meson: Add headerscheck and cpluspluscheck targets
Author: Miłosz Bieniek
Co-authored-by: Peter Eisentraut
Reviewed-by: Bilal Yavuz
Discussion:
https://www.postgresql.org/message-id/flat/CAMSWrt-PoQt4sHryWrB1ViuGBJF_PpbjoSGrWR2Ry47bHNLDqg%40mail.gmail.com
Branch
--
master
Details
---
meson: Add {perl|python}_includespec to generated Makefile.global
This is meant to help enable headerscheck under meson, but can also be
useful in general, for example for third-party extension that might
use these values.
Author: Nazir Bilal Yavuz
Discussion:
https://www.postgresql.org/message
Enable -Wstrict-prototypes and -Wold-style-definition by default
Those are available in all gcc and clang versions that support C11 and as C11
is required as of f5e0186f865c, then we can add them without capability test.
Having them enabled by default avoid having to chase these manually like
111
Fix pg_dump for CREATE FOREIGN DATA WRAPPER ... CONNECTION.
Discussion:
https://postgr.es/m/[email protected]
Branch
--
master
Details
---
https://git.postgresql.org/pg/commitdiff/b71bf3b8457027c445b5b3aa4914daa7e6718cf4
Modified Files
-
Exclude contrib/pg_plan_advice/pgpa_parser.h from headerscheck.
Like other Bison-written headers, it's not worth the trouble to
make this compilable standalone. (We might revisit this someday,
if we ever move up our minimum required Bison version.)
Branch
--
master
Details
---
https://g
ALTER SUBSCRIPTION ... SERVER test.
Test ALTER SUBSCRIPTION ... SERVER and ALTER SUBSCRIPTION
... CONNECTION, including invalidation.
Also run perltidy on the test file.
Discussion:
https://postgr.es/m/CAExHW5vV5znEvecX=ra2-v7ubj9-m6qvddzub78m-txbyd1...@mail.gmail.com
Suggested-by: Ashutosh Bap
Allow setting the collation strength in ICU tailoring rules
There was a bug that if you created an ICU collation with tailoring
rules, any strength specification inside the rules was ignored. This
was because we called ucol_openRules() with UCOL_DEFAULT_STRENGTH for
the strength argument, which o
ssl: Serverside SNI support for libpq
Support for SNI was added to clientside libpq in 5c55dc8b4733 with the
sslsni parameter, but there was no support for utilizing it serverside.
This adds support for serverside SNI such that certificate/key handling
is available per host. A new config file, $d
ssl: Add tests for client CA
These tests were originally written to test the SSL SNI patchset
but they have merit on their own since we lack coverage for these
scenarios in the non SNI case as well.
Author: Jacob Champion
Co-authored-by: Daniel Gustafsson
Discussion: https://postgr.es/m/1c81cd0
Update RELEASE_CHANGES
The existing instructions did not cover meson. Point to
src/common/unicode/README instead, where there is more information.
Reviewed-by: Chao Li
Discussion:
https://www.postgresql.org/message-id/flat/2a668979-ed92-49a3-abf9-a3ec2d460ec2%40eisentraut.org
Branch
--
ma
Implement unaccent Unicode data update in meson
The meson/ninja update-unicode target did not cover the required
updates in contrib/unaccent/. This is fixed now.
Reviewed-by: Chao Li
Reviewed-by: Alexander Borisov
Discussion:
https://www.postgresql.org/message-id/flat/2a668979-ed92-49a3-abf9-
pg_restore: Remove unnecessary strlen() calls in options parsing.
Unlike pg_dump and pg_dumpall, pg_restore first checks whether the
argument passed to --format, --host, and --port is empty before
setting the corresponding variable. Consequently, pg_restore does
not error if given an empty format
Fix -Wstrict-prototypes warning in ecpg_init_sqlca() declaration.
When headerscheck compiles ecpglib_extern.h, POSTGRES_ECPG_INTERNAL is
not defined, causing sqlca.h to expand "sqlca" as a macro
(*ECPGget_sqlca()). This causes the ecpg_init_sqlca() declaration to
trigger a -Wstrict-prototypes war
ssl: Skip passphrase reload tests in EXEC_BACKEND builds
SSL password command reloading must be enabled on Windows and in
EXEC_BACKEND builds due to them always reloading the context. The
new tests in commit 4f433025 skipped under Windows but missed the
EXEC_BACKEND check. Reported by buildfarm m
Add some const qualifiers enabled by typeof_unqual change on copyObject
The recent commit to change copyObject() to use typeof_unqual allows
cleaning up some APIs to take advantage of this improved qualifier
handling. EventTriggerCollectSimpleCommand() is a good example: It
takes a node tree and
test_saslprep: Test module for SASLprep()
This module includes two functions:
- test_saslprep(), that performs pg_saslprep on a bytea.
- test_saslprep_ranges(), able to check for all valid ranges of UTF-8
codepoints pg_saslprep() handles each one of them.
This provides a detailed coverage of our
Short-circuit row estimation in NOT IN containing NULL consts
ScalarArrayOpExpr used for either NOT IN or <>/= ALL, when the array
contains a NULL constant, will never evaluate to true. Here we add an
explicit short-circuit in scalararraysel() to account for this and return
0.0 rows when we see t
test_saslprep: Apply proper indentation
Noticed before koel has the idea to complain. Rebase thinko from commit
aa73838a5c94.
Branch
--
master
Details
---
https://git.postgresql.org/pg/commitdiff/015d32016d845f8a29b3ec3ab7fa98a69cea1a0f
Modified Files
--
.../modules/test_sa
Add commit 015d32016d84 to .git-blame-ignore-revs.
Branch
--
master
Details
---
https://git.postgresql.org/pg/commitdiff/46fb08aff69fd293961ac57e9339aef85692
Modified Files
--
.git-blame-ignore-revs | 3 +++
1 file changed, 3 insertions(+)
pg_plan_advice: Fix multiple copy-and-paste-errors in test case.
The second half of this file is meant to test feedback, not
generated advice, and is meant to use the statements that it
prepares, not leftover prepared statements from earlier in the
file.
These mistakes resulted in failures under
doc: Clarify BUFFERS behavior without ANALYZE in EXPLAIN
This commit clarifies the documentation for the BUFFERS option of EXPLAIN
by explicitly describing its behavior when ANALYZE is not specified.
Author: Ryo Matsumura
Reviewed-by: Fujii Masao
Discussion:
https://postgr.es/m/tyrpr01mb13457d
Add more debugging information for bgworker termination tests of worker_spi
widowbird has failed again after af8837a10bc7, with the same symptoms of
a backend still lying around when attempting a database rename with a
bgworker connected to the database being renamed.
We are still not sure yet ho
22 matches
Mail list logo
