Re: [HACKERS] Open 6.4 items

Bruce Momjian <maillist@candle.pha.pa.us> writes:
> But we might as well fix it now, because we can't ask for initdb after
> the final release.

> Can someone check that that fix does not affect the regression test
> results?  I doubt we do lseg not-equal tests in the regression suite.
> You will have to do an initdb to see the change. 

I just rebuilt and initdb'd.  Regression test state is the same
as before, except that the new inet test fails:

*** expected/inet.out    Tue Oct 27 14:34:02 1998
--- results/inet.out    Wed Oct 28 14:02:45 1998
***************
*** 53,66 ****   i as inet, network(i) as "network(inet)" FROM INET_TBL; eight|cidr        |network(cidr)|inet
 |network(inet) -----+------------+-------------+----------------+-------------
 
!      |192.168.1/24|  0.1.168.192|192.168.1.226/24|  0.1.168.192
!      |192.168.1/24|  0.1.168.192|192.168.1.226   |226.1.168.192
!      |10/8        |     0.0.0.10|10.1.2.3/8      |     0.0.0.10
!      |10.0.0.0/32 |     0.0.0.10|10.1.2.3/8      |     0.0.0.10
!      |10.1.2.3/32 |     3.2.1.10|10.1.2.3        |     3.2.1.10
!      |10.1.2/24   |     0.2.1.10|10.1.2.3/24     |     0.2.1.10
!      |10.1/16     |     0.0.1.10|10.1.2.3/16     |     0.0.1.10
!      |10/8        |     0.0.0.10|10.1.2.3/8      |     0.0.0.10 (8 rows)  QUERY: SELECT '' as eight, c as cidr,
masklen(c)as "masklen(cidr)",
 
--- 53,66 ----   i as inet, network(i) as "network(inet)" FROM INET_TBL; eight|cidr        |network(cidr)|inet
 |network(inet) -----+------------+-------------+----------------+-------------
 
!      |192.168.1/24|  192.168.1.0|192.168.1.226/24|  192.168.1.0
!      |192.168.1/24|  192.168.1.0|192.168.1.226   |192.168.1.226
!      |10/8        |     10.0.0.0|10.1.2.3/8      |     10.0.0.0
!      |10.0.0.0/32 |     10.0.0.0|10.1.2.3/8      |     10.0.0.0
!      |10.1.2.3/32 |     10.1.2.3|10.1.2.3        |     10.1.2.3
!      |10.1.2/24   |     10.1.2.0|10.1.2.3/24     |     10.1.2.0
!      |10.1/16     |     10.1.0.0|10.1.2.3/16     |     10.1.0.0
!      |10/8        |     10.0.0.0|10.1.2.3/8      |     10.0.0.0 (8 rows)  QUERY: SELECT '' as eight, c as cidr,
masklen(c)as "masklen(cidr)",
 

----------------------

Offhand I would say that it's the "expected" file that is broken.
Shouldn't those octets be coming out in the other order?
        regards, tom lane