From: | Andrew Dunstan <andrew(at)dunslane(dot)net> |
---|---|
To: | PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
Subject: | obscure plperl bug |
Date: | 2004-09-30 22:46:54 |
Message-ID: | 415C8CDE.4080401@dunslane.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
While trying to fix a problem for David Fetter, I found this rather
obscure plperl bug. It occurs, I think, in these 2 lines in
plperl.c::plperl_get_elem():
if (hv_exists_ent(hash, eval_pv(key, TRUE), FALSE))
svp = hv_fetch(hash, key, strlen(key), FALSE);
This works except where key is the name of a perl builtin function, when
it comprehensively breaks, because instead of treating it as a string
perl will return the value of the function from eval_pv(). I hope to
have a solution soon, but meanwhile it needs to go on the open items list.
cheers
andrew
From | Date | Subject | |
---|---|---|---|
Next Message | Joe Conway | 2004-09-30 22:56:02 | Re: output branches before infinite recursion using connectby()? |
Previous Message | Gaetano Mendola | 2004-09-30 22:14:20 | Re: Index locking considerations |