I encountered a problem while implementing new CREATE
CONVERSION. Since converion procs are dynamically invoked while doing
an encoding conversion, it might fail for some reasons:
(1) stale pg_conversion entry. If someone re-register that proc, the
oid might be changed and the reference from
[Cc:ed to hackers]
From: [EMAIL PROTECTED] (Neil Conway)
Subject: pgbench questions
Date: Sat, 13 Jul 2002 00:57:37 -0400
Message-ID: <[EMAIL PROTECTED]>
> Hi,
>
> I was looking at doing some performance profiling on PostgreSQL, and
> I had a few questions on pgbench.
>
> (1) Is there a reason
Is it at all a problem that several columns in pg_conversion have the same
name as columns in pg_constraint?
Should the ones in pg_conversion become: convname instead of conname, etc.
simply for clarity?
Chris
- Original Message -
> Log message:
> Second phase of committing Rod Taylor'
> > I prefer ...add constraint. After a while (release or 2) removal of
> > create unique index all together.
>
> Remove CREATE UNIQUE INDEX entirely? Why?
I was looking to encourage users to use core SQL as I spend more time
than I want converting between systems -- thanks in part to users wh
Rod Taylor wrote:
> I'm going to change the pg_dump command to pull these constraints out of
> pg_constaint where possible, creating the appropriate alter table add
> constraint command (see primary key).
>
>
> Should unique constraints created with 'create index' (no entry in
> pg_constraint) b
What about the TPC-H benchmark ?
I cant recall if it has more than 10 tables, but it seemed like the
queries were "quite good" for a benchmark. In addition it comes with a
data generator.
regards
Mark
>On Sat, 2002-07-13 at 04:05, Neil Conway wrote:
> I'd like to look at the performance of the
On Sat, Jul 13, 2002 at 11:18:14AM +0800, Christopher Kings-Lynne wrote:
> What about the OSDB benchmark? Does that contain a large dataset?
No -- it only uses 5 relations total, with the most complex query
only involving 4 joins.
Cheers,
Neil
--
Neil Conway <[EMAIL PROTECTED]>
PGP Key ID: D
What about the OSDB benchmark? Does that contain a large dataset?
Chris
- Original Message -
From: "Neil Conway" <[EMAIL PROTECTED]>
To: "PostgreSQL Hackers" <[EMAIL PROTECTED]>
Sent: Saturday, July 13, 2002 12:05 AM
Subject: [HACKERS] test data for query optimizer
> I'd like to look
I'm going to change the pg_dump command to pull these constraints out of
pg_constaint where possible, creating the appropriate alter table add
constraint command (see primary key).
Should unique constraints created with 'create index' (no entry in
pg_constraint) be re-created via alter table add
On Fri, 2002-07-12 at 15:17, Tom Lane wrote:
> Now that the pg_depend mechanism is mostly in there, it is no longer
> a good idea to delete things directly (for example, by calling
> heap_drop_with_catalog or even just heap_delete'ing a catalog tuple).
I noticed that SERIAL sequences aren't dropp
Thanks, TODO updated. I split out "Make constraints clearer in dump
file" into a foreign key version, which I marked as done, and a second
version which I left as undone.
Thanks. That's a heap of items completed.
---
Rod
> * Add pg_depend table for dependency recording; use sysrelid, oid,
> depend_sysrelid, depend_oid, name
> * Auto-destroy sequence on DROP of table with SERIAL; perhaps a separate
> SERIAL type
> * Have SERIAL generate non-colliding sequence names when we have
>
Tom Lane wrote:
> Now that the pg_depend mechanism is mostly in there, it is no longer
> a good idea to delete things directly (for example, by calling
> heap_drop_with_catalog or even just heap_delete'ing a catalog tuple).
>
> The correct thing to do is to call performDeletion() with a parameter
Now that the pg_depend mechanism is mostly in there, it is no longer
a good idea to delete things directly (for example, by calling
heap_drop_with_catalog or even just heap_delete'ing a catalog tuple).
The correct thing to do is to call performDeletion() with a parameter
specifying what it is you
Thomas Lockhart wrote:
> (crossposted to -hackers, should follow up on that list)
> OK, this is in the "can't do it what we have" category. Should we have
> it accept a regular expression rather than a simple string? In either
> case it should probably go into the main distro. Except that I se
[EMAIL PROTECTED] (Neil Conway) writes:
> I'd like to look at the performance of the query optimizer (both the
> traditional one and GEQO) when joining large numbers of tables: 10-15,
> or more. In order to do that (and to get meaningful results), I'll
> need to work with some data that actually r
Thomas Lockhart <[EMAIL PROTECTED]> writes:
> So, what should the behavior be of a constant declared as
> CHAR 'hi'
> ? Right now it fails, since SQL9x asks that the char type defaults to a
> length of one and our parser does not distinguish between usage as a
> constant declaration and as a colu
I'd like to look at the performance of the query optimizer (both the
traditional one and GEQO) when joining large numbers of tables: 10-15,
or more. In order to do that (and to get meaningful results), I'll
need to work with some data that actually requires joins of that
magnitude. Ideally, I'd li
So, what should the behavior be of a constant declared as
CHAR 'hi'
? Right now it fails, since SQL9x asks that the char type defaults to a
length of one and our parser does not distinguish between usage as a
constant declaration and as a column definition (where you would want
the "char(1)" to
On Fri, Jul 12, 2002 at 03:48:59PM +0200, Zeugswetter Andreas SB SD wrote:
> Imho the advantages of an automatic coercion would outweigh the few
> corner cases where the behavior would not be intuitive to
> everybody.
How then would one get the correct behaviour from char()?
A
--
Andrew S
"eutm" <[EMAIL PROTECTED]> writes:
> Dear Sirs!:)I encounted one small problem,working with
> PostgreSQL 7.3devel.It can look a
> bit strange,but i have to use whitespaces in names of databases,tables,fields
> and so on(like "roomno jk").It's possible to create them all and work with them
>
(crossposted to -hackers, should follow up on that list)
> Well, OVERLAY is defined as:
> overlay(string placing string from integer [for integer])
> and replace() is defined (by me at least) as:
> replace(inputstring, old-substr, new-substr)
OK.
> OVERLAY requires that I know the "fr
Florian Helmberger wrote:
> Hi.
>
> > Florian, I haven't seen this patch yet. Did you send it in?
>
> Yes, I sent it to Christopher for reviewing, as allready mentioned by
> himself :)
> I still had not the time to update the docs though, hope to get this done
> next week.
Yes, I had an email
> Is this possible ?
Sure.
- Thomas
---(end of broadcast)---
TIP 5: Have you checked our extensive FAQ?
http://www.postgresql.org/users-lounge/docs/faq.html
There is no comparison of varchar to char in Oracle too.
Scott provided cast cases are some unique features in psql,
each database MAY handle those casting differently.
In good design/application, char should be replaced by
varchar type unless you know the exact bytes. It would be
not bad idea t
On Fri, 12 Jul 2002, Tom Lane wrote:
> "Zeugswetter Andreas SB SD" <[EMAIL PROTECTED]> writes:
> >> If the conversion where varchar(5) --> char(5) then
> >> they would compare equal.
>
> > I am not sure, since, if the varchar stored 'S ' then the comparison
> > to a char 'S' should probably sti
> Has anyone studied how other DBMSs handle CHAR vs VARCHAR? Judging
> from the number of questions we get on this point, I have to wonder
> if we are not out of step with the way other systems do it.
Well, I already gave the Informix example, that compares them as equal.
(they obviously coerce
"Zeugswetter Andreas SB SD" <[EMAIL PROTECTED]> writes:
>> If the conversion where varchar(5) --> char(5) then
>> they would compare equal.
> I am not sure, since, if the varchar stored 'S ' then the comparison
> to a char 'S' should probably still fail,
There is no comparison of varchar to ch
I see this on the TODO list:
# Fix PL/PgSQL to handle quoted mixed-case identifiers
Perhaps you could make a view (alias the names with spaces) to work on?
On Fri, 2002-07-12 at 06:31, eutm wrote:
>
> Dear Sirs!:)I encounted one small problem,working with
> PostgreSQL 7.3devel.It can
On Fri, 12 Jul 2002, [EMAIL PROTECTED] wrote:
> Hi
> Were two doctoring students and we have a little problem to resolve.
> Were using Grass5pre3 and PostgreSQL 7.2 (Linux)to map vehicular
> pollution of our city. We have a map of the streets and we have to
> assign 24 values (+ the label) to e
.. inside of each other using ECPG ?
I have a situation where it would be advantages to open a cursor, retrieve
a tuple, then open another query based on the results of the first. Then
when that query has been processed return to the first query and get the
second tuple.
Is this possible ?
che
Dear Sirs!:)I encounted one small problem,working with
PostgreSQL 7.3devel.It can look a
bit strange,but i have to use whitespaces in names of databases,tables,fields
and so on(like "roomno jk").It's possible to create them all and work with them
(INSERT,DELETE,UPDATE),but PL/pgSQL parse
Hi
Were two doctoring students and we have a little problem to resolve.
Were using Grass5pre3 and PostgreSQL 7.2 (Linux)to map vehicular
pollution of our city. We have a map of the streets and we have to
assign 24 values (+ the label) to each street.
What would be a smart way to solve this probl
Hi.
> Florian, I haven't seen this patch yet. Did you send it in?
Yes, I sent it to Christopher for reviewing, as allready mentioned by
himself :)
I still had not the time to update the docs though, hope to get this done
next week.
Florian
---(end of broadcast)---
> I guess the strangest part was that both a.foo = 'S' and b.foo = 'S' but
> not a.foo=b.foo; (a.foo is varchar(5) , b.foo is char(5) )
>
> I guess that tha 'S' that b.foo gets compared to is converted to 'S'
> before comparison but when comparing varchar(5) and char(5) they are
> both comp
35 matches
Mail list logo
