Thread: head contrib is broken (crypto)

It seems that last contrib crypto changes broke buildfarm. You can see
problem e.g. here:

http://www.pgbuildfarm.org/cgi-bin/show_log.pl?nm=emperor_moth&dt=2009-08-04%2020:06:01

I don't have time to look on it now. But it should be reproducible
everywhere.

Zdenek

Zdenek Kotala wrote:
> It seems that last contrib crypto changes broke buildfarm. You can see
> problem e.g. here:
> 
> http://www.pgbuildfarm.org/cgi-bin/show_log.pl?nm=emperor_moth&dt=2009-08-04%2020:06:01
> 
> I don't have time to look on it now. But it should be reproducible
> everywhere.

It's not crypto changes that's the problem -- it's bytea now
being output in hex form (encrypt and decrypt both return bytea):

*** /store/postgres/buildfarm/emperor_moth/HEAD/pgsql.10027/contrib/pgcrypto/expected/3des.out    Tue Aug  4 22:06:20
2009
--- /store/postgres/buildfarm/emperor_moth/HEAD/pgsql.10027/contrib/pgcrypto/results/3des.out    Tue Aug  4 22:13:42
2009
***************
*** 37,44 **** -- decrypt select decrypt(encrypt('foo', '0123456', '3des'), '0123456', '3des');  decrypt 
! ---------
!  foo (1 row)  -- iv
--- 37,44 ---- -- decrypt select decrypt(encrypt('foo', '0123456', '3des'), '0123456', '3des');  decrypt  
! ----------
!  \x666f6f (1 row)  -- iv


-- 
Alvaro Herrera                                http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.

Alvaro Herrera <alvherre@commandprompt.com> writes:
> It's not crypto changes that's the problem -- it's bytea now
> being output in hex form (encrypt and decrypt both return bytea):

I was insufficiently careful about checking the non-core regression
tests :-(.  Mea culpa.
        regards, tom lane