Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.javascript > #23055

Re: block evaluation

Show all headers | View raw

Stefan Ram wrote:

>   I observe that under Firefox
> 
> console.log( eval( "{1;}" ));
> 
>   prints
> 
> 1

This should not only be so in Firefox, i.e. with Mozilla JavaScript, but 
with all ECMAScript implementations whose host environment or user code 
provide an adequate method.
 
>   . Now I'd like to simplify the block evaluation.
>   I'd like to get rid of the »eval«. However,
> 
> console.log( {1;} );
> 
>   gives an error
> 
> SyntaxError: missing : after property id

Works as designed.
 
>   . What is the simplemost console.log statement
>   that shows the value of the block?

I think you have found it already.  You appear to be unaware of the object 
initialiser [1], which is syntactically similar to a /Block/ statement [2].  
(BTW, support for block *scoping* can AFAIK be found only in Mozilla 
JavaScript 1.7 with the “let” keyword; it has not been standardized yet.)

With console.log(…), particularly within the /ArgumentList/ of the 
/CallExpression/ the parser is in /Expression/ context [3], and “{…}” is not 
going to be evaluated as a /Block/ statement, but as an /ObjectLiteral/.

This is different with eval(), because the *string* argument is evaluated as 
a /Program/; as the parser is initially not in /Expression/, but in 
/Statement/ context, “{…}” will be evaluated as a /Block/ statement if 
possible [4] (otherwise it is a syntax error).

For that reason you need to enforce /Expression/ context evaluation if you 
want to evaluate generated object initalisers; the most efficient way is to 
put the generated expression in parentheses [5]:

  var obj = eval("(" + foo + ")");

Whereas “foo” could have a value of, e.g., '{"bar": "baz"}'.

However, when evaluating JSON, the safest way is to use JSON.parse(), 
provided it is available (either built-in or through a library) [6][7].

___________
[1] <http://ecma-international.org/ecma-262/5.1/#sec-11.1.5>
[2] <http://ecma-international.org/ecma-262/5.1/#sec-12.1>
[3] <http://ecma-international.org/ecma-262/5.1/#sec-11.2>
[4] <http://ecma-international.org/ecma-262/5.1/#sec-14>
[5] <http://ecma-international.org/ecma-262/5.1/#sec-11.1>
[6] <http://ecma-international.org/ecma-262/5.1/#sec-15.12.2>
[7] <http://json.org/>
-- 
PointedEars

Twitter: @PointedEars2
Please do not Cc: me. / Bitte keine Kopien per E-Mail.

Back to comp.lang.javascript | Previous | Next | Find similar | Unroll thread

Thread

Re: block evaluation Thomas 'PointedEars' Lahn <PointedEars@web.de> - 2014-02-10 16:37 +0100

csiph-web