Path: csiph.com!usenet.pasdenom.info!news.albasani.net!weretis.net!feeder1.news.weretis.net!news.solani.org!.POSTED!not-for-mail
From: Thomas 'PointedEars' Lahn <PointedEars@web.de>
Newsgroups: comp.lang.javascript
Subject: Re: Can somebody explain this to me
Date: Thu, 08 Jan 2015 17:54:25 +0100
Organization: PointedEars Software (PES)
Lines: 106
Message-ID: <3105476.iMYZV00USQ@PointedEars.de>
References: <m8f7s7$lcv$1@dont-email.me> <1590597.KOVAYohEvS@PointedEars.de> <qj1oaahmnp3r6j45b27sm5ujsfhqu4igv6@4ax.com> <060120151734221023%timstreater@greenbee.net> <m8h81k$p88$1@solani.org> <vsiqaaha4ne28chagor402ccjpvvifn9dn@4ax.com> <m8jk3c$fhh$1@solani.org> <1676759.bbBSSFFLiy@PointedEars.de> <66nsaal9kcls42elbn16dkbv3fiille47o@4ax.com> <m8ltkq$rq3$1@solani.org>
Reply-To: Thomas 'PointedEars' Lahn <cljs@PointedEars.de>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8Bit
X-Trace: solani.org 1420736088 20846 eJwFwQkBwDAIA0BLfAmdHAbFv4TewansDIKBxbaspLHN1zTcUT57CnKBYJbmf3dOmMRcfngOJxCc (8 Jan 2015 16:54:48 GMT)
X-Complaints-To: abuse@news.solani.org
NNTP-Posting-Date: Thu, 8 Jan 2015 16:54:48 +0000 (UTC)
User-Agent: KNode/4.14.2
X-User-ID: eJwNwokRwDAIA7CVYlK+cSjg/UdoT9JrsPbH1B7lbzOQMrLtQZ+D1EA5AZYK3piWvROs6M6xOrunyKWZID9YnRY1
Cancel-Lock: sha1:9fO8rqjTV2e2K/WOG8cnGQq7l5c=
X-NNTP-Posting-Host: eJwFwYEBwCAIA7CXwFLAc1bF/09YQqTnqUhm8PGt7d4qzuyLe6atuKRjBukLB9wRDCmra/gDFNoQlA==
Xref: csiph.com comp.lang.javascript:26500

Christoph M. Becker wrote:

> John Harris wrote:
>> On Wed, 07 Jan 2015 17:22:40 +0100, Thomas 'PointedEars' Lahn
>> <PointedEars@web.de> wrote:
>>> It is because, as is obvious from the above, he has not realized yet
>>> that there are (at least) two ways to “define” a function in ECMAScript,
>>> of which one includes a *declaration*.
>> 
>> Both ways create a function object. One way gives it a global name,
>> the other need not. What else matters?
> 
> Most importantly scope rules.

The scope (chain) rules are the same with either syntax.

> A function declaration creates a function object that is passed in the 
> *variable* environment of the running execution context as environment;
> a function object that is created by a function expression is passed in 
> the *lexical* environment.

Objects are not "passed in environments".

<http://ecma-international.org/ecma-262/5.1/#sec-10.3>

| LexicalEnvironment  | Identifies the Lexical Environment
|                     | used to resolve identifier references made by code
|                     | within this execution context.
| --------------------+---------------------------------------------------
| VariableEnvironment | Identifies the Lexical Environment
|                     | whose environment record holds bindings created by
|                     | VariableStatements and FunctionDeclarations within 
|                     | this execution context.

(intentional word-wrap)

> Informally, one might say that function declarations are "hoisted".

Yes, but as with all informal descriptions, it is imprecise and potentially 
misleading.  For example, a so "hoisted" function can still access 
identifiers in the outer execution context even though they have only been 
defined (and assigned to) later, because it is a closure and because of the 
scope chain:

  var answer;

  function deepThought ()
  {
    return answer;
  }

  /* undefined */
  deepThought();

  answer = 42;

  /* 42 */
  deepThought();

> Another pitfall that one should be aware of is that the behavior of a
> function expression in a statement context is standardized, while the
> behavior of a function declaration (better called function statement
> here) in a statement context is not, even though it is allowed in
> several ECMAScript implementations.

True.  However, I think it is more important to realize that function 
expressions observe source code order (so I referred to that).  A function 
defined with a function expression is not defined before execution *passed* 
the latter.

  /* 42 */
  foo();

  function foo ()
  {
    return 42;
  }

  /* TypeError: undefined is not a function */
  bar();

  var bar = function () {
    return 42;
  };

[Per Specification, this is independent of whether the function *expression* 
contains an identifier (a function *declaration* must contain an identifier 
or it is parsed as a function expression).  There is a JScript bug, fixed in 
version 9+ when in IE 9+ standards Document Mode¹ (test with Firebug Lite or 
F12 – developer tools), that allows the bar() call above to be successful if 
you write

  var bar = function bar () {
    return 42;
  }; 

instead.  IOW, the function identifier can leak to the outer scope.

______
¹ Did I already say that there is no javascript/Javascript?]

-- 
PointedEars
FAQ: <http://PointedEars.de/faq> | SVN: <http://PointedEars.de/wsvn/>
Twitter: @PointedEars2 | ES Matrix: <http://PointedEars.de/es-matrix>
Please do not cc me. / Bitte keine Kopien per E-Mail.