Use match within match expression

I was wondering whether I could use a part of the match within the match expression.

I would need that for automatic program testing, where some random pattern is occurring twice in the output, like this:

kfvggfhuvgdnjhrk=kfvggfhuvgdnjhrk

So the first attempt was using (for expect):

-re "\r\n\[cbdefghijklnrtuv\]{16}=\[cbdefghijklnrtuv\]{16}\r\n"

This works; however the strings left and right of = may be different.

In Perl 5.26.1 I was playing with the debugger, I I found out this (with a simplified example):

main::(-e:1):   1
  DB<1> x 'a=a' =~ /^(a)=$1$/
  empty array
  DB<2> x 'a=a' =~ /^(a)=\g1$/
0  'a'
  DB<3> x 'ab=ab' =~ /^(ab)=\g1$/
0  'ab'
  DB<4> x 'ab=ab' =~ /^(ab)=\g1b$/
  empty array
  DB<5> q

So for some reason $1 does not work, but \g1 does!

Now I wonder if something similar exists for expect regular expressions.

If I understood correctly, if you want to match the same text as already captured text, you need to:

• use capturing group for that desired text,
• use \1 to reference first capturing group, \2 for second, and so on, in our case \1 will be enough.

So the pattern to use is

^([^=]+)=\1$

Explanation: ^ - assert start of text

(...) - capturing group

[^=]+ - match one or more characters other than =

= - match = literally

\1 - match the same text as in first capturing group

$ - assert end of text

Regex Demo