Handling a html-doc in a variable instead of array

[adelante]

Hello, when I open a html template, and want to replace the marks with content, I have to do it line by line, can't it be done differently?? Or doesn't it really matter. It doesn't feel slow, but my sense tells me that it should take more time the way I'm doing it now (line-by-line, instead of ONE go).

This is how I would like to do it:

#I would like to open the html as a variable
open (TEM, "<template.htm");
my $html=<TEM>;
close(TEM);

$html =~ s/#content#/$content/i;
print $html;

But this is how I have to do it:

#I have to open the html as an array
open (TEM, "<template.htm");
my @html=<TEM>;
close(TEM);

foreach my $line(@html){
   $line =~ s/#content#/$content/i;
   print $line;
   }

I guess it takes more CPU to do it line by line, doesn't it? instead of the whole thing in ONE go?

What do I need to do to let the first version work?

 

[KevinADC]

open (TEM, "<template.htm");
my $html= [b]do{local $/; <TEM>};
close(TEM);

$html =~ s/#content#/$content/i[b]gs[/b];
print $html;

but I think line by line might actually be a little more efficient.

------------------------------------------
- Kevin, perl coder unexceptional!

 

[adelante]

Thanks! I didn't understand much of line 2 :S

But if you say that "line by line" (foreach) is more efficient, then I will stick to that.

I would just think that ONE run, would be faster and use less CPU, instead of introducing "my $line" 200 times (if my html-doc has 200 lines).

Thanks alot!

 

[Kirsle]

This might be even more efficient:

#I have to open the html as an array
open (TEM, "<template.htm");

while (<TEM>) {
   $_ =~ s/#content#/$content/i;
   print $_;
   }

close(TEM);

Using an array or a scalar, your entire 200 line file has to be loaded into memory all at once. This way it'll only read one line at a time into memory, substitute #content# and print it.

-------------
Cuvou.com | The NEW Kirsle.net

 

[KevinADC]

>> But if you say that "line by line" (foreach) is more efficient, then I will stick to that.

I'm not positve, benchmarking the code with some test data would show one way or the other.

------------------------------------------
- Kevin, perl coder unexceptional!