pgbench: Add \syncpipeline
This change adds a new meta-command called \syncpipeline to pgbench,
able to send a sync message without flushing using the new libpq
function PQsendPipelineSync().
This meta-command is available within a block made of \startpipeline and
\endpipeline.
Author: Anthonin
Adjust populate_record_field() to handle errors softly
This adds a Node *escontext parameter to it and a bunch of functions
downstream to it, replacing any ereport()s in that path by either
errsave() or ereturn() as appropriate. This also adds code to those
functions where necessary to return
Refactor code used by jsonpath executor to fetch variables
Currently, getJsonPathVariable() directly extracts a named
variable/key from the source Jsonb value. This commit puts that
logic into a callback function called by getJsonPathVariable().
Other implementations of the callback may accept
Add soft error handling to some expression nodes
This adjusts the code for CoerceViaIO and CoerceToDomain expression
nodes to handle errors softly.
For CoerceViaIo, this adds a new ExprEvalStep opcode
EEOP_IOCOERCE_SAFE, which is implemented in the new accompanying
function
Fix ALTER TABLE .. ADD COLUMN with complex inheritance trees
This command, when used to add a column on a parent table with a complex
inheritance tree, tried to update multiple times the same tuple in
pg_attribute for a child table when incrementing attinhcount, causing
failures with "tuple
Fix ALTER TABLE .. ADD COLUMN with complex inheritance trees
This command, when used to add a column on a parent table with a complex
inheritance tree, tried to update multiple times the same tuple in
pg_attribute for a child table when incrementing attinhcount, causing
failures with "tuple
Fix ALTER TABLE .. ADD COLUMN with complex inheritance trees
This command, when used to add a column on a parent table with a complex
inheritance tree, tried to update multiple times the same tuple in
pg_attribute for a child table when incrementing attinhcount, causing
failures with "tuple
Fix ALTER TABLE .. ADD COLUMN with complex inheritance trees
This command, when used to add a column on a parent table with a complex
inheritance tree, tried to update multiple times the same tuple in
pg_attribute for a child table when incrementing attinhcount, causing
failures with "tuple
Fix ALTER TABLE .. ADD COLUMN with complex inheritance trees
This command, when used to add a column on a parent table with a complex
inheritance tree, tried to update multiple times the same tuple in
pg_attribute for a child table when incrementing attinhcount, causing
failures with "tuple
Fix ALTER TABLE .. ADD COLUMN with complex inheritance trees
This command, when used to add a column on a parent table with a complex
inheritance tree, tried to update multiple times the same tuple in
pg_attribute for a child table when incrementing attinhcount, causing
failures with "tuple
Fix crash in autoprewarm.
Commit abb0b4fc03 moved the shared state for autoprewarm to a
dynamic shared memory (DSM) segment, but it left apw_detach_shmem()
in the on_shmem_exit callback list for the autoprewarm leader
process. This is a problem because shmem_exit() detaches all the
DSM segments
Support shared libraries on Android (using make)
While the rest of the make build system maps Android to Linux, Android
uses unversioned shared libraries (like "libpq.so"). This patch makes
it so. (Meson already supported it.)
Reported-by: Matthias Kuhn
Discussion:
meson: Map host_system android to linux
This appears to be what is missing to support builds on Android (with
meson).
Author: Andres Freund
Discussion:
https://www.postgresql.org/message-id/flat/CAC7zN94TdsHhY88XkroJzSMx7E%3DBQpV9LKKjNSEnTM04ihoWCA%40mail.gmail.com
Branch
--
master
Peter Eisentraut writes:
> On 23.01.24 15:08, Tom Lane wrote:
>> Per guaibasaurus[1], this commit broke "make dist":
>> I'm not sure why --- the makefile changes treat these new files
>> exactly like other generated files. It doesn't repro here,
>> either, which is odd. It looks like a timing
Fix makefiles for newly added files
The new files added in 9b1a6f50b9 need to be mentioned in a few more
places in the makefiles.
Discussion:
https://www.postgresql.org/message-id/flat/E1rSAY2-002hIk-4y%40gemulon.postgresql.org
Branch
--
master
Details
---
On 23.01.24 15:08, Tom Lane wrote:
Peter Eisentraut writes:
Generate syscache info from catalog files
Per guaibasaurus[1], this commit broke "make dist":
...
rm -rf tmp_install/ portlock/
rm -f config.cache config.log config.status GNUmakefile
make[1]: Leaving directory
I wrote:
> Peter Eisentraut writes:
>> Generate syscache info from catalog files
> Per guaibasaurus[1], this commit broke "make dist":
> ...
> I'm not sure why --- the makefile changes treat these new files
> exactly like other generated files. It doesn't repro here,
> either, which is odd. It
Peter Eisentraut writes:
> Generate syscache info from catalog files
Per guaibasaurus[1], this commit broke "make dist":
...
rm -rf tmp_install/ portlock/
rm -f config.cache config.log config.status GNUmakefile
make[1]: Leaving directory
doc: Add acronym and glossary term for Access Method
AM was used throughout the documentation to denote Access Method, but
the acronym was not described. This adds an acronym entry as well as
a glossary term which the acronym links to. Each page which describe
AMs have the first occurrence with
meson: portname was set too early
portname is set to host_system, but host_system might still be changed
later in the file. As a consequence, platforms where host_system is
changed later wouldn't find some of their platform-specific files
(driven by portname), and the build would fail. Move the
meson: portname was set too early
portname is set to host_system, but host_system might still be changed
later in the file. As a consequence, platforms where host_system is
changed later wouldn't find some of their platform-specific files
(driven by portname), and the build would fail. Move the
Revert "libpqwalreceiver: Convert to libpq-be-fe-helpers.h"
This reverts commit 728f86fec65537eade8d9e751961782ddb527934.
The signal handling was a few bricks shy of a load in that commit,
which made the walreceiver non-responsive to SIGTERM while it was
waiting for the connection to be
Revert "libpqwalreceiver: Convert to libpq-be-fe-helpers.h"
This reverts commit 728f86fec65537eade8d9e751961782ddb527934.
The signal handling was a few bricks shy of a load in that commit,
which made the walreceiver non-responsive to SIGTERM while it was
waiting for the connection to be
23 matches
Mail list logo
