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

Groups > comp.lang.javascript > #34088 > unrolled thread

What is wrong with this code?

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

Contents

  What is wrong with this code? Bart Smissaert <bart.smissaert@gmail.com> - 2017-10-01 03:06 -0700
    Re: What is wrong with this code? John G Harris <niam@jghnorth.org.uk.invalid> - 2017-10-01 11:38 +0100
      Re: What is wrong with this code? Bart Smissaert <bart.smissaert@gmail.com> - 2017-10-01 03:49 -0700
      Re: What is wrong with this code? Bart Smissaert <bart.smissaert@gmail.com> - 2017-10-01 03:58 -0700
    Re: What is wrong with this code? Julio Di Egidio <julio@diegidio.name> - 2017-10-01 05:08 -0700
      Re: What is wrong with this code? Bart Smissaert <bart.smissaert@gmail.com> - 2017-10-01 06:32 -0700
      Re: What is wrong with this code? John G Harris <niam@jghnorth.org.uk.invalid> - 2017-10-01 15:04 +0100
        Re: What is wrong with this code? Julio Di Egidio <julio@diegidio.name> - 2017-10-01 07:33 -0700

#34088 — What is wrong with this code?

Having trouble getting the right syntax for JScript function.
I have run it through the online JSLint syntax checker, but not sure how
to correct the failings it found.
All is fine up to the very last block of code, so if I leave out that nested
loop I can register my function OK.

This is the JS code:

function JS_CountString(lParamCount, strParam1, lParam2, dParam3)

{

    var arrSearchVB = new VBArray(UDF.GetVariantArray(1));

    //no string to search so exit
    if (arrSearchVB.getItem(0) == -1) {
        return 0;
    }

    var arrFindVB = new VBArray(UDF.GetVariantArray(2));

    //no string to search for, so exit
    if (arrFindVB.getItem(0) == -1) {
        return 0;
    }

    var arrSearchJS = arrSearchVB.toArray();
    var arrFindJS = arrFindVB.toArray();

    //note we do not use new here
    var UBS = arrSearchJS.length;
    var UBF = arrFindJS.length;
    var i = 0;
    var n = 0;
    var lCount = 0;

    //error in this code block
    for (i = 0; i < UBS - UBF; i++) {
        for (n = 0; n < UBF; n++) {
           If (arrSearchJS[i + n] != arrFindJS[n]);
           { break; }  //no need for else here
           If (n == UBF)
           { lCount++; }
           }
        }
     }

return lCount;

}


-------------------------------------------------------------------------------
And these are the failings found by JSLint:

Problem at line 32 character 12: Missing 'new' prefix when invoking a constructor.

 If (arrSearchJS[i + n] != arrFindJS[n]);

Problem at line 33 character 14: Expected to see a statement and instead saw a block.

 { break; } //no need for else here

Problem at line 33 character 14: Stopping, unable to continue. (76% scanned).

 { break; } //no need for else here
-------------------------------------------------------------------------------



Any suggestions how this can be fixed?
Thanks in advance for any advice.


RBS

[toc] | [next] | [standalone]

#34089

On Sun, 1 Oct 2017 03:06:31 -0700 (PDT), Bart Smissaert
<bart.smissaert@gmail.com> wrote:

  <snip>
>    //error in this code block
>    for (i = 0; i < UBS - UBF; i++) {
>        for (n = 0; n < UBF; n++) {
>           If (arrSearchJS[i + n] != arrFindJS[n]);
>           { break; }  //no need for else here
>           If (n == UBF)
>           { lCount++; }
>           }
>        }
>     }

Re-indenting this gives :
    
    //error in this code block
for (i = 0; i < UBS - UBF; i++) 
  {
    for (n = 0; n < UBF; n++) 
      {
        If (arrSearchJS[i + n] != arrFindJS[n])
          ;
        { break; }  //no need for else here
        If (n == UBF)
          { lCount++; }
      }
  }     
}    // <== What is this close bracket supposed to close ?
     // and why do you have an unconditional break statement ?



  <snip>
>Problem at line 33 character 14: Expected to see a statement and instead saw a block.
>
> { break; } //no need for else here
  <snip>

Lint isn't making sense here. A block *is* a statement. The Standard
says so.

  John

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

#34090

On Sunday, 1 October 2017 11:38:52 UTC+1, John G Harris  wrote:
> On Sun, 1 Oct 2017 03:06:31 -0700 (PDT), Bart Smissaert
> <bart.smissaert@gmail.com> wrote:
> 
>   <snip>
> >    //error in this code block
> >    for (i = 0; i < UBS - UBF; i++) {
> >        for (n = 0; n < UBF; n++) {
> >           If (arrSearchJS[i + n] != arrFindJS[n]);
> >           { break; }  //no need for else here
> >           If (n == UBF)
> >           { lCount++; }
> >           }
> >        }
> >     }
> 
> Re-indenting this gives :
>     
>     //error in this code block
> for (i = 0; i < UBS - UBF; i++) 
>   {
>     for (n = 0; n < UBF; n++) 
>       {
>         If (arrSearchJS[i + n] != arrFindJS[n])
>           ;
>         { break; }  //no need for else here
>         If (n == UBF)
>           { lCount++; }
>       }
>   }     
> }    // <== What is this close bracket supposed to close ?
>      // and why do you have an unconditional break statement ?
> 
> 
> 
>   <snip>
> >Problem at line 33 character 14: Expected to see a statement and instead saw a block.
> >
> > { break; } //no need for else here
>   <snip>
> 
> Lint isn't making sense here. A block *is* a statement. The Standard
> says so.
> 
>   John

>> wat is this close bracket supposed to close ?
I am not familiar with JS at all and I thought it was matching the the first
opening bracket in this code block.

>> and why do you have an unconditional break statement ?
I thought the condition was the preceding line:
If (arrSearchJS[i + n] != arrFindJS[n])


RBS

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

#34091

On Sunday, 1 October 2017 11:38:52 UTC+1, John G Harris  wrote:
> On Sun, 1 Oct 2017 03:06:31 -0700 (PDT), Bart Smissaert
> <bart.smissaert@gmail.com> wrote:
> 
>   <snip>
> >    //error in this code block
> >    for (i = 0; i < UBS - UBF; i++) {
> >        for (n = 0; n < UBF; n++) {
> >           If (arrSearchJS[i + n] != arrFindJS[n]);
> >           { break; }  //no need for else here
> >           If (n == UBF)
> >           { lCount++; }
> >           }
> >        }
> >     }
> 
> Re-indenting this gives :
>     
>     //error in this code block
> for (i = 0; i < UBS - UBF; i++) 
>   {
>     for (n = 0; n < UBF; n++) 
>       {
>         If (arrSearchJS[i + n] != arrFindJS[n])
>           ;
>         { break; }  //no need for else here
>         If (n == UBF)
>           { lCount++; }
>       }
>   }     
> }    // <== What is this close bracket supposed to close ?
>      // and why do you have an unconditional break statement ?
> 
> 
> 
>   <snip>
> >Problem at line 33 character 14: Expected to see a statement and instead saw a block.
> >
> > { break; } //no need for else here
>   <snip>
> 
> Lint isn't making sense here. A block *is* a statement. The Standard
> says so.
> 
>   John

Yes, thanks, last closing curly brace was the problem. Should have seen that.
Also there is no need for the semi-colon after:
If (arrSearchJS[i + n] != arrFindJS[n])
But JSLint insisted on that.
Function registers fine now, just see now if it works.

RBS
But

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

#34092

On Sunday, October 1, 2017 at 12:06:41 PM UTC+2, Bart Smissaert wrote:

> Any suggestions how this can be fixed?

If and else, the keywords, are lower-case...

Julio

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

#34093

On Sunday, 1 October 2017 13:08:54 UTC+1, Julio Di Egidio  wrote:
> On Sunday, October 1, 2017 at 12:06:41 PM UTC+2, Bart Smissaert wrote:
> 
> > Any suggestions how this can be fixed?
> 
> If and else, the keywords, are lower-case...
> 
> Julio

Ah, OK, thanks.

RBS

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

#34094

On Sun, 1 Oct 2017 05:08:43 -0700 (PDT), Julio Di Egidio
<julio@diegidio.name> wrote:

>On Sunday, October 1, 2017 at 12:06:41 PM UTC+2, Bart Smissaert wrote:
>
>> Any suggestions how this can be fixed?
>
>If and else, the keywords, are lower-case...

I didn't notice that, but why didn't JSLint ?

  John

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

#34095

On Sunday, October 1, 2017 at 4:04:55 PM UTC+2, John G Harris wrote:
> On Sun, 1 Oct 2017 05:08:43 -0700 (PDT), Julio Di Egidio
> <julio@diegidio.name> wrote:
> 
> >On Sunday, October 1, 2017 at 12:06:41 PM UTC+2, Bart Smissaert wrote:
> >
> >> Any suggestions how this can be fixed?
> >
> >If and else, the keywords, are lower-case...
> 
> I didn't notice that, but why didn't JSLint ?

Because JSLint isn't *that* smart: keep in mind that If(something) by the
JS parser is simply a function call.  Hence all those seemingly weird
warnings.  Also keep in mind that JSLint is parsing, cross-referencing, then
heuristics: that warning about blocks may be slightly imprecise, but it is
warning that blocks per se do not make scope in JS so there is probably a
programming error lurking there (as indeed there was in our case).  But
JSLint can be tweeked to one's personal or a team's style: for example, I
love leveraging hoisting so I disable the rule about re-declared variables,
but mileages vary...

Julio

[toc] | [prev] | [standalone]

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

csiph-web