Mark Gibson <[EMAIL PROTECTED]> writes:
I have an idea, to break pg_dump into functions within PostgreSQL.
Philip Warner wrote:
However, there are some complications because pg_dump is also the
upgrade tool; the backed can only know how to describe itself for the
current dialect of SQL ac
.
This would help greatly when it comes to development, migrating
the schema, etc.
Would this be practical, would it be possible to break the pg_dump
code out into PostgreSQL functions?
Anyone have any ideas/thoughts/suggestions on this?
Cheers
--
Mark Gibson
Web Developer & Database Admin
Crom
Hi,
Is there any reason that the following shouldn't work:
CREATE TABLE my_new_table (LIKE my_composite_type);
I get the following error (in 7.4.2):
ERROR: "my_composite_type" is a composite type
With the improved handling of composite types,
will this work in 7.5?
Cheers
--
M
DOMAIN anotherdomain AS mydomain;
ERROR: "mydomain" is not a valid base type for a domain
And why isn't an array type created for a domain?
(Is there a way to create an array for a domain without resorting to an
array of the domain's base type?)
Cheers
--
Mark Gi
Mark Gibson wrote:
Are there any Gentoo users here?
Do you wish the PostgreSQL ebuild made use of SLOTS?
I've submitted a slotted ebuild for testing at:
http://bugs.gentoo.org/show_bug.cgi?id=42894
It's the first slotted ebuild i've attempted, so it may be total rubish!
But I have
it is determined that everything is fine.
At which point the latest rc is renamed as the final release
(ie. 7.4.2).
Unless you already do this, and I've completely missed it somehow
--
Mark Gibson
Web Developer & Database Admin
Cromwell Tools Ltd.
Leicester, England.
--
hread at:
http://bugs.gentoo.org/show_bug.cgi?id=42894
With a initial plan for relocation of files to support slotting.
Comments, suggestions, corrections, or messages to say it's a silly idea
(and why)
would be very much appreciated before I attempt this next week.
Cheers
--
Mark G
he mutable bits!
Giving you the best of both worlds.
I haven't tried this or thought it through very much - too busy - but
I'd thought I'd throw
it in for a bit o' head scratching, and chin stroking :)
Cheers
--
Mark Gibson
Web Developer & Database Admin
Crom
Joe Conway wrote:
Mark Gibson wrote:
I've found the problem, although I'm still a bit confused by it.
I hate to do this to you now, but after some thought I think I have a
better approach -- I'd be interested in opinions on that assessment.
The attached eliminates pgres
Mark Gibson wrote:
I've found that dblink works without the oid map, ie: dblink(... , ...
, false)
but returns the error when enabled, ie: dblink(... , ... , true)
Hello,
I've found the problem, although I'm still a bit confused by it.
When the call to SPI_finish(
Mark Gibson wrote:
Well, I've re-compiled PostgreSQL 7.4.1 with --enable-debug
(was previously working with a Gentoo ebuild of it).
And strangely I get a different error now:
ERROR: query-specified return row and actual function return row do
not match
After placing a few elog state
Tom Lane wrote:
Mark Gibson <[EMAIL PROTECTED]> writes:
Unfortunately, it doesn't work using the oid the map, whether custom types
are involved or not. All I get is the following message:
ERROR: unsupported byval length:
SPI is very new to me (like 2 days old ;).
Any sugges
27;hostaddr' + 'dbname'
+ 'port' ???
For testing i've used an db_id of type 'name' with the value: 'test'.
Unfortunately, it doesn't work using the oid the map, whether custom types
are involved or not. All I get is the following message:
ERROR:
13 matches
Mail list logo
