clared types if the
> caller didn't setup the FuncExpr node.
This will result in "plpgsql functions cannot return type anyarray"
which is at best misleading. It'd be better to have a specific error
message, say "could not determine actual return type for poly
as this. Why are you using a nonstandard syntax?
regards, tom lane
---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])
draft
version in our developer's FAQ, IIRC.
regards, tom lane
---(end of broadcast)---
TIP 6: Have you searched our list archives?
http://archives.postgresql.org
now that I look more closely at the new docs, there are some
classifications that seem pretty bizarre. For instance,
CLIENT_MIN_MESSAGES has nothing to do with the server log and doesn't
seem to belong in a section about logging ...
regards, tom lane
---
Josh Berkus <[EMAIL PROTECTED]> writes:
> What is CLIENT_MIN_MESSAGES for, then?
Controls which messages get sent to the client.
regards, tom lane
---(end of broadcast)---
TIP 9: the planner will ignore your
orting and Logging"?
regards, tom lane
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster
ubmit a patch for both runtime.sgml and postgresql.conf.sample?
I can fold that in while I'm doing the other stuff...
regards, tom lane
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
sitions in the argument type vector.
BTW, I can't see any need to include the return type in the hash key ---
wouldn't it be predetermined given the argument types?
regards, tom lane
---(end of broadcast)---
TIP 2:
nly shouldn't have a
name as generic as resolve_type.
regards, tom lane
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
Joe Conway <[EMAIL PROTECTED]> writes:
> Tom Lane wrote:
>> You're not providing enough info to resolve_type to let it handle all
>> four cases correctly.
> No, resolve_type() is not at all specific to polymorphic aggregates. It
> implements the rules of polymorph
rence from ANY is that multiple occurences of ANYELEMENT are all
constrained to stand for the same type.)
regards, tom lane
---(end of broadcast)---
TIP 3: if posting/reading through Usenet, please send an appropriate
su
on for it outside of deriving a polymorphic
aggregate's state type, though.
regards, tom lane
---(end of broadcast)---
TIP 3: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to [EM
ed with the current function cache entry. Then it would be
practical to recover the memory associated with an obsoleted cache
entry. But we've lived without this so far, so I guess it's not causing
many people problems.
regards, tom lane
-
tual assigned result type, and
then we can run the normal checking operations without any problem.
Applied patch attached, just FYI. (It still needs documentation
updates, which I trust you will supply later.)
Now back to looking at polymorphic aggregates...
regards, tom lane
des a
couple more flag bits to try to identify the behaviors of various
variables. (I've not checked Aizaz' settings for the flags though,
there might be some mistakes.)
regards, tom lane
---(end of broadcast)---
TIP 6: Have you searched our list archives?
http://archives.postgresql.org
ted to correct the
function signatures before anything else started to depend on them.
regards, tom lane
---(end of broadcast)---
TIP 8: explain analyze is your friend
someone apply the below patch from Jon?
Applied.
regards, tom lane
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
y identified a bug here; initializing
to NULL will just result in a guaranteed crash.
regards, tom lane
---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister You
n of GROUP BY
lists so that a CUBE operation could be distinguished from ordinary
grouping.
regards, tom lane
---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])
thers later? I also think the messages should be made more
> consistent in various ways, but that can be done later.
I made some small changes for improved consistency but I'm sure
considerably more editing could be done. Have at it.
regards, tom lane
--
will turn up ...
regards, tom lane
---(end of broadcast)---
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
ould
modify.) Or perhaps "functionname%type", although that seems subtly
wrong for reasons I can't quite put my finger on at this late hour.
Anyway, I'd like a representation that doesn't involve a phony variable.
regards, tom lane
ry context
> to ensure we don't leak memory when we don't want to.
Yah.
regards, tom lane
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
Joe Conway <[EMAIL PROTECTED]> writes:
> Tom Lane wrote:
>> I'm not excited about picking a notation for the long term on the
>> grounds that it takes the least code to implement today.
> I knew you wouldn't be ;-)
>> I admit I have not got a clearly-bett
t perhaps some will disagree.
3. What should be the set of tested values? I have it as
buffers: first to work of 1000 900 800 700 600 500 400 300 200 100 50
connections: first to work of 100 50 40 30 20 10
but we could certainly argue for different rules.
regards, tom
hey'l fix it.
regards, tom lane
---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])
't this called flex 3.0?
Or at least 2.6? Calling it a minor version bump has to violate a
truth-in-advertising law or two.
regards, tom lane
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
ould be
unable to cope with this.
I disagree with the concept of expecting someone to supply useful values
at install time anyway, since a newbie is the *least* likely to have any
idea what to say at that time. Heck, the experts can hardly agree on
what to use ...
erious
applications. initdb can't be expected to do this unless you want to
make it interactive, which would certainly make the RPM guys really
unhappy.
I'd rather see such considerations pushed off to a separate tool,
some kind of "configuration wizard" perhaps.
d changed in the meantime.
If the comment had been there, it would not have saved him from this
error, since I doubt it would have occurred to him to look for such a
comment.
regards, tom lane
---(end of broadcast)---
T
Weiping He <[EMAIL PROTECTED]> writes:
> seems the comment of macro ARR_LBOUND missed '(a)',
> here is the diff:
Applied, thanks!
regards, tom lane
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Tom Lane wrote:
>> This reminds me, someone (Barry?) was griping that SHOW DATESTYLE
>> doesn't produce a value that SET DATESTYLE will take. Did we agree
>> that it was OK to change the output to look like "ISO, US
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Tom Lane wrote:
>> No, it's a DONE.
> I just tried this with CVS:
> test=> show datestyle;
> DateStyle
> ---
>ISO with US (NonEur
Kenji Sugita <[EMAIL PROTECTED]> writes:
> This small patch eliminates relations in information_schema from oid2name
> listing.
Seems like it'd be a good idea to eliminate views and composite types as
well.
regards, tom lane
---
nd table names. As such, what it *ought* to be doing is
suppressing views, since those don't have disk files --- they are only
cluttering the listing with irrelevant data. If we put that in, there'd
be no need to special-case information_schema.
regards, tom lane
e should unlink the file that
it was originally told to unlink, regardless of what happens to those
globals.
If you are intent on spending code to free stuff just before the
postmaster exits, a better fix would be for UnlinkLockFile to free its
string argument after using it.
h doing, and I'll accept the change as a matter of
future-proofing that routine against such use. But on the argument of
preventing resource leakage today, this is just a waste of code space.
regards, tom lane
---(end of broadcast)-
t on translation work after that. There will doubtless be some
further tweaking, as Peter says.
regards, tom lane
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
Joe Conway <[EMAIL PROTECTED]> writes:
> I just noticed a new compile warning on date.c. Attached patch fixes it.
Wups, I shoulda looked closer at that code I was forward-porting from
7.2. Thanks for the timely (ahem) correction!
regards,
't keep a copy, and
I'm not having any luck finding that message in the archives right
now. Sumit, did you keep a copy?
regards, tom lane
---(end of broadcast)---
TIP 5: Have you checked our extensive F
e there are no other paths for parsing a date spec in the main
date/time code?
regards, tom lane
---(end of broadcast)---
TIP 6: Have you searched our list archives?
http://archives.postgresql.org
Kris Jurka <[EMAIL PROTECTED]> writes:
> On Fri, 25 Jul 2003, Tom Lane wrote:
>> I'm not having any luck finding that message in the archives right
>> now. Sumit, did you keep a copy?
> http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&selm=28882.105
uot; (accept dd/mm/yy) and a third case that accepts yy/mm/dd.
But I do not believe we can get away with deciding that common date
formats aren't common.
regards, tom lane
---(end of broadcast)---
TIP 6: Have you searched our list archives?
http://archives.postgresql.org
patches. The people who will be
screaming about it don't read -patches.
regards, tom lane
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster
's'? I think v,s,t are all relkinds to exclude here.
regards, tom lane
---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])
-parameter case of each function.
regards, tom lane
---(end of broadcast)---
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
what I changed,
except that while I was at it I renamed freeaddrinfo2 to the hopefully
more mnemonic freeaddrinfo_all). The change in miscinit should still
work though.
regards, tom lane
---(end of broadcast)---
TIP 1: subscribe
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Tom Lane wrote:
>> I think v,s,t are all relkinds to exclude here.
> Is 't' for toast tables? If so, we should allow 't', no? I wasn't sure
> about 's'?
Wups, you are right --- I was thinking &#
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Tom, how do I pass PG_FUNCTION_ARGS to another function, while adding a
> new parameter?
I wouldn't. Do the PG_GETARGS in the wrapper, and have the called
function take a normal C parameter list.
rite any ARGISNULL checks.
regards, tom lane
---(end of broadcast)---
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
ependently patching the same code with too long delay between patch
submission and application. The comment wouldn't have prevented the
error, because patch(1) can't read comments.
regards, tom lane
---(end of broadcast)---
TIP 8: explain analyze is your friend
ng for the constants should help you find
places you need to change when adding one.
regards, tom lane
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
Andreas Haumer <[EMAIL PROTECTED]> writes:
> But genbki.sh is not SMP safe: it uses a temporary file
> TMPFILE="$TMPDIR/genbkitmp.c"
Good catch. Applied for 7.4 --- thanks!
regards, tom lane
---(end of broadcast)---
h of settings.
In any case, we already have mechanisms for preventing specific users
from altering data: that's what GRANT/REVOKE are for. I don't think
anyone would have bothered with START TRANSACTION READ ONLY if it
weren't required by the SQL spec.
rega
avior is changed again.
regards, tom lane
---(end of broadcast)---
TIP 5: Have you checked our extensive FAQ?
http://www.postgresql.org/docs/faqs/FAQ.html
bjecting to using GUC for it. Send in a patch that, say, adds a
bool column to pg_shadow, and I'll be happy.
regards, tom lane
---(end of broadcast)---
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
ns when needed. We now need the additional
> prototype include patch from builtins.h.
Applied with some editorializing. In particular, I don't believe the
original did the right thing with (a - (b - c)).
regards, tom lane
--
UC for
this is just asking for trouble. Even if there is no security hole
today, it's very easy to imagine future changes in GUC that would
unintentionally create one.
regards, tom lane
---(end of broadcast)---
TIP
tip or a nightly
snapshot tarball should be pretty-durn-close-to-the-beta.
regards, tom lane
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
t have been visible according to the
original snapshot.
I don't understand what the diffs in postgres.c are supposed to
accomplish, but I can just about promise that they are misplaced too,
since they won't affect queries executed via functions or SPI.
ay, one is needed.
But I disagree with your opinion anyway --- _ext for "extended" made
perfect sense to me, especially for the get_indexdef one which adds more
than just prettyprint functionality.
regards, tom lane
---(end of broadcast)
sult, for example, just moves the core dump
somewhere else, unless you find reasonable fallbacks for *all* its
callers (including applications you don't have the source code for,
but in any case including every one of the calls in libpq).
regards, tom lane
ddition
is not associative.)
More: if the tree structure for ops of equal precedence looks like
a + (b + c), then it's a near certainty that the user wrote those
parentheses. Why would you think that removing them is pretty-printing?
regards, tom lane
---
sion=# select 1.0::float8 + ((-1.0::float8) + 1.0e-20::float8);
?column?
--
0
(1 row)
regards, tom lane
---(end of broadcast)---
TIP 5: Have you checked our extensive FAQ?
http://www.postgre
Bruce Momjian <[EMAIL PROTECTED]> writes:
> This has been saved for the 7.5 release:
> http:/momjian.postgresql.org/cgi-bin/pgpatches2
"Saved for"? It's not gonna be acceptable in its current form for 7.5,
either.
regards, tom lane
tcl to try to execute the unknown-module
load on every pltcl function call :-(. You really need two bits of
state if you are going to have separate postmaster-time and backend-time
initialization.
Will fix and commit.
regards, tom lane
---
[ followup to old message ]
Greg Stark <[EMAIL PROTECTED]> writes:
> This patch adds an autoconf test to check for the new incompatible version of
> flex.
Peter E. claims that CVS tip does work with flex 2.5.31 --- so we
shouldn't need this patch anymore.
he changes since 7.3...
regards, tom lane
---(end of broadcast)---
TIP 8: explain analyze is your friend
mp_var(&fact, &one);
Uh, why not use const_one?
regards, tom lane
---(end of broadcast)---
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
gic just deals with int.
We could leave the factorial(numeric) case open for a future
implementation that uses gamma, if anyone gets hot to do it.
regards, tom lane
---(end of broadcast)---
TIP 7: don't forget to incr
did ;-).
Sorry about the duplicated work ...
regards, tom lane
---(end of broadcast)---
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
Manfred Koizar <[EMAIL PROTECTED]> writes:
> On Fri, 04 Jul 2003 15:29:37 -0400, Tom Lane <[EMAIL PROTECTED]>
> wrote:
>> The attached patch shows how initdb can dynamically determine reasonable
>> shared_buffers and max_connections settings that will work on the
>
It might
work as-is for text fields, and for integers on bigendian machines, but
not for much else.)
We are not going back to the pre-7.4 format. Sorry.
regards, tom lane
---(end of broadcast)---
TIP 6: Have you searched
nt value; otherwise the patch looks fine.
> If this patch is acceptable, I'll make similar changes to plpython.
Please do.
regards, tom lane
---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])
ort. So the use of elog is sufficient documentation. IMHO
anyway.
regards, tom lane
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster
We are not going to back-patch this, since it creates portability
hazards that we can't risk without a proper beta test cycle.
Feel free to back-patch locally if it's in your way.
regards, tom lane
---(end of broadcast)--
ontrib module and putting earthdistance back with the rest,
myself. But you'll have to convince Marc ;-)
regards, tom lane
---(end of broadcast)---
TIP 8: explain analyze is your friend
heritance
cases (I think a child tuple should be said to be IS OF the tuple type
of its parent). And what about domains --- should we say a domain type
IS OF its base type?
regards, tom lane
---(end of broadcast)---
TIP 8: explain analyze is your friend
cking of my own.
regards, tom lane
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster
Joe Conway <[EMAIL PROTECTED]> writes:
> Tom Lane wrote:
>> I'm unconvinced that the parse-time-constant implementation Lockhart
>> started has anything whatever to do with the semantics the SQL99 spec
>> has in mind.
> Yeah - I've realized this is
t once.
I think that's a reasonable assumption; the question was just about
whether your code acts sanely if the assumption isn't true.
regards, tom lane
---(end of broadcast)---
TIP 8: explain analyze is your friend
eave pgsql as a module that contains only
pgsql-server, and thus not break anyone's checkout?
regards, tom lane
---(end of broadcast)---
TIP 8: explain analyze is your friend
t; docs to follow once
> I'm sure what actually ends up being committed.
You're on the hook for docs fixes...
regards, tom lane
---(end of broadcast)---
TIP 8: explain analyze is your friend
Joe Conway <[EMAIL PROTECTED]> writes:
> Here's a patch for the documentation updates. Please apply.
Applied with one small change --- I thought the last example failed
to make the point that the elements of an ARRAY[] construct could be
dissimilar.
reg
g us to remove functionality.
regards, tom lane
---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])
the 7.3 branch --- the rest of the changes didn't apply cleanly,
and seemed unrelated anyway. This fixes the immediate complaint, but
if there's some reason to apply the rest to 7.3.*, would you submit
an additional patch against that branch?
regards, tom lane
Karel Zak <[EMAIL PROTECTED]> writes:
> Hmm.. but the patch fix another small bug which I found today.
Okay. Please send a patch against 7.3 tip and I'll throw it in.
regards, tom lane
---(en
Karel Zak <[EMAIL PROTECTED]> writes:
>>> Hmm.. but the patch fix another small bug which I found today.
>>
>> Okay. Please send a patch against 7.3 tip and I'll throw it in.
> The patch for REL7_3_4 is attached.
Applied, thanks.
regards, tom lane
---(end of broadcast)---
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
this hint is fairly silly,
especially since it would pop out for simple misspellings of language
names.
regards, tom lane
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
Andrew Dunstan <[EMAIL PROTECTED]> writes:
> can you the missing space on the marked line, or do you need another
> patch from me?
Fixed.
regards, tom lane
---(end of broadcast)---
TIP 8: explain analyze is your friend
Pv6 setup here; please check that it does what you want.
regards, tom lane
*** src/backend/libpq/hba.c.origFri Sep 5 10:35:54 2003
--- src/backend/libpq/hba.c Fri Sep 5 16:24:41 2003
***
*** 673,685
if (cidr_slash)
*cidr_slash =
be valid.
*/
attStruct->atttypid = InvalidOid;
regards, tom lane
---(end of broadcast)---
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
Andreas Pflug <[EMAIL PROTECTED]> writes:
> Tom Lane wrote:
>> I thought this was still really messy, so I modified it to use a
>> separate "promote v4 address to v6" subroutine. I've applied the
>> attached patch (plus docs). It's not very well
ode either way I think. If you wanna try
to clean it up more, be my guest ...
regards, tom lane
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
ppercase though, no? Or were
you intending the variable definitions to be upper case and all the
references lower case? It's not real clear what the intended style is,
and I have to say that I thought Neil was heading in the intended
direction also.
regards, tom lane
this one yet.
regards, tom lane
---(end of broadcast)---
TIP 5: Have you checked our extensive FAQ?
http://www.postgresql.org/docs/faqs/FAQ.html
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Would you like me to conditionally add the IPv6 line to pg_hba.conf from
> initdb now?
I was going to tackle that tomorrow, but if you wanna do it, go for it.
regards, tom lane
---(end
ikely to work when building outside the source tree.
regards, tom lane
---(end of broadcast)---
TIP 6: Have you searched our list archives?
http://archives.postgresql.org
Bruce Momjian <[EMAIL PROTECTED]> writes:
> but top_srcdir isn't defined at this point.
Perhaps $srcdir? Peter would probably know what to use.
regards, tom lane
---(end of broadcast)---
TIP 3: if posting/r
1 - 100 of 4175 matches
Mail list logo
