> regression=# create index mv1 on t1 (f1);
...
> ERROR: materialized view "mv1" already exists
Is in fact the one I prefer ;^)
I come from a DBMS world where Tables and Indexes do not share the same name
space,
and have to change my mind !
Thanks you Tom for pointing that.
Regards
PAscal
legrand legrand writes:
> I thougth that thoses messages where using relation's relkind:
> ..
> wouldn't it be easier to read for beginners ?
I doubt it would be an improvement. Consider this example:
regression=# create table t1 (f1 int);
CREATE TABLE
regression=# create materialized view mv1
OK, that noted !
thank you for the quick answers
Regards
PAscal
--
Sent from: http://www.postgresql-archive.org/PostgreSQL-general-f1843780.html
On Sat, Mar 10, 2018 at 10:54 AM, legrand legrand <
legrand_legr...@hotmail.com> wrote:
> I thougth that thoses messages where using relation's relkind:
> r = ordinary table,
> i = index,
> S = sequence,
> t = TOAST table,
> v = view,
> m = materialized view,
> c = composite type,
>
I thougth that thoses messages where using relation's relkind:
r = ordinary table,
i = index,
S = sequence,
t = TOAST table,
v = view,
m = materialized view,
c = composite type,
f = foreign table,
p = partitioned table
wouldn't it be easier to read for beginners ?
Regard
On 03/10/2018 07:00 AM, legrand legrand wrote:
Hello,
When trying to create an already existing index (in pg 9.5)
SQL> create index if not exists NEWINDEX on SCHEMA.TABLE(COL);
> relation "NEWINDEX" already exists, skipping
message speaks about relation (and not index)
https://www.pos
