Time::Piece handling about dbdpg HOT 11 CLOSED

I see the same behavior with the latest DBI and DBD::Pg, of course:

# DBI                         Version 1.642
# DBD::Pg                     Version 3.10.0
# Perl                        Version 5.18.2
# OS                          linux
# PostgreSQL (compiled)       90614
# PostgreSQL (target)         90614
# PostgreSQL (reported)       PostgreSQL 9.6.14 on i686-pc-linux-gnu, compiled by gcc (GCC) 4.4.7 20120313 (Red Hat 4.4.7-23), 32-bit

from dbdpg.

Strangely though, on another system, the same script does not fail:

# DBI                         Version 1.642
# DBD::Pg                     Version 3.10.0
# Perl                        Version 5.18.2
# OS                          linux
# PostgreSQL (compiled)       90615
# PostgreSQL (target)         90615
# PostgreSQL (reported)       PostgreSQL 9.6.15 on x86_64-pc-linux-gnu (Ubuntu 9.6.15-1.pgdg16.04+1), compiled by gcc (Ubuntu 5.4.0-6ubuntu1~16.04.11) 5.4.0 20160609, 64-bit

32- vs 64-bit difference?

from dbdpg.

It turned out to be commit 3929990:

Fix bug where $DBD::Pg::DBDPG_DEFAULT not picked up as a magic string first
time it is used in a script. Basically, we move the !SvAMAGIC exception for not
binding references down a little further inside dbdimp.c. RT #112309

Also the big int tests in 04misc.t is failing on 32-bit systems with:

t/04misc.t .. 1/101 DBD::Pg::st execute failed: ERROR: invalid input syntax for
integer: "-9.22337203685478e+18" at t/04misc.t line 87.

from dbdpg.

@joshwilliams ?

from dbdpg.

@turnstep?

from dbdpg.

On holiday but have it on todo list for 2020

from dbdpg.

👍 thanks!

from dbdpg.

That second problem (t/04misc.t on 32-bit systems) should be fixed as of version 3.10.2.

Still looking into the first.

from dbdpg.

Should be fixed as of 15f58c2. Please test if you are able.

from dbdpg.

Working perfectly for me. Thanks!

from dbdpg.

For the record, this is fixed in version 3.10.4

from dbdpg.