And your code doesn't even compile, it's full of syntax errors. My thoughts are that you haven't even tried this in Perl, but instead transalted it here from some other programming language, just for the purpose of posting it here.

Update: My mistake, there wheren't as many errors as I first thought. I tried multiplying all numbers by 100, and it gives a different result now, more like you expected.

You've clearly been hit by the "floating point numbers are not exact" syndrome.

You are using "real" values, but what you really need is "string" values:

  $x = 1;   # integer
  $y = 6;   # integer
  for( 1 .. 10 )
  {
    $start_pt="$x.$y";   # string
    print " $start_pt ";
    $y++;
  }
  # 1.6  1.7  1.8  1.9  1.10  1.11  1.12  1.13  1.14  1.15
[download]

Hello,

thanks a lot for the reply.That seems to be a nice
and simple solution to the problem.I'll try it out.
Thanx

Hi,
Here's my changed code(part of it).



foreach $mitem(@missing) { 
  $result=$t->search(-forwards,"**",'end');
  #print"\nThe start pos of the pattern:";
  #print "\n$result";        #1.4
  
  $start=$result+.2;
  #print "\n$start";         #1.6
     
  $x=substr($start,0,1);# integer
  $y = substr($start,2,2);# integer
  #print $y;
  
  $z=$y + 1;
  #print $z;
    
  for( 1 .. 10 )
  {
    $start_pt="$x.$y";   # string
    #print " $start_pt ";
    $y++;
    $end_pt="$x.$z";   # string
    #print " $end_pt ";
    $z++;
    
    $char_read=$t->get("$start_pt","$end_pt");
    #print "\nThe charcter:";
    #print "\n$char_read"
    push @chars,$char_read;
    #print "\nThe word:";   
    #print "\n@chars";
  }
 
   #$x=1;
   #$y=6;
      
   #$l=length(@chars);
   #$fract=(($y+$l)-1);
   #$wd_end=$x.$fract;
                     
#$t->tagConfigure("wrong",-foreground=>"red");    
#$t->tagAdd("wrong","$start","$wd_end");
  
#print "\n@chars";
  
 }
[download]


I'm using the above code inside a subroutine, which is called when a button is clicked in a Tk text widget.

This for loop: for( 1 .. 10 ) {

needs to read the word immediately next to the **
(the first time this pattern is found by $t->search)
but here this loop reads two words.The loop should stop
reading characters once it encounters a whitespace.
I've tried putting a do loop outside the for loop like this



do { 
               
     $white_space=" ";

  for( 1 .. 10 )
   {
   .....
    stuff inside for loop.
   .....
   }
  
 } until ($char_read eq $white_space);
[download]


but this code just hangs when executed(Argh!).

What is the problem here?

Each time the outer foreach loop is executed,
the search function should look for the first
occurence of ** in the widget window.The word next
to the **'s needs to be read, and it's font color set to
red, using a tag,and the two **'s in front of the word
deleted.

The next time around the foreach loop the next ocurrence
of ** should be located, the next word extracted ,it's
colour changed, the **'s deleted, and so on.I hope this
works(!).



e.g. **word  **nextword
[download]


In the above example I need the for loop to read
'word' and stop once a whitespace is encountered.
The 'nextword' should be taken care of the next time
the foreach loop runs.

Of course we can read a word like this too instead
of char by char:



$char_read=$t->get("$start_pt","$start_pt wordend");
[download]


But I'm working on text that is in another language(font)
and the 'wordend' modifier does not work in my case, though
it works fine with text in an English font(I tried it)
So all the trouble.
Any help would be appreciated.
Thanx.

You wrote:

    $start=$result+.2;
    $x = substr($start,0,1);# integer
    $y = substr($start,2,2);# integer
[download]

It would be better to work in string mode all the time. If you had "1.8" you would get "2" instead of "1.10".

How about:

    ($x, $y) = split '\.' => $result;
    $y += 2;
[download]

Hi,
yes that helps.I would have messed up the indices(2
instead of 1.10 otherwise)

I still have the prob. of reading the word next to the
**'s and stopping at a whitespace, though.
:)

Beyond the commments about why the comparisons don't work as expected, I would like to point out that if you were to use if-elsif-else conditionals, you could dump the use of labels and last and make the code a bit shorter as well as more readable. You can check out perlsyn for more information on control structures in perl.

antirice    
The first rule of Perl club is - use Perl
The ith rule of Perl club is - follow rule i - 1 for i > 1