When I execute the following query:
SELECT DISTINCT title FROM division ORDER BY UPPER(title);
I get:
ERROR: For SELECT DISTINCT, ORDER BY expressions must appear in target list
If I remove DISTINCT, the query works fine.
Is this illegal or a known bug?
thanks for any help,
J.R.
PostgreSQL
Hi!
When I execute
SELECT min(id) FROM test;
PostgreSQL optimizer uses sequential scan for performing the query
even if any index on "id" exists. VACUUM ANALYZE just before the query
doesn't help. Runtime option to not use sequenatial scans doesn't help.
So, I suppose it's a real bug.
--
Best
I installed the sample startup script for postgres for FreeBSD, but it
reports the following error upon running:
# /usr/local/etc/rc.d/postgres.sh start
pg_ctl: no database directory or environment variable $PGDATA is specified
Try 'pg_ctl --help' for more information.
I see clearly that PGDATA
I just installed Postgres 7.1. At the end of make install, it said:
And, so that we have an idea of who is using what, please connect to the
following registration URL:
http://www.postgresql.org -> Helping Us -> Survey/Register
But there is no "Helping Us" menu/link to be found. I di
POSTGRESQL BUG REPORT TEMPLATE
Your name : Tomasz Wegrzanowski
Your email address : [EMAIL PROTE
Hi,
I'm not using pg_ctl, I'm using postmaster directly. So, in my case I
tried passing "-d0" to it, but it had no effect. Command line:
postmaster -i -d0 -D /var/pgsql/data -c syslog=2
Any ideas? I could patch the code to remove the excessive elog's in the
vacuum command, but I'd rather
Apparently readline-4.2 gratuitously breaks backward compatibility by
undefining filename_completion_function (now called
rl_filename_completion_function).
This causes the compilation of src/bin/psql/tab-complete.c to fail.
Perhaps configure should be updated to detect this.
Yours,
--
Fazal Maji
Hello,
Could you please make techdocs.postgresql.org available on the mirrors
instead of a centralized site similar to the User Docs..
http://postgresql.nextpath.com/users-lounge/docs/
It seems that www.postgresql.org is often unavailable as a result of
server troubles or backbone outages and t
I found an example when postgres while executing the same query uses null
values as sometimes bigger than everything and sometimes smaller.
And I think it's BAD.
Check this out:
---
DROP TABLE NUTKI ;
CREATE TABLE NUTKI (
ID INT
Got it. I recommend changing 'su -l' to 'su -m' to preserve the
environment. How does that sound?
> I installed the sample startup script for postgres for FreeBSD, but it
> reports the following error upon running:
>
> # /usr/local/etc/rc.d/postgres.sh start
> pg_ctl: no database directory o
"J.R. Onyschak" <[EMAIL PROTECTED]> writes:
> When I execute the following query:
> SELECT DISTINCT title FROM division ORDER BY UPPER(title);
> I get:
> ERROR: For SELECT DISTINCT, ORDER BY expressions must appear in target list
> If I remove DISTINCT, the query works fine.
> Is this illegal
> SELECT min(id) FROM test; uses sequential scan even if an index on "id" exists.
>VACUUM ANALYZE doesn't help.
This is not a bug. In the current implementation, min and max cannot make
use of an index directly.
--
Peter Eisentraut [EMAIL PROTECTED] http://funkturm.homeip.net/~peter
---
Hello:
In version 7.1 the man page for pg_ctl does not document -s option.
Thanks,
--
Arcady Genkin
---(end of broadcast)---
TIP 5: Have you checked our extensive FAQ?
http://www.postgresql.org/users-lounge/docs/faq.html
Thanks. Will be fixed in 7.2.
>
> POSTGRESQL BUG REPORT TEMPLATE
>
>
>
> Your name : Tomasz Wegrzanow
I think this will be fixed in 7.1.1.
[ Charset ISO-8859-1 unsupported, converting... ]
> Apparently readline-4.2 gratuitously breaks backward compatibility by
> undefining filename_completion_function (now called
> rl_filename_completion_function).
> This causes the compilation of src/bin/psql/
In the PostgreSQL 7.1 Tutorial, filename
relmodel-oper.html, the HTML special characters ∣ and ∖ are
used. They are not recognized by either Netscape 4.76 or IE 5.5 and I
cannot find them in any of my HTML reference books.
Bruce PatinSystem
AdministratorPADECS(215) 504-4910 x 511
"Fazal Majid" <[EMAIL PROTECTED]> writes:
> Apparently readline-4.2 gratuitously breaks backward compatibility
Yup :-(.
We have updated our configure checks for 7.1.1.
regards, tom lane
---(end of broadcast)---
TIP 4: Don'
Can you give me a few sample lines that you are seeing in the log?
> Hi,
> I'm not using pg_ctl, I'm using postmaster directly. So, in my case I
> tried passing "-d0" to it, but it had no effect. Command line:
>
> postmaster -i -d0 -D /var/pgsql/data -c syslog=2
>
> Any ideas? I could p
Vivek Khera writes:
> I see clearly that PGDATA is set and exported. I'm suspecting that
> the "su -l" causes the environment to disappear. My guess is that
> whomever wrote this script has PGDATA set in the ~/.profile (or
> equivalent) in the postgres user's home directory.
This is fixed in t
Thank you.
I hope the fix will implement polymorphism also, i.e., if a superclass and
subclass define an identically-named function, then calling the function on
all records of the superclass should transparently call the subclass's
function for the records corresponding to the subclass.
Regard
Jon writes:
> I'm not using pg_ctl, I'm using postmaster directly. So, in my case I
> tried passing "-d0" to it, but it had no effect. Command line:
>
> postmaster -i -d0 -D /var/pgsql/data -c syslog=2
>
> Any ideas? I could patch the code to remove the excessive elog's in the
> vacuum com
> When I execute the following query:
> SELECT DISTINCT title FROM division ORDER BY UPPER(title);
> I get:
> ERROR: For SELECT DISTINCT, ORDER BY expressions must appear in target list
> If I remove DISTINCT, the query works fine.
> Is this illegal or a known bug?
Illegal. I believe that allowi
Fazal Majid writes:
> Apparently readline-4.2 gratuitously breaks backward compatibility by
> undefining filename_completion_function (now called
> rl_filename_completion_function).
> This causes the compilation of src/bin/psql/tab-complete.c to fail.
> Perhaps configure should be updated to dete
Vivek Khera writes:
> > "PE" == Peter Eisentraut <[EMAIL PROTECTED]> writes:
>
> >> Also, the following line in the start action of the sript is useful:
> >>
> >> /sbin/ldconfig -m $prefix/lib
>
> PE> You ought to run this once, not every time the system starts.
>
> No, FreeBSD does *not* cac
OK, I think this is already fixed in 7.1.1. The fix was to add -D to
pg_ctl. Would you try 7.1.1 or the snapshot and let me know. Thanks.
New line shows:
su -l $PGUSER -c "$DAEMON start -D '$PGDATA' -s -l $PGLOG"
You will need an updated pg_ctl that understands -D.
>
> Got it. I
>From my point of view, NULL is neither bigger, nor smaller, you can't
compare it with a number.
So it just comes at the end if you sort at all.
(Perhaps you need to take a think about what NULL means in your data. Should
NULL sort as if it's 0?, +infinity?, -infinity? if so why?)
regards,
This is fixed in 7.1.1.
> Hello:
>
> In version 7.1 the man page for pg_ctl does not document -s option.
>
> Thanks,
> --
> Arcady Genkin
>
> ---(end of broadcast)---
> TIP 5: Have you checked our extensive FAQ?
>
> http://www.postgresql.org/u
Tomasz Wegrzanowski <[EMAIL PROTECTED]> writes:
> In psql typelist help (\dT) macaddr is described:
> macaddr | XX:XX:XX:XX:XX, MAC address
> 40 bits total.
> It should be:
> macaddr | XX:XX:XX:XX:XX:XX, MAC address
> 48 bits total.
Yup. I have changed the pg_
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Thanks. Will be fixed in 7.2.
I went ahead and changed the pg_description entry for 7.1.1; there
doesn't seem to be any harm in that ...
regards, tom lane
---(end of broadcast)--
Bruce Momjian writes:
> You will need an updated pg_ctl that understands -D.
All versions of pg_ctl understand -D.
--
Peter Eisentraut [EMAIL PROTECTED] http://funkturm.homeip.net/~peter
---(end of broadcast)---
TIP 1: subscribe and unsubsc
Ian Harding ([EMAIL PROTECTED]) reports a bug with a severity of 3
The lower the number the more severe it is.
Short Description
Error in Makefile.tcldefs
Long Description
'gmake' process blows up when it gets to compiling pl/tcl. The error message is
something to the effect of "flag '-Wl,-R/u
> >From my point of view, NULL is neither bigger, nor smaller, you can't
> compare it with a number.
>
> So it just comes at the end if you sort at all.
>
> (Perhaps you need to take a think about what NULL means in your data. Should
> NULL sort as if it's 0?, +infinity?, -infinity? if so why?
> Ian Harding ([EMAIL PROTECTED]) reports a bug with a severity of 3
> 'gmake' process blows up when it gets to compiling pl/tcl. The error message is
>something to the effect of "flag '-Wl,-R/usr/pgk/lib' is not recognized." I didn't
>see where -Wl was a valid flag in a quick scan of the man
[EMAIL PROTECTED] writes:
> 'gmake' process blows up when it gets to compiling pl/tcl. The error
> message is something to the effect of "flag '-Wl,-R/usr/pgk/lib' is
> not recognized."
My guess is that your Tcl installation is broken: tclConfig.sh contains
settings that aren't actually valid fo
Bruce Momjian <[EMAIL PROTECTED]> writes:
> We have a TODO item:
> * Make NULL's come out at the beginning or end depending on the
> ORDER BY direction
The tricky part of this is to know which direction you are talking
about, when all you've been given is an operator that might hav
kun xie ([EMAIL PROTECTED]) reports a bug with a severity of 3
The lower the number the more severe it is.
Short Description
error in make the example
Long Description
In postgresql7.1,I cann't "make" the examples(/postgresql7.1/src/test/example/),it
returns something like this:
undefined refe
kun xie ([EMAIL PROTECTED]) reports a bug with a severity of 3
The lower the number the more severe it is.
Short Description
error in make the example
Long Description
In postgresql7.1,I cann't "make" the examples(/postgresql7.1/src/test/example/),it
returns something like this:
undefined refe
37 matches
Mail list logo
