the formal JDBC spec, but the latter seems perhaps
safer to me.
Regards,
-Chap
olate a caller's expectation
that the whole outcome is known by that point.
Regards,
-Chap
On 2023-07-14 14:19, David G. Johnston wrote:
Is there some magic set of arguments I should be using besides: tcpdump
-Ar
filename ?
I opened it with Wireshark, which has a pgsql protocol decoder.
Regards,
-Chap
returned rows
in the suspended-cursor case before executeUpdate returns, but keeping
the cursor around for getGeneratedKeys() to use, so executeUpdate
returns -1 before even having seen the later command complete, and would
still do that even if the command complete message had the right count.
Regards,
-Chap
portal gets suspended,
then the later command completion reports a zero count?
Regards,
-Chap
f two different operations, say one
an INSERT RETURNING and the other something else?
Regards,
-Chap
an INSERT RETURNING?
execute() -> true
getResultSet() -> the rows
getMoreResults() -> false
getUpdateCount() -> number inserted?
It seems that would fit the portal's behavior easily enough.
Or is the JDBC spec insisting on some other order?
Regards,
-Chap
readable on the server and has
the same format that COPY FROM expects?
Regards,
-Chap
d hey presto, the VOLATILE ones all work,
and the IMMUTABLE ones aren't there yet.
Regards,
-Chap
x27;t IMMUTABLE, only because
it depends on a fetch /of its own implementation/, and overall
its behavior is better described by marking it IMMUTABLE.
Regards,
-Chap
rent in what that declaration
means to the optimizer. (And also influences what snapshot the
function is looking at, and therefore what it can see, which has
also struck me more as a tacked-on effect than something inherent
in the declaration's meaning.)
Regards,
-Chap
ut. Configuration may get
trickier, and there may be some incentive to to include, say,
sepgsql in the picture.
Sure, even access to a file API can be game over, depending on
what file you open, but that's already the risk for every PL
with an 'untrusted' flavor.
Regards,
-Chap
ckends (and, for Java purposes, how many JVMs) get
fired up, and have each sharing sessions that have something in
common.
Or, would that just require all the complexity of both
approaches to synchronization, with no sufficient benefit?
Regards,
-Chap
[1]
https://docs.oracle.com/en/java/javase/
, they are necessary to allow one inline
handler to serve more than one PL.
ok
I will probably be unable to produce a new patch for this CF.
Family obligation.
Regards,
-Chap
ry to allow one inline
handler to serve more than one PL.
Regards,
-Chap
e zone:
jshell> java.time.ZoneId.of("America/Toronto").
...> getRules().
...> nextTransition(java.time.Instant.parse("0101-01-01T00:00:00Z"))
$1 ==> Transition[Gap at 1895-01-01T00:00-05:17:32 to -05:00]
Regards,
-Chap
reflecting Toronto's actual longitude.
Regards,
-Chap
On 2022-02-22 12:59, Chapman Flack wrote:
It would have been painful to write documentation of get_func_trftypes
saying its result isn't what get_transform_{from.to}sql expect, so
patch 1 does add a get_call_trftypes that returns a List *.
Patch 2 then updates the docs as discussed in this threa
On 2022-04-07 15:04, Andres Freund wrote:
And done. Chap, could you confirm this fixes the issue for you?
Looks good from here. One installcheck-world with no failures;
previously,
it failed for me every time.
Regards,
-Chap
;-ggdb -Og -g3 -fno-omit-frame-pointer'
Regards,
-Chap
nch of others. So a2f433f seems to have silenced all the rest
of those, but not this one.
If I build from ad40166, installcheck-world passes. That's as far
as I have pursued it.
Regards,
-Chap
would both streamline the example to leave
out the pfree calls, and be an illustration of best practice in letting
the memory context machinery handle all the deallocation at once, where
there isn't a special need to free something, like an especially large
allocation, at retail.
Regards,
-Chap
nd sessions, and already has a
column
for current_timestamp - state_change.
Regards,
-Chap
23 matches
Mail list logo
