"Campbell, Lance" <lance@illinois.edu> writes:
> I don’t know the best way to do this. I need to do a find and replace in text fields. The value I need to find and
replacemay occur more than once in each field per record.
> The value I am trying to match on:
> Starts with a single { .
> Ends with a single } .
> In between these brackets can be the characters 0-9, a-z, A-Z, hyphens and underscores. But no spaces. These
characterscould be in any order.
> The replacement value on a match is the same as what was found except for double {{ at the beginning and double }} at
theend. Same values between the brackets as what was matched on.
Sounds like a job for regular expressions.
regression=# select regexp_replace('abc{foo1}def{goo_bug}a', '{([-_a-zA-Z0-9]*)}', '{{\1}}', 'g');
regexp_replace
----------------------------
abc{{foo1}}def{{goo_bug}}a
(1 row)
See
https://www.postgresql.org/docs/current/functions-matching.html#FUNCTIONS-POSIX-REGEXP
regards, tom lane