Re: configure's checks for --enable-tap-tests are insufficient - Mailing list pgsql-hackers

From Tom Lane
Subject Re: configure's checks for --enable-tap-tests are insufficient
Date
Msg-id 22462.1521572770@sss.pgh.pa.us
Whole thread Raw
In response to Re: configure's checks for --enable-tap-tests are insufficient  (Alvaro Herrera <alvherre@alvh.no-ip.org>)
List pgsql-hackers
Alvaro Herrera <alvherre@alvh.no-ip.org> writes:
> Tom Lane wrote:
>> Ah, thanks.  Not sure we need to make an explicit test for Opcode; we've
>> not heard of anyone not having that.

> In Debian stable, Opcode is in package libperl5.24.  This probably
> explains why nobody runs into trouble with it, since libperl is already
> checked by configure for other reasons.  Not sure about the RH world.

[ pokes around ]  Opcode is in the base perl package on RHEL6;
on Fedora 28, it's in perl-interpreter.  Either way, it's in the same
package as /usr/bin/perl, so ain't nobody not gonna have it.

> As for the others Dagfinn mentioned, most seem to be in perl-base, and a
> few in perl-modules-5.24, which is depended upon by libperl5.24, so I
> think the proposed set is good enough for now.

Sounds good.  We can always tweak the list some more if the situation
changes, or if we hear from someone using a different distro that slices
things up yet differently.  But I don't feel a need to expend configure
cycles on purely hypothetical configurations.  I imagine that most
off-the-beaten-track cases would involve hand-built Perl installations,
and those will almost certainly have all of "Perl core".  So I think
as long as we cover what the major Linux distros are doing, that's
probably good enough.

            regards, tom lane


pgsql-hackers by date:

Previous
From: Pavel Stehule
Date:
Subject: Re: [HACKERS] plpgsql - additional extra checks
Next
From: Peter Eisentraut
Date:
Subject: Re: PQHost() undefined behavior if connecting string contains bothhost and hostaddr types