The problem is precisely that \1 in a character class is not a backreference: it refers to the ASCII character chr(1), an abbreviation of the octal escape sequence \001.

You can achieve what you want with a slightly more complicated approach using negative lookahead:

  m{
    \A    # anchor to start
    (\W)  # open text
    (?: (?!\1) . )*  # anything that isn't the closer
    \1    # close text
    (\W)  # separator
  }xs
[download]

That works for the general case, when you simply want to match a bunch of stuff not containing a given substring. In this case though, you want to match "up to the first occurrence" of that substring, so it's much simpler - you just need a minimal match:

  m{
    \A    # anchor to start
    (\W)  # open text
    .*?   # anything contained, up to the ...
    \1    # ... close text
    (\W)  # separator
  }xs
[download]

(I've taken the liberty of replacing your '+' with '*', on the assumption that you want to allow empty fields.)

Hugo

$foo = qq{^snafu^|^foobar^\n};

$foo =~ m/\A(\W) # \A instead of ^ and match first non-word
          .+?    # +? Minimal match everything that isn't in \1
          \1(\W) # Match non-word following the 2nd \1
         /xms;
$TEXT_QUAL = $1;
$FIELD_SEP = $2;
[download]



Having now found the proper way to attempt to acquire delimiters, the following questions how to utilize these new found delimiters.
Instead of creating one large regex, I'd perfer to store them in scalars, which is the core of this particular problem.

$foo =~ /\G$TEXT_QUAL(.*?)$TEXT_QUAL[$FIELD_SEP\n]/xmsgc;
[download]

Fails to work since the qualifiers are metacharacters used in regular expressions.




$foo =~ /\G\$TEXT_QUAL(.*?)\$TEXT_QUAL[\$FIELD_SEP\n]/xmsgc;
[download]

Fails to work as \$ is a literal $ followed by the name.




$foo =~ /\G\\$TEXT_QUAL(.*?)\\$TEXT_QUAL[\\$FIELD_SEP\n]/xmsgc;
[download]

Also Fails to work as \\ is is a literal \ The only way I've found is




$LIT_TEXT_QUAL = qq{\\$TEXT_QUAL};
$LIT_FIELD_SEP = qq{\\$FIELD_SEP};

$foo =~ /\G$LIT_TEXT_QUAL(.*?)$LIT_TEXT_QUAL[$LIT_FIELD_SEP\n]/xmsgc;
[download]



I do have reasons for using all those flags as this thread continues, however with the intent to get discrete answers to smaller problems I'm hoping to reduce the amount of new information my brain will have to process.

Basically this post is looking for a way to use any variable in a regex that may or may not contain metachacters. Edit:: OK yeah missed the boat on this one, answer is just quotemeta function, from CB thanx guys!

I think the problem is simply that the [^\1]+ is greedy. Changing it to [^\1]+? seems to work:

$foo = qq{^snafu^|^foobar^\n};

$foo =~ m/\A(\W)  # \A instead of ^ and match first non-word
          [^\1]+? # Match everything that isn't in \1
          \1(\W)  # Match non-word following the 2nd \1
         /xms;

$text_qual = $1;
$field_sep = $2;

print "[$text_qual]\n";
print "[$field_sep]\n";
[download]

[^]
[|]
[download]

Update: Nope I think I'm at least partially wrong too. Greedy or not, if \1 is ^, then [^\1]+ should stop at the first ^.

Changing [^\1]+ to the hardcoded [^^]+ shows that greedyness doesn't matter. So it does appear to have something to do with \1 inside character classes.

Update 2: I think .*? instead of [^\1]+ might work, but it contains the dreaded dot-star. I assume the real solution uses some form of lookahead, but I've never been good with those.

$foo = qq{^snafu^|^foobar^\n};

$foo =~ m/(\W)([^\1]+)\1(\W)/;
$text_qual = $1;
$field_sep = $3;
print "text: $text_qual\n";
print "field: $field_sep\n";
print $2;

print "\n2nd try\n";
$foo2 = qq{^snafu1|^foobar^\n};
$foo2 =~ m/(\W)([^\1]+)\1(\W)/;
$text_qual = $1;
$field_sep = $3;
print "text: $text_qual\n";
print "field: $field_sep\n";
print $2;
[download]

H:\script>perl majingz.pl
text: ^
field:

snafu^|^foobar
2nd try
text: ^
field:

snafu1|^foobar
[download]