Re: [SQL] Operator <> for lseg,lseg bug
От | Bruce Momjian |
---|---|
Тема | Re: [SQL] Operator <> for lseg,lseg bug |
Дата | |
Msg-id | 199810281605.LAA17100@candle.pha.pa.us обсуждение исходный текст |
Ответ на | Operator <> for lseg,lseg bug (Guido Weber <guido@castor.atlas.de>) |
Список | pgsql-sql |
> Hi > > I was testing some things with the geometric datatype in Postgres (6.3) > and got some results out of selects that I did not expect. Doing a \do > in psql gave me: > > <--SNIP--> > <> |bool |float8 |float8 |not equal > <> |bool |int2 |int2 |not equal > <> |bool |int4 |int4 |not equal > <> |bool |lseg |lseg |equals > <> |bool |money |money |not equal > <> |bool |name |name |not equal > <> |bool |oid |oid |not equal > <--SNIP--> > > This looks like a wrong entry in one of the system tables. Any ideas which > one it is, and how to correct this ? Good eye. Yes, it was wrong. To fix it, edit /pgsql/src/include/catalog/pg_operator.h. Search for the strings 'lseq_eq'. The first instance is correct, but the second instance should be 'lseg_neq'. Then you must run initdb to reload the system tables. Another way to do this would be to do an update on 'pg_operator'. I believe the oid is 1586, and the field is oprcode, so you can try: update pg_operator set oprcode = 'lseg_neq' where oid = 1586 \do should then show the proper values. -- Bruce Momjian | http://www.op.net/~candle maillist@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
В списке pgsql-sql по дате отправления: