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

Groups > comp.lang.javascript > #25942 > unrolled thread

JS Array problem

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

Contents

  JS Array problem "Mel Smith" <syntel@cox.net> - 2014-10-20 15:11 -0600
    Re: JS Array problem Denis McMahon <denismfmcmahon@gmail.com> - 2014-10-20 22:25 +0000
      Re: JS Array problem "Mel Smith" <syntel@cox.net> - 2014-10-21 08:33 -0600
    Re: JS Array problem danca <cyberdanny@tiscalinet.it> - 2014-10-21 09:08 +0200
      Re: JS Array problem Une Bévue <unbewusst.sein@fai.invalid> - 2014-10-21 09:42 +0200
        Re: JS Array problem "Evertjan." <exxjxw.hannivoort@inter.nl.net> - 2014-10-21 09:47 +0200
          Re: JS Array problem Une Bévue <unbewusst.sein@fai.invalid> - 2014-10-21 09:51 +0200
            Re: JS Array problem "Evertjan." <exxjxw.hannivoort@inter.nl.net> - 2014-10-21 09:53 +0200
          Re: JS Array problem Une Bévue <unbewusst.sein@fai.invalid> - 2014-10-21 09:52 +0200
            Re: JS Array problem "Evertjan." <exxjxw.hannivoort@inter.nl.net> - 2014-10-21 10:36 +0200
        Re: JS Array problem "Mel Smith" <syntel@cox.net> - 2014-10-21 08:36 -0600
      Re: JS Array problem "Mel Smith" <syntel@cox.net> - 2014-10-21 08:35 -0600
        Re: JS Array problem Thomas 'PointedEars' Lahn <PointedEars@web.de> - 2014-10-21 19:02 +0200
    Re: JS Array problem "Christoph M. Becker" <cmbecker69@arcor.de> - 2014-10-21 13:12 +0200
      Re: JS Array problem "Mel Smith" <syntel@cox.net> - 2014-10-21 08:38 -0600
    Re: JS Array problem "Mel Smith" <syntel@cox.net> - 2014-10-21 08:29 -0600
      My 'actual' code "Mel Smith" <syntel@cox.net> - 2014-10-21 11:20 -0600
        Re: My 'actual' code Ben Bacarisse <ben.usenet@bsb.me.uk> - 2014-10-21 19:52 +0100
          Re: My 'actual' code Thomas 'PointedEars' Lahn <PointedEars@web.de> - 2014-10-21 21:23 +0200
    Re: JS Array problem Violet Deng <dengviolet@gmail.com> - 2014-10-28 02:03 -0700
      Re: JS Array problem "Evertjan." <exxjxw.hannivoort@inter.nl.net> - 2014-10-28 10:08 +0100

Page 1 of 2  [1] 2  Next page →

#25942 — JS Array problem

Problem:

   I have seveal undefined objects/vrbls (?) in an array.  I wish each of 
these objects to
   acquire a new value in a loop thru the array and
   my brain has gone dead again :(

// Assume :

var a,b,c,d ;

var abcd = [a,b,c,d] ;

var somestr = "" ;

for (var i= 0; i<abcd.length; i++) {
   somestr = i.toString() ;
   // I wish to place the current value of somestr in each
   // of the named but undefined objects/vrbls that are carried in the abcd 
array

   // e.g.,
   abcd[i] = somestr ;  // but this, of course, ruins everything :(

   // what is the correct statement please ?
   }

Result I wish here is:

// a has value '0'
// b has value '1'
// c has value '2'
// d has value '3'


Thanks for (another) kick-in-the-butt.

-Mel Smith

[toc] | [next] | [standalone]

#25943

On Mon, 20 Oct 2014 15:11:45 -0600, Mel Smith wrote:

> var abcd = [a,b,c,d] ;

This creates an array that contains the values of the 4 variables a, b, c 
and d. It does not create a connection between the array members and the 
original variables such that editing the array values will change the 
variables.

Why do you need the four values to be addressable as a,b,c,d as well as 
abcd[n]?

-- 
Denis McMahon, denismfmcmahon@gmail.com

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

#25959

"Denis McMahon" <denismfmcmahon@gmail.com> wrote in message 
news:m2424p$mdn$1@dont-email.me...
> On Mon, 20 Oct 2014 15:11:45 -0600, Mel Smith wrote:
>
>> var abcd = [a,b,c,d] ;
>
> This creates an array that contains the values of the 4 variables a, b, c
> and d. It does not create a connection between the array members and the
> original variables such that editing the array values will change the
> variables.
>
> Why do you need the four values to be addressable as a,b,c,d as well as
> abcd[n]?
>

In my 'real' app, there are consistency probs where those vrbls and many 
others are built and returned to the server under those vrbl names a few 
lines below that 'collection area'.

Thanks, and please not my reply to myself

-Mel

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

#25946

Il 20/10/2014 23:11, Mel Smith ha scritto:
> Problem:
>
>     I have seveal undefined objects/vrbls (?) in an array.  I wish each of
> these objects to
>     acquire a new value in a loop thru the array and
>     my brain has gone dead again :(
>
> // Assume :
>
> var a,b,c,d ;
>
> var abcd = [a,b,c,d] ;
>
> var somestr = "" ;
>
> for (var i= 0; i<abcd.length; i++) {
>     somestr = i.toString() ;
>     // I wish to place the current value of somestr in each
>     // of the named but undefined objects/vrbls that are carried in the abcd
> array
>
>     // e.g.,
>     abcd[i] = somestr ;  // but this, of course, ruins everything :(
>
>     // what is the correct statement please ?
>     }
>
> Result I wish here is:
>
> // a has value '0'
> // b has value '1'
> // c has value '2'
> // d has value '3'
>
>
> Thanks for (another) kick-in-the-butt.
>
> -Mel Smith
>
>
>
Maybe something like this?

var a,b,c,d ;
var abcd = ["a","b","c","d"] ;
for (var i= 0; i<abcd.length; i++) {
      eval(abcd[i]+"="+i);
}
Dan

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

#25947

Le 21/10/14 09:08, danca a écrit :
> var abcd = ["a","b","c","d"] ;

or :
var abcd = 'a b c d'.split(' ');

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

#25948

=?ISO-8859-15?Q?Une_B=E9vue?= <unbewusst.sein@fai.invalid> wrote on 21 okt 
2014 in comp.lang.javascript:

> Le 21/10/14 09:08, danca a écrit :
>> var abcd = ["a","b","c","d"] ;
> 
> or :
> var abcd = 'a b c d'.split(' ');

Skip the whitespace noise:

var abcd = 'abcd'.split('');

-- 
Evertjan.
The Netherlands.
(Please change the x'es to dots in my emailaddress)

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

#25949

Le 21/10/14 09:47, Evertjan. a écrit :
> Skip the whitespace noise:
>
> var abcd = 'abcd'.split('');

fine !

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

#25951

Une Bévue <unbewusst.sein@fai.invalid> wrote on 21 okt 2014 in 
comp.lang.javascript:

> Le 21/10/14 09:47, Evertjan. a écrit :
>> Skip the whitespace noise:
>>
>> var abcd = 'abcd'.split('');
> 
> fine !

or more generic:

var abcd = [];
for (i=97;i<101;i++)
   abcd.push(String.fromCharCode(i));



-- 
Evertjan.
The Netherlands.
(Please change the x'es to dots in my emailaddress)

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

#25950

Le 21/10/14 09:47, Evertjan. a écrit :
> Skip the whitespace noise:
>
> var abcd = 'abcd'.split('');

I'm using for words then, i need a separator.

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

#25952

Une Bévue <unbewusst.sein@fai.invalid> wrote on 21 okt 2014 in 
comp.lang.javascript:

> Le 21/10/14 09:47, Evertjan. a écrit :
>> Skip the whitespace noise:
>>
>> var abcd = 'abcd'.split('');
> 
> I'm using for words then, i need a separator.

Not needed if all words have the same length:

var months = 'janfebmaraprmayjunetc'.match(/.{3}/g);

==================================

having different length words yes, a separator is usefull:

var months = 'january/february/march/april/may/june/etc'.split('/');

var months = 'january/february/march/april/may/june/etc'.match(/[^\/]+/g);

==================================

but even then not always necessary:

var months = 'JanuaryFebruaryMarchAprilMayJuneEtc'.match(/[A-Z][a-z]+/g);


-- 
Evertjan.
The Netherlands.
(Please change the x'es to dots in my emailaddress)

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

#25961

"Une Bévue" <unbewusst.sein@fai.invalid> wrote in message 
news:m252oq$gcb$1@shakotay.alphanet.ch...
> Le 21/10/14 09:08, danca a écrit :
>> var abcd = ["a","b","c","d"] ;
>
> or :
> var abcd = 'a b c d'.split(' ');

????
Don't understand

-Mel

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

#25960

"danca" <cyberdanny@tiscalinet.it> wrote in message 
news:m250og$r1l$1@dont-email.me...
> Il 20/10/2014 23:11, Mel Smith ha scritto:
>>
>>
> Maybe something like this?
>
> var a,b,c,d ;
> var abcd = ["a","b","c","d"] ;
> for (var i= 0; i<abcd.length; i++) {
>      eval(abcd[i]+"="+i);
> }


Hi Dan:

    Yes, That's what I was looking for.

Thank you !

-Mel 

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

#25965

Mel Smith wrote:

> "danca" <cyberdanny@tiscalinet.it> wrote […]:
>> Il 20/10/2014 23:11, Mel Smith ha scritto:
>> Maybe something like this?
>>
>> var a,b,c,d ;
>> var abcd = ["a","b","c","d"] ;
>> for (var i= 0; i<abcd.length; i++) {
>>      eval(abcd[i]+"="+i);
>> }
> 
> […]
>     Yes, That's what I was looking for.

Hardly, as it is harder to debug and more error-prone.  It will not work in 
strict mode (the scope chain of eval code is empty then), which you should 
be using in order to avoid common programming errors.

Also,

  for (var i = 0, len = abcd.length; i < len; ++i)

is more efficient when “abcd.length” does not change;

  for (var i = abcd.length; i--;)

is even more efficient if the array is not sparse and order does not matter.  

If the array-like object is sparse, the most efficient way, where order 
matters and non-index own properties are not expected, is probably

  var indexes = Object.keys(abcd);
  for (var i = 0, len = indexes.length; i < len; ++i)
  {
    var index = indexes[i];
    abcd[index] = …;
  }

(Array indexes are names of own enumerable properties.  If non-index own 
properties are expected, an additional test of the property name is 
required.)

-- 
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.

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

#25956

Mel Smith wrote:

> var a,b,c,d ;
> 
> var abcd = [a,b,c,d] ;
> 
> var somestr = "" ;
> 
> for (var i= 0; i<abcd.length; i++) {
>    somestr = i.toString() ;
>    // I wish to place the current value of somestr in each
>    // of the named but undefined objects/vrbls that are carried in the abcd 
> array
> 
>    // e.g.,
>    abcd[i] = somestr ;  // but this, of course, ruins everything :(
> 
>    // what is the correct statement please ?

As you only want *undefined* values to be replaced, you have to check
for that:

     if (typeof(abcd[i]) == "undefined") {
         abcd[i] = somestr;
     }

>    }

-- 
Christoph M. Becker

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

#25962

Hi All:

    Thank you for the responses.

    I'm off and running again  :))

-Mel

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

#25958

I said:

> // Assume :
>
> var a,b,c,d ;
>
> var abcd = [a,b,c,d] ;

    BEFORE I read any of the many replies to my post, I want to say that I 
now realize my question/problem was based on not realizing that *values* not 
*references/pointers* were placed in my abcd vrbl.  I *should* (I think) 
have inserted referenes to the vars a,b,c,d in the abcd array *not* the 
undefined value of those vars.

    My 'insights' always seem to occur *after* I post my problem :((

    Now I'm going to read the rest of the many posts.


-Mel

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

#25966 — My 'actual' code

Hi All:

   FYI, below is the actual code I'm currently working on.
   I've included Danca's Eval() code suggestions and
   thought about Evertjan's improvement.

   btw, it works great !


    Also, I'll be examining Thomas Lahn's other suggestions this afternoon. 
In my 'for' loops, I usually pre-calc the length instead of using .length in 
the header).  I didn't bother this time but will change it later.

Thanks to all

-Mel


<snipped about 60 lines about in this func>
//vars i,j are local vars from above of course
// the values I'm gathering are single-char values in a 5 row, 23 column 
input table, where there are only 18 input elements in each row.
// As the user inputs single values, he is 'kicked' to the next element 
(either verticall or horizontally), and the read-only fields in tge row are 
updated
// there will be five (for now) strings submitted to the server - one string 
from each row
**************************
// now have to code the table for all internal values

var hscells,ptcells,ftcells,dlcells,rlcells ;
var hsstr,ptstr,ftstr,dlstr,rlstr ;

hsstr = ptstr =  ftstr = dlstr =  rlstr = "" ;

hscells = getobj("ihsrow").cells ;
ptcells = getobj("iptrow").cells ;
ftcells = getobj("iftrow").cells ;
dlcells = getobj("idlrow").cells ;
rlcells = getobj("irlrow").cells ;

var allrows = [hscells,ptcells,ftcells,dlcells,rlcells] ;
var allstrs = ["hsstr","ptstr","ftstr","dlstr","rlstr"] ;   // could use 
.split() maybe


var rowcells,rowstrval ;
var numcells = hscells.length ; // each row has same number of cells
var cellobj,cellid,cellval ;

for (j=0;j<allrows.length;j++) {
   rowcells = allrows[j] ;
   rowstrval = "" ;

   for (i=1;i<numcells;i++) {

      if (i === 10) continue ;

      cellobj = rowcells[i].firstChild ;
      cellid = cellobj.id ;
      cellval = getobj(cellid).value ;

      if (empty(cellval)) cellval = "X" ;

      // do work on building a string here
      if (i > 19) break ;
      rowstrval += cellval +"^" ;
      }

   // select string reference and place composite value in it
   // next is suggestion by Dan and later Evertjan from my JS ng on Oct 
21/14
   eval(allstrs[j] + '= "' + rowstrval+'"') ;

   }

// testing code here
alert("hsstr="+hsstr) ;
alert("ptstr="+ptstr) ;
alert("ftstr="+ftstr) ;
alert("dlstr="+dlstr) ;
alert("rlstr="+rlstr) ;


// much more to be done but ...
return false ; // for now ...

// much more work to be done below

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

#25969 — Re: My 'actual' code

"Mel Smith" <syntel@cox.net> writes:

>    FYI, below is the actual code I'm currently working on.

As is so often the case, the actual code shows an altogether different
way to do things that avoids the problems in the question as asked!

<snip>
> <snipped about 60 lines about in this func>
> //vars i,j are local vars from above of course
> // the values I'm gathering are single-char values in a 5 row, 23 column 
> input table, where there are only 18 input elements in each row.
> // As the user inputs single values, he is 'kicked' to the next element 
> (either verticall or horizontally), and the read-only fields in tge row are 
> updated
> // there will be five (for now) strings submitted to the server - one string 
> from each row
> **************************
> // now have to code the table for all internal values
>
> var hscells,ptcells,ftcells,dlcells,rlcells ;
> var hsstr,ptstr,ftstr,dlstr,rlstr ;
>
> hsstr = ptstr =  ftstr = dlstr =  rlstr = "" ;
>
> hscells = getobj("ihsrow").cells ;
> ptcells = getobj("iptrow").cells ;
> ftcells = getobj("iftrow").cells ;
> dlcells = getobj("idlrow").cells ;
> rlcells = getobj("irlrow").cells ;
>
> var allrows = [hscells,ptcells,ftcells,dlcells,rlcells] ;
> var allstrs = ["hsstr","ptstr","ftstr","dlstr","rlstr"] ;   // could use 
> .split() maybe

It looks like you've sliced an object into different variables.  I'd put
those objects back together again:

  var all = [
      { cells: getobj("ihsrow").cells,
        string: ""
      },
      { cells: getobj("ipsrow").cells,
        string: ""
      },
      { cells: getobj("ifsrow").cells,
        string: ""
      },
      { cells: getobj("idsrow").cells,
        string: ""
      },
      { cells: getobj("irsrow").cells,
        string: ""
      }
  ];

though I'd be tempted to build these using a function -- only one letter
is needed to create the different objects.

Now when you iterate through the array, you have the cells the string
you are building linked together.  No need to link them by a name and an
eval call (unless I've missed some key part from the code I cut -- I did
scan it in rather a hurry).

<snip code>
-- 
Ben.

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

#25970 — Re: My 'actual' code

Ben Bacarisse wrote:

>   var all = [
>       { cells: getobj("ihsrow").cells,
>         string: ""
>       },
>       { cells: getobj("ipsrow").cells,
>         string: ""
>       },
>       { cells: getobj("ifsrow").cells,
>         string: ""
>       },
>       { cells: getobj("idsrow").cells,
>         string: ""
>       },
>       { cells: getobj("irsrow").cells,
>         string: ""
>       }
>   ];

This is asking for a DRY LART.
 
> though I'd be tempted to build these using a function -- only one letter
> is needed to create the different objects.

A function is not necessary:

  var names = ["ihsrow", "ipsrow", "ifsrow", "idsrow", "irsrow"];
  var all = [];

  for (var i = 0, len = names.length; i < len; ++i)
  {
    all.push({cells: getobj(names[i]).cells, string: ""});
  }

But (with an implementation of ECMAScript Ed. 5’s Array.prototype.map) with 
regard to code length at least it can be optimized to

  var all = names.map(function (name) {
    return {cells: getobj(name).cells, string: ""};
  });

and maybe even (longer, though)

  var all = "hpfds".split("").map(function (letter) {
    return {cells: getobj("i" + letter + "srow").cells, string: ""};
  });
 
I suggest avoiding the identifier “all”, though: it is not self-explanatory, 
and already used by a deprecated property of the object referred to by 
“document”, which could interfere in an HTML context.

-- 
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.

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

#26003

eval can solve this problem, but it's not recommended for poor security problem. you can try
for (var i= 0; i<abcd.length; i++) { 
   window[i] = "whatever you want";
} 
this will be available if the a, b, c, d is belonged to window, you can also change window to any other namespace.

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

Page 1 of 2  [1] 2  Next page →

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

csiph-web