Re: Character Text Delimiters

[Sue D. Nymme]

Have you considered splitting the string on "), XX("? Something like:

@chunks = split /\), ([A-Z][A-Z])\(/, $input;

That'll give you all the values interspersed with the prefixes (with the first and last elements being wacky because you matched on the stuff that goes between each value).

In fact, you could make the split pattern a bit more complex, to separate out the first "AA(" and the last ")":

@chunks = split /(?: (?:^|\),\s) ([A-Z][A-Z])\( | \)$)/x, $input;
shift @chunks;  # remove extra blank element at front

Then you can use part from List::MoreUtils to partition the list into prefixes and values, or you could just assign it to a hash and use it directly:

use List::MoreUtils qw(part);
my @chunks = ...;  # as above
shift @chunks;     # as above
my $ix = 0;
my ($prefixes,$texts) = part {++$ix % 2} @chunks;
# or
my %text = @chunks;
say "Text value AF is $text{AF}";

[Ninthwave]

I did try something similar but on the last pattern the text value in one case was

AC(Departament d'Astronomia i Astrofísica. Universitat de València. C/Dr. Moliner 50, 46100 Burjassot (València), Spain)

And was having problems with the (Valencia) portion and if I used the | was getting too much but I will try this or variant to see if it can be consistent. Sorry I don't have the exact pattern from the previous attempt that was similar to this split.