Hi guys,
I am looking for a way to get list of supported types in Postgres
together with information whether type can have optional size (like
varchar for example), whether it can have precision (like decimal for
example), and whether it can come as value of sequence (like all integer
types f
On 08/15/13 05:23, Michael Paquier napisa:
On Thu, Aug 15, 2013 at 6:00 AM, Bruce Momjian wrote:
Try psql -E, and run the \dT command to see the query it uses.
You have also the following commands:
- ¥dT+, all types with addition information like its size ('var' is
for example variable length)
On 08/15/13 16:30, Adrian Klaver napisa:
On 08/15/2013 02:33 AM, Ivan Radovanovic wrote:
On 08/15/13 05:23, Michael Paquier napisa:
On Thu, Aug 15, 2013 at 6:00 AM, Bruce Momjian wrote:
Try psql -E, and run the \dT command to see the query it uses.
You have also the following commands:
- ¥dT
On 08/15/13 16:49, Adrian Klaver napisa:
On 08/15/2013 07:37 AM, Ivan Radovanovic wrote:
On 08/15/13 16:30, Adrian Klaver napisa:
On 08/15/2013 02:33 AM, Ivan Radovanovic wrote:
On 08/15/13 05:23, Michael Paquier napisa:
On Thu, Aug 15, 2013 at 6:00 AM, Bruce Momjian
wrote:
Try psql -E, and
On 08/15/13 16:59, Adrian Klaver napisa:
On 08/15/2013 07:53 AM, Ivan Radovanovic wrote:
Now I just need to find out which types can be indexed (and which types
can be part of PK)
http://www.postgresql.org/docs/9.2/interactive/indexes.html
doesn't list which types can be indexe
On 08/15/13 17:15, Tom Lane napisa:
Ivan Radovanovic writes:
Thanks Adrian, but question was how to decide which types are indexable
A little bit of research in the system-catalogs documentation will show
you how to find the types that can be accepted by some index opclass
(hint
On 08/15/13 17:27, Adrian Klaver napisa:
On 08/15/2013 08:07 AM, Ivan Radovanovic wrote:
On 08/15/13 16:59, Adrian Klaver napisa:
On 08/15/2013 07:53 AM, Ivan Radovanovic wrote:
Now I just need to find out which types can be indexed (and which
types
can be part of PK)
http
Hello,
I was checking for way to get object comments, and it seems that \dd has
bug when it comes to extracting descriptions for constraints. Relevant
part of query psql is executing is:
SELECT DISTINCT tt.nspname AS "Schema", tt.name AS "Name", tt.object AS
"Object", d.description AS "Descr
On 08/21/13 16:03, Tom Lane napisa:
Ivan Radovanovic writes:
I was checking for way to get object comments, and it seems that \dd has
bug when it comes to extracting descriptions for constraints.
That code looks right to me, and it works according to a simple test:
d1=# create table foo (f1
On 08/21/13 16:34, Ivan Radovanovic napisa:
On 08/21/13 16:03, Tom Lane napisa:
Ivan Radovanovic writes:
I was checking for way to get object comments, and it seems that \dd has
bug when it comes to extracting descriptions for constraints.
That code looks right to me, and it works according
Just to verify:
- when unique constraint is created using appropriate syntax rows are
added to tables pg_constraint and pg_index (pg_constraint with type 'u'
and referring to index with indisunique set to true)
- when unique index is created row is added only to pg_index table but
not to pg_con
Hello,
I couldn't find anything mentioned about this in documentation and
googling didn't help either:
- if I connect to database as user who doesn't have permission to access
all schemas then querying information_schema.schemata returns no rows
(querying information_schema.tables returns only
Hello,
I need to log access to certain data in database in some log (I prefer
to have that both in syslog and table in database), and I find it easy
to write to syslog, but I can't solve the problem of writing this to
database table.
If this protected data is read only using postgres functio
On 01/25/12 18:38, Greg Sabino Mullane napisa:
You would need to break out of the transaction somehow within that
function and make a new call to the database, for example using dblink
or plperlu. I've done the latter before and it wasn't too painful.
The general idea is:
- ---
$dbh = DBI->conn
On 01/25/12 20:02, Misa Simic napisa:
Thanks Bill,
Make sense... db_link is probably then solution... Everything depends on
concrete problem...
But I still think security should be reconsidered (I would use db_link
just in case there is no other options - if we must let users to have
direct ac
15 matches
Mail list logo
