BUG #5034: plperlu problem with gethostbyname

Started by Diego de Limaover 16 years ago4 messagesbugs
Jump to latest
#1Diego de Lima
diego_de_lima@hotmail.com

The following bug has been logged online:

Bug reference: 5034
Logged by: Diego de Lima
Email address: diego_de_lima@hotmail.com
PostgreSQL version: 8.3.7
Operating system: Linux Fedora 10
Description: plperlu problem with gethostbyname
Details:

The gethostbyname Perl internal function doesn't work properly on 8.3.7,
returning "1" instead of the desired data array.

On Postgres 8.2.x it works perfectly.

Using Perl v5.10.0.

Using Fedora 10 default rpm packages, all updated last month via yum.

#2Robert Haas
robertmhaas@gmail.com
In reply to: Diego de Lima (#1)
Re: BUG #5034: plperlu problem with gethostbyname

On Thu, Sep 3, 2009 at 5:21 PM, Diego de Lima<diego_de_lima@hotmail.com> wrote:

The following bug has been logged online:

Bug reference:      5034
Logged by:          Diego de Lima
Email address:      diego_de_lima@hotmail.com
PostgreSQL version: 8.3.7
Operating system:   Linux Fedora 10
Description:        plperlu problem with gethostbyname
Details:

The gethostbyname Perl internal function doesn't work properly on 8.3.7,
returning "1" instead of the desired data array.

On Postgres 8.2.x it works perfectly.

Using Perl v5.10.0.

Using Fedora 10 default rpm packages, all updated last month via yum.

I can't reproduce this on a clean build of 8.3.7 or on a clean build
of CVS HEAD, both against perl 5.10.0, so I don't think this is a
PostgreSQL bug. I'm guessing you have a bug in your code someplace.
Here's the test case I used.

create or replace function test() returns varchar as $$use
Data::Dumper; Dumper(gethostbyname("www.google.com"));$$ language
plperlu;

select test();

...Robert

#3Tom Lane
tgl@sss.pgh.pa.us
In reply to: Robert Haas (#2)
Re: BUG #5034: plperlu problem with gethostbyname

Robert Haas <robertmhaas@gmail.com> writes:

On Thu, Sep 3, 2009 at 5:21 PM, Diego de Lima<diego_de_lima@hotmail.com> wrote:

Using Fedora 10 default rpm packages, all updated last month via yum.

I can't reproduce this on a clean build of 8.3.7 or on a clean build
of CVS HEAD, both against perl 5.10.0, so I don't think this is a
PostgreSQL bug. I'm guessing you have a bug in your code someplace.
Here's the test case I used.

create or replace function test() returns varchar as $$use
Data::Dumper; Dumper(gethostbyname("www.google.com"));$$ language
plperlu;

select test();

Don't have Fedora 10 installed anymore, but the F-11 packages should
be equivalent, and on F-11 I get this:

test
-----------------------------
$VAR1 = 'www.l.google.com';
$VAR2 = 'www.google.com';
$VAR3 = 2;
$VAR4 = 4;
$VAR5 = 'J}';
$VAR6 = 'J}c';
$VAR7 = 'J}g';
$VAR8 = 'J}h';
$VAR9 = 'J}i';
$VAR10 = 'J}j';

(1 row)

I don't know enough about either gethostbyname or Data::Dumper
to figure out if this is sane or not.

regards, tom lane

#4Robert Haas
robertmhaas@gmail.com
In reply to: Tom Lane (#3)
Re: BUG #5034: plperlu problem with gethostbyname

On Sep 5, 2009, at 6:52 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:

Robert Haas <robertmhaas@gmail.com> writes:

On Thu, Sep 3, 2009 at 5:21 PM, Diego de Lima<diego_de_lima@hotmail.com

wrote:
Using Fedora 10 default rpm packages, all updated last month via
yum.

I can't reproduce this on a clean build of 8.3.7 or on a clean build
of CVS HEAD, both against perl 5.10.0, so I don't think this is a
PostgreSQL bug. I'm guessing you have a bug in your code someplace.
Here's the test case I used.

create or replace function test() returns varchar as $$use
Data::Dumper; Dumper(gethostbyname("www.google.com"));$$ language
plperlu;

select test();

Don't have Fedora 10 installed anymore, but the F-11 packages should
be equivalent, and on F-11 I get this:

test
-----------------------------
$VAR1 = 'www.l.google.com';
$VAR2 = 'www.google.com';
$VAR3 = 2;
$VAR4 = 4;
$VAR5 = 'J}';
$VAR6 = 'J}c';
$VAR7 = 'J}g';
$VAR8 = 'J}h';
$VAR9 = 'J}i';
$VAR10 = 'J}j';

(1 row)

I don't know enough about either gethostbyname or Data::Dumper
to figure out if this is sane or not.

I think so. It's certainly an array rather than just the scalar 1, as
the OP alleged we were producing.

...Robert