On Mon, 2024-03-04 at 18:10 +, Tom Lane wrote:
> Further further fix pg_upgrade crossversion test for adminpack.
Ever since this commit the cross-version upgrade test is failing (for
me, at least) with:
# Running: psql -X -v ON_ERROR_STOP=1 -c drop database if exists
Document units of "timeout" in ConditionVariableTimedSleep()
The timeout is passed down to WaitLatch() as milliseconds.
Author: Shveta Malik
Discussion:
https://postgr.es/m/CAJpy0uC=xibqd1wapgyyvoiytap6uljaakld867zzxqu9ty...@mail.gmail.com
Branch
--
master
Details
---
On Fri, 2024-03-08 at 10:24 +0200, Heikki Linnakangas wrote:
> On 07/03/2024 21:18, Jeff Davis wrote:
> > Unicode case mapping tables and functions.
>
> With -Wtype-limits, I'm seeing this warning:
Thank you, fixed. Somehow I lost that flag from my script.
Can you please add some recommended
Fix type signedness error in commit 5c40364dd6.
Use ssize_t instead of size_t.
Discussion: https://postgr.es/m/b20d6d97-7338-48ea-ba33-837a1c8ef...@iki.fi
Reported-by: Heikki Linnakangas
Branch
--
master
Details
---
Replace perror with custom postgres logging
perror() is not used in postgres anymore out of policy, this replaces
the final callsites with the custom postgres logging framework.
Reviewed-by: Tom Lane
Reviewed-by: Peter Eisentraut
Discussion:
Fix errorhandling for reading from a pipe
When reading a line from a pipe failed on no data being read, the
errorhandling was erroneously logging with %m even thoug no error
description is available for %m to print. This flaw accidentally
introduced in 5c7038d70bb.
Reported-by: Peter Eisentraut
Improve WIN32 waiting logic in psql's \watch command.
do_watch had some leftover logic for enabling siglongjmp out of
waiting for input. That's never done anything on Windows (cf.
psql_cancel_callback), and do_watch no longer relies on it for
non-Windows, so let's drop it.
Also, when the user
Admit deferrable PKs into rd_pkindex, but flag them as such
... and in particular don't return them as replica identity.
The motivation for this change is letting the primary keys be seen by
code that derives NOT NULL constraints from them, when creating
inheritance children; before this change,
Avoid stack overflow in ShowTransactionStateRec()
The function recurses, but didn't perform stack-depth checks. It's
just a debugging aid, so instead of the usual check_stack_depth()
call, stop the printing if we'd risk stack overflow.
Here's an example of how to test this:
(n=100;
Avoid recursion in MemoryContext functions
You might run out of stack space with recursion, which is not nice in
functions that might be used e.g. at cleanup after transaction
abort. MemoryContext contains pointer to parent and siblings, so we
can traverse a tree of contexts iteratively, without
Turn tail recursion into iteration in CommitTransactionCommand()
Usually the compiler will optimize away the tail recursion anyway, but
if it doesn't, you can drive the function into stack overflow. For
example:
(n=100; printf "BEGIN;"; for ((i=1;i<=$n;i++)); do printf "SAVEPOINT
s$i;";
On 07/03/2024 21:18, Jeff Davis wrote:
Unicode case mapping tables and functions.
With -Wtype-limits, I'm seeing this warning:
unicode_case.c: In function ‘convert_case’:
unicode_case.c:107:47: warning: comparison of unsigned expression in ‘<
0’ is always false [-Wtype-limits]
107 |
12 matches
Mail list logo
