Search for <n>th occurrence of regex

The following form searches for the <n>th occurence of a regex:

/^(?:.*?\Kregex){<n>}/
[download]

This is based on a regex I discovered during my Vim hacking days.

Vim regex provides the atom \zs to set the start of a regex match. It wasn't until recent Perl 5.10+ that the \K construct was introduced.

# only Perl 5.10+
my $a;
my $ref = "abc x abc y abc z abc\n";

# search 3rd occurence of 'abc' from start-of-line
($a = $ref) =~ s/^(?:.*?\Kabc){3}/___/;
print $a;

# search 4th occurence of 'abc' from start-of-line
($a = $ref) =~ s/^(?:.*?\Kabc){4}/___/;
print $a;

# omit ^ to search every 2nd occurence of 'abc'
($a = $ref) =~ s/(?:.*?\Kabc){2}/___/g;
print $a;

# omit ^ to search every 1st occurence of 'abc'
($a = $ref) =~ s/(?:.*?\Kabc){1}/___/g;
print $a;

__END__

abc x abc y ___ z abc
abc x abc y abc z ___
abc x ___ y abc z ___
___ x ___ y ___ z ___
[download]

If appropriate, this could be added as an update to: How do I change the Nth occurrence of something?

Comment on Search for <n>th occurrence of regex Select or Download Code

Replies are listed 'Best First'.
Re: Search for <n>th occurrence of regex by ikegami (Patriarch) on Nov 05, 2008 at 20:49 UTC
As a corollary, if "`abc`" is placed before "`\K`", the substitution appends instead of replaces. `# Appends to the 3rd occurence of 'abc' from start-of-line ($a = $ref) =~ s/^(?:.?abc\K){3}/___/; print $a; # Appends to the 4th occurence of 'abc' from start-of-line ($a = $ref) =~ s/^(?:.?abc\K){4}/___/; print $a; # omit ^ to append to every 2nd occurence of 'abc' ($a = $ref) =~ s/(?:.?abc\K){2}/___/g; print $a; # omit ^ to append to every 1st occurence of 'abc' ($a = $ref) =~ s/(?:.?abc\K){1}/___/g; print $a;` [download] Of course, the "`{1}`" is optional in the last pattern.	[reply] [d/l] [select]

Replies are listed 'Best First'.

Re: Search for <n>th occurrence of regex
by ikegami (Patriarch) on Nov 05, 2008 at 20:49 UTC

abc

\K

# Appends to the 3rd occurence of 'abc' from start-of-line
($a = $ref) =~ s/^(?:.*?abc\K){3}/___/;
print $a;

# Appends to the 4th occurence of 'abc' from start-of-line
($a = $ref) =~ s/^(?:.*?abc\K){4}/___/;
print $a;

# omit ^ to append to every 2nd occurence of 'abc'
($a = $ref) =~ s/(?:.*?abc\K){2}/___/g;
print $a;

# omit ^ to append to every 1st occurence of 'abc'
($a = $ref) =~ s/(?:.*?abc\K){1}/___/g;
print $a;
[download]

Of course, the "{1}" is optional in the last pattern.

[reply]
[d/l]
[select]