Re: Add on_trusted_init and on_untrusted_init to plperl UPDATED [PATCH] - Mailing list pgsql-hackers

From Alex Hunsaker
Subject Re: Add on_trusted_init and on_untrusted_init to plperl UPDATED [PATCH]
Date
Msg-id 34d269d41002022113t13abb1aeic9afefeb0fd164a9@mail.gmail.com
Whole thread Raw
In response to Re: Add on_trusted_init and on_untrusted_init to plperl UPDATED [PATCH]  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: Add on_trusted_init and on_untrusted_init to plperl UPDATED [PATCH]
List pgsql-hackers
On Tue, Feb 2, 2010 at 21:38, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Alex Hunsaker <badalex@gmail.com> writes:
>> Yeah the both is gross.  How about:
>> plperl.on_plperl_init
>> plperl.on_plperlu_init
>> plperl.on_init ?
>
> I like the first two.  The problem of selecting a good name for the
> third one is easily solved: don't have it.  What would it be except
> a headache and a likely security problem?

Well its already in.  (FYI its also PGC_SIGHUP)  I included it here
because I wanted to make sure it made sense in context of the other
new plperl GUCs.

I *think* its there so people can:
-"use" the same modules in both
-profile both plperl and plperlu code easily (which is really the same point...)

The main difference between on_init and the other two is the later get
eval()ed in while the former is treated as "perl -e". (Did I get that
right Tim? I did not really look over the first patch).  Im not sure
if there are different consequences for that style...  But I would
venture its done that way so we can profile any perl interpreter
startup stuff we do in plperl.c or the src/pl/plperl/*.pl files.

So there might be a reason for it...


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Add on_trusted_init and on_untrusted_init to plperl UPDATED [PATCH]
Next
From: Tom Lane
Date:
Subject: Re: Add on_trusted_init and on_untrusted_init to plperl UPDATED [PATCH]