多次运行grunt任务直到失败 [英] Running grunt task multiple times until it fails

问题描述

---

我是不是经常连续运行同一个grunt任务的规范方法， d喜欢将问题保留为泛型，但这里有一个具体的用例：

我们有一大批用Protractor编写的端到端测试，我们通过它运行 grunt 借助 grunt-protractor-runner 和 咕噜-的contrib-连接 。我们想要做的是保持连接任务运行（Web服务器从 dist 目录服务） ），同时遍历量角器，直到它失败（或/和高达 N 次）。任务：

  connect：{
 test：{
 options：{
 base： 'dist'，
端口：9001 
} 
}，
}，
 
量角器：{
选项：{
 keepAlive：true，
 noColor：false 
}，
 local：{
 options：{
 configFile：test / e2e / config / local.conf.js 
 
 
 
 $ b grunt.registerTask（'e2e：local'，[
'connect：test'，
'量角器：本地'
]）; 
  

在这里，我们要执行 connect：test 一次和量角器：本地多次。

解决方案

Turn就像将任务添加到任务数组N次一样简单（仍然不确定是否可以避免指定 N 并无限期地运行任务直到失败），因为默认 grunt 在fail-fast模式下工作，所以整个任务在第一次失败时会失败。在我们的例子中：

  grunt.registerTask（'e2e：local'，function（）{
 var tasks = [ 'connect：test']; 
 
 var N = 100; 
 for（var i = 0; i< N; i ++）{
 tasks.push（'protractor：本地'）
} 
 
 grunt.task.run（任务）; 
}）; 
  

N 这里是硬编码的任务，但可以作为 grunt选项从外部传递。

What is the canonical way to run the same grunt task continuously, multiple times until it fails?

---

I'd like to keep the question generic, but here is a specific use case:

We have a huge set of end-to-end tests written in Protractor which we run via grunt with the help of grunt-protractor-runner and grunt-contrib-connect. What we'd like to do is to keep the connect task running (the web-server serving from a dist directory) while looping over the protractor until it fails (or/and up to N times). The tasks:

connect: {
    test: {
        options: {
            base: 'dist',
            port: 9001
        }
    },
},

protractor: {
    options: {
        keepAlive: true,
        noColor: false
    },
    local: {
        options: {
            configFile: "test/e2e/config/local.conf.js"
        }
    }
},

grunt.registerTask('e2e:local', [
    'connect:test',
    'protractor:local'
]);

Here, we'd like to execute connect:test once and protractor:local multiple times.

解决方案

Turns out to be as simple as adding the same task to the array of tasks N times (still not sure if we can somehow avoid specifying the N and run the task indefinitely until it fails) and, since, by default grunt works in the "fail-fast" mode, the whole task would fail on the first fail. In our case:

grunt.registerTask('e2e:local', function () {
    var tasks = ['connect:test'];

    var N = 100;
    for (var i = 0; i < N; i++) {
        tasks.push('protractor:local')
    }

    grunt.task.run(tasks);
});

N here is hardcoded inside the task, but could be passed "from outside" as a grunt option.

这篇关于多次运行grunt任务直到失败的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！