@icyfenix-dmla/cli 2026.5.2-2339 → 2026.5.3-1019

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@icyfenix-dmla/cli",
-  "version": "2026.5.2-2339",
+  "version": "2026.5.3-1019",
   "description": "DMLA 沙箱服务命令行工具",
   "type": "module",
   "main": "src/index.js",
@@ -22,13 +22,14 @@
     "README.md"
   ],
   "dependencies": {
-    "commander": "^12.1.0",
+    "@icyfenix-dmla/install": "*",
+    "adm-zip": "^0.5.17",
     "chalk": "^5.3.0",
-    "enquirer": "^2.4.1",
-    "dockerode": "^5.0.0",
-    "express": "^4.21.2",
+    "commander": "^12.1.0",
     "cors": "^2.8.5",
-    "@icyfenix-dmla/install": "*"
+    "dockerode": "^5.0.0",
+    "enquirer": "^2.4.1",
+    "express": "^4.21.2"
   },
   "devDependencies": {
     "jest": "^29.7.0"

package/src/commands/data.js

@@ -10,6 +10,7 @@ import path from 'path'
 import os from 'os'
 import { spawn } from 'child_process'
 import { execSync } from 'child_process'
+import AdmZip from 'adm-zip'
 
 // 配置文件路径
 const DMLA_CONFIG_DIR = path.join(os.homedir(), '.dmla')
@@ -27,7 +28,11 @@ const DATASETS = [
     size: '247MB',
     format: 'git',
     targetDir: 'datasets/tiny-imagenet-200',
-    source: 'ModelScope (icyfenix)'
+    source: 'ModelScope (icyfenix)',
+    // git clone 后需要解压的 zip 文件
+    zipFile: 'tiny-imagenet-200.zip',
+    // zip 内部的顶层目录名（解压后需要将此目录内容移到上层）
+    zipInnerDir: 'tiny-imagenet-200'
   },
   {
     id: 'cifar-10',
@@ -526,6 +531,58 @@ async function downloadDataset(dataPath, dataset) {
       console.log()
       console.log(chalk.green('下载完成'))
 
+      // 解压数据集内的 zip 文件（如果有）
+      if (dataset.zipFile) {
+        const zipPath = path.join(targetDir, dataset.zipFile)
+
+        if (fs.existsSync(zipPath)) {
+          console.log()
+          console.log(chalk.gray(`解压 ${dataset.zipFile}...`))
+
+          try {
+            const zip = new AdmZip(zipPath)
+
+            // 解压到临时目录
+            const tempDir = path.join(targetDir, '_extract_temp')
+            zip.extractAllTo(tempDir, true)
+
+            // 将 zip 内部目录内容移到目标目录
+            const innerDir = dataset.zipInnerDir
+              ? path.join(tempDir, dataset.zipInnerDir)
+              : tempDir
+
+            if (fs.existsSync(innerDir)) {
+              // 移动内部目录的所有内容到目标目录
+              const items = fs.readdirSync(innerDir)
+              for (const item of items) {
+                const srcPath = path.join(innerDir, item)
+                const destPath = path.join(targetDir, item)
+
+                // 如果目标已存在且不是 zip 文件，跳过
+                if (fs.existsSync(destPath) && item !== dataset.zipFile) {
+                  continue
+                }
+
+                fs.cpSync(srcPath, destPath, { recursive: true, force: true })
+              }
+
+              // 清理临时目录
+              fs.rmSync(tempDir, { recursive: true, force: true })
+
+              // 删除 zip 文件
+              fs.rmSync(zipPath, { force: true })
+
+              console.log(chalk.green('解压完成'))
+            } else {
+              console.log(chalk.yellow(`  ⚠ zip 内部目录 ${dataset.zipInnerDir} 不存在`))
+            }
+          } catch (err) {
+            console.log(chalk.red(`解压失败: ${err.message}`))
+            console.log(chalk.yellow(`请手动解压: ${zipPath}`))
+          }
+        }
+      }
+
     } else {
       // 原有的 curl/wget 下载逻辑（保留兼容性）
       const cacheDir = path.join(dataPath, 'cache', 'downloads')
@@ -569,17 +626,17 @@ async function downloadDataset(dataPath, dataset) {
       console.log(chalk.gray('正在解压...'))
 
       if (dataset.format === 'zip') {
-        const unzipDir = path.join(cacheDir, dataset.id)
-        fs.mkdirSync(unzipDir, { recursive: true })
-        execSync(`unzip -o "${downloadFile}" -d "${unzipDir}"`, { stdio: 'inherit' })
-
-        const extractedDir = path.join(unzipDir, dataset.id)
-        if (fs.existsSync(extractedDir)) {
-          fs.cpSync(extractedDir, targetDir, { recursive: true })
-          fs.rmSync(unzipDir, { recursive: true, force: true })
+        try {
+          const zip = new AdmZip(downloadFile)
+          zip.extractAllTo(targetDir, true)  // overwrite = true
+          console.log(chalk.green('解压完成'))
+        } catch (err) {
+          console.log(chalk.red(`解压失败: ${err.message}`))
+          throw err
         }
 
       } else if (dataset.format === 'tar.gz') {
+        // tar.gz 文件仍使用系统命令（adm-zip 不支持）
         execSync(`tar -xzf "${downloadFile}" -C "${path.join(dataPath, 'datasets')}"`, { stdio: 'inherit' })
       }
 

package/src/server/routes/sandbox.js

@@ -2,9 +2,11 @@
  * 沙箱 API 路由
  */
 import { Router } from 'express'
+import Docker from 'dockerode'
 import sandbox, { runPythonCode, checkImageExists, checkGPUAvailable, checkCUDACompatibility } from '../sandbox.js'
 
 const { SANDBOX_CONFIG } = sandbox
+const docker = new Docker()
 
 const router = Router()
 
@@ -147,8 +149,6 @@ router.get('/progress', async (req, res) => {
   try {
     // 进度文件在容器内的 /workspace/progress.json
     // 需要通过 docker exec 读取
-    const Docker = require('dockerode').default || require('dockerode')
-    const docker = new Docker()
 
     // 查找运行中的沙箱容器
     const containers = await docker.listContainers({ filters: { status: ['running'] } })

package/src/server/sandbox.js

@@ -471,12 +471,13 @@ export async function runPythonCode(code, useGpu = false, imageOverride = null,
     container = await docker.createContainer(containerConfig)
     log(`Container created: ${container.id}`)
 
-    // 设置超时
+    // 设置超时（使用动态计算的超时时间，unlimited 时为 24 小时）
+    const containerTimeoutMs = timeoutSeconds * 1000 + 10000  // 转换为毫秒，额外 10 秒用于清理
     const timeoutPromise = new Promise((_, reject) => {
       timeoutId = setTimeout(() => {
         log('Execution timeout triggered')
         reject(new Error('Execution timeout'))
-      }, SANDBOX_CONFIG.timeout + 10000)  // 额外 10 秒用于清理
+      }, containerTimeoutMs)
     })
 
     // 启动容器

package/version.json

@@ -1,4 +1,4 @@
 {
-  "buildTime": "2026-05-02T15:40:30.403Z",
-  "cliVersion": "2026.5.2-2339"
+  "buildTime": "2026-05-03T02:20:30.472Z",
+  "cliVersion": "2026.5.3-1019"
 }