+              const paramName = paramNameMatch[1].trim();
+              const paramDesc = cleanHtmlTags(paramDescMatch[1].trim());
+              markdownContent += `- \`${paramName}\`: ${paramDesc}\n`;
+            }
+          }
+          markdownContent += '\n';
+        }
+      }
+    }
+  }
+
+  if (outputDir) {
+    // 创建输出目录（如果不存在）
+    if (!fs.existsSync(outputDir)) {
+      fs.mkdirSync(outputDir, { recursive: true });
+    }
+
+    // 生成输出文件名
+    const outputFileName = path.basename(htmlFilePath, '.html') + '.md';
+    const outputPath = path.join(outputDir, outputFileName);
+
+    // 写入 Markdown 文件
+    fs.writeFileSync(outputPath, markdownContent);
+    console.log(`已转换: ${htmlFilePath} -> ${outputPath}`);
+  }
+  return { content: markdownContent, name: packageName + '.' + className1 };
+}
+/**
+ * 提取方法详细资料部分
+ * @param {string} htmlContent HTML 内容
+ * @returns {string|null} 方法详细资料部分的内容
+ */
+function extractMethodDetails(htmlContent) {
+  // 查找方法详细资料部分
+  const methodDetailsStart = htmlContent.indexOf('<h3>方法详细资料</h3>');
+  if (methodDetailsStart === -1) return null;
+
+  // 查找方法详细资料部分的结束
+  let methodDetailsEnd = htmlContent.indexOf('<h3>', methodDetailsStart + 1);
+  if (methodDetailsEnd === -1) {
+    methodDetailsEnd = htmlContent.indexOf('</div><!-- ========= END OF CLASS DATA ========= -->', methodDetailsStart);
+    if (methodDetailsEnd === -1) {
+      methodDetailsEnd = htmlContent.length;
+    }
+  }
+
+  return htmlContent.substring(methodDetailsStart, methodDetailsEnd);
+}
+
+/**
+ * 从 HTML 片段中提取纯文本
+ * @param {string} html HTML 片段
+ * @returns {string} 纯文本
+ */
+function extractTextFromHtml(html) {
+  return html
+    .replace(/<[^>]+>/g, '') // 移除所有 HTML 标签
+    .replace(/\s+/g, ' ') // 将多个空格合并为一个
+    .trim();
+}
+
+/**
+ * 清理 HTML 标签，但保留一些基本格式
+ * @param {string} html HTML 片段
+ * @returns {string} 清理后的文本
+ */
+function cleanHtmlTags(html) {
+  return html
+    .replace(/<code>/g, '`')
+    .replace(/<\/code>/g, '`')
+    .replace(/<strong>/g, '**')
+    .replace(/<\/strong>/g, '**')
+    .replace(/<em>/g, '*')
+    .replace(/<\/em>/g, '*')
+    .replace(/<[^>]+>/g, '') // 移除所有其他 HTML 标签
+    .replace(/\s+/g, ' ') // 将多个空格合并为一个
+    .trim();
+}
+
+function getModName(key) {
+  var modName = '未知';
+  if (key.startsWith('com.wsgjp.ct')) {
+    var arr = key.split('.');
+    modName = arr[3]; // shell
+  }
+  return modName;
+}

package/tool/start.js

@@ -315,6 +315,15 @@ class Start {
     require('./iconfont.js').start();
   }
 
+  convertJavadoc() {
+    var r = require('./convertJavadoc.js');
+    if (config.filePath) {
+      r.testFile();
+    } else {
+      r.start();
+    }
+  }
+
   // 校验项目js语法
   checkAllJs() {
     start.checkWebPath(); // 兼容老平台