如何使用 grunt-contrib-coffee 避免多个帮助器实现,例如 __indexOf()
我有一个 Gruntfile.coffee
有一个 grunt-contrib-咖啡 配置如下:
I have a Gruntfile.coffee
that has a grunt-contrib-coffee configuration like this:
coffee:
compile:
files:
'public/assets/application.js': [
'multiple/files' #, ...
]
options:
bare: true
问题在于,它会生成多个辅助方法的实现,例如
The Problem is, that it generates multiple implementations of helper methods like
var __indexOf = [].indexOf || function(item) { for (var i = 0, l = this.length; i < l; i++) { if (i in this && this[i] === item) return i; } return -1; };
如何让它识别这些实现已经编译?
How to make it recognize that these implementations are already compiled?
原因很明显是编译后的 JavaScript 在编译之后被连接起来.如果将 CoffeeScript 连接在编译它是可以避免的.如果这不是我缺少的 grunt-contrib-coffee 的配置选项,我会在github上提交一个问题.
The reason is obviously that the compiled JavaScript gets concatenated after it got compiled. It could be avoided if the CoffeeScript would be concatenated before compiling it. If this is not a configuration option of grunt-contrib-coffee that I'm missing, I'll submit an issue on github.
那么剩下的问题是:如何在编译之前连接 CoffeeScript,以避免多个帮助器实现?
So the question that remains is: how to concatenate the CoffeeScript before compiling it, to avoid multiple helper implementations?
有一种解决方法,它涉及使用 concat
任务手动连接源文件.这需要一个临时文件,然后可以将其从 CoffeeScript 编译为 JavaScript.我创建的问题中提到了它.
There is a workaround, which involves manual concatenation of the source files with the concat
task. This requires a temporary file, that can then be compiled from CoffeeScript to JavaScript. It is mentioned in the issue I created.
将来,任务将能够将它们的东西作为缓冲区传递给彼此(参见 问题).
In future, tasks will be able to pass their stuff as buffers to each other (see the issue).
推荐答案
从 grunt-contrib-coffee-0.6.1 开始,这适用于 join
选项:
As of grunt-contrib-coffee-0.6.1 this works with the join
option:
coffee:
compile:
files:
'public/assets/application.js': [
'multiple/files' #, ...
]
options:
bare: true
join: true # concatenate coffee files before compiling. default is false
相关文章