Re: [JSMentors] Blog post on closures in JavaScript
Thanks! I'd just been watching some SICP videos when I came up with the idea for the post, so I had that kind of explanation on my mind (but I wanted to make it accessible). Nick On 26 April 2011 19:47, Peter van der Zee jsment...@qfox.nl wrote: On Tue, Apr 26, 2011 at 10:48 AM, Nick Morgan skilldr...@gmail.com wrote: Hi all I've written a new post on my blog, Closures explained with JavaScript. http://skilldrick.co.uk/2011/04/closures-explained-with-javascript/ Mentioning it here for two reasons (a) to help the mentees understand this crucial (and often confusing) aspect of JS, and (b) to ask the mentors if they have any feedback/criticism/clarifications... Cheers -- Nick Morgan http://skilldrick.co.uk @skilldrick I like the article and have little to say about it otherwise :) The free variable approach is something you don't see very often. - peter -- To view archived discussions from the original JSMentors Mailman list: http://www.mail-archive.com/jsmentors@jsmentors.com/ To search via a non-Google archive, visit here: http://www.mail-archive.com/jsmentors@googlegroups.com/ To unsubscribe from this group, send email to jsmentors+unsubscr...@googlegroups.com -- Nick Morgan http://skilldrick.co.uk @skilldrick -- To view archived discussions from the original JSMentors Mailman list: http://www.mail-archive.com/jsmentors@jsmentors.com/ To search via a non-Google archive, visit here: http://www.mail-archive.com/jsmentors@googlegroups.com/ To unsubscribe from this group, send email to jsmentors+unsubscr...@googlegroups.com
Re: [JSMentors] Blog post on closures in JavaScript
On Tue, 26 Apr 2011 17:48:45 +0200, Nick Morgan skilldr...@gmail.com wrote: Hi all I've written a new post on my blog, Closures explained with JavaScript. http://skilldrick.co.uk/2011/04/closures-explained-with-javascript/ Mentioning it here for two reasons (a) to help the mentees understand this crucial (and often confusing) aspect of JS, and (b) to ask the mentors if they have any feedback/criticism/clarifications... Minor nitpicks only. I like the description. The ability to nest functions gives us closures. I'd say that even top-level function expression evaluate to closures. The scope they capture might just be the global scope, but it's still there. In a language with first-class functions and static scope, all functions are (possibly trivial) closures, where the trivial case is a function with no free variables (which can arguably be said to not be a closure). As you have said, a closure is code combined with an environment where the free variables are looked up. A top-level function is that, with the global scope as its environment. A variable is free in any particular scope if it is defined within an enclosing scope. I would normally not separate a nested scope from the surrounding scope - it's more like an extension of the outer scope. I would prefer to talk about variables being free in an expression/statement/function declaration, i.e., at the syntactic level. However, it still makes sense as written, so that's just me preferring my preference :) Apart from that, in most other languages, I would agree with the statement, because it's equivalent to being a variable that's not bound by a declaration in the function declaration we are focusing on. Not so in Javascript. In Javascript you can create functions with free variables that aren't defined anywhere (yet), because the scope chain is mutable. You can add global, and even local, variable declarations after you have created the function object. So, I'd say something like: A variable is free in a function's code, if it's not bound by a variable declaration inside that function. Resolving that variable will use the scope where the function occurs. or something to that effect, which I think you can make more readable than I can. (Although that still ignores the case where the variable may or may not be declared dynamically inside the function). It's much harder to describe this in Javascript than in, say, Scheme, because variables in Javascript are resolved every time they are accessed, and the scope chain may change between each access. Closures are functions that retain a reference to their free variables Again, in other languages this would be the whole truth. In Javascript, the closure retains a reference to the environment where variables will be looked up. That might just be a little too tricky for this blog post, though. No need to confuse people more than necessary :) I really like the summary. It sums up the general concept very well. /L ' http://fiddle.jshell.net/Gmwuw/ ' -- Lasse Reichstein Holst Nielsen - reichsteinatw...@gmail.com -- To view archived discussions from the original JSMentors Mailman list: http://www.mail-archive.com/jsmentors@jsmentors.com/ To search via a non-Google archive, visit here: http://www.mail-archive.com/jsmentors@googlegroups.com/ To unsubscribe from this group, send email to jsmentors+unsubscr...@googlegroups.com
Re: [JSMentors] Blog post on closures in JavaScript
2011/4/26 Lasse Reichstein reichsteinatw...@gmail.com: In Javascript you can create functions with free variables that aren't defined anywhere (yet), because the scope chain is mutable. In JavaScript there is thing called hoisting, so ever variable that is defined in the function will be defined at the begining of the function, the position of the definition doesnt really matter. So I dont really understand what you mean by this. -- Poetro -- To view archived discussions from the original JSMentors Mailman list: http://www.mail-archive.com/jsmentors@jsmentors.com/ To search via a non-Google archive, visit here: http://www.mail-archive.com/jsmentors@googlegroups.com/ To unsubscribe from this group, send email to jsmentors+unsubscr...@googlegroups.com
Re: [JSMentors] Blog post on closures in JavaScript
On Tue, 26 Apr 2011 21:31:16 +0200, Poetro poe...@gmail.com wrote:
2011/4/26 Lasse Reichstein reichsteinatw...@gmail.com:
In Javascript you can create functions with free variables that aren't
defined
anywhere (yet), because the scope chain is mutable.
In JavaScript there is thing called hoisting, so ever variable that is
defined in the function will be defined at the begining of the
function, the position of the definition doesnt really matter. So I
dont really understand what you mean by this.
You can add properties and variables to the scope chain after it has
been created, and remove them again, e.g., by adding and removing
properties on the global object or on with objects.
You can even add and remove variable declarations in function scopes
by using eval.
Example: http://fiddle.jshell.net/Gmwuw/
This shows the same function (geta) referencing different a-variables
over time, as it adds and then removes declarations of a at different
levels of the scope chain.
It's a tricky question, whether a is free in the function:
function geta(s) {
if (s) eval(s);
return a;
}
Depending on the passed value of s, a may or may not be bound inside
the function itself.
/L
--
Lasse Reichstein Holst Nielsen - reichsteinatw...@gmail.com
--
To view archived discussions from the original JSMentors Mailman list:
http://www.mail-archive.com/jsmentors@jsmentors.com/
To search via a non-Google archive, visit here:
http://www.mail-archive.com/jsmentors@googlegroups.com/
To unsubscribe from this group, send email to
jsmentors+unsubscr...@googlegroups.com
