This is not the case in regex though, $str =~ qq/$ x/; will use the value of $x as a regular expression, while $str =~ qr/$ x/; just doesn't match anything ($ is the position before \n, so the characther after that can neither be a space nor x). This means whitespace is not ignored when doing so yields a non-sensical expression :P

perl -E "$x = 'Hello'; say $ x =~ qr/$ x/ ? 'Match': 'No match'"
No match

perl -E "$x = 'Hello'; say $ x =~ qq/$ x/ ? 'Match': 'No match'"
Match
[download]

Edit: it's whitespace, not blankspace, got the wrong word :)

Good example of why you should always use warnings;?

$ perl -wE '$name = q(Fred); $debt = 50; say qq(You pay me ${debt}$ 00
+ cents, $name, or else!)'

Use of uninitialized value $00 in concatenation (.) or string at -e li
+ne 1.
You pay me 50 cents, Fred, or else!
[download]

Update: Also a good demo of why interpolation can be dangerous and the case for using sprintf?

$ perl -E '$name = q(Fred); $debt = 50; say sprintf q(You pay me %s$ 0
+0 cents, %s, or else!), $debt, $name'

You pay me 50$ 00 cents, Fred, or else!
[download]

This is a good thing for you, infact you can print a boublequoted string as multiline expression.

If you by other hand insert (even if allowed) white spaces between sigils and identifiers, you are somehow shooting on your own feet.

Anyway perlintro states:

> Whitespace is irrelevant:

print
         "Hello, world"
         ;
[download]

L*

Whitespace is irrelevant

p r i n t "Hello, world"; doesn't print "Hello, world" though, so you can't just put whitespace anywhere.
Also $x = "Hello"; print "$ x" will print "Hello" but @x = "Hello"; print "@ x" will print "@ x", and @$ x, $# x and $ #x are all syntax errors unless you remove the spaces.

#Edited: added those examples


use feature 'say';

our $x = "Hello";
our @y = "Hi";
our $z = ["Bonjour"];
while (<DATA>)
{
  chomp;
  print "$_ \t";
  no warnings;
  eval "$_; 1" or say $@ =~ tr/\n/ /r;
}
__DATA__
say "$x";
say "$ x";
say "@y";
say "@ y";
say  @ y;
say $#y;
say $# y;
say $ #y;
say @ $z;
say @$ z;
[download]

say "$x";       Hello
say "$ x";      Hello
say "@y";       Hi
say "@ y";      @ y
say  @ y;       Hi
say $#y;        0
say $# y;       syntax error at (eval 6) line 2, at EOF   (Might be a 
+runaway multi-line ;; string starting on line 1)
say $ #y;       syntax error at (eval 7) line 1, at EOF
say @ $z;       Bonjour
say @$ z;       syntax error at (eval 9) line 1, near "@$ z"
[download]

If you [...] insert [...] white spaces between sigils and identifiers, you are somehow shooting on your own feet.

If you do it on purpose yes, but it's pretty easy to do it by mistake, either because of a typo, or because you expected spaces to be relevant inside a string, or even in some cases without noticing what's wrong, and perl may not warn you:

use v5.14;
use strict;
use warnings;

use constant MASK => 0xF0;
use constant VAL  => 0x0F;

sub get_val1   { VAL }
sub get_val2() { VAL }
sub get_mask   { MASK }

# Is & a function sigil or the bitwise and?
say VAL & get_mask;
say get_val1 & get_mask;
say get_val1 & MASK;
say get_val2 & MASK;
say get_val1() & MASK;
[download]

0
15
15
0
0
[download]

> Whitespace is irrelevant

Yes indeed, and don't I over-abuse this fact with non-standard indentation, building geometric figures from my source, to help me better read and remember what it was, in 1-2 years. But I never saw sigils standing separate, even in code meant for puzzling and obfuscation, e.g. this, have you?

chromatic said

First of all, there's no "BNF grammar" in the interpreter. There's a parser and a lexer which you could, if you were enterprising enough, potentially represent in Backus-Naur Form, but it's not expressed as such anywhere within the Perl source code, and there is no singular representation of grammar within the Perl source.

I don't know much about the subject, but I'd be willing to believe that chromatic does.

You say "especially in use declarations", are you actually making a reference to the thread this answer came from?