Re: [HACKERS] Failures in 'rules' regression test

Tom Lane wrote:
> 
> Is anyone else seeing failure of the "rules" regression test with
> current CVS sources, or is it just me?

"me too"

Though my output is in a slightly different order again, ie., different
system, so your qsort() theory seems good.

BTW, the error messages seem to have changed (running NetBSD-current),
so apart from rules, everything passes.

float8 .. failed
geometry .. failed
misc .. failed
rules .. failed

*** expected/float8-NetBSD.out  Sat Feb  6 19:53:55 1999
--- results/float8.out  Sun Feb 14 14:16:38 1999
***************
*** 209,217 **** (5 rows)  QUERY: INSERT INTO FLOAT8_TBL(f1) VALUES ('10e400');
! ERROR:  Bad float8 input format '10e400' QUERY: INSERT INTO FLOAT8_TBL(f1) VALUES ('-10e400');
! ERROR:  Bad float8 input format '-10e400' QUERY: INSERT INTO FLOAT8_TBL(f1) VALUES ('10e-400'); QUERY: INSERT INTO
FLOAT8_TBL(f1)VALUES ('-10e-400'); QUERY: DELETE FROM FLOAT8_TBL;
 
--- 209,217 ---- (5 rows)  QUERY: INSERT INTO FLOAT8_TBL(f1) VALUES ('10e400');
! ERROR:  Input '10e400' is out of range for float8 QUERY: INSERT INTO FLOAT8_TBL(f1) VALUES ('-10e400');
! ERROR:  Input '-10e400' is out of range for float8 QUERY: INSERT INTO FLOAT8_TBL(f1) VALUES ('10e-400'); QUERY:
INSERTINTO FLOAT8_TBL(f1) VALUES ('-10e-400'); QUERY: DELETE FROM FLOAT8_TBL;
 

*** expected/geometry-NetBSD.out        Sat Feb  6 19:53:55 1999
--- results/geometry.out        Sun Feb 14 14:16:40 1999
***************
*** 87,93 ****  QUERY: SELECT '' AS count, p.f1, l.s, l.s # p.f1 AS intersection    FROM LSEG_TBL l, POINT_TBL p;
! ERROR:  There is more than one possible operator '#' for types 'lseg' and 'point'       You will have to retype this
queryusing an explicit cast QUERY: SELECT '' AS thirty, p.f1, l.s, p.f1 ## l.s AS closest    FROM LSEG_TBL l, POINT_TBL
p;
--- 87,93 ----  QUERY: SELECT '' AS count, p.f1, l.s, l.s # p.f1 AS intersection    FROM LSEG_TBL l, POINT_TBL p;
! ERROR:  Unable to identify an operator '#' for types 'lseg' and 'point'       You will have to retype this query
usingan explicit cast QUERY: SELECT '' AS thirty, p.f1, l.s, p.f1 ## l.s AS closest    FROM LSEG_TBL l, POINT_TBL p;
 

*** expected/misc.out   Sun Feb 14 14:16:25 1999
--- results/misc.out    Sun Feb 14 14:18:42 1999
***************
*** 6,19 ****    SET stringu1 = reverse_name(onek.stringu1)    WHERE onek.stringu1 = 'JBAAAA' and         onek.stringu1
=tmp.stringu1;
 
- NOTICE:  Non-functional update, only first update is performed
- NOTICE:  Non-functional update, only first update is performed QUERY: UPDATE tmp    SET stringu1 =
reverse_name(onek2.stringu1)   WHERE onek2.stringu1 = 'JCAAAA' and         onek2.stringu1 = tmp.stringu1;
 
- NOTICE:  Non-functional update, only first update is performed
- NOTICE:  Non-functional update, only first update is performed QUERY: DROP TABLE tmp; QUERY: COPY onek TO
'/home/prlw1/pgsql/src/test/regress/input/../results/onek.data';QUERY: DELETE FROM onek;
 
--- 6,15 ----


Cheers,

Patrick