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

Groups > comp.lang.javascript > #8416 > unrolled thread

"" + some form value

Back to article view | Back to comp.lang.javascript

Contents

  "" + some form value Gene Wirchenko <genew@ocis.net> - 2011-11-18 15:07 -0800
    Re: "" + some form value "Richard Cornford" <Richard@litotes.demon.co.uk> - 2011-11-19 03:17 +0000
      Re: "" + some form value Thomas 'PointedEars' Lahn <PointedEars@web.de> - 2011-11-19 17:09 +0100
      Re: "" + some form value Gene Wirchenko <genew@ocis.net> - 2011-11-20 00:07 -0800
    Re: "" + some form value Dr J R Stockton <reply1146@merlyn.demon.co.uk> - 2011-11-19 18:53 +0000

#8416 — "" + some form value

Dear JavaScripters:

     What does "" + some form value do?  For example,
          ""+document.Contest.Last.value
Won't document.Contest.Last.value be string regardless?  I tried
forcing one such to a number, and after the assignment, typeof()
reported it was a string.

     Is this ""+ a bit of superstition, or am I missing something?

Sincerely,

Gene Wirchenko

[toc] | [next] | [standalone]

#8420

Gene Wirchenko wrote:
> Dear JavaScripters:
>
>     What does "" + some form value do?  For example,

The addition operator does string concatenation if either of its 
argument is a string primitive, as in this case. However, when it does 
string concatenation and non-string argument (the right hand side in 
this case) is type-converted into a string primitive (by applying the 
language's type conversion rules). It is this type-conversion that 
usually motivates the concatenation of an empty string, that is, the 
result is just the result of type converting the right hand side into a 
string primitive, as concatenating an empty string has no effect on that 
result.

>          ""+document.Contest.Last.value

Assuming that - document.Contest.Last.value - if a reference to the - 
value - property of a form element then the concatenation is actually 
worthless as such - value - property values are already string 
primitives. Unfortunately forced type-conversion methods often ends up 
being applied as a mystical incantation, and so in contexts where they 
actually have no effect.

Incidentally, to force type conversion to a string primitive the String 
constructor can be used without the - new - operator, as in:-

String(document.Contest.Last.value)

- Which has been both recommended against on the grounds that it could 
be confused with a miss-typed - new String(s); - and recommended as a 
self-documenting and explicit means of forcing type-conversion to a 
string. I am leaning towards the latter. Concatenating the empty string 
has been shown to be the most runtime efficient type-conversion to 
string primitive operation.

> Won't document.Contest.Last.value be string regardless?

Yep.

> I tried forcing one such to a number,

How? Strings can be type-converted into number primitive values using, 
say,  +string  or  Number(string)  (and commonly in the past or 
string*1 , or  string-0 ).


> and after the assignment, typeof()
> reported it was a string.
>
>     Is this ""+ a bit of superstition, or am I missing something?

No, it appears to be superstition in this case. Though there is a slight 
possibility that - document.Contest.Last - was not referring to a form 
element in this case (then you only have evidence of poor code design 
rather than a mystical incantation).

Richard. 

[toc] | [prev] | [next] | [standalone]

#8424

Richard Cornford wrote:

> Gene Wirchenko wrote:
>> Dear JavaScripters:
>>
>>     What does "" + some form value do?  For example,
>>          ""+document.Contest.Last.value
> 
> Assuming that - document.Contest.Last.value - if a reference to the -
> value - property of a form element then the concatenation is actually
> worthless as such - value - property values are already string
> primitives. Unfortunately forced type-conversion methods often ends up
> being applied as a mystical incantation, and so in contexts where they
> actually have no effect.
> 
> Incidentally, to force type conversion to a string primitive the String
> constructor can be used without the - new - operator, as in:-
> 
> String(document.Contest.Last.value)
> 
> - Which has been both recommended against on the grounds that it could
> be confused with a miss-typed - new String(s); - and recommended as a
> self-documenting and explicit means of forcing type-conversion to a
> string. I am leaning towards the latter. Concatenating the empty string
> has been shown to be the most runtime efficient type-conversion to
> string primitive operation.

There is no consensus on what is most runtime-efficient here.  In fact, the 
results at <http://jsperf.com/string-conversion-speed> vary so much between 
implementations, runtime environments, and even consecutive tests in the 
very same runtime environment, and String() is indeed self-documenting, that 
I am going to keep my original, ECMAScript-supported approach¹ of using 
String() when I do not know the type of the value and if it(s object 
representation) has a toString() method.


PointedEars
___________
¹) the `+' operator requires more algorithmic steps for the type conversion
   than String()
-- 
Prototype.js was written by people who don't know javascript for people
who don't know javascript. People who don't know javascript are not
the best source of advice on designing systems that use javascript.
  -- Richard Cornford, cljs, <f806at$ail$1$8300dec7@news.demon.co.uk>

[toc] | [prev] | [next] | [standalone]

#8433

On Sat, 19 Nov 2011 03:17:21 -0000, "Richard Cornford"
<Richard@litotes.demon.co.uk> wrote:

>Gene Wirchenko wrote:

[snip]

>Incidentally, to force type conversion to a string primitive the String 
>constructor can be used without the - new - operator, as in:-
>
>String(document.Contest.Last.value)
>
>- Which has been both recommended against on the grounds that it could 
>be confused with a miss-typed - new String(s); - and recommended as a 
>self-documenting and explicit means of forcing type-conversion to a 
>string. I am leaning towards the latter. Concatenating the empty string 

     I think I will, too.

>has been shown to be the most runtime efficient type-conversion to 
>string primitive operation.
>
>> Won't document.Contest.Last.value be string regardless?
>
>Yep.
>
>> I tried forcing one such to a number,
>
>How? Strings can be type-converted into number primitive values using, 
>say,  +string  or  Number(string)  (and commonly in the past or 
>string*1 , or  string-0 ).

     With an assignment statement.  The following statement was an
alert() to get the typeof() and reported string.

[snip]

>>     Is this ""+ a bit of superstition, or am I missing something?
>
>No, it appears to be superstition in this case. Though there is a slight 
>possibility that - document.Contest.Last - was not referring to a form 
>element in this case (then you only have evidence of poor code design 
>rather than a mystical incantation).

     No, definitely a form element.  That is why I was puzzled about
it.

Sincerely,

Gene Wirchenko

[toc] | [prev] | [next] | [standalone]

#8429

In comp.lang.javascript message <fuodc799nhk0c9ofihe817rjtkjid2cq79@4ax.
com>, Fri, 18 Nov 2011 15:07:56, Gene Wirchenko <genew@ocis.net> posted:

>     What does "" + some form value do?  For example,
>          ""+document.Contest.Last.value
>Won't document.Contest.Last.value be string regardless?

Not inevitably.

        document.Contest = {} ;
        document.Contest.Last = {} ;
        document.Contest.Last.value = new Date() ;
        99 ;
        document.Contest.Last.value.getHours() ;

gives me 18 (YMMV)

>  I tried
>forcing one such to a number, and after the assignment, typeof()
>reported it was a string.
>
>     Is this ""+ a bit of superstition, or am I missing something?

It ensures a string value; for example, this

        var U ; typeof ( ''+U)          // gives 'string'
                typeof (+''+U)          // gives 'number'
                typeof (''+ +U)         // gives 'string'

Some coders find it easier to force a string rather than to become
confident that it is and always will be inevitably a string already.

-- 
 (c) John Stockton, nr London, UK.    ?@merlyn.demon.co.uk     Turnpike v6.05.
 Website  <http://www.merlyn.demon.co.uk/> - w. FAQish topics, links, acronyms
 PAS EXE etc. : <http://www.merlyn.demon.co.uk/programs/> - see in 00index.htm
 Dates - miscdate.htm estrdate.htm js-dates.htm pas-time.htm critdate.htm etc.

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.javascript

csiph-web