Reputation: 107950
I have just started writing my own JavaScript Framework (just for the learning experience), and have prefixed some private members with a _, like such:
var _isFireBugEnabled = function () {
return (window.console && window.console.firebug);
};
When I ran my code against Crockford's JSLint (as always), with Recommended Options
on, I was told about not using a _ as an identifier.
My question is, Why does JSLint warn me as regards not using a _ as an identifier?
Are there some side effects or implications I am missing here?
PS. As far as I could scan just now, this is not documented in the book
Upvotes: 1
Views: 990
Reputation: 211980
The reason is that Douglas Crockford hates about 78% of Javascript*. Many people think he's a bit strict, and in fact many libraries do use leading underscores in production code. I don't see much wrong with it. There are no side effects.
Additionally, the '$', not the underscore, was the symbol set aside for "system" code by the ECMA spec.
from ECMA 262, section 7.6:
This standard specifies one departure from the grammar given in the Unicode standard: The dollar sign ($) and the underscore (_) are permitted anywhere in an identifier. The dollar sign is intended for use only in mechanically generated code.
*Note: I'm being facetious. He really only hates about half, and he typically has good reason. I'd disagree with Crockford here, but he's usually very right.
Upvotes: 13
Reputation: 11402
I've actually e-mailed Crockford on this. This was his response:
I think _ should be reserved for system code implementation, and not used by applications.
I disagree with him somewhat, I tend to use _ to prefix truly private members in my own classes, because it makes it clear to me what is private. Google's Caja has some rules regarding the use of _, but nothing that should cause problems with what you're describing.
Upvotes: 8