feat: 将word转pdf (#1380)

Co-authored-by: bai.zixv <bai.zixv@foxmail.com>
Co-authored-by: sealday <sealday@gmail.com>
Reviewed-on: daoyoucloud/tachybase#1380
Reviewed-by: sealday <zhanglin@daoyoucloud.com>
Co-authored-by: yoona <1486343814@qq.com>
Co-committed-by: yoona <1486343814@qq.com>

packages/plugins/@tachybase/plugin-print-template/package.json

@@ -3,10 +3,20 @@
   "version": "0.21.77",
   "main": "dist/server/index.js",
   "dependencies": {
+    "@bull-board/api": "^5.21.1",
+    "@bull-board/koa": "^5.21.1",
+    "bullmq": "^5.10.3",
     "docxtemplater": "^3.48.0",
+    "koa-bodyparser": "^4.4.1",
+    "koa-router": "^10.1.1",
     "pizzip": "^3.1.7",
+    "ts-node": "9.1.1",
+    "typescript": "^5.5.4",
     "xml2js": "^0.6.2"
   },
+  "devDependencies": {
+    "@types/koa-router": "^7.4.8"
+  },
   "peerDependencies": {
     "@tachybase/client": "workspace:*",
     "@tachybase/server": "workspace:*",

packages/plugins/@tachybase/plugin-print-template/src/server/actions/printTemplates.ts

@@ -1,21 +1,20 @@
-import fs from 'fs';
-import path from 'path';
-import { Readable } from 'stream';
-import { Context } from '@tachybase/actions';
-import { Model } from '@tachybase/database';
+import fs from 'fs'; //导入Node.js的文件系统模块，用于文件操作
+import path from 'path'; //导入Node.js的路径模块，用于处理文件路径
+import { Readable } from 'stream'; // 从'stream'模块导入Readable类，用于创建可读流
+import { Context } from '@tachybase/actions'; //一个自定义库，用于处理请求上下文
+import { Model } from '@tachybase/database'; //用于数据库操作的模型
+
+import Docxtemplater from 'docxtemplater'; //用于处理word文版模板
 
-import Docxtemplater from 'docxtemplater';
 // 导入 Docxtemplater 的 InspectModule 类
-import InspectModule from 'docxtemplater/js/inspect-module';
-import PizZip from 'pizzip';
+import InspectModule from 'docxtemplater/js/inspect-module'; //Docxtemplater使用它来处理Word文档。
+import PizZip from 'pizzip'; //导入PizZip库，用于处理ZIP文件，Docxtemplater使用它来处理word文档
 
 export const generate = async (ctx: Context) => {
   const id = ctx.action.params.id;
   const repo = ctx.db.getRepository('templateManage');
   const template: Model = await repo.findOne({
-    filter: {
-      id,
-    },
+    filter: { id },
     appends: ['template'],
   });
 
@@ -24,17 +23,19 @@ export const generate = async (ctx: Context) => {
   const data = rawTemplate.testData;
 
   try {
-    // 生成定制的 docx 文件
     const buffer = generateDocxFromTemplate(templatePath, data);
     ctx.withoutDataWrapping = true;
     ctx.body = buffer;
+    return buffer;
   } catch (error) {
     console.error('Error:', error);
     ctx.status = 500;
     ctx.body = 'Error generating files: ' + error.message;
+    return { filePath: '' }; // 返回空路径以防止进一步的错误
   }
 };
 
+//用于获取文档中的所有占位符标签
 export const getTags = async (ctx: Context) => {
   const id = ctx.action.params.id;
   const repo = ctx.db.getRepository('templateManage');
@@ -48,13 +49,14 @@ export const getTags = async (ctx: Context) => {
   const templatePath = path.join(process.env.PWD, rawTemplate.template[0].get().url);
 
   try {
-    const content = fs.readFileSync(templatePath, 'binary');
-    const zip = new PizZip(content);
-    const iModule = new InspectModule();
-    const doc = new Docxtemplater(zip, { modules: [iModule] });
+    const content = fs.readFileSync(templatePath, 'binary'); //同步读取模板文件的内容
+    const zip = new PizZip(content); //用读取的内容创建一个PizZip对象
+    const iModule = new InspectModule(); //创建一个InspectModule实例
+    const doc = new Docxtemplater(zip, { modules: [iModule] }); //使用PizZip对象和InspectModule创建一个Docxtemplater实例
 
     // 获取所有占位符
-    const allTags = iModule.getAllTags();
+    const allTags = iModule.getAllTags(); //InspectModule获取文档中的所有标签
+    console.log('All tags in the document:', allTags);
 
     // 将 tags 转换为 JSON 格式的字符串
     const tagsJson = JSON.stringify(allTags, null, 2);
@@ -78,15 +80,17 @@ export const getTags = async (ctx: Context) => {
   }
 };
 
+//用于生成word文档的二进制数据
 function generateDocxFromTemplate(templatePath, data): Buffer {
   try {
     const content = fs.readFileSync(templatePath, 'binary');
     const zip = new PizZip(content);
-    const doc = new Docxtemplater(zip);
+    const doc = new Docxtemplater(zip); //Docxtemplater 是一个用于处理DOCX模板并填充数据的库。
 
     doc.setData(data);
-    doc.render();
+    doc.render(); //渲染模板，将数据填充到模板的占位符中
 
+    //调用 getZip 方法获取处理后的ZIP对象，然后调用 generate 方法生成DOCX文件的二进制数据。{ type: 'nodebuffer' } 选项指定输出类型为Node.js的Buffer。buffer是一个缓存区，用于存取二进制数据。
     return doc.getZip().generate({ type: 'nodebuffer' });
   } catch (error) {
     console.error('Error generating DOCX from template:', error);

packages/plugins/@tachybase/plugin-print-template/src/server/actions/producer.ts

@@ -0,0 +1,63 @@
+import fs from 'fs';
+import os from 'os';
+import path from 'path';
+import { Context } from '@tachybase/actions';
+
+import { Queue } from 'bullmq';
+
+import { generate } from './printTemplates';
+
+// 用于将 Buffer 保存为临时文件并返回文件路径
+function saveBufferToTempFile(buffer: Buffer): string {
+  const tempDir = os.tmpdir();
+  const tempFilePath = path.join(tempDir, `document-${Date.now()}.docx`);
+  fs.writeFileSync(tempFilePath, buffer);
+  return tempFilePath;
+}
+
+// 创建队列
+const redisOptions = {
+  port: Number(process.env.REDIS_PORT || 6379),
+  host: process.env.REDIS_HOST || 'localhost',
+  password: process.env.REDIS_PASSWORD || '',
+  // 移除 tls 属性
+};
+
+const pdfConversionQueue = new Queue('default', { connection: redisOptions });
+
+export const addConversionJob = async (ctx: Context) => {
+  try {
+    if (!ctx.action || !ctx.action.params) {
+      throw new Error('Missing action or action params in context');
+    }
+
+    // 生成文件路径并添加到队列
+    const result = await generate(ctx);
+
+    // 检查 result 类型
+    let filePath: string;
+    if (Buffer.isBuffer(result)) {
+      filePath = saveBufferToTempFile(result);
+    } else if (result && typeof result === 'object' && 'filePath' in result) {
+      filePath = (result as { filePath: string }).filePath;
+    } else {
+      throw new Error('Invalid result from generate function');
+    }
+
+    console.log('Generated DOCX file path:', filePath);
+
+    const outputDir = '/root/tachybase/storage/uploads';
+
+    // 添加作业到队列
+    const job = await pdfConversionQueue.add('convert', { wordFilePath: filePath, outputDir });
+
+    // 返回作业 ID 和 PDF 文件路径
+    ctx.body = { jobId: job.id };
+  } catch (error) {
+    console.error('Error adding conversion job:', error);
+    ctx.status = 500;
+    ctx.body = 'Error adding conversion job: ' + error.message;
+  }
+};
+
+export default addConversionJob;

packages/plugins/@tachybase/plugin-print-template/src/server/actions/template-params.ts

@@ -1,54 +0,0 @@
-import fs from 'fs';
-
-import PizZip from 'pizzip';
-import { parseStringPromise } from 'xml2js';
-
-export async function getTemplateParams(templatePath) {
-  try {
-    const content = fs.readFileSync(templatePath, 'binary');
-    const zip = new PizZip(content);
-
-    const getXmlContent = (fileName) => {
-      const file = zip.file(fileName);
-      return file ? file.asText() : '';
-    };
-
-    const docXml = getXmlContent('word/document.xml');
-
-    const parseXml = async (xml) => {
-      try {
-        const result = await parseStringPromise(xml, { explicitArray: false, ignoreAttrs: true });
-        const body = result['w:document']['w:body'];
-        return Array.isArray(body['w:p']) ? body['w:p'] : [body['w:p']];
-      } catch (error) {
-        console.error('Error parsing XML:', error);
-        throw error;
-      }
-    };
-
-    const xmlContent = await parseXml(docXml);
-
-    const extractTags = (nodes) => {
-      const tags = new Set();
-      nodes.forEach((node) => {
-        if (node['w:r'] && Array.isArray(node['w:r'])) {
-          node['w:r'].forEach((r) => {
-            if (r['w:t'] && r['w:t']['#']) {
-              const matches = r['w:t']['#'].match(/\{[\w-]+\.}/g);
-              if (matches) {
-                matches.forEach((tag) => tags.add(tag));
-              }
-            }
-          });
-        }
-      });
-      return tags;
-    };
-
-    const tags = extractTags(xmlContent);
-    return Array.from(tags); // Ensure the return value is an array
-  } catch (error) {
-    console.error('Error in getTemplateParams:', error);
-    throw error;
-  }
-}

packages/plugins/@tachybase/plugin-print-template/src/server/plugin.ts

@@ -1,6 +1,53 @@
+import { exec } from 'child_process';
+import path from 'path';
+import util from 'util';
 import { Plugin } from '@tachybase/server';
 
+import { Worker } from 'bullmq';
+
 import { generate, getTags } from './actions/printTemplates';
+import { addConversionJob } from './actions/producer';
+
+const execPromise = util.promisify(exec);
+
+const redisOptions = {
+  port: Number(process.env.REDIS_PORT || 6379),
+  host: process.env.REDIS_HOST || 'localhost',
+  password: process.env.REDIS_PASSWORD || '',
+};
+
+async function convertDocxToPdf(wordFilePath: string, outputDir: string, job: any): Promise<string> {
+  try {
+    const fileName = path.basename(wordFilePath, '.docx') + '.pdf';
+    const pdfFilePath = path.join(outputDir, fileName);
+
+    console.log(`Converting ${wordFilePath} to PDF at ${pdfFilePath}`);
+    const command = `libreoffice --headless --convert-to pdf --outdir "${outputDir}" "${wordFilePath}"`;
+    console.log('commandcommandcommand:', command);
+
+    // 更新进度到 20%：转换开始
+    job.updateProgress(20);
+
+    const { stdout, stderr } = await execPromise(command);
+    console.log('stdout', stdout);
+    console.log('stderr', stderr);
+
+    // 更新进度到 80%：转换进行中
+    job.updateProgress(80);
+
+    console.log(`PDF created successfully: ${pdfFilePath}`);
+
+    // 更新进度到 100%：转换完成
+    job.updateProgress(100);
+
+    return pdfFilePath;
+  } catch (error) {
+    console.error('Error during conversion:', error);
+    // 更新进度到 0% 或其他表示失败的值
+    job.updateProgress(0);
+    throw error;
+  }
+}
 
 export class PluginPrintTemplateServer extends Plugin {
   async afterAdd() {}
@@ -8,24 +55,47 @@ export class PluginPrintTemplateServer extends Plugin {
   async beforeLoad() {}
 
   async load() {
-    // localhost:3000/api/printTemplates:generate
     this.app.resourcer.define({
       name: 'printTemplates',
       actions: {
         generate,
         getTags,
+        addConversionJob,
       },
     });
     this.app.acl.allow('printTemplates', '*', 'public');
+
+    this.app.use(require('koa-bodyparser')());
+
+    const worker = new Worker(
+      process.env.MSG_QUEUE_NAME || 'default',
+      async (job) => {
+        if (!job.data.wordFilePath || !job.data.outputDir) {
+          console.error('Invalid job data:', job.data);
+          throw new Error('Missing required job data properties');
+        }
+
+        const { wordFilePath, outputDir } = job.data;
+
+        // 自动更新进度
+        const pdfFilePath = await convertDocxToPdf(wordFilePath, outputDir, job);
+
+        // 返回 PDF 文件路径
+        return { pdfFilePath };
+      },
+      { connection: redisOptions },
+    );
+
+    worker.on('completed', (job, returnValue) => {
+      console.log(`Job ${job.id} completed with result: ${JSON.stringify(returnValue)}`);
+    });
+
+    worker.on('failed', (job, err) => {
+      console.error(`Job ${job.id} failed with error: ${err.message}`);
+    });
   }
 
-  async install() {}
-
-  async afterEnable() {}
-
-  async afterDisable() {}
-
-  async remove() {}
+  async beforeDestroy() {}
 }
 
 export default PluginPrintTemplateServer;