external
/
citus
mirror of https://github.com/citusdata/citus.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'release-13.0' into fix-issue-7676

pull/7914/head
Colm 2025-03-07 08:47:09 +00:00 committed by GitHub
parent 10c953a883 1d0b15723b
commit b201eee280
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
252 changed files with 9349 additions and 33423 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
.devcontainer/Dockerfile
View File

@ -68,7 +68,7 @@ USER citus
# build postgres versions separately for effective parrallelism and caching of already built versions when changing only certain versions
FROM base AS pg14
RUN MAKEFLAGS="-j $(nproc)" pgenv build 14.14
RUN MAKEFLAGS="-j $(nproc)" pgenv build 14.15
RUN rm .pgenv/src/*.tar*
RUN make -C .pgenv/src/postgresql-*/ clean
RUN make -C .pgenv/src/postgresql-*/src/include install
@ -80,7 +80,7 @@ RUN cp -r .pgenv/src .pgenv/pgsql-* .pgenv/config .pgenv-staging/
RUN rm .pgenv-staging/config/default.conf
FROM base AS pg15
RUN MAKEFLAGS="-j $(nproc)" pgenv build 15.9
RUN MAKEFLAGS="-j $(nproc)" pgenv build 15.10
RUN rm .pgenv/src/*.tar*
RUN make -C .pgenv/src/postgresql-*/ clean
RUN make -C .pgenv/src/postgresql-*/src/include install
@ -92,7 +92,7 @@ RUN cp -r .pgenv/src .pgenv/pgsql-* .pgenv/config .pgenv-staging/
RUN rm .pgenv-staging/config/default.conf
FROM base AS pg16
RUN MAKEFLAGS="-j $(nproc)" pgenv build 16.5
RUN MAKEFLAGS="-j $(nproc)" pgenv build 16.6
RUN rm .pgenv/src/*.tar*
RUN make -C .pgenv/src/postgresql-*/ clean
RUN make -C .pgenv/src/postgresql-*/src/include install
@ -104,7 +104,7 @@ RUN cp -r .pgenv/src .pgenv/pgsql-* .pgenv/config .pgenv-staging/
RUN rm .pgenv-staging/config/default.conf
FROM base AS pg17
RUN MAKEFLAGS="-j $(nproc)" pgenv build 17.1
RUN MAKEFLAGS="-j $(nproc)" pgenv build 17.2
RUN rm .pgenv/src/*.tar*
RUN make -C .pgenv/src/postgresql-*/ clean
RUN make -C .pgenv/src/postgresql-*/src/include install
@ -223,7 +223,7 @@ COPY --chown=citus:citus .psqlrc .
RUN sudo chown --from=root:root citus:citus -R ~
# sets default pg version
RUN pgenv switch 17.1
RUN pgenv switch 17.2
# make connecting to the coordinator easy
ENV PGPORT=9700

2
.gitattributes vendored
View File

@ -25,8 +25,6 @@ configure -whitespace
# except these exceptions...
src/backend/distributed/utils/citus_outfuncs.c -citus-style
src/backend/distributed/deparser/ruleutils_13.c -citus-style
src/backend/distributed/deparser/ruleutils_14.c -citus-style
src/backend/distributed/deparser/ruleutils_15.c -citus-style
src/backend/distributed/deparser/ruleutils_16.c -citus-style
src/backend/distributed/deparser/ruleutils_17.c -citus-style

2
.github/actions/save_logs_and_results/action.yml vendored
View File

@ -6,7 +6,7 @@ inputs:
runs:
using: composite
steps:
- uses: actions/upload-artifact@v3.1.1
- uses: actions/upload-artifact@v4.6.0
name: Upload logs
with:
name: ${{ inputs.folder }}

2
.github/actions/setup_extension/action.yml vendored
View File

@ -17,7 +17,7 @@ runs:
echo "PG_MAJOR=${{ inputs.pg_major }}" >> $GITHUB_ENV
fi
shell: bash
- uses: actions/download-artifact@v3.0.1
- uses: actions/download-artifact@v4.1.8
with:
name: build-${{ env.PG_MAJOR }}
- name: Install Extension

4
.github/actions/upload_coverage/action.yml vendored
View File

@ -21,7 +21,7 @@ runs:
mkdir -p /tmp/codeclimate
cc-test-reporter format-coverage -t lcov -o /tmp/codeclimate/${{ inputs.flags }}.json lcov.info
shell: bash
- uses: actions/upload-artifact@v3.1.1
- uses: actions/upload-artifact@v4.6.0
with:
path: "/tmp/codeclimate/*.json"
name: codeclimate
name: codeclimate-${{ inputs.flags }}

88
.github/workflows/build_and_test.yml vendored
View File

@ -26,13 +26,13 @@ jobs:
pgupgrade_image_name: "ghcr.io/citusdata/pgupgradetester"
style_checker_image_name: "ghcr.io/citusdata/stylechecker"
style_checker_tools_version: "0.8.18"
sql_snapshot_pg_version: "17.1"
image_suffix: "-v84c0cf8"
pg14_version: '{ "major": "14", "full": "14.14" }'
pg15_version: '{ "major": "15", "full": "15.9" }'
pg16_version: '{ "major": "16", "full": "16.5" }'
pg17_version: '{ "major": "17", "full": "17.1" }'
upgrade_pg_versions: "14.14-15.9-16.5-17.1"
sql_snapshot_pg_version: "17.2"
image_suffix: "-v889e4c1"
image_suffix_citus_upgrade: "-dev-2ad1f90"
pg15_version: '{ "major": "15", "full": "15.10" }'
pg16_version: '{ "major": "16", "full": "16.6" }'
pg17_version: '{ "major": "17", "full": "17.2" }'
upgrade_pg_versions: "14.15-15.10-16.6-17.2"
steps:
# Since GHA jobs need at least one step we use a noop step here.
- name: Set up parameters
@ -44,7 +44,7 @@ jobs:
image: ${{ needs.params.outputs.build_image_name }}:${{ needs.params.outputs.sql_snapshot_pg_version }}${{ needs.params.outputs.image_suffix }}
options: --user root
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
- name: Check Snapshots
run: |
git config --global --add safe.directory ${GITHUB_WORKSPACE}
@ -58,7 +58,7 @@ jobs:
- name: Check Snapshots
run: |
git config --global --add safe.directory ${GITHUB_WORKSPACE}
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Check C Style
@ -106,7 +106,6 @@ jobs:
image_suffix:
- ${{ needs.params.outputs.image_suffix}}
pg_version:
- ${{ needs.params.outputs.pg14_version }}
- ${{ needs.params.outputs.pg15_version }}
- ${{ needs.params.outputs.pg16_version }}
- ${{ needs.params.outputs.pg17_version }}
@ -115,14 +114,14 @@ jobs:
image: "${{ matrix.image_name }}:${{ fromJson(matrix.pg_version).full }}${{ matrix.image_suffix }}"
options: --user root
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
- name: Expose $PG_MAJOR to Github Env
run: echo "PG_MAJOR=${PG_MAJOR}" >> $GITHUB_ENV
shell: bash
- name: Build
run: "./ci/build-citus.sh"
shell: bash
- uses: actions/upload-artifact@v3.1.1
- uses: actions/upload-artifact@v4.6.0
with:
name: build-${{ env.PG_MAJOR }}
path: |-
@ -138,7 +137,6 @@ jobs:
image_name:
- ${{ needs.params.outputs.test_image_name }}
pg_version:
- ${{ needs.params.outputs.pg14_version }}
- ${{ needs.params.outputs.pg15_version }}
- ${{ needs.params.outputs.pg16_version }}
- ${{ needs.params.outputs.pg17_version }}
@ -159,10 +157,6 @@ jobs:
- check-enterprise-isolation-logicalrep-2
- check-enterprise-isolation-logicalrep-3
include:
- make: check-failure
pg_version: ${{ needs.params.outputs.pg14_version }}
suite: regress
image_name: ${{ needs.params.outputs.fail_test_image_name }}
- make: check-failure
pg_version: ${{ needs.params.outputs.pg15_version }}
suite: regress
@ -175,10 +169,6 @@ jobs:
pg_version: ${{ needs.params.outputs.pg17_version }}
suite: regress
image_name: ${{ needs.params.outputs.fail_test_image_name }}
- make: check-enterprise-failure
pg_version: ${{ needs.params.outputs.pg14_version }}
suite: regress
image_name: ${{ needs.params.outputs.fail_test_image_name }}
- make: check-enterprise-failure
pg_version: ${{ needs.params.outputs.pg15_version }}
suite: regress
@ -191,10 +181,6 @@ jobs:
pg_version: ${{ needs.params.outputs.pg17_version }}
suite: regress
image_name: ${{ needs.params.outputs.fail_test_image_name }}
- make: check-pytest
pg_version: ${{ needs.params.outputs.pg14_version }}
suite: regress
image_name: ${{ needs.params.outputs.fail_test_image_name }}
- make: check-pytest
pg_version: ${{ needs.params.outputs.pg15_version }}
suite: regress
@ -219,10 +205,6 @@ jobs:
suite: cdc
image_name: ${{ needs.params.outputs.test_image_name }}
pg_version: ${{ needs.params.outputs.pg17_version }}
- make: check-query-generator
pg_version: ${{ needs.params.outputs.pg14_version }}
suite: regress
image_name: ${{ needs.params.outputs.fail_test_image_name }}
- make: check-query-generator
pg_version: ${{ needs.params.outputs.pg15_version }}
suite: regress
@ -246,7 +228,7 @@ jobs:
- params
- build
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
- uses: "./.github/actions/setup_extension"
- name: Run Test
run: gosu circleci make -C src/test/${{ matrix.suite }} ${{ matrix.make }}
@ -275,13 +257,12 @@ jobs:
image_name:
- ${{ needs.params.outputs.fail_test_image_name }}
pg_version:
- ${{ needs.params.outputs.pg14_version }}
- ${{ needs.params.outputs.pg15_version }}
- ${{ needs.params.outputs.pg16_version }}
- ${{ needs.params.outputs.pg17_version }}
parallel: [0,1,2,3,4,5] # workaround for running 6 parallel jobs
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
- uses: "./.github/actions/setup_extension"
- name: Test arbitrary configs
run: |-
@ -303,10 +284,12 @@ jobs:
check-arbitrary-configs parallel=4 CONFIGS=$TESTS
- uses: "./.github/actions/save_logs_and_results"
if: always()
with:
folder: ${{ env.PG_MAJOR }}_arbitrary_configs_${{ matrix.parallel }}
- uses: "./.github/actions/upload_coverage"
if: always()
with:
flags: ${{ env.pg_major }}_upgrade
flags: ${{ env.PG_MAJOR }}_arbitrary_configs_${{ matrix.parallel }}
codecov_token: ${{ secrets.CODECOV_TOKEN }}
test-pg-upgrade:
name: PG${{ matrix.old_pg_major }}-PG${{ matrix.new_pg_major }} - check-pg-upgrade
@ -321,23 +304,17 @@ jobs:
fail-fast: false
matrix:
include:
- old_pg_major: 14
new_pg_major: 15
- old_pg_major: 15
new_pg_major: 16
- old_pg_major: 14
new_pg_major: 16
- old_pg_major: 16
new_pg_major: 17
- old_pg_major: 15
new_pg_major: 17
- old_pg_major: 14
new_pg_major: 17
env:
old_pg_major: ${{ matrix.old_pg_major }}
new_pg_major: ${{ matrix.new_pg_major }}
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
- uses: "./.github/actions/setup_extension"
with:
pg_major: "${{ env.old_pg_major }}"
@ -360,22 +337,24 @@ jobs:
if: failure()
- uses: "./.github/actions/save_logs_and_results"
if: always()
with:
folder: ${{ env.old_pg_major }}_${{ env.new_pg_major }}_upgrade
- uses: "./.github/actions/upload_coverage"
if: always()
with:
flags: ${{ env.old_pg_major }}_${{ env.new_pg_major }}_upgrade
codecov_token: ${{ secrets.CODECOV_TOKEN }}
test-citus-upgrade:
name: PG${{ fromJson(needs.params.outputs.pg14_version).major }} - check-citus-upgrade
name: PG${{ fromJson(needs.params.outputs.pg15_version).major }} - check-citus-upgrade
runs-on: ubuntu-20.04
container:
image: "${{ needs.params.outputs.citusupgrade_image_name }}:${{ fromJson(needs.params.outputs.pg14_version).full }}${{ needs.params.outputs.image_suffix }}"
image: "${{ needs.params.outputs.citusupgrade_image_name }}:${{ fromJson(needs.params.outputs.pg15_version).full }}${{ needs.params.outputs.image_suffix_citus_upgrade }}"
options: --user root
needs:
- params
- build
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
- uses: "./.github/actions/setup_extension"
with:
skip_installation: true
@ -405,10 +384,12 @@ jobs:
done;
- uses: "./.github/actions/save_logs_and_results"
if: always()
with:
folder: ${{ env.PG_MAJOR }}_citus_upgrade
- uses: "./.github/actions/upload_coverage"
if: always()
with:
flags: ${{ env.pg_major }}_upgrade
flags: ${{ env.PG_MAJOR }}_citus_upgrade
codecov_token: ${{ secrets.CODECOV_TOKEN }}
upload-coverage:
if: always()
@ -424,10 +405,11 @@ jobs:
- test-citus-upgrade
- test-pg-upgrade
steps:
- uses: actions/download-artifact@v3.0.1
- uses: actions/download-artifact@v4.1.8
with:
name: "codeclimate"
path: "codeclimate"
pattern: codeclimate*
path: codeclimate
merge-multiple: true
- name: Upload coverage results to Code Climate
run: |-
cc-test-reporter sum-coverage codeclimate/*.json -o total.json
@ -439,7 +421,7 @@ jobs:
needs:
- build
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
- uses: azure/login@v1
with:
creds: ${{ secrets.AZURE_CREDENTIALS }}
@ -457,7 +439,7 @@ jobs:
needs:
- build
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
- uses: azure/login@v1
with:
creds: ${{ secrets.AZURE_CREDENTIALS }}
@ -476,7 +458,7 @@ jobs:
outputs:
json: ${{ steps.parallelization.outputs.json }}
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
- uses: "./.github/actions/parallelization"
id: parallelization
with:
@ -489,7 +471,7 @@ jobs:
outputs:
tests: ${{ steps.detect-regression-tests.outputs.tests }}
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Detect regression tests need to be ran
@ -524,8 +506,8 @@ jobs:
fail-fast: false
matrix: ${{ fromJson(needs.prepare_parallelization_matrix_32.outputs.json) }}
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/download-artifact@v3.0.1
- uses: actions/checkout@v4
- uses: actions/download-artifact@v4.1.8
- uses: "./.github/actions/setup_extension"
- name: Run minimal tests
run: |-

2
.github/workflows/codeql.yml vendored
View File

@ -21,7 +21,7 @@ jobs:
steps:
- name: Checkout repository
uses: actions/checkout@v3
uses: actions/checkout@v4
- name: Initialize CodeQL
uses: github/codeql-action/init@v2

8
.github/workflows/flaky_test_debugging.yml vendored
View File

@ -28,13 +28,13 @@ jobs:
image: ${{ vars.build_image_name }}:${{ vars.pg15_version }}${{ vars.image_suffix }}
options: --user root
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
- name: Configure, Build, and Install
run: |
echo "PG_MAJOR=${PG_MAJOR}" >> $GITHUB_ENV
./ci/build-citus.sh
shell: bash
- uses: actions/upload-artifact@v3.1.1
- uses: actions/upload-artifact@v4.6.0
with:
name: build-${{ env.PG_MAJOR }}
path: |-
@ -46,7 +46,7 @@ jobs:
outputs:
json: ${{ steps.parallelization.outputs.json }}
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
- uses: "./.github/actions/parallelization"
id: parallelization
with:
@ -67,7 +67,7 @@ jobs:
fail-fast: false
matrix: ${{ fromJson(needs.prepare_parallelization_matrix.outputs.json) }}
steps:
- uses: actions/checkout@v3.5.0
- uses: actions/checkout@v4
- uses: "./.github/actions/setup_extension"
- name: Run minimal tests
run: |-

3
.github/workflows/packaging-test-pipelines.yml vendored
View File

@ -115,7 +115,6 @@ jobs:
# for each deb based image and we use POSTGRES_VERSION to set
# PG_CONFIG variable in each of those runs.
packaging_docker_image:
- debian-buster-all
- debian-bookworm-all
- debian-bullseye-all
- ubuntu-focal-all
@ -129,7 +128,7 @@ jobs:
steps:
- name: Checkout repository
uses: actions/checkout@v3
uses: actions/checkout@v4
- name: Set pg_config path and python parameters for deb based distros
run: |

42
CHANGELOG.md
View File

@ -1,3 +1,45 @@
### citus v13.0.1 (February 4th, 2025) ###
* Drops support for PostgreSQL 14 (#7753)
### citus v13.0.0 (January 22, 2025) ###
* Adds support for PostgreSQL 17 (#7699, #7661)
* Adds `JSON_TABLE()` support in distributed queries (#7816)
* Propagates `MERGE ... WHEN NOT MATCHED BY SOURCE` (#7807)
* Propagates `MEMORY` and `SERIALIZE` options of `EXPLAIN` (#7802)
* Adds support for identity columns in distributed partitioned tables (#7785)
* Allows specifying an access method for distributed partitioned tables (#7818)
* Allows exclusion constraints on distributed partitioned tables (#7733)
* Allows configuring sslnegotiation using `citus.node_conn_info` (#7821)
* Avoids wal receiver timeouts during large shard splits (#7229)
* Fixes a bug causing incorrect writing of data to target `MERGE` repartition
command (#7659)
* Fixes a crash that happens because of unsafe catalog access when re-assigning
the global pid after `application_name` changes (#7791)
* Fixes incorrect `VALID UNTIL` setting assumption made for roles when syncing
them to new nodes (#7534)
* Fixes segfault when calling distributed procedure with a parameterized
distribution argument (#7242)
* Fixes server crash when trying to execute `activate_node_snapshot()` on a
single-node cluster (#7552)
* Improves `citus_move_shard_placement()` to fail early if there is a new node
without reference tables yet (#7467)
### citus v12.1.5 (July 17, 2024) ###
* Adds support for MERGE commands with single shard distributed target tables

2
ci/check_gucs_are_alphabetically_sorted.sh
View File

@ -5,6 +5,6 @@ set -euo pipefail
source ci/ci_helpers.sh
# extract citus gucs in the form of "citus.X"
grep -o -E "(\.*\"citus.\w+\")," src/backend/distributed/shared_library_init.c > gucs.out
grep -o -E "(\.*\"citus\.\w+\")," src/backend/distributed/shared_library_init.c > gucs.out
sort -c gucs.out
rm gucs.out

20
configure vendored
View File

@ -1,6 +1,6 @@
#! /bin/sh
# Guess values for system-dependent variables and create Makefiles.
# Generated by GNU Autoconf 2.69 for Citus 13.0.0.
# Generated by GNU Autoconf 2.69 for Citus 13.0.1.
#
#
# Copyright (C) 1992-1996, 1998-2012 Free Software Foundation, Inc.
@ -579,8 +579,8 @@ MAKEFLAGS=
# Identity of this package.
PACKAGE_NAME='Citus'
PACKAGE_TARNAME='citus'
PACKAGE_VERSION='13.0.0'
PACKAGE_STRING='Citus 13.0.0'
PACKAGE_VERSION='13.0.1'
PACKAGE_STRING='Citus 13.0.1'
PACKAGE_BUGREPORT=''
PACKAGE_URL=''
@ -1262,7 +1262,7 @@ if test "$ac_init_help" = "long"; then
# Omit some internal or obsolete options to make the list less imposing.
# This message is too long to be a string in the A/UX 3.1 sh.
cat <<_ACEOF
\`configure' configures Citus 13.0.0 to adapt to many kinds of systems.
\`configure' configures Citus 13.0.1 to adapt to many kinds of systems.
Usage: $0 [OPTION]... [VAR=VALUE]...
@ -1324,7 +1324,7 @@ fi
if test -n "$ac_init_help"; then
case $ac_init_help in
short | recursive ) echo "Configuration of Citus 13.0.0:";;
short | recursive ) echo "Configuration of Citus 13.0.1:";;
esac
cat <<\_ACEOF
@ -1429,7 +1429,7 @@ fi
test -n "$ac_init_help" && exit $ac_status
if $ac_init_version; then
cat <<\_ACEOF
Citus configure 13.0.0
Citus configure 13.0.1
generated by GNU Autoconf 2.69
Copyright (C) 2012 Free Software Foundation, Inc.
@ -1912,7 +1912,7 @@ cat >config.log <<_ACEOF
This file contains any messages produced by compilers while
running configure, to aid debugging if configure makes a mistake.
It was created by Citus $as_me 13.0.0, which was
It was created by Citus $as_me 13.0.1, which was
generated by GNU Autoconf 2.69. Invocation command line was
$ $0 $@
@ -2588,7 +2588,7 @@ fi
if test "$with_pg_version_check" = no; then
{ $as_echo "$as_me:${as_lineno-$LINENO}: building against PostgreSQL $version_num (skipped compatibility check)" >&5
$as_echo "$as_me: building against PostgreSQL $version_num (skipped compatibility check)" >&6;}
elif test "$version_num" != '14' -a "$version_num" != '15' -a "$version_num" != '16' -a "$version_num" != '17'; then
elif test "$version_num" != '15' -a "$version_num" != '16' -a "$version_num" != '17'; then
as_fn_error $? "Citus is not compatible with the detected PostgreSQL version ${version_num}." "$LINENO" 5
else
{ $as_echo "$as_me:${as_lineno-$LINENO}: building against PostgreSQL $version_num" >&5
@ -5393,7 +5393,7 @@ cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1
# report actual input values of CONFIG_FILES etc. instead of their
# values after options handling.
ac_log="
This file was extended by Citus $as_me 13.0.0, which was
This file was extended by Citus $as_me 13.0.1, which was
generated by GNU Autoconf 2.69. Invocation command line was
CONFIG_FILES = $CONFIG_FILES
@ -5455,7 +5455,7 @@ _ACEOF
cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1
ac_cs_config="`$as_echo "$ac_configure_args" | sed 's/^ //; s/[\\""\`\$]/\\\\&/g'`"
ac_cs_version="\\
Citus config.status 13.0.0
Citus config.status 13.0.1
configured by $0, generated by GNU Autoconf 2.69,
with options \\"\$ac_cs_config\\"

4
configure.ac
View File

@ -5,7 +5,7 @@
# everyone needing autoconf installed, the resulting files are checked
# into the SCM.
AC_INIT([Citus], [13.0.0])
AC_INIT([Citus], [13.0.1])
AC_COPYRIGHT([Copyright (c) Citus Data, Inc.])
# we'll need sed and awk for some of the version commands
@ -80,7 +80,7 @@ AC_SUBST(with_pg_version_check)
if test "$with_pg_version_check" = no; then
AC_MSG_NOTICE([building against PostgreSQL $version_num (skipped compatibility check)])
elif test "$version_num" != '14' -a "$version_num" != '15' -a "$version_num" != '16' -a "$version_num" != '17'; then
elif test "$version_num" != '15' -a "$version_num" != '16' -a "$version_num" != '17'; then
AC_MSG_ERROR([Citus is not compatible with the detected PostgreSQL version ${version_num}.])
else
AC_MSG_NOTICE([building against PostgreSQL $version_num])

133
gucs.out Normal file
View File

@ -0,0 +1,133 @@
"citus.all_modifications_commutative",
"citus.allow_modifications_from_workers_to_replicated_tables",
"citus.allow_nested_distributed_execution",
"citus.allow_unsafe_constraints",
"citus.allow_unsafe_locks_from_workers",
"citus.background_task_queue_interval",
"citus.check_available_space_before_move",
"citus.cluster_name",
"citus.coordinator_aggregation_strategy",
"citus.copy_switchover_threshold",
"citus.count_distinct_error_rate",
"citus.cpu_priority",
"citus.cpu_priority_for_logical_replication_senders",
"citus.create_object_propagation",
"citus.defer_drop_after_shard_move",
"citus.defer_drop_after_shard_split",
"citus.defer_shard_delete_interval",
"citus.desired_percent_disk_available_after_move",
"citus.distributed_deadlock_detection_factor",
"citus.enable_alter_database_owner",
"citus.enable_alter_role_propagation",
"citus.enable_alter_role_set_propagation",
"citus.enable_binary_protocol",
"citus.enable_change_data_capture",
"citus.enable_cluster_clock",
"citus.enable_cost_based_connection_establishment",
"citus.enable_create_role_propagation",
"citus.enable_create_type_propagation",
"citus.enable_ddl_propagation",
"citus.enable_deadlock_prevention",
"citus.enable_fast_path_router_planner",
"citus.enable_local_execution",
"citus.enable_local_reference_table_foreign_keys",
"citus.enable_manual_changes_to_shards",
"citus.enable_manual_metadata_changes_for_user",
"citus.enable_metadata_sync",
"citus.enable_non_colocated_router_query_pushdown",
"citus.enable_repartition_joins",
"citus.enable_repartitioned_insert_select",
"citus.enable_router_execution",
"citus.enable_schema_based_sharding",
"citus.enable_single_hash_repartition_joins",
"citus.enable_statistics_collection",
"citus.enable_unique_job_ids",
"citus.enable_unsafe_triggers",
"citus.enable_unsupported_feature_messages",
"citus.enable_version_checks",
"citus.enforce_foreign_key_restrictions",
"citus.enforce_object_restrictions_for_local_objects",
"citus.executor_slow_start_interval",
"citus.explain_all_tasks",
"citus.explain_analyze_sort_method",
"citus.explain_distributed_queries",
"citus.force_max_query_parallelization",
"citus.function_opens_transaction_block",
"citus.grep_remote_commands",
"citus.hide_citus_dependent_objects",
"citus.hide_shards_from_app_name_prefixes",
"citus.isolation_test_session_process_id",
"citus.isolation_test_session_remote_process_id",
"citus.limit_clause_row_fetch_count",
"citus.local_copy_flush_threshold",
"citus.local_hostname",
"citus.local_shared_pool_size",
"citus.local_table_join_policy",
"citus.log_distributed_deadlock_detection",
"citus.log_intermediate_results",
"citus.log_local_commands",
"citus.log_multi_join_order",
"citus.log_remote_commands",
"citus.logical_replication_timeout",
"citus.main_db",
"citus.max_adaptive_executor_pool_size",
"citus.max_background_task_executors",
"citus.max_background_task_executors_per_node",
"citus.max_cached_connection_lifetime",
"citus.max_cached_conns_per_worker",
"citus.max_client_connections",
"citus.max_high_priority_background_processes",
"citus.max_intermediate_result_size",
"citus.max_matview_size_to_auto_recreate",
"citus.max_rebalancer_logged_ignored_moves",
"citus.max_shared_pool_size",
"citus.max_worker_nodes_tracked",
"citus.metadata_sync_interval",
"citus.metadata_sync_mode",
"citus.metadata_sync_retry_interval",
"citus.mitmfifo",
"citus.multi_shard_modify_mode",
"citus.multi_task_query_log_level",
"citus.next_cleanup_record_id",
"citus.next_operation_id",
"citus.next_placement_id",
"citus.next_shard_id",
"citus.node_connection_timeout",
"citus.node_conninfo",
"citus.override_table_visibility",
"citus.prevent_incomplete_connection_establishment",
"citus.propagate_session_settings_for_loopback_connection",
"citus.propagate_set_commands",
"citus.rebalancer_by_disk_size_base_cost",
"citus.recover_2pc_interval",
"citus.remote_copy_flush_threshold",
"citus.remote_task_check_interval",
"citus.repartition_join_bucket_count_per_node",
"citus.replicate_reference_tables_on_activate",
"citus.replication_model",
"citus.running_under_citus_test_suite",
"citus.select_opens_transaction_block",
"citus.shard_count",
"citus.shard_replication_factor",
"citus.show_shards_for_app_name_prefixes",
"citus.skip_advisory_lock_permission_checks",
"citus.skip_constraint_validation",
"citus.skip_jsonb_validation_in_copy",
"citus.sort_returning",
"citus.stat_statements_max",
"citus.stat_statements_purge_interval",
"citus.stat_statements_track",
"citus.stat_tenants_limit",
"citus.stat_tenants_log_level",
"citus.stat_tenants_period",
"citus.stat_tenants_track",
"citus.stat_tenants_untracked_sample_rate",
"citus.subquery_pushdown",
"citus.task_assignment_policy",
"citus.task_executor_type",
"citus.use_citus_managed_tables",
"citus.use_secondary_nodes",
"citus.values_materialization_threshold",
"citus.version",
"citus.worker_min_messages",
"citus.writable_standby_coordinator",

12
src/backend/columnar/columnar_customscan.c
View File

@ -1051,6 +1051,15 @@ FindCandidateRelids(PlannerInfo *root, RelOptInfo *rel, List *joinClauses)
candidateRelids = bms_del_members(candidateRelids, rel->relids);
candidateRelids = bms_del_members(candidateRelids, rel->lateral_relids);
/*
* For the relevant PG16 commit requiring this addition:
* postgres/postgres@2489d76
*/
#if PG_VERSION_NUM >= PG_VERSION_16
candidateRelids = bms_del_members(candidateRelids, root->outer_join_rels);
#endif
return candidateRelids;
}
@ -1312,11 +1321,8 @@ AddColumnarScanPath(PlannerInfo *root, RelOptInfo *rel, RangeTblEntry *rte,
cpath->methods = &ColumnarScanPathMethods;
#if (PG_VERSION_NUM >= PG_VERSION_15)
/* necessary to avoid extra Result node in PG15 */
cpath->flags = CUSTOMPATH_SUPPORT_PROJECTION;
#endif
/*
* populate generic path information

2
src/backend/columnar/columnar_metadata.c
View File

@ -1686,7 +1686,7 @@ DeleteTupleAndEnforceConstraints(ModifyState *state, HeapTuple heapTuple)
simple_heap_delete(state->rel, tid);
/* execute AFTER ROW DELETE Triggers to enforce constraints */
ExecARDeleteTriggers_compat(estate, resultRelInfo, tid, NULL, NULL, false);
ExecARDeleteTriggers(estate, resultRelInfo, tid, NULL, NULL, false);
}

15
src/backend/columnar/columnar_tableam.c
View File

@ -877,7 +877,7 @@ columnar_relation_set_new_filelocator(Relation rel,
*freezeXid = RecentXmin;
*minmulti = GetOldestMultiXactId();
SMgrRelation srel = RelationCreateStorage_compat(*newrlocator, persistence, true);
SMgrRelation srel = RelationCreateStorage(*newrlocator, persistence, true);
ColumnarStorageInit(srel, ColumnarMetadataNewStorageId());
InitColumnarOptions(rel->rd_id);
@ -2245,7 +2245,6 @@ ColumnarProcessAlterTable(AlterTableStmt *alterTableStmt, List **columnarOptions
columnarRangeVar = alterTableStmt->relation;
}
}
#if PG_VERSION_NUM >= PG_VERSION_15
else if (alterTableCmd->subtype == AT_SetAccessMethod)
{
if (columnarRangeVar || *columnarOptions)
@ -2265,7 +2264,6 @@ ColumnarProcessAlterTable(AlterTableStmt *alterTableStmt, List **columnarOptions
DeleteColumnarTableOptions(RelationGetRelid(rel), true);
}
}
#endif /* PG_VERSION_15 */
}
relation_close(rel, NoLock);
@ -2649,21 +2647,12 @@ ColumnarCheckLogicalReplication(Relation rel)
return;
}
#if PG_VERSION_NUM >= PG_VERSION_15
{
PublicationDesc pubdesc;
RelationBuildPublicationDesc(rel, &pubdesc);
pubActionInsert = pubdesc.pubactions.pubinsert;
}
#else
if (rel->rd_pubactions == NULL)
{
GetRelationPublicationActions(rel);
Assert(rel->rd_pubactions != NULL);
}
pubActionInsert = rel->rd_pubactions->pubinsert;
#endif
if (pubActionInsert)
{
@ -3040,6 +3029,8 @@ AvailableExtensionVersionColumnar(void)
ereport(ERROR, (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
errmsg("citus extension is not found")));
return NULL; /* keep compiler happy */
}

11
src/backend/distributed/clock/causal_clock.c
View File

@ -145,17 +145,6 @@ LogicalClockShmemSize(void)
void
InitializeClusterClockMem(void)
{
/* On PG 15 and above, we use shmem_request_hook_type */
#if PG_VERSION_NUM < PG_VERSION_15
/* allocate shared memory for pre PG-15 versions */
if (!IsUnderPostmaster)
{
RequestAddinShmemSpace(LogicalClockShmemSize());
}
#endif
prev_shmem_startup_hook = shmem_startup_hook;
shmem_startup_hook = LogicalClockShmemInit;
}

75
src/backend/distributed/commands/alter_table.c
View File

@ -209,12 +209,9 @@ static void ReplaceTable(Oid sourceId, Oid targetId, List *justBeforeDropCommand
static bool HasAnyGeneratedStoredColumns(Oid relationId);
static List * GetNonGeneratedStoredColumnNameList(Oid relationId);
static void CheckAlterDistributedTableConversionParameters(TableConversionState *con);
static char * CreateWorkerChangeSequenceDependencyCommand(char *sequenceSchemaName,
char *sequenceName,
char *sourceSchemaName,
char *sourceName,
char *targetSchemaName,
char *targetName);
static char * CreateWorkerChangeSequenceDependencyCommand(char *qualifiedSequeceName,
char *qualifiedSourceName,
char *qualifiedTargetName);
static void ErrorIfMatViewSizeExceedsTheLimit(Oid matViewOid);
static char * CreateMaterializedViewDDLCommand(Oid matViewOid);
static char * GetAccessMethodForMatViewIfExists(Oid viewOid);
@ -791,13 +788,15 @@ ConvertTableInternal(TableConversionState *con)
justBeforeDropCommands = lappend(justBeforeDropCommands, detachFromParentCommand);
}
char *qualifiedRelationName = quote_qualified_identifier(con->schemaName,
con->relationName);
if (PartitionedTable(con->relationId))
{
if (!con->suppressNoticeMessages)
{
ereport(NOTICE, (errmsg("converting the partitions of %s",
quote_qualified_identifier(con->schemaName,
con->relationName))));
qualifiedRelationName)));
}
List *partitionList = PartitionList(con->relationId);
@ -870,9 +869,7 @@ ConvertTableInternal(TableConversionState *con)
if (!con->suppressNoticeMessages)
{
ereport(NOTICE, (errmsg("creating a new table for %s",
quote_qualified_identifier(con->schemaName,
con->relationName))));
ereport(NOTICE, (errmsg("creating a new table for %s", qualifiedRelationName)));
}
TableDDLCommand *tableCreationCommand = NULL;
@ -999,8 +996,6 @@ ConvertTableInternal(TableConversionState *con)
{
continue;
}
char *qualifiedRelationName = quote_qualified_identifier(con->schemaName,
con->relationName);
TableConversionParameters cascadeParam = {
.relationId = colocatedTableId,
@ -1750,9 +1745,7 @@ CreateMaterializedViewDDLCommand(Oid matViewOid)
{
StringInfo query = makeStringInfo();
char *viewName = get_rel_name(matViewOid);
char *schemaName = get_namespace_name(get_rel_namespace(matViewOid));
char *qualifiedViewName = quote_qualified_identifier(schemaName, viewName);
char *qualifiedViewName = generate_qualified_relation_name(matViewOid);
/* here we need to get the access method of the view to recreate it */
char *accessMethodName = GetAccessMethodForMatViewIfExists(matViewOid);
@ -1801,9 +1794,8 @@ ReplaceTable(Oid sourceId, Oid targetId, List *justBeforeDropCommands,
bool suppressNoticeMessages)
{
char *sourceName = get_rel_name(sourceId);
char *targetName = get_rel_name(targetId);
Oid schemaId = get_rel_namespace(sourceId);
char *schemaName = get_namespace_name(schemaId);
char *qualifiedSourceName = generate_qualified_relation_name(sourceId);
char *qualifiedTargetName = generate_qualified_relation_name(targetId);
StringInfo query = makeStringInfo();
@ -1811,8 +1803,7 @@ ReplaceTable(Oid sourceId, Oid targetId, List *justBeforeDropCommands,
{
if (!suppressNoticeMessages)
{
ereport(NOTICE, (errmsg("moving the data of %s",
quote_qualified_identifier(schemaName, sourceName))));
ereport(NOTICE, (errmsg("moving the data of %s", qualifiedSourceName)));
}
if (!HasAnyGeneratedStoredColumns(sourceId))
@ -1822,8 +1813,7 @@ ReplaceTable(Oid sourceId, Oid targetId, List *justBeforeDropCommands,
* "INSERT INTO .. SELECT *"".
*/
appendStringInfo(query, "INSERT INTO %s SELECT * FROM %s",
quote_qualified_identifier(schemaName, targetName),
quote_qualified_identifier(schemaName, sourceName));
qualifiedTargetName, qualifiedSourceName);
}
else
{
@ -1838,9 +1828,8 @@ ReplaceTable(Oid sourceId, Oid targetId, List *justBeforeDropCommands,
char *insertColumnString = StringJoin(nonStoredColumnNameList, ',');
appendStringInfo(query,
"INSERT INTO %s (%s) OVERRIDING SYSTEM VALUE SELECT %s FROM %s",
quote_qualified_identifier(schemaName, targetName),
insertColumnString, insertColumnString,
quote_qualified_identifier(schemaName, sourceName));
qualifiedTargetName, insertColumnString,
insertColumnString, qualifiedSourceName);
}
ExecuteQueryViaSPI(query->data, SPI_OK_INSERT);
@ -1864,14 +1853,11 @@ ReplaceTable(Oid sourceId, Oid targetId, List *justBeforeDropCommands,
*/
if (ShouldSyncTableMetadata(targetId))
{
Oid sequenceSchemaOid = get_rel_namespace(sequenceOid);
char *sequenceSchemaName = get_namespace_name(sequenceSchemaOid);
char *sequenceName = get_rel_name(sequenceOid);
char *qualifiedSequenceName = generate_qualified_relation_name(sequenceOid);
char *workerChangeSequenceDependencyCommand =
CreateWorkerChangeSequenceDependencyCommand(sequenceSchemaName,
sequenceName,
schemaName, sourceName,
schemaName, targetName);
CreateWorkerChangeSequenceDependencyCommand(qualifiedSequenceName,
qualifiedSourceName,
qualifiedTargetName);
SendCommandToWorkersWithMetadata(workerChangeSequenceDependencyCommand);
}
else if (ShouldSyncTableMetadata(sourceId))
@ -1894,25 +1880,23 @@ ReplaceTable(Oid sourceId, Oid targetId, List *justBeforeDropCommands,
if (!suppressNoticeMessages)
{
ereport(NOTICE, (errmsg("dropping the old %s",
quote_qualified_identifier(schemaName, sourceName))));
ereport(NOTICE, (errmsg("dropping the old %s", qualifiedSourceName)));
}
resetStringInfo(query);
appendStringInfo(query, "DROP %sTABLE %s CASCADE",
IsForeignTable(sourceId) ? "FOREIGN " : "",
quote_qualified_identifier(schemaName, sourceName));
qualifiedSourceName);
ExecuteQueryViaSPI(query->data, SPI_OK_UTILITY);
if (!suppressNoticeMessages)
{
ereport(NOTICE, (errmsg("renaming the new table to %s",
quote_qualified_identifier(schemaName, sourceName))));
ereport(NOTICE, (errmsg("renaming the new table to %s", qualifiedSourceName)));
}
resetStringInfo(query);
appendStringInfo(query, "ALTER TABLE %s RENAME TO %s",
quote_qualified_identifier(schemaName, targetName),
qualifiedTargetName,
quote_identifier(sourceName));
ExecuteQueryViaSPI(query->data, SPI_OK_UTILITY);
}
@ -2172,18 +2156,13 @@ CheckAlterDistributedTableConversionParameters(TableConversionState *con)
* worker_change_sequence_dependency query with the parameters.
*/
static char *
CreateWorkerChangeSequenceDependencyCommand(char *sequenceSchemaName, char *sequenceName,
char *sourceSchemaName, char *sourceName,
char *targetSchemaName, char *targetName)
CreateWorkerChangeSequenceDependencyCommand(char *qualifiedSequeceName,
char *qualifiedSourceName,
char *qualifiedTargetName)
{
char *qualifiedSchemaName = quote_qualified_identifier(sequenceSchemaName,
sequenceName);
char *qualifiedSourceName = quote_qualified_identifier(sourceSchemaName, sourceName);
char *qualifiedTargetName = quote_qualified_identifier(targetSchemaName, targetName);
StringInfo query = makeStringInfo();
appendStringInfo(query, "SELECT worker_change_sequence_dependency(%s, %s, %s)",
quote_literal_cstr(qualifiedSchemaName),
quote_literal_cstr(qualifiedSequeceName),
quote_literal_cstr(qualifiedSourceName),
quote_literal_cstr(qualifiedTargetName));

8
src/backend/distributed/commands/citus_add_local_table_to_metadata.c
View File

@ -1160,9 +1160,7 @@ DropIdentitiesOnTable(Oid relationId)
if (attributeForm->attidentity)
{
char *tableName = get_rel_name(relationId);
char *schemaName = get_namespace_name(get_rel_namespace(relationId));
char *qualifiedTableName = quote_qualified_identifier(schemaName, tableName);
char *qualifiedTableName = generate_qualified_relation_name(relationId);
StringInfo dropCommand = makeStringInfo();
@ -1222,9 +1220,7 @@ DropViewsOnTable(Oid relationId)
Oid viewId = InvalidOid;
foreach_declared_oid(viewId, reverseOrderedViews)
{
char *viewName = get_rel_name(viewId);
char *schemaName = get_namespace_name(get_rel_namespace(viewId));
char *qualifiedViewName = quote_qualified_identifier(schemaName, viewName);
char *qualifiedViewName = generate_qualified_relation_name(viewId);
StringInfo dropCommand = makeStringInfo();
appendStringInfo(dropCommand, "DROP %sVIEW IF EXISTS %s",

30
src/backend/distributed/commands/collation.c
View File

@ -68,8 +68,6 @@ CreateCollationDDLInternal(Oid collationId, Oid *collowner, char **quotedCollati
char *collcollate;
char *collctype;
#if PG_VERSION_NUM >= PG_VERSION_15
/*
* In PG15, there is an added option to use ICU as global locale provider.
* pg_collation has three locale-related fields: collcollate and collctype,
@ -112,16 +110,6 @@ CreateCollationDDLInternal(Oid collationId, Oid *collowner, char **quotedCollati
}
Assert((collcollate && collctype) || colllocale);
#else
/*
* In versions before 15, collcollate and collctype were type "name". Use
* pstrdup() to match the interface of 15 so that we consistently free the
* result later.
*/
collcollate = pstrdup(NameStr(collationForm->collcollate));
collctype = pstrdup(NameStr(collationForm->collctype));
#endif
if (collowner != NULL)
{
@ -147,7 +135,6 @@ CreateCollationDDLInternal(Oid collationId, Oid *collowner, char **quotedCollati
"CREATE COLLATION %s (provider = '%s'",
*quotedCollationName, providerString);
#if PG_VERSION_NUM >= PG_VERSION_15
if (colllocale)
{
appendStringInfo(&collationNameDef,
@ -173,24 +160,7 @@ CreateCollationDDLInternal(Oid collationId, Oid *collowner, char **quotedCollati
pfree(collcollate);
pfree(collctype);
}
#else
if (strcmp(collcollate, collctype) == 0)
{
appendStringInfo(&collationNameDef,
", locale = %s",
quote_literal_cstr(collcollate));
}
else
{
appendStringInfo(&collationNameDef,
", lc_collate = %s, lc_ctype = %s",
quote_literal_cstr(collcollate),
quote_literal_cstr(collctype));
}
pfree(collcollate);
pfree(collctype);
#endif
#if PG_VERSION_NUM >= PG_VERSION_16
char *collicurules = NULL;
datum = SysCacheGetAttr(COLLOID, heapTuple, Anum_pg_collation_collicurules, &isnull);

14
src/backend/distributed/commands/create_distributed_table.c
View File

@ -170,12 +170,10 @@ static void EnsureDistributedSequencesHaveOneType(Oid relationId,
static void CopyLocalDataIntoShards(Oid distributedTableId);
static List * TupleDescColumnNameList(TupleDesc tupleDescriptor);
#if (PG_VERSION_NUM >= PG_VERSION_15)
static bool DistributionColumnUsesNumericColumnNegativeScale(TupleDesc relationDesc,
Var *distributionColumn);
static int numeric_typmod_scale(int32 typmod);
static bool is_valid_numeric_typmod(int32 typmod);
#endif
static bool DistributionColumnUsesGeneratedStoredColumn(TupleDesc relationDesc,
Var *distributionColumn);
@ -1325,10 +1323,7 @@ CreateCitusTable(Oid relationId, CitusTableType tableType,
{
List *partitionList = PartitionList(relationId);
Oid partitionRelationId = InvalidOid;
Oid namespaceId = get_rel_namespace(relationId);
char *schemaName = get_namespace_name(namespaceId);
char *relationName = get_rel_name(relationId);
char *parentRelationName = quote_qualified_identifier(schemaName, relationName);
char *parentRelationName = generate_qualified_relation_name(relationId);
/*
* when there are many partitions, each call to CreateDistributedTable
@ -2117,8 +2112,6 @@ EnsureRelationCanBeDistributed(Oid relationId, Var *distributionColumn,
"AS (...) STORED.")));
}
#if (PG_VERSION_NUM >= PG_VERSION_15)
/* verify target relation is not distributed by a column of type numeric with negative scale */
if (distributionMethod != DISTRIBUTE_BY_NONE &&
DistributionColumnUsesNumericColumnNegativeScale(relationDesc,
@ -2129,7 +2122,6 @@ EnsureRelationCanBeDistributed(Oid relationId, Var *distributionColumn,
errdetail("Distribution column must not use numeric type "
"with negative scale")));
}
#endif
/* check for support function needed by specified partition method */
if (distributionMethod == DISTRIBUTE_BY_HASH)
@ -2847,8 +2839,6 @@ TupleDescColumnNameList(TupleDesc tupleDescriptor)
}
#if (PG_VERSION_NUM >= PG_VERSION_15)
/*
* is_valid_numeric_typmod checks if the typmod value is valid
*
@ -2898,8 +2888,6 @@ DistributionColumnUsesNumericColumnNegativeScale(TupleDesc relationDesc,
}
#endif
/*
* DistributionColumnUsesGeneratedStoredColumn returns whether a given relation uses
* GENERATED ALWAYS AS (...) STORED on distribution column

5
src/backend/distributed/commands/database.c
View File

@ -185,8 +185,6 @@ PreprocessAlterDatabaseStmt(Node *node, const char *queryString,
}
#if PG_VERSION_NUM >= PG_VERSION_15
/*
* PreprocessAlterDatabaseSetStmt is executed before the statement is applied to the local
* postgres instance.
@ -217,9 +215,6 @@ PreprocessAlterDatabaseRefreshCollStmt(Node *node, const char *queryString,
}
#endif
/*
* GetDatabaseAddressFromDatabaseName gets the database name and returns the ObjectAddress
* of the database.

24
src/backend/distributed/commands/distribute_object_ops.c
View File

@ -364,6 +364,15 @@ static DistributeObjectOps Any_Rename = {
.address = NULL,
.markDistributed = false,
};
static DistributeObjectOps Any_SecLabel = {
.deparse = DeparseSecLabelStmt,
.qualify = NULL,
.preprocess = NULL,
.postprocess = PostprocessSecLabelStmt,
.operationType = DIST_OPS_ALTER,
.address = SecLabelStmtObjectAddress,
.markDistributed = false,
};
static DistributeObjectOps Attribute_Rename = {
.deparse = DeparseRenameAttributeStmt,
.qualify = QualifyRenameAttributeStmt,
@ -456,7 +465,6 @@ static DistributeObjectOps Database_Alter = {
.markDistributed = false,
};
#if PG_VERSION_NUM >= PG_VERSION_15
static DistributeObjectOps Database_RefreshColl = {
.deparse = DeparseAlterDatabaseRefreshCollStmt,
.qualify = NULL,
@ -467,7 +475,6 @@ static DistributeObjectOps Database_RefreshColl = {
.address = NULL,
.markDistributed = false,
};
#endif
static DistributeObjectOps Domain_Alter = {
.deparse = DeparseAlterDomainStmt,
@ -828,7 +835,6 @@ static DistributeObjectOps Sequence_AlterOwner = {
.address = AlterSequenceOwnerStmtObjectAddress,
.markDistributed = false,
};
#if (PG_VERSION_NUM >= PG_VERSION_15)
static DistributeObjectOps Sequence_AlterPersistence = {
.deparse = DeparseAlterSequencePersistenceStmt,
.qualify = QualifyAlterSequencePersistenceStmt,
@ -838,7 +844,6 @@ static DistributeObjectOps Sequence_AlterPersistence = {
.address = AlterSequencePersistenceStmtObjectAddress,
.markDistributed = false,
};
#endif
static DistributeObjectOps Sequence_Drop = {
.deparse = DeparseDropSequenceStmt,
.qualify = QualifyDropSequenceStmt,
@ -1290,7 +1295,7 @@ static DistributeObjectOps View_Rename = {
static DistributeObjectOps Trigger_Rename = {
.deparse = NULL,
.qualify = NULL,
.preprocess = PreprocessAlterTriggerRenameStmt,
.preprocess = NULL,
.operationType = DIST_OPS_ALTER,
.postprocess = PostprocessAlterTriggerRenameStmt,
.address = NULL,
@ -1312,13 +1317,11 @@ GetDistributeObjectOps(Node *node)
return &Database_Alter;
}
#if PG_VERSION_NUM >= PG_VERSION_15
case T_AlterDatabaseRefreshCollStmt:
{
return &Database_RefreshColl;
}
#endif
case T_AlterDomainStmt:
{
return &Domain_Alter;
@ -1603,7 +1606,6 @@ GetDistributeObjectOps(Node *node)
case OBJECT_SEQUENCE:
{
#if (PG_VERSION_NUM >= PG_VERSION_15)
ListCell *cmdCell = NULL;
foreach(cmdCell, stmt->cmds)
{
@ -1631,7 +1633,6 @@ GetDistributeObjectOps(Node *node)
}
}
}
#endif
/*
* Prior to PG15, the only Alter Table statement
@ -1991,6 +1992,11 @@ GetDistributeObjectOps(Node *node)
return &Vacuum_Analyze;
}
case T_SecLabelStmt:
{
return &Any_SecLabel;
}
case T_RenameStmt:
{
RenameStmt *stmt = castNode(RenameStmt, node);

2
src/backend/distributed/commands/foreign_constraint.c
View File

@ -467,7 +467,6 @@ ForeignKeyGetDefaultingAttrs(HeapTuple pgConstraintTuple)
}
List *onDeleteSetDefColumnList = NIL;
#if PG_VERSION_NUM >= PG_VERSION_15
Datum onDeleteSetDefColumnsDatum = SysCacheGetAttr(CONSTROID, pgConstraintTuple,
Anum_pg_constraint_confdelsetcols,
&isNull);
@ -482,7 +481,6 @@ ForeignKeyGetDefaultingAttrs(HeapTuple pgConstraintTuple)
onDeleteSetDefColumnList =
IntegerArrayTypeToList(DatumGetArrayTypeP(onDeleteSetDefColumnsDatum));
}
#endif
if (list_length(onDeleteSetDefColumnList) == 0)
{

8
src/backend/distributed/commands/multi_copy.c
View File

@ -2549,12 +2549,8 @@ ShardIdForTuple(CitusCopyDestReceiver *copyDest, Datum *columnValues, bool *colu
if (columnNulls[partitionColumnIndex])
{
Oid relationId = copyDest->distributedRelationId;
char *relationName = get_rel_name(relationId);
Oid schemaOid = get_rel_namespace(relationId);
char *schemaName = get_namespace_name(schemaOid);
char *qualifiedTableName = quote_qualified_identifier(schemaName,
relationName);
char *qualifiedTableName = generate_qualified_relation_name(
copyDest->distributedRelationId);
ereport(ERROR, (errcode(ERRCODE_NULL_VALUE_NOT_ALLOWED),
errmsg("the partition column of table %s cannot be NULL",

30
src/backend/distributed/commands/publication.c
View File

@ -33,11 +33,9 @@
static CreatePublicationStmt * BuildCreatePublicationStmt(Oid publicationId);
#if (PG_VERSION_NUM >= PG_VERSION_15)
static PublicationObjSpec * BuildPublicationRelationObjSpec(Oid relationId,
Oid publicationId,
bool tableOnly);
#endif
static void AppendPublishOptionList(StringInfo str, List *strings);
static char * AlterPublicationOwnerCommand(Oid publicationId);
static bool ShouldPropagateCreatePublication(CreatePublicationStmt *stmt);
@ -154,7 +152,6 @@ BuildCreatePublicationStmt(Oid publicationId)
ReleaseSysCache(publicationTuple);
#if (PG_VERSION_NUM >= PG_VERSION_15)
List *schemaIds = GetPublicationSchemas(publicationId);
Oid schemaId = InvalidOid;
@ -170,7 +167,6 @@ BuildCreatePublicationStmt(Oid publicationId)
createPubStmt->pubobjects = lappend(createPubStmt->pubobjects, publicationObject);
}
#endif
List *relationIds = GetPublicationRelations(publicationId,
publicationForm->pubviaroot ?
@ -184,7 +180,6 @@ BuildCreatePublicationStmt(Oid publicationId)
foreach_declared_oid(relationId, relationIds)
{
#if (PG_VERSION_NUM >= PG_VERSION_15)
bool tableOnly = false;
/* since postgres 15, tables can have a column list and filter */
@ -192,15 +187,6 @@ BuildCreatePublicationStmt(Oid publicationId)
BuildPublicationRelationObjSpec(relationId, publicationId, tableOnly);
createPubStmt->pubobjects = lappend(createPubStmt->pubobjects, publicationObject);
#else
/* before postgres 15, only full tables are supported */
char *schemaName = get_namespace_name(get_rel_namespace(relationId));
char *tableName = get_rel_name(relationId);
RangeVar *rangeVar = makeRangeVar(schemaName, tableName, -1);
createPubStmt->tables = lappend(createPubStmt->tables, rangeVar);
#endif
if (IsCitusTable(relationId))
{
@ -276,8 +262,6 @@ AppendPublishOptionList(StringInfo str, List *options)
}
#if (PG_VERSION_NUM >= PG_VERSION_15)
/*
* BuildPublicationRelationObjSpec returns a PublicationObjSpec that
* can be included in a CREATE or ALTER PUBLICATION statement.
@ -357,9 +341,6 @@ BuildPublicationRelationObjSpec(Oid relationId, Oid publicationId,
}
#endif
/*
* PreprocessAlterPublicationStmt handles ALTER PUBLICATION statements
* in a way that is mostly similar to PreprocessAlterDistributedObjectStmt,
@ -458,7 +439,6 @@ GetAlterPublicationTableDDLCommand(Oid publicationId, Oid relationId,
ReleaseSysCache(pubTuple);
#if (PG_VERSION_NUM >= PG_VERSION_15)
bool tableOnly = !isAdd;
/* since postgres 15, tables can have a column list and filter */
@ -467,16 +447,6 @@ GetAlterPublicationTableDDLCommand(Oid publicationId, Oid relationId,
alterPubStmt->pubobjects = lappend(alterPubStmt->pubobjects, publicationObject);
alterPubStmt->action = isAdd ? AP_AddObjects : AP_DropObjects;
#else
/* before postgres 15, only full tables are supported */
char *schemaName = get_namespace_name(get_rel_namespace(relationId));
char *tableName = get_rel_name(relationId);
RangeVar *rangeVar = makeRangeVar(schemaName, tableName, -1);
alterPubStmt->tables = lappend(alterPubStmt->tables, rangeVar);
alterPubStmt->tableAction = isAdd ? DEFELEM_ADD : DEFELEM_DROP;
#endif
/* we take the WHERE clause from the catalog where it is already transformed */
bool whereClauseNeedsTransform = false;

99
src/backend/distributed/commands/role.c
View File

@ -22,6 +22,7 @@
#include "catalog/pg_auth_members.h"
#include "catalog/pg_authid.h"
#include "catalog/pg_db_role_setting.h"
#include "catalog/pg_shseclabel.h"
#include "catalog/pg_type.h"
#include "commands/dbcommands.h"
#include "nodes/makefuncs.h"
@ -65,6 +66,7 @@ static DefElem * makeDefElemBool(char *name, bool value);
static List * GenerateRoleOptionsList(HeapTuple tuple);
static List * GenerateGrantRoleStmtsFromOptions(RoleSpec *roleSpec, List *options);
static List * GenerateGrantRoleStmtsOfRole(Oid roleid);
static List * GenerateSecLabelOnRoleStmts(Oid roleid, char *rolename);
static void EnsureSequentialModeForRoleDDL(void);
static char * GetRoleNameFromDbRoleSetting(HeapTuple tuple,
@ -491,18 +493,17 @@ GenerateRoleOptionsList(HeapTuple tuple)
options = lappend(options, makeDefElem("password", NULL, -1));
}
/* load valid unitl data from the heap tuple, use default of infinity if not set */
/* load valid until data from the heap tuple */
Datum rolValidUntilDatum = SysCacheGetAttr(AUTHNAME, tuple,
Anum_pg_authid_rolvaliduntil, &isNull);
char *rolValidUntil = "infinity";
if (!isNull)
{
rolValidUntil = pstrdup((char *) timestamptz_to_str(rolValidUntilDatum));
}
char *rolValidUntil = pstrdup((char *) timestamptz_to_str(rolValidUntilDatum));
Node *validUntilStringNode = (Node *) makeString(rolValidUntil);
DefElem *validUntilOption = makeDefElem("validUntil", validUntilStringNode, -1);
options = lappend(options, validUntilOption);
Node *validUntilStringNode = (Node *) makeString(rolValidUntil);
DefElem *validUntilOption = makeDefElem("validUntil", validUntilStringNode, -1);
options = lappend(options, validUntilOption);
}
return options;
}
@ -517,13 +518,14 @@ GenerateCreateOrAlterRoleCommand(Oid roleOid)
{
HeapTuple roleTuple = SearchSysCache1(AUTHOID, ObjectIdGetDatum(roleOid));
Form_pg_authid role = ((Form_pg_authid) GETSTRUCT(roleTuple));
char *rolename = pstrdup(NameStr(role->rolname));
CreateRoleStmt *createRoleStmt = NULL;
if (EnableCreateRolePropagation)
{
createRoleStmt = makeNode(CreateRoleStmt);
createRoleStmt->stmt_type = ROLESTMT_ROLE;
createRoleStmt->role = pstrdup(NameStr(role->rolname));
createRoleStmt->role = rolename;
createRoleStmt->options = GenerateRoleOptionsList(roleTuple);
}
@ -534,7 +536,7 @@ GenerateCreateOrAlterRoleCommand(Oid roleOid)
alterRoleStmt->role = makeNode(RoleSpec);
alterRoleStmt->role->roletype = ROLESPEC_CSTRING;
alterRoleStmt->role->location = -1;
alterRoleStmt->role->rolename = pstrdup(NameStr(role->rolname));
alterRoleStmt->role->rolename = rolename;
alterRoleStmt->action = 1;
alterRoleStmt->options = GenerateRoleOptionsList(roleTuple);
}
@ -546,7 +548,7 @@ GenerateCreateOrAlterRoleCommand(Oid roleOid)
{
/* add a worker_create_or_alter_role command if any of them are set */
char *createOrAlterRoleQuery = CreateCreateOrAlterRoleCommand(
pstrdup(NameStr(role->rolname)),
rolename,
createRoleStmt,
alterRoleStmt);
@ -568,6 +570,20 @@ GenerateCreateOrAlterRoleCommand(Oid roleOid)
{
completeRoleList = lappend(completeRoleList, DeparseTreeNode(stmt));
}
/*
* append SECURITY LABEL ON ROLE commands for this specific user
* When we propagate user creation, we also want to make sure that we propagate
* all the security labels it has been given. For this, we check pg_shseclabel
* for the ROLE entry corresponding to roleOid, and generate the relevant
* SecLabel stmts to be run in the new node.
*/
List *secLabelOnRoleStmts = GenerateSecLabelOnRoleStmts(roleOid, rolename);
stmt = NULL;
foreach_declared_ptr(stmt, secLabelOnRoleStmts)
{
completeRoleList = lappend(completeRoleList, DeparseTreeNode(stmt));
}
}
return completeRoleList;
@ -897,6 +913,54 @@ GenerateGrantRoleStmtsOfRole(Oid roleid)
}
/*
* GenerateSecLabelOnRoleStmts generates the SecLabelStmts for the role
* whose oid is roleid.
*/
static List *
GenerateSecLabelOnRoleStmts(Oid roleid, char *rolename)
{
List *secLabelStmts = NIL;
/*
* Note that roles are shared database objects, therefore their
* security labels are stored in pg_shseclabel instead of pg_seclabel.
*/
Relation pg_shseclabel = table_open(SharedSecLabelRelationId, AccessShareLock);
ScanKeyData skey[1];
ScanKeyInit(&skey[0], Anum_pg_shseclabel_objoid, BTEqualStrategyNumber, F_OIDEQ,
ObjectIdGetDatum(roleid));
SysScanDesc scan = systable_beginscan(pg_shseclabel, SharedSecLabelObjectIndexId,
true, NULL, 1, &skey[0]);
HeapTuple tuple = NULL;
while (HeapTupleIsValid(tuple = systable_getnext(scan)))
{
SecLabelStmt *secLabelStmt = makeNode(SecLabelStmt);
secLabelStmt->objtype = OBJECT_ROLE;
secLabelStmt->object = (Node *) makeString(pstrdup(rolename));
Datum datumArray[Natts_pg_shseclabel];
bool isNullArray[Natts_pg_shseclabel];
heap_deform_tuple(tuple, RelationGetDescr(pg_shseclabel), datumArray,
isNullArray);
secLabelStmt->provider = TextDatumGetCString(
datumArray[Anum_pg_shseclabel_provider - 1]);
secLabelStmt->label = TextDatumGetCString(
datumArray[Anum_pg_shseclabel_label - 1]);
secLabelStmts = lappend(secLabelStmts, secLabelStmt);
}
systable_endscan(scan);
table_close(pg_shseclabel, AccessShareLock);
return secLabelStmts;
}
/*
* PreprocessCreateRoleStmt creates a worker_create_or_alter_role query for the
* role that is being created. With that query we can create the role in the
@ -963,13 +1027,8 @@ makeStringConst(char *str, int location)
{
A_Const *n = makeNode(A_Const);
#if PG_VERSION_NUM >= PG_VERSION_15
n->val.sval.type = T_String;
n->val.sval.sval = str;
#else
n->val.type = T_String;
n->val.val.str = str;
#endif
n->location = location;
return (Node *) n;
@ -989,13 +1048,8 @@ makeIntConst(int val, int location)
{
A_Const *n = makeNode(A_Const);
#if PG_VERSION_NUM >= PG_VERSION_15
n->val.ival.type = T_Integer;
n->val.ival.ival = val;
#else
n->val.type = T_Integer;
n->val.val.ival = val;
#endif
n->location = location;
return (Node *) n;
@ -1012,13 +1066,8 @@ makeFloatConst(char *str, int location)
{
A_Const *n = makeNode(A_Const);
#if PG_VERSION_NUM >= PG_VERSION_15
n->val.fval.type = T_Float;
n->val.fval.fval = str;
#else
n->val.type = T_Float;
n->val.val.str = str;
#endif
n->location = location;
return (Node *) n;

125
src/backend/distributed/commands/seclabel.c Normal file
View File

@ -0,0 +1,125 @@
/*-------------------------------------------------------------------------
*
* seclabel.c
*
* This file contains the logic of SECURITY LABEL statement propagation.
*
* Copyright (c) Citus Data, Inc.
*
*-------------------------------------------------------------------------
*/
#include "postgres.h"
#include "distributed/commands.h"
#include "distributed/commands/utility_hook.h"
#include "distributed/coordinator_protocol.h"
#include "distributed/deparser.h"
#include "distributed/log_utils.h"
#include "distributed/metadata/distobject.h"
#include "distributed/metadata_sync.h"
/*
* PostprocessSecLabelStmt prepares the commands that need to be run on all workers to assign
* security labels on distributed objects, currently supporting just Role objects.
* It also ensures that all object dependencies exist on all
* nodes for the object in the SecLabelStmt.
*/
List *
PostprocessSecLabelStmt(Node *node, const char *queryString)
{
if (!ShouldPropagate())
{
return NIL;
}
SecLabelStmt *secLabelStmt = castNode(SecLabelStmt, node);
List *objectAddresses = GetObjectAddressListFromParseTree(node, false, true);
if (!IsAnyObjectDistributed(objectAddresses))
{
return NIL;
}
if (secLabelStmt->objtype != OBJECT_ROLE)
{
/*
* If we are not in the coordinator, we don't want to interrupt the security
* label command with notices, the user expects that from the worker node
* the command will not be propagated
*/
if (EnableUnsupportedFeatureMessages && IsCoordinator())
{
ereport(NOTICE, (errmsg("not propagating SECURITY LABEL commands whose "
"object type is not role"),
errhint("Connect to worker nodes directly to manually "
"run the same SECURITY LABEL command.")));
}
return NIL;
}
if (!EnableCreateRolePropagation)
{
return NIL;
}
EnsureCoordinator();
EnsureAllObjectDependenciesExistOnAllNodes(objectAddresses);
const char *sql = DeparseTreeNode((Node *) secLabelStmt);
List *commandList = list_make3(DISABLE_DDL_PROPAGATION,
(void *) sql,
ENABLE_DDL_PROPAGATION);
return NodeDDLTaskList(NON_COORDINATOR_NODES, commandList);
}
/*
* SecLabelStmtObjectAddress returns the object address of the object on
* which this statement operates (secLabelStmt->object). Note that it has no limitation
* on the object type being OBJECT_ROLE. This is intentionally implemented like this
* since it is fairly simple to implement and we might extend SECURITY LABEL propagation
* in the future to include more object types.
*/
List *
SecLabelStmtObjectAddress(Node *node, bool missing_ok, bool isPostprocess)
{
SecLabelStmt *secLabelStmt = castNode(SecLabelStmt, node);
Relation rel = NULL;
ObjectAddress address = get_object_address(secLabelStmt->objtype,
secLabelStmt->object, &rel,
AccessShareLock, missing_ok);
if (rel != NULL)
{
relation_close(rel, AccessShareLock);
}
ObjectAddress *addressPtr = palloc0(sizeof(ObjectAddress));
*addressPtr = address;
return list_make1(addressPtr);
}
/*
* citus_test_object_relabel is a dummy function for check_object_relabel_type hook.
* It is meant to be used in tests combined with citus_test_register_label_provider
*/
void
citus_test_object_relabel(const ObjectAddress *object, const char *seclabel)
{
if (seclabel == NULL ||
strcmp(seclabel, "citus_unclassified") == 0 ||
strcmp(seclabel, "citus_classified") == 0 ||
strcmp(seclabel, "citus '!unclassified") == 0)
{
return;
}
ereport(ERROR,
(errcode(ERRCODE_INVALID_NAME),
errmsg("'%s' is not a valid security label for Citus tests.", seclabel)));
}

5
src/backend/distributed/commands/sequence.c
View File

@ -735,8 +735,6 @@ PostprocessAlterSequenceOwnerStmt(Node *node, const char *queryString)
}
#if (PG_VERSION_NUM >= PG_VERSION_15)
/*
* PreprocessAlterSequencePersistenceStmt is called for change of persistence
* of sequences before the persistence is changed on the local instance.
@ -847,9 +845,6 @@ PreprocessSequenceAlterTableStmt(Node *node, const char *queryString,
}
#endif
/*
* PreprocessGrantOnSequenceStmt is executed before the statement is applied to the local
* postgres instance.

15
src/backend/distributed/commands/table.c
View File

@ -1153,7 +1153,6 @@ PreprocessAlterTableStmt(Node *node, const char *alterTableCommand,
{
AlterTableStmt *stmtCopy = copyObject(alterTableStatement);
stmtCopy->objtype = OBJECT_SEQUENCE;
#if (PG_VERSION_NUM >= PG_VERSION_15)
/*
* it must be ALTER TABLE .. OWNER TO ..
@ -1163,16 +1162,6 @@ PreprocessAlterTableStmt(Node *node, const char *alterTableCommand,
*/
return PreprocessSequenceAlterTableStmt((Node *) stmtCopy, alterTableCommand,
processUtilityContext);
#else
/*
* it must be ALTER TABLE .. OWNER TO .. command
* since this is the only ALTER command of a sequence that
* passes through an AlterTableStmt
*/
return PreprocessAlterSequenceOwnerStmt((Node *) stmtCopy, alterTableCommand,
processUtilityContext);
#endif
}
else if (relKind == RELKIND_VIEW)
{
@ -3673,9 +3662,8 @@ ErrorIfUnsupportedAlterTableStmt(AlterTableStmt *alterTableStatement)
"are currently unsupported.")));
break;
}
#endif
#if PG_VERSION_NUM >= PG_VERSION_15
case AT_SetAccessMethod:
{
/*
@ -3695,7 +3683,6 @@ ErrorIfUnsupportedAlterTableStmt(AlterTableStmt *alterTableStatement)
break;
}
#endif
case AT_SetNotNull:
case AT_ReplicaIdentity:
case AT_ChangeOwner:

95
src/backend/distributed/commands/trigger.c
View File

@ -57,9 +57,6 @@ static void ExtractDropStmtTriggerAndRelationName(DropStmt *dropTriggerStmt,
static void ErrorIfDropStmtDropsMultipleTriggers(DropStmt *dropTriggerStmt);
static char * GetTriggerNameById(Oid triggerId);
static int16 GetTriggerTypeById(Oid triggerId);
#if (PG_VERSION_NUM < PG_VERSION_15)
static void ErrorOutIfCloneTrigger(Oid tgrelid, const char *tgname);
#endif
/* GUC that overrides trigger checks for distributed tables and reference tables */
@ -404,40 +401,6 @@ CreateTriggerEventExtendNames(CreateTrigStmt *createTriggerStmt, char *schemaNam
}
/*
* PreprocessAlterTriggerRenameStmt is called before a ALTER TRIGGER RENAME
* command has been executed by standard process utility. This function errors
* out if we are trying to rename a child trigger on a partition of a distributed
* table. In PG15, this is not allowed anyway.
*/
List *
PreprocessAlterTriggerRenameStmt(Node *node, const char *queryString,
ProcessUtilityContext processUtilityContext)
{
#if (PG_VERSION_NUM < PG_VERSION_15)
RenameStmt *renameTriggerStmt = castNode(RenameStmt, node);
Assert(renameTriggerStmt->renameType == OBJECT_TRIGGER);
RangeVar *relation = renameTriggerStmt->relation;
bool missingOk = false;
Oid relationId = RangeVarGetRelid(relation, ALTER_TRIGGER_LOCK_MODE, missingOk);
if (!IsCitusTable(relationId))
{
return NIL;
}
EnsureCoordinator();
ErrorOutForTriggerIfNotSupported(relationId);
ErrorOutIfCloneTrigger(relationId, renameTriggerStmt->subname);
#endif
return NIL;
}
/*
* PostprocessAlterTriggerRenameStmt is called after a ALTER TRIGGER RENAME
* command has been executed by standard process utility. This function errors
@ -759,64 +722,6 @@ ErrorIfRelationHasUnsupportedTrigger(Oid relationId)
}
#if (PG_VERSION_NUM < PG_VERSION_15)
/*
* ErrorOutIfCloneTrigger is a helper function to error
* out if we are trying to rename a child trigger on a
* partition of a distributed table.
* A lot of this code is borrowed from PG15 because
* renaming clone triggers isn't allowed in PG15 anymore.
*/
static void
ErrorOutIfCloneTrigger(Oid tgrelid, const char *tgname)
{
HeapTuple tuple;
ScanKeyData key[2];
Relation tgrel = table_open(TriggerRelationId, RowExclusiveLock);
/*
* Search for the trigger to modify.
*/
ScanKeyInit(&key[0],
Anum_pg_trigger_tgrelid,
BTEqualStrategyNumber, F_OIDEQ,
ObjectIdGetDatum(tgrelid));
ScanKeyInit(&key[1],
Anum_pg_trigger_tgname,
BTEqualStrategyNumber, F_NAMEEQ,
CStringGetDatum(tgname));
SysScanDesc tgscan = systable_beginscan(tgrel, TriggerRelidNameIndexId, true,
NULL, 2, key);
if (HeapTupleIsValid(tuple = systable_getnext(tgscan)))
{
Form_pg_trigger trigform = (Form_pg_trigger) GETSTRUCT(tuple);
/*
* If the trigger descends from a trigger on a parent partitioned
* table, reject the rename.
* Appended shard ids to find the trigger on the partition's shards
* are not correct. Hence we would fail to find the trigger on the
* partition's shard.
*/
if (OidIsValid(trigform->tgparentid))
{
ereport(ERROR, (
errmsg(
"cannot rename child triggers on distributed partitions")));
}
}
systable_endscan(tgscan);
table_close(tgrel, RowExclusiveLock);
}
#endif
/*
* GetDropTriggerStmtRelation takes a DropStmt for a trigger object and returns
* RangeVar for the relation that owns the trigger.

4
src/backend/distributed/commands/view.c
View File

@ -392,9 +392,7 @@ CreateViewDDLCommand(Oid viewOid)
static void
AppendQualifiedViewNameToCreateViewCommand(StringInfo buf, Oid viewOid)
{
char *viewName = get_rel_name(viewOid);
char *schemaName = get_namespace_name(get_rel_namespace(viewOid));
char *qualifiedViewName = quote_qualified_identifier(schemaName, viewName);
char *qualifiedViewName = generate_qualified_relation_name(viewOid);
appendStringInfo(buf, "%s ", qualifiedViewName);
}

10
src/backend/distributed/connection/shared_connection_stats.c
View File

@ -614,16 +614,6 @@ WaitForSharedConnection(void)
void
InitializeSharedConnectionStats(void)
{
/* on PG 15, we use shmem_request_hook_type */
#if PG_VERSION_NUM < PG_VERSION_15
/* allocate shared memory */
if (!IsUnderPostmaster)
{
RequestAddinShmemSpace(SharedConnectionStatsShmemSize());
}
#endif
prev_shmem_startup_hook = shmem_startup_hook;
shmem_startup_hook = SharedConnectionStatsShmemInit;
}

4
src/backend/distributed/deparser/citus_ruleutils.c
View File

@ -258,10 +258,8 @@ pg_get_sequencedef_string(Oid sequenceRelationId)
char *typeName = format_type_be(pgSequenceForm->seqtypid);
char *sequenceDef = psprintf(CREATE_SEQUENCE_COMMAND,
#if (PG_VERSION_NUM >= PG_VERSION_15)
get_rel_persistence(sequenceRelationId) ==
RELPERSISTENCE_UNLOGGED ? "UNLOGGED " : "",
#endif
qualifiedSequenceName,
typeName,
pgSequenceForm->seqincrement, pgSequenceForm->seqmin,
@ -857,12 +855,10 @@ deparse_shard_index_statement(IndexStmt *origStmt, Oid distrelid, int64 shardid,
appendStringInfoString(buffer, ") ");
}
#if PG_VERSION_NUM >= PG_VERSION_15
if (indexStmt->nulls_not_distinct)
{
appendStringInfoString(buffer, "NULLS NOT DISTINCT ");
}
#endif /* PG_VERSION_15 */
if (indexStmt->options != NIL)
{

4
src/backend/distributed/deparser/deparse_database_stmts.c
View File

@ -159,7 +159,6 @@ DeparseAlterDatabaseStmt(Node *node)
}
#if PG_VERSION_NUM >= PG_VERSION_15
char *
DeparseAlterDatabaseRefreshCollStmt(Node *node)
{
@ -174,6 +173,3 @@ DeparseAlterDatabaseRefreshCollStmt(Node *node)
return str.data;
}
#endif

136
src/backend/distributed/deparser/deparse_publication_stmts.c
View File

@ -32,7 +32,6 @@
static void AppendCreatePublicationStmt(StringInfo buf, CreatePublicationStmt *stmt,
bool whereClauseNeedsTransform,
bool includeLocalTables);
#if (PG_VERSION_NUM >= PG_VERSION_15)
static bool AppendPublicationObjects(StringInfo buf, List *publicationObjects,
bool whereClauseNeedsTransform,
bool includeLocalTables);
@ -40,10 +39,6 @@ static void AppendWhereClauseExpression(StringInfo buf, RangeVar *tableName,
Node *whereClause,
bool whereClauseNeedsTransform);
static void AppendAlterPublicationAction(StringInfo buf, AlterPublicationAction action);
#else
static bool AppendTables(StringInfo buf, List *tables, bool includeLocalTables);
static void AppendDefElemAction(StringInfo buf, DefElemAction action);
#endif
static bool AppendAlterPublicationStmt(StringInfo buf, AlterPublicationStmt *stmt,
bool whereClauseNeedsTransform,
bool includeLocalTables);
@ -108,7 +103,6 @@ AppendCreatePublicationStmt(StringInfo buf, CreatePublicationStmt *stmt,
{
appendStringInfoString(buf, " FOR ALL TABLES");
}
#if (PG_VERSION_NUM >= PG_VERSION_15)
else if (stmt->pubobjects != NIL)
{
bool hasObjects = false;
@ -146,32 +140,6 @@ AppendCreatePublicationStmt(StringInfo buf, CreatePublicationStmt *stmt,
includeLocalTables);
}
}
#else
else if (stmt->tables != NIL)
{
bool hasTables = false;
RangeVar *rangeVar = NULL;
/*
* Check whether there are tables to propagate, mainly to know whether
* we should include "FOR".
*/
foreach_declared_ptr(rangeVar, stmt->tables)
{
if (includeLocalTables || IsCitusTableRangeVar(rangeVar, NoLock, false))
{
hasTables = true;
break;
}
}
if (hasTables)
{
appendStringInfoString(buf, " FOR");
AppendTables(buf, stmt->tables, includeLocalTables);
}
}
#endif
if (stmt->options != NIL)
{
@ -182,8 +150,6 @@ AppendCreatePublicationStmt(StringInfo buf, CreatePublicationStmt *stmt,
}
#if (PG_VERSION_NUM >= PG_VERSION_15)
/*
* AppendPublicationObjects appends a string representing a list of publication
* objects to a buffer.
@ -320,57 +286,6 @@ AppendWhereClauseExpression(StringInfo buf, RangeVar *tableName,
}
#else
/*
* AppendPublicationObjects appends a string representing a list of publication
* objects to a buffer.
*
* For instance: TABLE users, departments
*/
static bool
AppendTables(StringInfo buf, List *tables, bool includeLocalTables)
{
RangeVar *rangeVar = NULL;
bool appendedObject = false;
foreach_declared_ptr(rangeVar, tables)
{
if (!includeLocalTables &&
!IsCitusTableRangeVar(rangeVar, NoLock, false))
{
/* do not propagate local tables */
continue;
}
char *schemaName = rangeVar->schemaname;
char *tableName = rangeVar->relname;
if (schemaName != NULL)
{
/* qualified table name */
appendStringInfo(buf, "%s %s",
appendedObject ? "," : " TABLE",
quote_qualified_identifier(schemaName, tableName));
}
else
{
/* unqualified table name */
appendStringInfo(buf, "%s %s",
appendedObject ? "," : " TABLE",
quote_identifier(tableName));
}
appendedObject = true;
}
return appendedObject;
}
#endif
/*
* DeparseAlterPublicationSchemaStmt builds and returns a string representing
* an AlterPublicationStmt.
@ -439,19 +354,12 @@ AppendAlterPublicationStmt(StringInfo buf, AlterPublicationStmt *stmt,
return true;
}
#if (PG_VERSION_NUM >= PG_VERSION_15)
AppendAlterPublicationAction(buf, stmt->action);
return AppendPublicationObjects(buf, stmt->pubobjects, whereClauseNeedsTransform,
includeLocalTables);
#else
AppendDefElemAction(buf, stmt->tableAction);
return AppendTables(buf, stmt->tables, includeLocalTables);
#endif
}
#if (PG_VERSION_NUM >= PG_VERSION_15)
/*
* AppendAlterPublicationAction appends a string representing an AlterPublicationAction
* to a buffer.
@ -487,46 +395,6 @@ AppendAlterPublicationAction(StringInfo buf, AlterPublicationAction action)
}
#else
/*
* AppendDefElemAction appends a string representing a DefElemAction
* to a buffer.
*/
static void
AppendDefElemAction(StringInfo buf, DefElemAction action)
{
switch (action)
{
case DEFELEM_ADD:
{
appendStringInfoString(buf, " ADD");
break;
}
case DEFELEM_DROP:
{
appendStringInfoString(buf, " DROP");
break;
}
case DEFELEM_SET:
{
appendStringInfoString(buf, " SET");
break;
}
default:
{
ereport(ERROR, (errmsg("unrecognized publication action: %d", action)));
}
}
}
#endif
/*
* DeparseDropPublicationStmt builds and returns a string representing the DropStmt
*/
@ -651,11 +519,7 @@ AppendPublicationOptions(StringInfo stringBuffer, List *optionList)
appendStringInfo(stringBuffer, "%s = ",
quote_identifier(optionName));
#if (PG_VERSION_NUM >= PG_VERSION_15)
if (valueType == T_Integer || valueType == T_Float || valueType == T_Boolean)
#else
if (valueType == T_Integer || valueType == T_Float)
#endif
{
/* string escaping is unnecessary for numeric types and can cause issues */
appendStringInfo(stringBuffer, "%s", optionValue);

79
src/backend/distributed/deparser/deparse_seclabel_stmts.c Normal file
View File

@ -0,0 +1,79 @@
/*-------------------------------------------------------------------------
*
* deparse_seclabel_stmts.c
* All routines to deparse SECURITY LABEL statements.
*
* Copyright (c), Citus Data, Inc.
*
*-------------------------------------------------------------------------
*/
#include "postgres.h"
#include "nodes/parsenodes.h"
#include "utils/builtins.h"
#include "distributed/deparser.h"
static void AppendSecLabelStmt(StringInfo buf, SecLabelStmt *stmt);
/*
* DeparseSecLabelStmt builds and returns a string representing of the
* SecLabelStmt for application on a remote server.
*/
char *
DeparseSecLabelStmt(Node *node)
{
SecLabelStmt *secLabelStmt = castNode(SecLabelStmt, node);
StringInfoData buf = { 0 };
initStringInfo(&buf);
AppendSecLabelStmt(&buf, secLabelStmt);
return buf.data;
}
/*
* AppendSecLabelStmt generates the string representation of the
* SecLabelStmt and appends it to the buffer.
*/
static void
AppendSecLabelStmt(StringInfo buf, SecLabelStmt *stmt)
{
appendStringInfoString(buf, "SECURITY LABEL ");
if (stmt->provider != NULL)
{
appendStringInfo(buf, "FOR %s ", quote_identifier(stmt->provider));
}
appendStringInfoString(buf, "ON ");
switch (stmt->objtype)
{
case OBJECT_ROLE:
{
appendStringInfo(buf, "ROLE %s ", quote_identifier(strVal(stmt->object)));
break;
}
/* normally, we shouldn't reach this */
default:
{
ereport(ERROR, (errmsg("unsupported security label statement for"
" deparsing")));
}
}
appendStringInfoString(buf, "IS ");
if (stmt->label != NULL)
{
appendStringInfo(buf, "%s", quote_literal_cstr(stmt->label));
}
else
{
appendStringInfoString(buf, "NULL");
}
}

7
src/backend/distributed/deparser/deparse_sequence_stmts.c
View File

@ -28,9 +28,7 @@ static void AppendSequenceNameList(StringInfo buf, List *objects, ObjectType obj
static void AppendRenameSequenceStmt(StringInfo buf, RenameStmt *stmt);
static void AppendAlterSequenceSchemaStmt(StringInfo buf, AlterObjectSchemaStmt *stmt);
static void AppendAlterSequenceOwnerStmt(StringInfo buf, AlterTableStmt *stmt);
#if (PG_VERSION_NUM >= PG_VERSION_15)
static void AppendAlterSequencePersistenceStmt(StringInfo buf, AlterTableStmt *stmt);
#endif
static void AppendGrantOnSequenceStmt(StringInfo buf, GrantStmt *stmt);
static void AppendGrantOnSequenceSequences(StringInfo buf, GrantStmt *stmt);
@ -262,8 +260,6 @@ AppendAlterSequenceOwnerStmt(StringInfo buf, AlterTableStmt *stmt)
}
#if (PG_VERSION_NUM >= PG_VERSION_15)
/*
* DeparseAlterSequencePersistenceStmt builds and returns a string representing
* the AlterTableStmt consisting of changing the persistence of a sequence
@ -349,9 +345,6 @@ AppendAlterSequencePersistenceStmt(StringInfo buf, AlterTableStmt *stmt)
}
#endif
/*
* DeparseGrantOnSequenceStmt builds and returns a string representing the GrantOnSequenceStmt
*/

2
src/backend/distributed/deparser/deparse_table_stmts.c
View File

@ -193,12 +193,10 @@ AppendAlterTableCmdConstraint(StringInfo buf, Constraint *constraint,
{
appendStringInfoString(buf, " UNIQUE");
#if (PG_VERSION_NUM >= PG_VERSION_15)
if (constraint->nulls_not_distinct == true)
{
appendStringInfoString(buf, " NULLS NOT DISTINCT");
}
#endif
}
if (subtype == AT_AddConstraint)

34
src/backend/distributed/deparser/qualify_publication_stmt.c
View File

@ -19,11 +19,7 @@
#include "distributed/deparser.h"
#include "distributed/listutils.h"
#if (PG_VERSION_NUM >= PG_VERSION_15)
static void QualifyPublicationObjects(List *publicationObjects);
#else
static void QualifyTables(List *tables);
#endif
static void QualifyPublicationRangeVar(RangeVar *publication);
@ -36,16 +32,10 @@ QualifyCreatePublicationStmt(Node *node)
{
CreatePublicationStmt *stmt = castNode(CreatePublicationStmt, node);
#if (PG_VERSION_NUM >= PG_VERSION_15)
QualifyPublicationObjects(stmt->pubobjects);
#else
QualifyTables(stmt->tables);
#endif
}
#if (PG_VERSION_NUM >= PG_VERSION_15)
/*
* QualifyPublicationObjects ensures all table names in a list of
* publication objects are fully qualified.
@ -68,26 +58,6 @@ QualifyPublicationObjects(List *publicationObjects)
}
#else
/*
* QualifyTables ensures all table names in a list are fully qualified.
*/
static void
QualifyTables(List *tables)
{
RangeVar *rangeVar = NULL;
foreach_declared_ptr(rangeVar, tables)
{
QualifyPublicationRangeVar(rangeVar);
}
}
#endif
/*
* QualifyPublicationObjects ensures all table names in a list of
* publication objects are fully qualified.
@ -97,11 +67,7 @@ QualifyAlterPublicationStmt(Node *node)
{
AlterPublicationStmt *stmt = castNode(AlterPublicationStmt, node);
#if (PG_VERSION_NUM >= PG_VERSION_15)
QualifyPublicationObjects(stmt->pubobjects);
#else
QualifyTables(stmt->tables);
#endif
}

5
src/backend/distributed/deparser/qualify_sequence_stmt.c
View File

@ -52,8 +52,6 @@ QualifyAlterSequenceOwnerStmt(Node *node)
}
#if (PG_VERSION_NUM >= PG_VERSION_15)
/*
* QualifyAlterSequencePersistenceStmt transforms a
* ALTER SEQUENCE .. SET LOGGED/UNLOGGED
@ -80,9 +78,6 @@ QualifyAlterSequencePersistenceStmt(Node *node)
}
#endif
/*
* QualifyAlterSequenceSchemaStmt transforms a
* ALTER SEQUENCE .. SET SCHEMA ..

8638
src/backend/distributed/deparser/ruleutils_14.c
View File

File diff suppressed because it is too large Load Diff

25
src/backend/distributed/executor/adaptive_executor.c
View File

@ -720,10 +720,8 @@ static void RebuildWaitEventSetForSessions(DistributedExecution *execution);
static void AddLatchWaitEventToExecution(DistributedExecution *execution);
static void ProcessWaitEvents(DistributedExecution *execution, WaitEvent *events, int
eventCount, bool *cancellationReceived);
#if PG_VERSION_NUM >= PG_VERSION_15
static void RemoteSocketClosedForAnySession(DistributedExecution *execution);
static void ProcessWaitEventsForSocketClosed(WaitEvent *events, int eventCount);
#endif
static long MillisecondsBetweenTimestamps(instr_time startTime, instr_time endTime);
static uint64 MicrosecondsBetweenTimestamps(instr_time startTime, instr_time endTime);
static int WorkerPoolCompare(const void *lhsKey, const void *rhsKey);
@ -1769,11 +1767,8 @@ FindOrCreateWorkerSession(WorkerPool *workerPool, MultiConnection *connection)
session->commandsSent = 0;
session->waitEventSetIndex = WAIT_EVENT_SET_INDEX_NOT_INITIALIZED;
#if PG_VERSION_NUM >= PG_VERSION_15
/* always detect closed sockets */
UpdateConnectionWaitFlags(session, WL_SOCKET_CLOSED);
#endif
dlist_init(&session->pendingTaskQueue);
dlist_init(&session->readyTaskQueue);
@ -1817,7 +1812,6 @@ FindOrCreateWorkerSession(WorkerPool *workerPool, MultiConnection *connection)
* the events, even ignores cancellation events. Future callers of this
* function should consider its limitations.
*/
#if PG_VERSION_NUM >= PG_VERSION_15
static void
RemoteSocketClosedForAnySession(DistributedExecution *execution)
{
@ -1835,9 +1829,6 @@ RemoteSocketClosedForAnySession(DistributedExecution *execution)
}
#endif
/*
* SequentialRunDistributedExecution gets a distributed execution and
* executes each individual task in the execution sequentially, one
@ -2173,8 +2164,6 @@ ProcessWaitEvents(DistributedExecution *execution, WaitEvent *events, int eventC
}
#if PG_VERSION_NUM >= PG_VERSION_15
/*
* ProcessWaitEventsForSocketClosed mainly checks for WL_SOCKET_CLOSED event.
* If WL_SOCKET_CLOSED is found, the function sets the underlying connection's
@ -2207,9 +2196,6 @@ ProcessWaitEventsForSocketClosed(WaitEvent *events, int eventCount)
}
#endif
/*
* ManageWorkerPool ensures the worker pool has the appropriate number of connections
* based on the number of pending tasks.
@ -2704,7 +2690,6 @@ OpenNewConnections(WorkerPool *workerPool, int newConnectionCount,
* Instead, we prefer this slight difference, which in effect has almost no
* difference, but doing things in different points in time.
*/
#if PG_VERSION_NUM >= PG_VERSION_15
/* we added new connections, rebuild the waitEventSet */
RebuildWaitEventSetForSessions(execution);
@ -2724,9 +2709,6 @@ OpenNewConnections(WorkerPool *workerPool, int newConnectionCount,
* of the execution.
*/
AddLatchWaitEventToExecution(execution);
#else
execution->rebuildWaitEventSet = true;
#endif
WorkerSession *session = NULL;
foreach_declared_ptr(session, newSessionsList)
@ -3663,13 +3645,8 @@ UpdateConnectionWaitFlags(WorkerSession *session, int waitFlags)
return;
}
#if PG_VERSION_NUM >= PG_VERSION_15
/* always detect closed sockets */
connection->waitFlags = waitFlags | WL_SOCKET_CLOSED;
#else
connection->waitFlags = waitFlags;
#endif
/* without signalling the execution, the flag changes won't be reflected */
execution->waitFlagsChanged = true;
@ -3694,13 +3671,11 @@ CheckConnectionReady(WorkerSession *session)
return false;
}
#if PG_VERSION_NUM >= PG_VERSION_15
if ((session->latestUnconsumedWaitEvents & WL_SOCKET_CLOSED) != 0)
{
connection->connectionState = MULTI_CONNECTION_LOST;
return false;
}
#endif
/* try to send all pending data */
int sendStatus = PQflush(connection->pgConn);

7
src/backend/distributed/executor/insert_select_executor.c
View File

@ -143,15 +143,10 @@ NonPushableInsertSelectExecScan(CustomScanState *node)
targetRelation->partitionColumn);
if (distributionColumnIndex == -1)
{
char *relationName = get_rel_name(targetRelationId);
Oid schemaOid = get_rel_namespace(targetRelationId);
char *schemaName = get_namespace_name(schemaOid);
ereport(ERROR, (errcode(ERRCODE_NULL_VALUE_NOT_ALLOWED),
errmsg(
"the partition column of table %s should have a value",
quote_qualified_identifier(schemaName,
relationName))));
generate_qualified_relation_name(targetRelationId))));
}
TargetEntry *selectPartitionTE = list_nth(selectQuery->targetList,

19
src/backend/distributed/executor/merge_executor.c
View File

@ -219,6 +219,7 @@ ExecuteSourceAtCoordAndRedistribution(CitusScanState *scanState)
copyObject(distributedPlan->selectPlanForModifyViaCoordinatorOrRepartition);
char *intermediateResultIdPrefix = distributedPlan->intermediateResultIdPrefix;
bool hasReturning = distributedPlan->expectResults;
bool hasNotMatchedBySource = HasMergeNotMatchedBySource(mergeQuery);
int partitionColumnIndex = distributedPlan->sourceResultRepartitionColumnIndex;
/*
@ -233,7 +234,7 @@ ExecuteSourceAtCoordAndRedistribution(CitusScanState *scanState)
ereport(DEBUG1, (errmsg("Collect source query results on coordinator")));
List *prunedTaskList = NIL;
List *prunedTaskList = NIL, *emptySourceTaskList = NIL;
HTAB *shardStateHash =
ExecuteMergeSourcePlanIntoColocatedIntermediateResults(
targetRelationId,
@ -255,7 +256,8 @@ ExecuteSourceAtCoordAndRedistribution(CitusScanState *scanState)
* We cannot actually execute MERGE INTO ... tasks that read from
* intermediate results that weren't created because no rows were
* written to them. Prune those tasks out by only including tasks
* on shards with connections.
* on shards with connections; however, if the MERGE INTO includes
* a NOT MATCHED BY SOURCE clause we need to include the task.
*/
Task *task = NULL;
foreach_declared_ptr(task, taskList)
@ -268,6 +270,19 @@ ExecuteSourceAtCoordAndRedistribution(CitusScanState *scanState)
{
prunedTaskList = lappend(prunedTaskList, task);
}
else if (hasNotMatchedBySource)
{
emptySourceTaskList = lappend(emptySourceTaskList, task);
}
}
if (emptySourceTaskList != NIL)
{
ereport(DEBUG1, (errmsg("MERGE has NOT MATCHED BY SOURCE clause, "
"execute MERGE on all shards")));
AdjustTaskQueryForEmptySource(targetRelationId, mergeQuery, emptySourceTaskList,
intermediateResultIdPrefix);
prunedTaskList = list_concat(prunedTaskList, emptySourceTaskList);
}
if (prunedTaskList == NIL)

13
src/backend/distributed/executor/query_stats.c
View File

@ -140,19 +140,6 @@ static void CitusQueryStatsRemoveExpiredEntries(HTAB *existingQueryIdHash);
void
InitializeCitusQueryStats(void)
{
/* on PG 15, we use shmem_request_hook_type */
#if PG_VERSION_NUM < PG_VERSION_15
/* allocate shared memory */
if (!IsUnderPostmaster)
{
RequestAddinShmemSpace(CitusQueryStatsSharedMemSize());
elog(LOG, "requesting named LWLockTranch for %s", STATS_SHARED_MEM_NAME);
RequestNamedLWLockTranche(STATS_SHARED_MEM_NAME, 1);
}
#endif
/* Install hook */
prev_shmem_startup_hook = shmem_startup_hook;
shmem_startup_hook = CitusQueryStatsShmemStartup;

121
src/backend/distributed/executor/repartition_executor.c
View File

@ -17,6 +17,7 @@
#include "nodes/parsenodes.h"
#include "distributed/citus_custom_scan.h"
#include "distributed/deparse_shard_query.h"
#include "distributed/intermediate_results.h"
#include "distributed/listutils.h"
#include "distributed/multi_physical_planner.h"
@ -101,6 +102,40 @@ IsRedistributablePlan(Plan *selectPlan)
}
/*
* HasMergeNotMatchedBySource returns true if the MERGE query has a
* WHEN NOT MATCHED BY SOURCE clause. If it does, we need to execute
* the MERGE query on all shards of the target table, regardless of
* whether or not the source shard has any rows.
*/
bool
HasMergeNotMatchedBySource(Query *query)
{
if (!IsMergeQuery(query))
{
return false;
}
bool haveNotMatchedBySource = false;
#if PG_VERSION_NUM >= PG_VERSION_17
ListCell *lc;
foreach(lc, query->mergeActionList)
{
MergeAction *action = lfirst_node(MergeAction, lc);
if (action->matchKind == MERGE_WHEN_NOT_MATCHED_BY_SOURCE)
{
haveNotMatchedBySource = true;
break;
}
}
#endif
return haveNotMatchedBySource;
}
/*
* GenerateTaskListWithColocatedIntermediateResults generates a list of tasks
* for a query that inserts into a target relation and selects from a set of
@ -200,6 +235,61 @@ GenerateTaskListWithColocatedIntermediateResults(Oid targetRelationId,
}
/*
* AdjustTaskQueryForEmptySource adjusts the query for tasks that read from an
* intermediate result to instead read from an empty relation. This ensures that
* the MERGE query is executed on all shards of the target table, because it has
* a NOT MATCHED BY SOURCE clause, which will be true for all target shards where
* the source shard has no rows.
*/
void
AdjustTaskQueryForEmptySource(Oid targetRelationId,
Query *mergeQuery,
List *tasks,
char *resultIdPrefix)
{
Query *mergeQueryCopy = copyObject(mergeQuery);
RangeTblEntry *selectRte = ExtractSourceResultRangeTableEntry(mergeQueryCopy);
RangeTblEntry *mergeRte = ExtractResultRelationRTE(mergeQueryCopy);
List *targetList = selectRte->subquery->targetList;
ListCell *taskCell = NULL;
foreach(taskCell, tasks)
{
Task *task = lfirst(taskCell);
uint64 shardId = task->anchorShardId;
StringInfo queryString = makeStringInfo();
StringInfo resultId = makeStringInfo();
appendStringInfo(resultId, "%s_" UINT64_FORMAT, resultIdPrefix, shardId);
/* Generate a query for an empty relation */
selectRte->subquery = BuildEmptyResultQuery(targetList, resultId->data);
/* setting an alias simplifies deparsing of RETURNING */
if (mergeRte->alias == NULL)
{
Alias *alias = makeAlias(CITUS_TABLE_ALIAS, NIL);
mergeRte->alias = alias;
}
/*
* Generate a query string for the query that merges into a shard and reads
* from an empty relation.
*
* Since CTEs have already been converted to intermediate results, they need
* to removed from the query. Otherwise, worker queries include both
* intermediate results and CTEs in the query.
*/
mergeQueryCopy->cteList = NIL;
deparse_shard_query(mergeQueryCopy, targetRelationId, shardId, queryString);
ereport(DEBUG2, (errmsg("distributed statement: %s", queryString->data)));
SetTaskQueryString(task, queryString->data);
}
}
/*
* GenerateTaskListWithRedistributedResults returns a task list to insert given
* redistributedResults into the given target relation.
@ -223,6 +313,7 @@ GenerateTaskListWithRedistributedResults(Query *modifyQueryViaCoordinatorOrRepar
Query *modifyResultQuery = copyObject(modifyQueryViaCoordinatorOrRepartition);
RangeTblEntry *insertRte = ExtractResultRelationRTE(modifyResultQuery);
Oid targetRelationId = targetRelation->relationId;
bool hasNotMatchedBySource = HasMergeNotMatchedBySource(modifyResultQuery);
int shardCount = targetRelation->shardIntervalArrayLength;
int shardOffset = 0;
@ -242,19 +333,33 @@ GenerateTaskListWithRedistributedResults(Query *modifyQueryViaCoordinatorOrRepar
StringInfo queryString = makeStringInfo();
/* skip empty tasks */
if (resultIdList == NIL)
if (resultIdList == NIL && !hasNotMatchedBySource)
{
continue;
}
/* sort result ids for consistent test output */
List *sortedResultIds = SortList(resultIdList, pg_qsort_strcmp);
Query *fragmentSetQuery = NULL;
/* generate the query on the intermediate result */
Query *fragmentSetQuery = BuildReadIntermediateResultsArrayQuery(selectTargetList,
NIL,
sortedResultIds,
useBinaryFormat);
if (resultIdList != NIL)
{
/* sort result ids for consistent test output */
List *sortedResultIds = SortList(resultIdList, pg_qsort_strcmp);
/* generate the query on the intermediate result */
fragmentSetQuery = BuildReadIntermediateResultsArrayQuery(selectTargetList,
NIL,
sortedResultIds,
useBinaryFormat);
}
else
{
/* No source data, but MERGE query has NOT MATCHED BY SOURCE */
StringInfo emptyFragmentId = makeStringInfo();
appendStringInfo(emptyFragmentId, "%s_" UINT64_FORMAT, "temp_empty_rel_",
shardId);
fragmentSetQuery = BuildEmptyResultQuery(selectTargetList,
emptyFragmentId->data);
}
/* put the intermediate result query in the INSERT..SELECT */
selectRte->subquery = fragmentSetQuery;

2
src/backend/distributed/executor/tuple_destination.c
View File

@ -109,7 +109,7 @@ TupleStoreTupleDestPutTuple(TupleDestination *self, Task *task,
uint64 tupleSize = tupleLibpqSize;
if (tupleSize == 0)
{
tupleSize = HeapTupleHeaderGetDatumLength(heapTuple);
tupleSize = heapTuple->t_len;
}
/*

2
src/backend/distributed/metadata/dependency.c
View File

@ -1717,13 +1717,11 @@ ExpandCitusSupportedTypes(ObjectAddressCollector *collector, ObjectAddress targe
/*
* As of PostgreSQL 15, the same applies to schemas.
*/
#if PG_VERSION_NUM >= PG_VERSION_15
List *schemaIdList =
GetPublicationSchemas(publicationId);
List *schemaDependencyList =
CreateObjectAddressDependencyDefList(NamespaceRelationId, schemaIdList);
result = list_concat(result, schemaDependencyList);
#endif
break;
}

16
src/backend/distributed/metadata/metadata_cache.c
View File

@ -379,7 +379,7 @@ EnsureModificationsCanRun(void)
/*
* EnsureModificationsCanRunOnRelation firsts calls into EnsureModificationsCanRun() and
* EnsureModificationsCanRunOnRelation first calls into EnsureModificationsCanRun() and
* then does one more additional check. The additional check is to give a proper error
* message if any relation that is modified is replicated, as replicated tables use
* 2PC and 2PC cannot happen when recovery is in progress.
@ -660,6 +660,18 @@ GetTableTypeName(Oid tableId)
bool
IsCitusTable(Oid relationId)
{
/*
* PostgreSQL's OID generator assigns user operation OIDs starting
* from FirstNormalObjectId. This means no user object can have
* an OID lower than FirstNormalObjectId. Therefore, if the
* relationId is less than FirstNormalObjectId
* (i.e. in PostgreSQL's reserved range), we can immediately
* return false, since such objects cannot be Citus tables.
*/
if (relationId < FirstNormalObjectId)
{
return false;
}
return LookupCitusTableCacheEntry(relationId) != NULL;
}
@ -2521,6 +2533,8 @@ AvailableExtensionVersion(void)
ereport(ERROR, (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
errmsg("citus extension is not found")));
return NULL; /* keep compiler happy */
}

43
src/backend/distributed/metadata/metadata_sync.c
View File

@ -1739,48 +1739,6 @@ GetSequencesFromAttrDef(Oid attrdefOid)
}
#if PG_VERSION_NUM < PG_VERSION_15
/*
* Given a pg_attrdef OID, return the relation OID and column number of
* the owning column (represented as an ObjectAddress for convenience).
*
* Returns InvalidObjectAddress if there is no such pg_attrdef entry.
*/
ObjectAddress
GetAttrDefaultColumnAddress(Oid attrdefoid)
{
ObjectAddress result = InvalidObjectAddress;
ScanKeyData skey[1];
HeapTuple tup;
Relation attrdef = table_open(AttrDefaultRelationId, AccessShareLock);
ScanKeyInit(&skey[0],
Anum_pg_attrdef_oid,
BTEqualStrategyNumber, F_OIDEQ,
ObjectIdGetDatum(attrdefoid));
SysScanDesc scan = systable_beginscan(attrdef, AttrDefaultOidIndexId, true,
NULL, 1, skey);
if (HeapTupleIsValid(tup = systable_getnext(scan)))
{
Form_pg_attrdef atdform = (Form_pg_attrdef) GETSTRUCT(tup);
result.classId = RelationRelationId;
result.objectId = atdform->adrelid;
result.objectSubId = atdform->adnum;
}
systable_endscan(scan);
table_close(attrdef, AccessShareLock);
return result;
}
#endif
/*
* GetAttrDefsFromSequence returns a list of attrdef OIDs that have
* a dependency on the given sequence
@ -3011,7 +2969,6 @@ SyncNodeMetadataToNodesMain(Datum main_arg)
PopActiveSnapshot();
CommitTransactionCommand();
ProcessCompletedNotifies();
if (syncedAllNodes)
{

3
src/backend/distributed/metadata/node_metadata.c
View File

@ -217,6 +217,9 @@ citus_set_coordinator_host(PG_FUNCTION_ARGS)
EnsureTransactionalMetadataSyncMode();
}
/* prevent concurrent modification */
LockRelationOid(DistNodeRelationId(), RowExclusiveLock);
bool isCoordinatorInMetadata = false;
WorkerNode *coordinatorNode = PrimaryNodeForGroup(COORDINATOR_GROUP_ID,
&isCoordinatorInMetadata);

4
src/backend/distributed/metadata/pg_get_object_address_13_14_15.c
View File

@ -283,9 +283,7 @@ PgGetObjectAddress(char *ttype, ArrayType *namearr, ArrayType *argsarr)
case OBJECT_FDW:
case OBJECT_FOREIGN_SERVER:
case OBJECT_LANGUAGE:
#if PG_VERSION_NUM >= PG_VERSION_15
case OBJECT_PARAMETER_ACL:
#endif
case OBJECT_PUBLICATION:
case OBJECT_ROLE:
case OBJECT_SCHEMA:
@ -323,9 +321,7 @@ PgGetObjectAddress(char *ttype, ArrayType *namearr, ArrayType *argsarr)
break;
}
#if PG_VERSION_NUM >= PG_VERSION_15
case OBJECT_PUBLICATION_NAMESPACE:
#endif
case OBJECT_USER_MAPPING:
{
objnode = (Node *) list_make2(linitial(name), linitial(args));

3
src/backend/distributed/operations/shard_rebalancer.c
View File

@ -319,7 +319,7 @@ PG_FUNCTION_INFO_V1(citus_rebalance_start);
PG_FUNCTION_INFO_V1(citus_rebalance_stop);
PG_FUNCTION_INFO_V1(citus_rebalance_wait);
bool RunningUnderIsolationTest = false;
bool RunningUnderCitusTestSuite = false;
int MaxRebalancerLoggedIgnoredMoves = 5;
int RebalancerByDiskSizeBaseCost = 100 * 1024 * 1024;
bool PropagateSessionSettingsForLoopbackConnection = false;
@ -384,6 +384,7 @@ CheckRebalanceStateInvariants(const RebalanceState *state)
Assert(shardCost->cost <= prevShardCost->cost);
}
totalCost += shardCost->cost;
prevShardCost = shardCost;
}
/* Check that utilization field is up to date. */

17
src/backend/distributed/operations/shard_transfer.c
View File

@ -294,6 +294,17 @@ citus_move_shard_placement(PG_FUNCTION_ARGS)
CheckCitusVersion(ERROR);
EnsureCoordinator();
List *referenceTableIdList = NIL;
if (HasNodesWithMissingReferenceTables(&referenceTableIdList))
{
ereport(ERROR, (errmsg("there are missing reference tables on some nodes"),
errhint("Copy reference tables first with "
"replicate_reference_tables() or use "
"citus_rebalance_start() that will do it automatically."
)));
}
int64 shardId = PG_GETARG_INT64(0);
char *sourceNodeName = text_to_cstring(PG_GETARG_TEXT_P(1));
int32 sourceNodePort = PG_GETARG_INT32(2);
@ -1940,11 +1951,7 @@ ConstructQualifiedShardName(ShardInterval *shardInterval)
static List *
RecreateTableDDLCommandList(Oid relationId)
{
const char *relationName = get_rel_name(relationId);
Oid relationSchemaId = get_rel_namespace(relationId);
const char *relationSchemaName = get_namespace_name(relationSchemaId);
const char *qualifiedRelationName = quote_qualified_identifier(relationSchemaName,
relationName);
const char *qualifiedRelationName = generate_qualified_relation_name(relationId);
StringInfo dropCommand = makeStringInfo();

3
src/backend/distributed/planner/combine_query_planner.c
View File

@ -136,11 +136,8 @@ CreateCitusCustomScanPath(PlannerInfo *root, RelOptInfo *relOptInfo,
path->custom_path.path.pathtarget = relOptInfo->reltarget;
path->custom_path.path.parent = relOptInfo;
#if (PG_VERSION_NUM >= PG_VERSION_15)
/* necessary to avoid extra Result node in PG15 */
path->custom_path.flags = CUSTOMPATH_SUPPORT_PROJECTION;
#endif
/*
* The 100k rows we put on the cost of the path is kind of arbitrary and could be

54
src/backend/distributed/planner/distributed_planner.c
View File

@ -151,7 +151,10 @@ static RouterPlanType GetRouterPlanType(Query *query,
bool hasUnresolvedParams);
static void ConcatenateRTablesAndPerminfos(PlannedStmt *mainPlan,
PlannedStmt *concatPlan);
static bool CheckPostPlanDistribution(bool isDistributedQuery,
Query *origQuery,
List *rangeTableList,
Query *plannedQuery);
/* Distributed planner hook */
PlannedStmt *
@ -272,6 +275,11 @@ distributed_planner(Query *parse,
planContext.plan = standard_planner(planContext.query, NULL,
planContext.cursorOptions,
planContext.boundParams);
needsDistributedPlanning = CheckPostPlanDistribution(needsDistributedPlanning,
planContext.originalQuery,
rangeTableList,
planContext.query);
if (needsDistributedPlanning)
{
result = PlanDistributedStmt(&planContext, rteIdCounter);
@ -703,6 +711,7 @@ DissuadePlannerFromUsingPlan(PlannedStmt *plan)
* Arbitrarily high cost, but low enough that it can be added up
* without overflowing by choose_custom_plan().
*/
Assert(plan != NULL);
plan->planTree->total_cost = FLT_MAX / 100000000;
}
@ -1441,13 +1450,8 @@ FinalizePlan(PlannedStmt *localPlan, DistributedPlan *distributedPlan)
customScan->custom_private = list_make1(distributedPlanData);
#if (PG_VERSION_NUM >= PG_VERSION_15)
/* necessary to avoid extra Result node in PG15 */
customScan->flags = CUSTOMPATH_SUPPORT_BACKWARD_SCAN | CUSTOMPATH_SUPPORT_PROJECTION;
#else
customScan->flags = CUSTOMPATH_SUPPORT_BACKWARD_SCAN;
#endif
/*
* Fast path queries cannot have any subplans by definition, so skip
@ -2733,3 +2737,41 @@ WarnIfListHasForeignDistributedTable(List *rangeTableList)
}
}
}
static bool
CheckPostPlanDistribution(bool isDistributedQuery,
Query *origQuery, List *rangeTableList,
Query *plannedQuery)
{
if (isDistributedQuery)
{
Node *origQuals = origQuery->jointree->quals;
Node *plannedQuals = plannedQuery->jointree->quals;
#if PG_VERSION_NUM >= PG_VERSION_17
if (IsMergeQuery(origQuery))
{
origQuals = origQuery->mergeJoinCondition;
plannedQuals = plannedQuery->mergeJoinCondition;
}
#endif
/*
* The WHERE quals have been eliminated by the Postgres planner, possibly by
* an OR clause that was simplified to TRUE. In such cases, we need to check
* if the planned query still requires distributed planning.
*/
if (origQuals != NULL && plannedQuals == NULL)
{
List *rtesPostPlan = ExtractRangeTableEntryList(plannedQuery);
if (list_length(rtesPostPlan) < list_length(rangeTableList))
{
isDistributedQuery = ListContainsDistributedTableRTE(
rtesPostPlan, NULL);
}
}
}
return isDistributedQuery;
}

12
src/backend/distributed/planner/function_call_delegation.c
View File

@ -531,8 +531,16 @@ ShardPlacementForFunctionColocatedWithDistTable(DistObjectCacheEntry *procedure,
if (partitionParam->paramkind == PARAM_EXTERN)
{
/* Don't log a message, we should end up here again without a parameter */
DissuadePlannerFromUsingPlan(plan);
/*
* Don't log a message, we should end up here again without a
* parameter.
* Note that "plan" can be null, for example when a CALL statement
* is prepared.
*/
if (plan)
{
DissuadePlannerFromUsingPlan(plan);
}
return NULL;
}
}

2
src/backend/distributed/planner/insert_select_planner.c
View File

@ -1810,6 +1810,8 @@ CastExpr(Expr *expr, Oid sourceType, Oid targetType, Oid targetCollation,
ereport(ERROR, (errmsg("could not find a conversion path from type %d to %d",
sourceType, targetType)));
}
return NULL; /* keep compiler happy */
}

45
src/backend/distributed/planner/merge_planner.c
View File

@ -38,8 +38,6 @@
#include "distributed/shard_pruning.h"
#include "distributed/shared_library_init.h"
#if PG_VERSION_NUM >= PG_VERSION_15
static int SourceResultPartitionColumnIndex(Query *mergeQuery,
List *sourceTargetList,
CitusTableCacheEntry *targetRelation);
@ -100,8 +98,6 @@ static char * MergeCommandResultIdPrefix(uint64 planId);
static void ErrorIfMergeHasReturningList(Query *query);
static Node * GetMergeJoinCondition(Query *mergeQuery);
#endif
/*
* CreateMergePlan
@ -118,13 +114,6 @@ CreateMergePlan(uint64 planId, Query *originalQuery, Query *query,
PlannerRestrictionContext *plannerRestrictionContext,
ParamListInfo boundParams)
{
/* function is void for pre-15 versions of Postgres */
#if PG_VERSION_NUM < PG_VERSION_15
ereport(ERROR, (errmsg("MERGE is not supported in pre-15 Postgres versions")));
#else
Oid targetRelationId = ModifyQueryResultRelationId(originalQuery);
/*
@ -153,8 +142,6 @@ CreateMergePlan(uint64 planId, Query *originalQuery, Query *query,
}
return distributedPlan;
#endif
}
@ -184,9 +171,6 @@ GetMergeJoinTree(Query *mergeQuery)
}
#if PG_VERSION_NUM >= PG_VERSION_15
/*
* GetMergeJoinCondition returns the quals of the ON condition
*/
@ -904,7 +888,7 @@ ConvertRelationRTEIntoSubquery(Query *mergeQuery, RangeTblEntry *sourceRte,
newRangeTableRef->rtindex = SINGLE_RTE_INDEX;
sourceResultsQuery->jointree = makeFromExpr(list_make1(newRangeTableRef), NULL);
sourceResultsQuery->targetList =
CreateAllTargetListForRelation(sourceRte->relid, requiredAttributes);
CreateFilteredTargetListForRelation(sourceRte->relid, requiredAttributes);
List *restrictionList =
GetRestrictInfoListForRelation(sourceRte, plannerRestrictionContext);
List *copyRestrictionList = copyObject(restrictionList);
@ -1443,9 +1427,6 @@ SourceResultPartitionColumnIndex(Query *mergeQuery, List *sourceTargetList,
}
#endif
/*
* ExtractMergeSourceRangeTableEntry returns the range table entry of source
* table or source query in USING clause.
@ -1453,13 +1434,6 @@ SourceResultPartitionColumnIndex(Query *mergeQuery, List *sourceTargetList,
RangeTblEntry *
ExtractMergeSourceRangeTableEntry(Query *query, bool joinSourceOk)
{
/* function is void for pre-15 versions of Postgres */
#if PG_VERSION_NUM < PG_VERSION_15
ereport(ERROR, (errmsg("MERGE is not supported in pre-15 Postgres versions")));
#else
Assert(IsMergeQuery(query));
List *fromList = query->jointree->fromlist;
@ -1498,8 +1472,6 @@ ExtractMergeSourceRangeTableEntry(Query *query, bool joinSourceOk)
RangeTblEntry *subqueryRte = rt_fetch(reference->rtindex, query->rtable);
return subqueryRte;
#endif
}
@ -1516,13 +1488,6 @@ ExtractMergeSourceRangeTableEntry(Query *query, bool joinSourceOk)
Var *
FetchAndValidateInsertVarIfExists(Oid targetRelationId, Query *query)
{
/* function is void for pre-15 versions of Postgres */
#if PG_VERSION_NUM < PG_VERSION_15
ereport(ERROR, (errmsg("MERGE is not supported in pre-15 Postgres versions")));
#else
Assert(IsMergeQuery(query));
if (!IsCitusTableType(targetRelationId, DISTRIBUTED_TABLE))
@ -1546,8 +1511,8 @@ FetchAndValidateInsertVarIfExists(Oid targetRelationId, Query *query)
continue;
}
/* NOT MATCHED can have either INSERT or DO NOTHING */
if (action->commandType == CMD_NOTHING)
/* NOT MATCHED can have either INSERT, DO NOTHING or UPDATE(PG17) */
if (action->commandType == CMD_NOTHING || action->commandType == CMD_UPDATE)
{
return NULL;
}
@ -1593,8 +1558,6 @@ FetchAndValidateInsertVarIfExists(Oid targetRelationId, Query *query)
}
return NULL;
#endif
}
@ -1620,7 +1583,7 @@ IsLocalTableModification(Oid targetRelationId, Query *query, uint64 shardId,
return true;
}
if (shardId == INVALID_SHARD_ID && ContainsOnlyLocalTables(rteProperties))
if (shardId == INVALID_SHARD_ID && ContainsOnlyLocalOrReferenceTables(rteProperties))
{
return true;
}

653
src/backend/distributed/planner/multi_explain.c
View File

@ -95,14 +95,24 @@ typedef struct
bool wal;
bool timing;
bool summary;
#if PG_VERSION_NUM >= PG_VERSION_17
bool memory;
ExplainSerializeOption serialize;
#endif
ExplainFormat format;
} ExplainOptions;
/* EXPLAIN flags of current distributed explain */
#if PG_VERSION_NUM >= PG_VERSION_17
static ExplainOptions CurrentDistributedQueryExplainOptions = {
0, 0, 0, 0, 0, 0, 0, EXPLAIN_SERIALIZE_NONE, EXPLAIN_FORMAT_TEXT
};
#else
static ExplainOptions CurrentDistributedQueryExplainOptions = {
0, 0, 0, 0, 0, 0, EXPLAIN_FORMAT_TEXT
};
#endif
/* Result for a single remote EXPLAIN command */
typedef struct RemoteExplainPlan
@ -124,6 +134,59 @@ typedef struct ExplainAnalyzeDestination
TupleDesc lastSavedExplainAnalyzeTupDesc;
} ExplainAnalyzeDestination;
#if PG_VERSION_NUM >= PG_VERSION_17
/*
* Various places within need to convert bytes to kilobytes. Round these up
* to the next whole kilobyte.
* copied from explain.c
*/
#define BYTES_TO_KILOBYTES(b) (((b) + 1023) / 1024)
/* copied from explain.c */
/* Instrumentation data for SERIALIZE option */
typedef struct SerializeMetrics
{
uint64 bytesSent; /* # of bytes serialized */
instr_time timeSpent; /* time spent serializing */
BufferUsage bufferUsage; /* buffers accessed during serialization */
} SerializeMetrics;
/* copied from explain.c */
static bool peek_buffer_usage(ExplainState *es, const BufferUsage *usage);
static void show_buffer_usage(ExplainState *es, const BufferUsage *usage);
static void show_memory_counters(ExplainState *es,
const MemoryContextCounters *mem_counters);
static void ExplainIndentText(ExplainState *es);
static void ExplainPrintSerialize(ExplainState *es,
SerializeMetrics *metrics);
static SerializeMetrics GetSerializationMetrics(DestReceiver *dest);
/*
* DestReceiver functions for SERIALIZE option
*
* A DestReceiver for query tuples, that serializes passed rows into RowData
* messages while measuring the resources expended and total serialized size,
* while never sending the data to the client. This allows measuring the
* overhead of deTOASTing and datatype out/sendfuncs, which are not otherwise
* exercisable without actually hitting the network.
*
* copied from explain.c
*/
typedef struct SerializeDestReceiver
{
DestReceiver pub;
ExplainState *es; /* this EXPLAIN statement's ExplainState */
int8 format; /* text or binary, like pq wire protocol */
TupleDesc attrinfo; /* the output tuple desc */
int nattrs; /* current number of columns */
FmgrInfo *finfos; /* precomputed call info for output fns */
MemoryContext tmpcontext; /* per-row temporary memory context */
StringInfoData buf; /* buffer to hold the constructed message */
SerializeMetrics metrics; /* collected metrics */
} SerializeDestReceiver;
#endif
/* Explain functions for distributed queries */
static void ExplainSubPlans(DistributedPlan *distributedPlan, ExplainState *es);
@ -144,14 +207,27 @@ static void ExplainTaskPlacement(ShardPlacement *taskPlacement, List *explainOut
ExplainState *es);
static StringInfo BuildRemoteExplainQuery(char *queryString, ExplainState *es);
static const char * ExplainFormatStr(ExplainFormat format);
#if PG_VERSION_NUM >= PG_VERSION_17
static const char * ExplainSerializeStr(ExplainSerializeOption serializeOption);
#endif
static void ExplainWorkerPlan(PlannedStmt *plannedStmt, DestReceiver *dest,
ExplainState *es,
const char *queryString, ParamListInfo params,
QueryEnvironment *queryEnv,
const instr_time *planduration,
#if PG_VERSION_NUM >= PG_VERSION_17
const BufferUsage *bufusage,
const MemoryContextCounters *mem_counters,
#endif
double *executionDurationMillisec);
static ExplainFormat ExtractFieldExplainFormat(Datum jsonbDoc, const char *fieldName,
ExplainFormat defaultValue);
#if PG_VERSION_NUM >= PG_VERSION_17
static ExplainSerializeOption ExtractFieldExplainSerialize(Datum jsonbDoc,
const char *fieldName,
ExplainSerializeOption
defaultValue);
#endif
static TupleDestination * CreateExplainAnlyzeDestination(Task *task,
TupleDestination *taskDest);
static void ExplainAnalyzeDestPutTuple(TupleDestination *self, Task *task,
@ -190,6 +266,14 @@ PG_FUNCTION_INFO_V1(worker_save_query_explain_analyze);
void
CitusExplainScan(CustomScanState *node, List *ancestors, struct ExplainState *es)
{
#if PG_VERSION_NUM >= PG_VERSION_16
if (es->generic)
{
ereport(ERROR, (errmsg(
"EXPLAIN GENERIC_PLAN is currently not supported for Citus tables")));
}
#endif
CitusScanState *scanState = (CitusScanState *) node;
DistributedPlan *distributedPlan = scanState->distributedPlan;
EState *executorState = ScanStateGetExecutorState(scanState);
@ -1017,24 +1101,30 @@ BuildRemoteExplainQuery(char *queryString, ExplainState *es)
{
StringInfo explainQuery = makeStringInfo();
const char *formatStr = ExplainFormatStr(es->format);
#if PG_VERSION_NUM >= PG_VERSION_17
const char *serializeStr = ExplainSerializeStr(es->serialize);
#endif
appendStringInfo(explainQuery,
"EXPLAIN (ANALYZE %s, VERBOSE %s, "
"COSTS %s, BUFFERS %s, WAL %s, "
#if PG_VERSION_NUM >= PG_VERSION_16
"GENERIC_PLAN %s, "
"TIMING %s, SUMMARY %s, "
#if PG_VERSION_NUM >= PG_VERSION_17
"MEMORY %s, SERIALIZE %s, "
#endif
"TIMING %s, SUMMARY %s, FORMAT %s) %s",
"FORMAT %s) %s",
es->analyze ? "TRUE" : "FALSE",
es->verbose ? "TRUE" : "FALSE",
es->costs ? "TRUE" : "FALSE",
es->buffers ? "TRUE" : "FALSE",
es->wal ? "TRUE" : "FALSE",
#if PG_VERSION_NUM >= PG_VERSION_16
es->generic ? "TRUE" : "FALSE",
#endif
es->timing ? "TRUE" : "FALSE",
es->summary ? "TRUE" : "FALSE",
#if PG_VERSION_NUM >= PG_VERSION_17
es->memory ? "TRUE" : "FALSE",
serializeStr,
#endif
formatStr,
queryString);
@ -1073,6 +1163,42 @@ ExplainFormatStr(ExplainFormat format)
}
#if PG_VERSION_NUM >= PG_VERSION_17
/*
* ExplainSerializeStr converts the given explain serialize option to string.
*/
static const char *
ExplainSerializeStr(ExplainSerializeOption serializeOption)
{
switch (serializeOption)
{
case EXPLAIN_SERIALIZE_NONE:
{
return "none";
}
case EXPLAIN_SERIALIZE_TEXT:
{
return "text";
}
case EXPLAIN_SERIALIZE_BINARY:
{
return "binary";
}
default:
{
return "none";
}
}
}
#endif
/*
* worker_last_saved_explain_analyze returns the last saved EXPLAIN ANALYZE output of
* a worker task query. It returns NULL if nothing has been saved yet.
@ -1132,6 +1258,11 @@ worker_save_query_explain_analyze(PG_FUNCTION_ARGS)
es->verbose = ExtractFieldBoolean(explainOptions, "verbose", es->verbose);
es->timing = ExtractFieldBoolean(explainOptions, "timing", es->timing);
es->format = ExtractFieldExplainFormat(explainOptions, "format", es->format);
#if PG_VERSION_NUM >= PG_VERSION_17
es->memory = ExtractFieldBoolean(explainOptions, "memory", es->memory);
es->serialize = ExtractFieldExplainSerialize(explainOptions, "serialize",
es->serialize);
#endif
TupleDesc tupleDescriptor = NULL;
Tuplestorestate *tupleStore = SetupTuplestore(fcinfo, &tupleDescriptor);
@ -1158,8 +1289,8 @@ worker_save_query_explain_analyze(PG_FUNCTION_ARGS)
}
/* resolve OIDs of unknown (user-defined) types */
Query *analyzedQuery = parse_analyze_varparams_compat(parseTree, queryString,
&paramTypes, &numParams, NULL);
Query *analyzedQuery = parse_analyze_varparams(parseTree, queryString,
&paramTypes, &numParams, NULL);
/* pg_rewrite_query is a wrapper around QueryRewrite with some debugging logic */
List *queryList = pg_rewrite_query(analyzedQuery);
@ -1177,6 +1308,36 @@ worker_save_query_explain_analyze(PG_FUNCTION_ARGS)
/* plan query and record planning stats */
instr_time planStart;
instr_time planDuration;
#if PG_VERSION_NUM >= PG_VERSION_17
BufferUsage bufusage_start,
bufusage;
MemoryContextCounters mem_counters;
MemoryContext planner_ctx = NULL;
MemoryContext saved_ctx = NULL;
if (es->memory)
{
/*
* Create a new memory context to measure planner's memory consumption
* accurately. Note that if the planner were to be modified to use a
* different memory context type, here we would be changing that to
* AllocSet, which might be undesirable. However, we don't have a way
* to create a context of the same type as another, so we pray and
* hope that this is OK.
*
* copied from explain.c
*/
planner_ctx = AllocSetContextCreate(CurrentMemoryContext,
"explain analyze planner context",
ALLOCSET_DEFAULT_SIZES);
saved_ctx = MemoryContextSwitchTo(planner_ctx);
}
if (es->buffers)
{
bufusage_start = pgBufferUsage;
}
#endif
INSTR_TIME_SET_CURRENT(planStart);
@ -1185,9 +1346,32 @@ worker_save_query_explain_analyze(PG_FUNCTION_ARGS)
INSTR_TIME_SET_CURRENT(planDuration);
INSTR_TIME_SUBTRACT(planDuration, planStart);
#if PG_VERSION_NUM >= PG_VERSION_17
if (es->memory)
{
MemoryContextSwitchTo(saved_ctx);
MemoryContextMemConsumed(planner_ctx, &mem_counters);
}
/* calc differences of buffer counters. */
if (es->buffers)
{
memset(&bufusage, 0, sizeof(BufferUsage));
BufferUsageAccumDiff(&bufusage, &pgBufferUsage, &bufusage_start);
}
/* do the actual EXPLAIN ANALYZE */
ExplainWorkerPlan(plan, tupleStoreDest, es, queryString, boundParams, NULL,
&planDuration,
(es->buffers ? &bufusage : NULL),
(es->memory ? &mem_counters : NULL),
&executionDurationMillisec);
#else
/* do the actual EXPLAIN ANALYZE */
ExplainWorkerPlan(plan, tupleStoreDest, es, queryString, boundParams, NULL,
&planDuration, &executionDurationMillisec);
#endif
ExplainEndOutput(es);
@ -1256,6 +1440,50 @@ ExtractFieldExplainFormat(Datum jsonbDoc, const char *fieldName, ExplainFormat
}
#if PG_VERSION_NUM >= PG_VERSION_17
/*
* ExtractFieldExplainSerialize gets value of fieldName from jsonbDoc, or returns
* defaultValue if it doesn't exist.
*/
static ExplainSerializeOption
ExtractFieldExplainSerialize(Datum jsonbDoc, const char *fieldName, ExplainSerializeOption
defaultValue)
{
Datum jsonbDatum = 0;
bool found = ExtractFieldJsonbDatum(jsonbDoc, fieldName, &jsonbDatum);
if (!found)
{
return defaultValue;
}
const char *serializeStr = DatumGetCString(DirectFunctionCall1(jsonb_out,
jsonbDatum));
if (pg_strcasecmp(serializeStr, "\"none\"") == 0)
{
return EXPLAIN_SERIALIZE_NONE;
}
else if (pg_strcasecmp(serializeStr, "\"off\"") == 0)
{
return EXPLAIN_SERIALIZE_NONE;
}
else if (pg_strcasecmp(serializeStr, "\"text\"") == 0)
{
return EXPLAIN_SERIALIZE_TEXT;
}
else if (pg_strcasecmp(serializeStr, "\"binary\"") == 0)
{
return EXPLAIN_SERIALIZE_BINARY;
}
ereport(ERROR, (errmsg("Invalid explain analyze serialize: %s", serializeStr)));
return 0;
}
#endif
/*
* CitusExplainOneQuery is the executor hook that is called when
* postgres wants to explain a query.
@ -1273,6 +1501,10 @@ CitusExplainOneQuery(Query *query, int cursorOptions, IntoClause *into,
CurrentDistributedQueryExplainOptions.summary = es->summary;
CurrentDistributedQueryExplainOptions.timing = es->timing;
CurrentDistributedQueryExplainOptions.format = es->format;
#if PG_VERSION_NUM >= PG_VERSION_17
CurrentDistributedQueryExplainOptions.memory = es->memory;
CurrentDistributedQueryExplainOptions.serialize = es->serialize;
#endif
/* rest is copied from ExplainOneQuery() */
instr_time planstart,
@ -1595,11 +1827,18 @@ WrapQueryForExplainAnalyze(const char *queryString, TupleDesc tupleDesc,
StringInfo explainOptions = makeStringInfo();
appendStringInfo(explainOptions,
"{\"verbose\": %s, \"costs\": %s, \"buffers\": %s, \"wal\": %s, "
#if PG_VERSION_NUM >= PG_VERSION_17
"\"memory\": %s, \"serialize\": \"%s\", "
#endif
"\"timing\": %s, \"summary\": %s, \"format\": \"%s\"}",
CurrentDistributedQueryExplainOptions.verbose ? "true" : "false",
CurrentDistributedQueryExplainOptions.costs ? "true" : "false",
CurrentDistributedQueryExplainOptions.buffers ? "true" : "false",
CurrentDistributedQueryExplainOptions.wal ? "true" : "false",
#if PG_VERSION_NUM >= PG_VERSION_17
CurrentDistributedQueryExplainOptions.memory ? "true" : "false",
ExplainSerializeStr(CurrentDistributedQueryExplainOptions.serialize),
#endif
CurrentDistributedQueryExplainOptions.timing ? "true" : "false",
CurrentDistributedQueryExplainOptions.summary ? "true" : "false",
ExplainFormatStr(CurrentDistributedQueryExplainOptions.format));
@ -1824,7 +2063,12 @@ ExplainOneQuery(Query *query, int cursorOptions,
static void
ExplainWorkerPlan(PlannedStmt *plannedstmt, DestReceiver *dest, ExplainState *es,
const char *queryString, ParamListInfo params, QueryEnvironment *queryEnv,
const instr_time *planduration, double *executionDurationMillisec)
const instr_time *planduration,
#if PG_VERSION_NUM >= PG_VERSION_17
const BufferUsage *bufusage,
const MemoryContextCounters *mem_counters,
#endif
double *executionDurationMillisec)
{
QueryDesc *queryDesc;
instr_time starttime;
@ -1893,6 +2137,32 @@ ExplainWorkerPlan(PlannedStmt *plannedstmt, DestReceiver *dest, ExplainState *es
/* Create textual dump of plan tree */
ExplainPrintPlan(es, queryDesc);
#if PG_VERSION_NUM >= PG_VERSION_17
/* Show buffer and/or memory usage in planning */
if (peek_buffer_usage(es, bufusage) || mem_counters)
{
ExplainOpenGroup("Planning", "Planning", true, es);
if (es->format == EXPLAIN_FORMAT_TEXT)
{
ExplainIndentText(es);
appendStringInfoString(es->str, "Planning:\n");
es->indent++;
}
if (bufusage)
show_buffer_usage(es, bufusage);
if (mem_counters)
show_memory_counters(es, mem_counters);
if (es->format == EXPLAIN_FORMAT_TEXT)
es->indent--;
ExplainCloseGroup("Planning", "Planning", true, es);
}
#endif
if (es->summary && planduration)
{
double plantime = INSTR_TIME_GET_DOUBLE(*planduration);
@ -1913,6 +2183,23 @@ ExplainWorkerPlan(PlannedStmt *plannedstmt, DestReceiver *dest, ExplainState *es
if (es->costs)
ExplainPrintJITSummary(es, queryDesc);
#if PG_VERSION_NUM >= PG_VERSION_17
if (es->serialize != EXPLAIN_SERIALIZE_NONE)
{
/* the SERIALIZE option requires its own tuple receiver */
DestReceiver *dest_serialize = CreateExplainSerializeDestReceiver(es);
/* grab serialization metrics before we destroy the DestReceiver */
SerializeMetrics serializeMetrics = GetSerializationMetrics(dest_serialize);
/* call the DestReceiver's destroy method even during explain */
dest_serialize->rDestroy(dest_serialize);
/* Print info about serialization of output */
ExplainPrintSerialize(es, &serializeMetrics);
}
#endif
/*
* Close down the query and free resources. Include time for this in the
* total execution time (although it should be pretty minimal).
@ -1961,3 +2248,351 @@ elapsed_time(instr_time *starttime)
INSTR_TIME_SUBTRACT(endtime, *starttime);
return INSTR_TIME_GET_DOUBLE(endtime);
}
#if PG_VERSION_NUM >= PG_VERSION_17
/*
* Return whether show_buffer_usage would have anything to print, if given
* the same 'usage' data. Note that when the format is anything other than
* text, we print even if the counters are all zeroes.
*
* Copied from explain.c.
*/
static bool
peek_buffer_usage(ExplainState *es, const BufferUsage *usage)
{
bool has_shared;
bool has_local;
bool has_temp;
bool has_shared_timing;
bool has_local_timing;
bool has_temp_timing;
if (usage == NULL)
return false;
if (es->format != EXPLAIN_FORMAT_TEXT)
return true;
has_shared = (usage->shared_blks_hit > 0 ||
usage->shared_blks_read > 0 ||
usage->shared_blks_dirtied > 0 ||
usage->shared_blks_written > 0);
has_local = (usage->local_blks_hit > 0 ||
usage->local_blks_read > 0 ||
usage->local_blks_dirtied > 0 ||
usage->local_blks_written > 0);
has_temp = (usage->temp_blks_read > 0 ||
usage->temp_blks_written > 0);
has_shared_timing = (!INSTR_TIME_IS_ZERO(usage->shared_blk_read_time) ||
!INSTR_TIME_IS_ZERO(usage->shared_blk_write_time));
has_local_timing = (!INSTR_TIME_IS_ZERO(usage->local_blk_read_time) ||
!INSTR_TIME_IS_ZERO(usage->local_blk_write_time));
has_temp_timing = (!INSTR_TIME_IS_ZERO(usage->temp_blk_read_time) ||
!INSTR_TIME_IS_ZERO(usage->temp_blk_write_time));
return has_shared || has_local || has_temp || has_shared_timing ||
has_local_timing || has_temp_timing;
}
/*
* Show buffer usage details. This better be sync with peek_buffer_usage.
*
* Copied from explain.c.
*/
static void
show_buffer_usage(ExplainState *es, const BufferUsage *usage)
{
if (es->format == EXPLAIN_FORMAT_TEXT)
{
bool has_shared = (usage->shared_blks_hit > 0 ||
usage->shared_blks_read > 0 ||
usage->shared_blks_dirtied > 0 ||
usage->shared_blks_written > 0);
bool has_local = (usage->local_blks_hit > 0 ||
usage->local_blks_read > 0 ||
usage->local_blks_dirtied > 0 ||
usage->local_blks_written > 0);
bool has_temp = (usage->temp_blks_read > 0 ||
usage->temp_blks_written > 0);
bool has_shared_timing = (!INSTR_TIME_IS_ZERO(usage->shared_blk_read_time) ||
!INSTR_TIME_IS_ZERO(usage->shared_blk_write_time));
bool has_local_timing = (!INSTR_TIME_IS_ZERO(usage->local_blk_read_time) ||
!INSTR_TIME_IS_ZERO(usage->local_blk_write_time));
bool has_temp_timing = (!INSTR_TIME_IS_ZERO(usage->temp_blk_read_time) ||
!INSTR_TIME_IS_ZERO(usage->temp_blk_write_time));
/* Show only positive counter values. */
if (has_shared || has_local || has_temp)
{
ExplainIndentText(es);
appendStringInfoString(es->str, "Buffers:");
if (has_shared)
{
appendStringInfoString(es->str, " shared");
if (usage->shared_blks_hit > 0)
appendStringInfo(es->str, " hit=%lld",
(long long) usage->shared_blks_hit);
if (usage->shared_blks_read > 0)
appendStringInfo(es->str, " read=%lld",
(long long) usage->shared_blks_read);
if (usage->shared_blks_dirtied > 0)
appendStringInfo(es->str, " dirtied=%lld",
(long long) usage->shared_blks_dirtied);
if (usage->shared_blks_written > 0)
appendStringInfo(es->str, " written=%lld",
(long long) usage->shared_blks_written);
if (has_local || has_temp)
appendStringInfoChar(es->str, ',');
}
if (has_local)
{
appendStringInfoString(es->str, " local");
if (usage->local_blks_hit > 0)
appendStringInfo(es->str, " hit=%lld",
(long long) usage->local_blks_hit);
if (usage->local_blks_read > 0)
appendStringInfo(es->str, " read=%lld",
(long long) usage->local_blks_read);
if (usage->local_blks_dirtied > 0)
appendStringInfo(es->str, " dirtied=%lld",
(long long) usage->local_blks_dirtied);
if (usage->local_blks_written > 0)
appendStringInfo(es->str, " written=%lld",
(long long) usage->local_blks_written);
if (has_temp)
appendStringInfoChar(es->str, ',');
}
if (has_temp)
{
appendStringInfoString(es->str, " temp");
if (usage->temp_blks_read > 0)
appendStringInfo(es->str, " read=%lld",
(long long) usage->temp_blks_read);
if (usage->temp_blks_written > 0)
appendStringInfo(es->str, " written=%lld",
(long long) usage->temp_blks_written);
}
appendStringInfoChar(es->str, '\n');
}
/* As above, show only positive counter values. */
if (has_shared_timing || has_local_timing || has_temp_timing)
{
ExplainIndentText(es);
appendStringInfoString(es->str, "I/O Timings:");
if (has_shared_timing)
{
appendStringInfoString(es->str, " shared");
if (!INSTR_TIME_IS_ZERO(usage->shared_blk_read_time))
appendStringInfo(es->str, " read=%0.3f",
INSTR_TIME_GET_MILLISEC(usage->shared_blk_read_time));
if (!INSTR_TIME_IS_ZERO(usage->shared_blk_write_time))
appendStringInfo(es->str, " write=%0.3f",
INSTR_TIME_GET_MILLISEC(usage->shared_blk_write_time));
if (has_local_timing || has_temp_timing)
appendStringInfoChar(es->str, ',');
}
if (has_local_timing)
{
appendStringInfoString(es->str, " local");
if (!INSTR_TIME_IS_ZERO(usage->local_blk_read_time))
appendStringInfo(es->str, " read=%0.3f",
INSTR_TIME_GET_MILLISEC(usage->local_blk_read_time));
if (!INSTR_TIME_IS_ZERO(usage->local_blk_write_time))
appendStringInfo(es->str, " write=%0.3f",
INSTR_TIME_GET_MILLISEC(usage->local_blk_write_time));
if (has_temp_timing)
appendStringInfoChar(es->str, ',');
}
if (has_temp_timing)
{
appendStringInfoString(es->str, " temp");
if (!INSTR_TIME_IS_ZERO(usage->temp_blk_read_time))
appendStringInfo(es->str, " read=%0.3f",
INSTR_TIME_GET_MILLISEC(usage->temp_blk_read_time));
if (!INSTR_TIME_IS_ZERO(usage->temp_blk_write_time))
appendStringInfo(es->str, " write=%0.3f",
INSTR_TIME_GET_MILLISEC(usage->temp_blk_write_time));
}
appendStringInfoChar(es->str, '\n');
}
}
else
{
ExplainPropertyInteger("Shared Hit Blocks", NULL,
usage->shared_blks_hit, es);
ExplainPropertyInteger("Shared Read Blocks", NULL,
usage->shared_blks_read, es);
ExplainPropertyInteger("Shared Dirtied Blocks", NULL,
usage->shared_blks_dirtied, es);
ExplainPropertyInteger("Shared Written Blocks", NULL,
usage->shared_blks_written, es);
ExplainPropertyInteger("Local Hit Blocks", NULL,
usage->local_blks_hit, es);
ExplainPropertyInteger("Local Read Blocks", NULL,
usage->local_blks_read, es);
ExplainPropertyInteger("Local Dirtied Blocks", NULL,
usage->local_blks_dirtied, es);
ExplainPropertyInteger("Local Written Blocks", NULL,
usage->local_blks_written, es);
ExplainPropertyInteger("Temp Read Blocks", NULL,
usage->temp_blks_read, es);
ExplainPropertyInteger("Temp Written Blocks", NULL,
usage->temp_blks_written, es);
if (track_io_timing)
{
ExplainPropertyFloat("Shared I/O Read Time", "ms",
INSTR_TIME_GET_MILLISEC(usage->shared_blk_read_time),
3, es);
ExplainPropertyFloat("Shared I/O Write Time", "ms",
INSTR_TIME_GET_MILLISEC(usage->shared_blk_write_time),
3, es);
ExplainPropertyFloat("Local I/O Read Time", "ms",
INSTR_TIME_GET_MILLISEC(usage->local_blk_read_time),
3, es);
ExplainPropertyFloat("Local I/O Write Time", "ms",
INSTR_TIME_GET_MILLISEC(usage->local_blk_write_time),
3, es);
ExplainPropertyFloat("Temp I/O Read Time", "ms",
INSTR_TIME_GET_MILLISEC(usage->temp_blk_read_time),
3, es);
ExplainPropertyFloat("Temp I/O Write Time", "ms",
INSTR_TIME_GET_MILLISEC(usage->temp_blk_write_time),
3, es);
}
}
}
/*
* Indent a text-format line.
*
* We indent by two spaces per indentation level. However, when emitting
* data for a parallel worker there might already be data on the current line
* (cf. ExplainOpenWorker); in that case, don't indent any more.
*
* Copied from explain.c.
*/
static void
ExplainIndentText(ExplainState *es)
{
Assert(es->format == EXPLAIN_FORMAT_TEXT);
if (es->str->len == 0 || es->str->data[es->str->len - 1] == '\n')
appendStringInfoSpaces(es->str, es->indent * 2);
}
/*
* Show memory usage details.
*
* Copied from explain.c.
*/
static void
show_memory_counters(ExplainState *es, const MemoryContextCounters *mem_counters)
{
int64 memUsedkB = BYTES_TO_KILOBYTES(mem_counters->totalspace -
mem_counters->freespace);
int64 memAllocatedkB = BYTES_TO_KILOBYTES(mem_counters->totalspace);
if (es->format == EXPLAIN_FORMAT_TEXT)
{
ExplainIndentText(es);
appendStringInfo(es->str,
"Memory: used=" INT64_FORMAT "kB allocated=" INT64_FORMAT "kB",
memUsedkB, memAllocatedkB);
appendStringInfoChar(es->str, '\n');
}
else
{
ExplainPropertyInteger("Memory Used", "kB", memUsedkB, es);
ExplainPropertyInteger("Memory Allocated", "kB", memAllocatedkB, es);
}
}
/*
* ExplainPrintSerialize -
* Append information about query output volume to es->str.
*
* Copied from explain.c.
*/
static void
ExplainPrintSerialize(ExplainState *es, SerializeMetrics *metrics)
{
const char *format;
/* We shouldn't get called for EXPLAIN_SERIALIZE_NONE */
if (es->serialize == EXPLAIN_SERIALIZE_TEXT)
format = "text";
else
{
Assert(es->serialize == EXPLAIN_SERIALIZE_BINARY);
format = "binary";
}
ExplainOpenGroup("Serialization", "Serialization", true, es);
if (es->format == EXPLAIN_FORMAT_TEXT)
{
ExplainIndentText(es);
if (es->timing)
appendStringInfo(es->str, "Serialization: time=%.3f ms output=" UINT64_FORMAT "kB format=%s\n",
1000.0 * INSTR_TIME_GET_DOUBLE(metrics->timeSpent),
BYTES_TO_KILOBYTES(metrics->bytesSent),
format);
else
appendStringInfo(es->str, "Serialization: output=" UINT64_FORMAT "kB format=%s\n",
BYTES_TO_KILOBYTES(metrics->bytesSent),
format);
if (es->buffers && peek_buffer_usage(es, &metrics->bufferUsage))
{
es->indent++;
show_buffer_usage(es, &metrics->bufferUsage);
es->indent--;
}
}
else
{
if (es->timing)
ExplainPropertyFloat("Time", "ms",
1000.0 * INSTR_TIME_GET_DOUBLE(metrics->timeSpent),
3, es);
ExplainPropertyUInteger("Output Volume", "kB",
BYTES_TO_KILOBYTES(metrics->bytesSent), es);
ExplainPropertyText("Format", format, es);
if (es->buffers)
show_buffer_usage(es, &metrics->bufferUsage);
}
ExplainCloseGroup("Serialization", "Serialization", true, es);
}
/*
* GetSerializationMetrics - collect metrics
*
* We have to be careful here since the receiver could be an IntoRel
* receiver if the subject statement is CREATE TABLE AS. In that
* case, return all-zeroes stats.
*
* Copied from explain.c.
*/
static SerializeMetrics
GetSerializationMetrics(DestReceiver *dest)
{
SerializeMetrics empty;
if (dest->mydest == DestExplainSerialize)
return ((SerializeDestReceiver *) dest)->metrics;
memset(&empty, 0, sizeof(SerializeMetrics));
INSTR_TIME_SET_ZERO(empty.timeSpent);
return empty;
}
#endif

36
src/backend/distributed/planner/multi_logical_optimizer.c
View File

@ -1557,9 +1557,10 @@ MasterAggregateMutator(Node *originalNode, MasterAggregateWalkerContext *walkerC
}
else if (IsA(originalNode, Var))
{
Var *newColumn = copyObject((Var *) originalNode);
newColumn->varno = masterTableId;
newColumn->varattno = walkerContext->columnId;
Var *origColumn = (Var *) originalNode;
Var *newColumn = makeVar(masterTableId, walkerContext->columnId,
origColumn->vartype, origColumn->vartypmod,
origColumn->varcollid, origColumn->varlevelsup);
walkerContext->columnId++;
newNode = (Node *) newColumn;
@ -4753,22 +4754,35 @@ WorkerLimitCount(Node *limitCount, Node *limitOffset, OrderByLimitReference
if (workerLimitNode != NULL && limitOffset != NULL)
{
Const *workerLimitConst = (Const *) workerLimitNode;
Const *workerOffsetConst = (Const *) limitOffset;
int64 workerLimitCount = DatumGetInt64(workerLimitConst->constvalue);
int64 workerOffsetCount = DatumGetInt64(workerOffsetConst->constvalue);
workerLimitCount = workerLimitCount + workerOffsetCount;
workerLimitNode = (Node *) MakeIntegerConstInt64(workerLimitCount);
/* Only update the worker limit if the const is not null.*/
if (!workerLimitConst->constisnull)
{
Const *workerOffsetConst = (Const *) limitOffset;
int64 workerLimitCount = DatumGetInt64(workerLimitConst->constvalue);
/* If the offset is null, it defaults to 0 when cast to int64. */
int64 workerOffsetCount = DatumGetInt64(workerOffsetConst->constvalue);
workerLimitCount = workerLimitCount + workerOffsetCount;
workerLimitNode = (Node *) MakeIntegerConstInt64(workerLimitCount);
}
}
/* display debug message on limit push down */
if (workerLimitNode != NULL)
{
Const *workerLimitConst = (Const *) workerLimitNode;
int64 workerLimitCount = DatumGetInt64(workerLimitConst->constvalue);
if (!workerLimitConst->constisnull)
{
int64 workerLimitCount = DatumGetInt64(workerLimitConst->constvalue);
ereport(DEBUG1, (errmsg("push down of limit count: " INT64_FORMAT,
workerLimitCount)));
ereport(DEBUG1, (errmsg("push down of limit count: " INT64_FORMAT,
workerLimitCount)));
}
else
{
ereport(DEBUG1, (errmsg("push down of limit count: ALL")));
}
}
return workerLimitNode;

3
src/backend/distributed/planner/multi_logical_planner.c
View File

@ -1170,7 +1170,8 @@ HasComplexRangeTableType(Query *queryTree)
if (rangeTableEntry->rtekind != RTE_RELATION &&
rangeTableEntry->rtekind != RTE_SUBQUERY &&
rangeTableEntry->rtekind != RTE_FUNCTION &&
rangeTableEntry->rtekind != RTE_VALUES)
rangeTableEntry->rtekind != RTE_VALUES &&
!IsJsonTableRTE(rangeTableEntry))
{
hasComplexRangeTableType = true;
}

10
src/backend/distributed/planner/multi_router_planner.c
View File

@ -2556,13 +2556,15 @@ AllShardsColocated(List *relationShardList)
/*
* ContainsOnlyLocalTables returns true if there is only
* local tables and not any distributed or reference table.
* ContainsOnlyLocalOrReferenceTables returns true if there are no distributed
* tables in the query. In other words, the query might reference only local
* tables and/or reference tables, but no fully distributed tables.
*/
bool
ContainsOnlyLocalTables(RTEListProperties *rteProperties)
ContainsOnlyLocalOrReferenceTables(RTEListProperties *rteProperties)
{
return !rteProperties->hasDistributedTable && !rteProperties->hasReferenceTable;
/* If hasDistributedTable is false, then all tables are either local or reference. */
return !rteProperties->hasDistributedTable;
}

4
src/backend/distributed/planner/query_colocation_checker.c
View File

@ -45,8 +45,6 @@
static RangeTblEntry * AnchorRte(Query *subquery);
static List * UnionRelationRestrictionLists(List *firstRelationList,
List *secondRelationList);
static List * CreateFilteredTargetListForRelation(Oid relationId,
List *requiredAttributes);
static List * CreateDummyTargetList(Oid relationId, List *requiredAttributes);
static TargetEntry * CreateTargetEntryForColumn(Form_pg_attribute attributeTuple, Index
rteIndex,
@ -378,7 +376,7 @@ CreateAllTargetListForRelation(Oid relationId, List *requiredAttributes)
* only the required columns of the given relation. If there is not required
* columns then a dummy NULL column is put as the only entry.
*/
static List *
List *
CreateFilteredTargetListForRelation(Oid relationId, List *requiredAttributes)
{
Relation relation = relation_open(relationId, AccessShareLock);

66
src/backend/distributed/planner/query_pushdown_planning.c
View File

@ -61,7 +61,8 @@ typedef enum RecurringTuplesType
RECURRING_TUPLES_FUNCTION,
RECURRING_TUPLES_EMPTY_JOIN_TREE,
RECURRING_TUPLES_RESULT_FUNCTION,
RECURRING_TUPLES_VALUES
RECURRING_TUPLES_VALUES,
RECURRING_TUPLES_JSON_TABLE
} RecurringTuplesType;
/*
@ -347,7 +348,8 @@ IsFunctionOrValuesRTE(Node *node)
RangeTblEntry *rangeTblEntry = (RangeTblEntry *) node;
if (rangeTblEntry->rtekind == RTE_FUNCTION ||
rangeTblEntry->rtekind == RTE_VALUES)
rangeTblEntry->rtekind == RTE_VALUES ||
IsJsonTableRTE(rangeTblEntry))
{
return true;
}
@ -700,6 +702,13 @@ DeferErrorIfFromClauseRecurs(Query *queryTree)
"the FROM clause contains VALUES", NULL,
NULL);
}
else if (recurType == RECURRING_TUPLES_JSON_TABLE)
{
return DeferredError(ERRCODE_FEATURE_NOT_SUPPORTED,
"correlated subqueries are not supported when "
"the FROM clause contains JSON_TABLE", NULL,
NULL);
}
/*
@ -1204,7 +1213,8 @@ DeferErrorIfUnsupportedTableCombination(Query *queryTree)
*/
if (rangeTableEntry->rtekind == RTE_RELATION ||
rangeTableEntry->rtekind == RTE_SUBQUERY ||
rangeTableEntry->rtekind == RTE_RESULT)
rangeTableEntry->rtekind == RTE_RESULT ||
IsJsonTableRTE(rangeTableEntry))
{
/* accepted */
}
@ -1372,6 +1382,13 @@ DeferErrorIfUnsupportedUnionQuery(Query *subqueryTree)
"VALUES is not supported within a "
"UNION", NULL);
}
else if (recurType == RECURRING_TUPLES_JSON_TABLE)
{
return DeferredError(ERRCODE_FEATURE_NOT_SUPPORTED,
"cannot push down this subquery",
"JSON_TABLE is not supported within a "
"UNION", NULL);
}
return NULL;
}
@ -1477,6 +1494,11 @@ RecurringTypeDescription(RecurringTuplesType recurType)
return "a VALUES clause";
}
case RECURRING_TUPLES_JSON_TABLE:
{
return "a JSON_TABLE";
}
case RECURRING_TUPLES_INVALID:
{
/*
@ -1673,7 +1695,8 @@ DeferredErrorIfUnsupportedLateralSubquery(PlannerInfo *plannerInfo,
* strings anyway.
*/
if (recurType != RECURRING_TUPLES_VALUES &&
recurType != RECURRING_TUPLES_RESULT_FUNCTION)
recurType != RECURRING_TUPLES_RESULT_FUNCTION &&
recurType != RECURRING_TUPLES_JSON_TABLE)
{
recurTypeDescription = psprintf("%s (%s)", recurTypeDescription,
recurringRangeTableEntry->eref->
@ -1750,6 +1773,26 @@ ContainsRecurringRangeTable(List *rangeTable, RecurringTuplesType *recurType)
}
/*
* IsJsonTableRTE checks whether the RTE refers to a JSON_TABLE
* table function, which was introduced in PostgreSQL 17.
*/
bool
IsJsonTableRTE(RangeTblEntry *rte)
{
#if PG_VERSION_NUM >= PG_VERSION_17
if (rte == NULL)
{
return false;
}
return (rte->rtekind == RTE_TABLEFUNC &&
rte->tablefunc->functype == TFT_JSON_TABLE);
#endif
return false;
}
/*
* HasRecurringTuples returns whether any part of the expression will generate
* the same set of tuples in every query on shards when executing a distributed
@ -1811,6 +1854,11 @@ HasRecurringTuples(Node *node, RecurringTuplesType *recurType)
*recurType = RECURRING_TUPLES_VALUES;
return true;
}
else if (IsJsonTableRTE(rangeTableEntry))
{
*recurType = RECURRING_TUPLES_JSON_TABLE;
return true;
}
return false;
}
@ -2049,6 +2097,16 @@ CreateSubqueryTargetListAndAdjustVars(List *columnList)
*/
column->varno = 1;
column->varattno = resNo;
/*
* 1 subquery means there is one range table entry so with Postgres 16+ we need
* to ensure that column's varnullingrels - the set of join rels that can null
* the var - is empty. Otherwise, when given the query, the Postgres planner
* may attempt to access a non-existent range table and segfault, as in #7787.
*/
#if PG_VERSION_NUM >= PG_VERSION_16
column->varnullingrels = NULL;
#endif
}
return subqueryTargetEntryList;

123
src/backend/distributed/planner/recursive_planning.c
View File

@ -2291,6 +2291,129 @@ BuildReadIntermediateResultsArrayQuery(List *targetEntryList,
}
/*
* For the given target list, build an empty relation with the same target list.
* For example, if the target list is (a, b, c), and resultId is "empty", then
* it returns a Query object for this SQL:
* SELECT a, b, c FROM (VALUES (NULL, NULL, NULL)) AS empty(a, b, c) WHERE false;
*/
Query *
BuildEmptyResultQuery(List *targetEntryList, char *resultId)
{
List *targetList = NIL;
ListCell *targetEntryCell = NULL;
List *colTypes = NIL;
List *colTypMods = NIL;
List *colCollations = NIL;
List *colNames = NIL;
List *valueConsts = NIL;
List *valueTargetList = NIL;
List *valueColNames = NIL;
int targetIndex = 1;
/* build the target list and column lists needed */
foreach(targetEntryCell, targetEntryList)
{
TargetEntry *targetEntry = (TargetEntry *) lfirst(targetEntryCell);
Node *targetExpr = (Node *) targetEntry->expr;
char *columnName = targetEntry->resname;
Oid columnType = exprType(targetExpr);
Oid columnTypMod = exprTypmod(targetExpr);
Oid columnCollation = exprCollation(targetExpr);
if (targetEntry->resjunk)
{
continue;
}
Var *tgtVar = makeVar(1, targetIndex, columnType, columnTypMod, columnCollation,
0);
TargetEntry *tgtEntry = makeTargetEntry((Expr *) tgtVar, targetIndex, columnName,
false);
Const *valueConst = makeConst(columnType, columnTypMod, columnCollation, 0,
(Datum) 0, true, false);
StringInfoData *columnString = makeStringInfo();
appendStringInfo(columnString, "column%d", targetIndex);
TargetEntry *valueTgtEntry = makeTargetEntry((Expr *) tgtVar, targetIndex,
columnString->data, false);
valueConsts = lappend(valueConsts, valueConst);
valueTargetList = lappend(valueTargetList, valueTgtEntry);
valueColNames = lappend(valueColNames, makeString(columnString->data));
colNames = lappend(colNames, makeString(columnName));
colTypes = lappend_oid(colTypes, columnType);
colTypMods = lappend_oid(colTypMods, columnTypMod);
colCollations = lappend_oid(colCollations, columnCollation);
targetList = lappend(targetList, tgtEntry);
targetIndex++;
}
/* Build a RangeTable Entry for the VALUES relation */
RangeTblEntry *valuesRangeTable = makeNode(RangeTblEntry);
valuesRangeTable->rtekind = RTE_VALUES;
valuesRangeTable->values_lists = list_make1(valueConsts);
valuesRangeTable->colcollations = colCollations;
valuesRangeTable->coltypes = colTypes;
valuesRangeTable->coltypmods = colTypMods;
valuesRangeTable->alias = NULL;
valuesRangeTable->eref = makeAlias("*VALUES*", valueColNames);
valuesRangeTable->inFromCl = true;
RangeTblRef *valuesRTRef = makeNode(RangeTblRef);
valuesRTRef->rtindex = 1;
FromExpr *valuesJoinTree = makeNode(FromExpr);
valuesJoinTree->fromlist = list_make1(valuesRTRef);
/* build the VALUES query */
Query *valuesQuery = makeNode(Query);
valuesQuery->canSetTag = true;
valuesQuery->commandType = CMD_SELECT;
valuesQuery->rtable = list_make1(valuesRangeTable);
#if PG_VERSION_NUM >= PG_VERSION_16
valuesQuery->rteperminfos = NIL;
#endif
valuesQuery->jointree = valuesJoinTree;
valuesQuery->targetList = valueTargetList;
/* build the relation selecting from the VALUES */
RangeTblEntry *emptyRangeTable = makeNode(RangeTblEntry);
emptyRangeTable->rtekind = RTE_SUBQUERY;
emptyRangeTable->subquery = valuesQuery;
emptyRangeTable->alias = makeAlias(resultId, colNames);
emptyRangeTable->eref = emptyRangeTable->alias;
emptyRangeTable->inFromCl = true;
/* build the SELECT query */
Query *resultQuery = makeNode(Query);
resultQuery->commandType = CMD_SELECT;
resultQuery->canSetTag = true;
resultQuery->rtable = list_make1(emptyRangeTable);
#if PG_VERSION_NUM >= PG_VERSION_16
resultQuery->rteperminfos = NIL;
#endif
RangeTblRef *rangeTableRef = makeNode(RangeTblRef);
rangeTableRef->rtindex = 1;
/* insert a FALSE qual to ensure 0 rows returned */
FromExpr *joinTree = makeNode(FromExpr);
joinTree->fromlist = list_make1(rangeTableRef);
joinTree->quals = makeBoolConst(false, false);
resultQuery->jointree = joinTree;
resultQuery->targetList = targetList;
return resultQuery;
}
/*
* BuildReadIntermediateResultsQuery is the common code for generating
* queries to read from result files. It is used by

21
src/backend/distributed/replication/multi_logical_replication.c
View File

@ -1143,7 +1143,7 @@ ConflictWithIsolationTestingBeforeCopy(void)
const bool sessionLock = false;
const bool dontWait = false;
if (RunningUnderIsolationTest)
if (RunningUnderCitusTestSuite)
{
SET_LOCKTAG_ADVISORY(tag, MyDatabaseId,
SHARD_MOVE_ADVISORY_LOCK_SECOND_KEY,
@ -1177,7 +1177,7 @@ ConflictWithIsolationTestingAfterCopy(void)
const bool sessionLock = false;
const bool dontWait = false;
if (RunningUnderIsolationTest)
if (RunningUnderCitusTestSuite)
{
SET_LOCKTAG_ADVISORY(tag, MyDatabaseId,
SHARD_MOVE_ADVISORY_LOCK_FIRST_KEY,
@ -1882,14 +1882,15 @@ WaitForGroupedLogicalRepTargetsToCatchUp(XLogRecPtr sourcePosition,
GetCurrentTimestamp(),
logicalReplicationProgressReportTimeout))
{
ereport(LOG, (errmsg(
"The LSN of the target subscriptions on node %s:%d have "
"increased from %ld to %ld at %s where the source LSN is %ld ",
superuserConnection->hostname,
superuserConnection->port, previousTargetBeforeThisLoop,
targetPosition,
timestamptz_to_str(previousLSNIncrementTime),
sourcePosition)));
ereport(LOG, (errmsg("The LSN of the target subscriptions on node %s:%d "
"has increased from %X/%X to %X/%X at %s where the "
"source LSN is %X/%X ",
superuserConnection->hostname,
superuserConnection->port,
LSN_FORMAT_ARGS(previousTargetBeforeThisLoop),
LSN_FORMAT_ARGS(targetPosition),
timestamptz_to_str(previousLSNIncrementTime),
LSN_FORMAT_ARGS(sourcePosition))));
previousReportTime = GetCurrentTimestamp();
}

42
src/backend/distributed/shardsplit/shardsplit_decoder.c
View File

@ -94,6 +94,42 @@ replication_origin_filter_cb(LogicalDecodingContext *ctx, RepOriginId origin_id)
}
/*
* update_replication_progress is copied from Postgres 15. We use it to send keepalive
* messages when we are filtering out the wal changes resulting from the initial copy.
* If we do not send out messages long enough, wal reciever will time out.
* Postgres 16 has refactored this code such that keepalive messages are sent during
* reordering phase which is above change_cb. So we do not need to send keepalive in
* change_cb.
*/
#if (PG_VERSION_NUM < PG_VERSION_16)
static void
update_replication_progress(LogicalDecodingContext *ctx, bool skipped_xact)
{
static int changes_count = 0;
/*
* We don't want to try sending a keepalive message after processing each
* change as that can have overhead. Tests revealed that there is no
* noticeable overhead in doing it after continuously processing 100 or so
* changes.
*/
#define CHANGES_THRESHOLD 100
/*
* After continuously processing CHANGES_THRESHOLD changes, we
* try to send a keepalive message if required.
*/
if (ctx->end_xact || ++changes_count >= CHANGES_THRESHOLD)
{
OutputPluginUpdateProgress(ctx, skipped_xact);
changes_count = 0;
}
}
#endif
/*
* shard_split_change_cb function emits the incoming tuple change
* to the appropriate destination shard.
@ -112,6 +148,12 @@ shard_split_change_cb(LogicalDecodingContext *ctx, ReorderBufferTXN *txn,
return;
}
#if (PG_VERSION_NUM < PG_VERSION_16)
/* Send replication keepalive. */
update_replication_progress(ctx, false);
#endif
/* check if the relation is publishable.*/
if (!is_publishable_relation(relation))
{

63
src/backend/distributed/shared_library_init.c
View File

@ -27,6 +27,7 @@
#include "catalog/pg_extension.h"
#include "commands/explain.h"
#include "commands/extension.h"
#include "commands/seclabel.h"
#include "common/string.h"
#include "executor/executor.h"
#include "libpq/auth.h"
@ -173,15 +174,11 @@ static bool FinishedStartupCitusBackend = false;
static object_access_hook_type PrevObjectAccessHook = NULL;
#if PG_VERSION_NUM >= PG_VERSION_15
static shmem_request_hook_type prev_shmem_request_hook = NULL;
#endif
void _PG_init(void);
#if PG_VERSION_NUM >= PG_VERSION_15
static void citus_shmem_request(void);
#endif
static void CitusObjectAccessHook(ObjectAccessType access, Oid classId, Oid objectId, int
subId, void *arg);
static void DoInitialCleanup(void);
@ -474,10 +471,8 @@ _PG_init(void)
original_client_auth_hook = ClientAuthentication_hook;
ClientAuthentication_hook = CitusAuthHook;
#if PG_VERSION_NUM >= PG_VERSION_15
prev_shmem_request_hook = shmem_request_hook;
shmem_request_hook = citus_shmem_request;
#endif
InitializeMaintenanceDaemon();
@ -572,6 +567,16 @@ _PG_init(void)
INIT_COLUMNAR_SYMBOL(PGFunction, columnar_storage_info);
INIT_COLUMNAR_SYMBOL(PGFunction, columnar_store_memory_stats);
INIT_COLUMNAR_SYMBOL(PGFunction, test_columnar_storage_write_new_page);
/*
* This part is only for SECURITY LABEL tests
* mimicking what an actual security label provider would do
*/
if (RunningUnderCitusTestSuite)
{
register_label_provider("citus '!tests_label_provider",
citus_test_object_relabel);
}
}
@ -591,8 +596,6 @@ AdjustDynamicLibraryPathForCdcDecoders(void)
}
#if PG_VERSION_NUM >= PG_VERSION_15
/*
* Requests any additional shared memory required for citus.
*/
@ -613,9 +616,6 @@ citus_shmem_request(void)
}
#endif
/*
* DoInitialCleanup does cleanup at start time.
* Currently it:
@ -2293,13 +2293,14 @@ RegisterCitusConfigVariables(void)
WarnIfReplicationModelIsSet, NULL, NULL);
DefineCustomBoolVariable(
"citus.running_under_isolation_test",
"citus.running_under_citus_test_suite",
gettext_noop(
"Only useful for testing purposes, when set to true, Citus does some "
"tricks to implement useful isolation tests with rebalancing. Should "
"tricks to implement useful isolation tests with rebalancing. It also "
"registers a dummy label provider for SECURITY LABEL tests. Should "
"never be set to true on production systems "),
gettext_noop("for details of the tricks implemented, refer to the source code"),
&RunningUnderIsolationTest,
&RunningUnderCitusTestSuite,
false,
PGC_SUSET,
GUC_SUPERUSER_ONLY | GUC_NO_SHOW_ALL | GUC_NOT_IN_SAMPLE,
@ -2853,14 +2854,27 @@ ApplicationNameAssignHook(const char *newval, void *extra)
DetermineCitusBackendType(newval);
/*
* AssignGlobalPID might read from catalog tables to get the the local
* nodeid. But ApplicationNameAssignHook might be called before catalog
* access is available to the backend (such as in early stages of
* authentication). We use StartupCitusBackend to initialize the global pid
* after catalogs are available. After that happens this hook becomes
* responsible to update the global pid on later application_name changes.
* So we set the FinishedStartupCitusBackend flag in StartupCitusBackend to
* indicate when this responsibility handoff has happened.
* We use StartupCitusBackend to initialize the global pid after catalogs
* are available. After that happens this hook becomes responsible to update
* the global pid on later application_name changes. So we set the
* FinishedStartupCitusBackend flag in StartupCitusBackend to indicate when
* this responsibility handoff has happened.
*
* Also note that when application_name changes, we don't actually need to
* try re-assigning the global pid for external client backends and
* background workers because application_name doesn't affect the global
* pid for such backends - note that !IsExternalClientBackend() check covers
* both types of backends. Plus,
* trying to re-assign the global pid for such backends would unnecessarily
* cause performing a catalog access when the cached local node id is
* invalidated. However, accessing to the catalog tables is dangerous in
* certain situations like when we're not in a transaction block. And for
* the other types of backends, i.e., the Citus internal backends, we need
* to re-assign the global pid when the application_name changes because for
* such backends we simply extract the global pid inherited from the
* originating backend from the application_name -that's specified by
* originating backend when openning that connection- and this doesn't require
* catalog access.
*
* Another solution to the catalog table acccess problem would be to update
* global pid lazily, like we do for HideShards. But that's not possible
@ -2870,7 +2884,7 @@ ApplicationNameAssignHook(const char *newval, void *extra)
* as reasonably possible, which is also why we extract global pids in the
* AuthHook already (extracting doesn't require catalog access).
*/
if (FinishedStartupCitusBackend)
if (FinishedStartupCitusBackend && !IsExternalClientBackend())
{
AssignGlobalPID(newval);
}
@ -2905,6 +2919,9 @@ NodeConninfoGucCheckHook(char **newval, void **extra, GucSource source)
"sslcrl",
"sslkey",
"sslmode",
#if PG_VERSION_NUM >= PG_VERSION_17
"sslnegotiation",
#endif
"sslrootcert",
"tcp_user_timeout",
};

1
src/backend/distributed/sql/citus--12.1-1--13.0-1.sql
View File

@ -2,3 +2,4 @@
-- bump version to 13.0-1
#include "udfs/citus_prepare_pg_upgrade/13.0-1.sql"
#include "udfs/create_time_partitions/13.0-1.sql"

2
src/backend/distributed/sql/downgrades/citus--13.0-1--12.1-1.sql
View File

@ -1,2 +1,4 @@
-- citus--13.0-1--12.1-1
-- this is an empty downgrade path since citus--12.1-1--13.0-1.sql is empty
#include "../udfs/create_time_partitions/10.2-1.sql"

58
src/backend/distributed/sql/udfs/create_time_partitions/13.0-1.sql generated Normal file
View File

@ -0,0 +1,58 @@
CREATE OR REPLACE FUNCTION pg_catalog.create_time_partitions(
table_name regclass,
partition_interval INTERVAL,
end_at timestamptz,
start_from timestamptz DEFAULT now())
returns boolean
LANGUAGE plpgsql
AS $$
DECLARE
-- partitioned table name
schema_name_text name;
table_name_text name;
-- record for to-be-created partition
missing_partition_record record;
-- result indiciates whether any partitions were created
partition_created bool := false;
BEGIN
IF start_from >= end_at THEN
RAISE 'start_from (%) must be older than end_at (%)', start_from, end_at;
END IF;
IF NOT isfinite(partition_interval) THEN
RAISE 'Partition interval must be a finite value';
END IF;
SELECT nspname, relname
INTO schema_name_text, table_name_text
FROM pg_class JOIN pg_namespace ON pg_class.relnamespace = pg_namespace.oid
WHERE pg_class.oid = table_name::oid;
-- Get missing partition range info using the get_missing_partition_ranges
-- and create partitions using that info.
FOR missing_partition_record IN
SELECT *
FROM get_missing_time_partition_ranges(table_name, partition_interval, end_at, start_from)
LOOP
EXECUTE format('CREATE TABLE %I.%I PARTITION OF %I.%I FOR VALUES FROM (%L) TO (%L)',
schema_name_text,
missing_partition_record.partition_name,
schema_name_text,
table_name_text,
missing_partition_record.range_from_value,
missing_partition_record.range_to_value);
partition_created := true;
END LOOP;
RETURN partition_created;
END;
$$;
COMMENT ON FUNCTION pg_catalog.create_time_partitions(
table_name regclass,
partition_interval INTERVAL,
end_at timestamptz,
start_from timestamptz)
IS 'create time partitions for the given range';

4
src/backend/distributed/sql/udfs/create_time_partitions/latest.sql
View File

@ -21,6 +21,10 @@ BEGIN
RAISE 'start_from (%) must be older than end_at (%)', start_from, end_at;
END IF;
IF NOT isfinite(partition_interval) THEN
RAISE 'Partition interval must be a finite value';
END IF;
SELECT nspname, relname
INTO schema_name_text, table_name_text
FROM pg_class JOIN pg_namespace ON pg_class.relnamespace = pg_namespace.oid

2
src/backend/distributed/test/fake_am.c
View File

@ -310,7 +310,7 @@ fake_relation_set_new_filenode(Relation rel,
*/
*minmulti = GetOldestMultiXactId();
SMgrRelation srel = RelationCreateStorage_compat(*newrnode, persistence, true);
SMgrRelation srel = RelationCreateStorage(*newrnode, persistence, true);
/*
* If required, set up an init fork for an unlogged table so that it can

7
src/backend/distributed/test/metadata_sync.c
View File

@ -50,6 +50,13 @@ activate_node_snapshot(PG_FUNCTION_ARGS)
* so we are using first primary worker node just for test purposes.
*/
WorkerNode *dummyWorkerNode = GetFirstPrimaryWorkerNode();
if (dummyWorkerNode == NULL)
{
ereport(ERROR, (errmsg("no worker nodes found"),
errdetail("Function activate_node_snapshot is meant to be "
"used when running tests on a multi-node cluster "
"with workers.")));
}
/*
* Create MetadataSyncContext which is used throughout nodes' activation.

4
src/backend/distributed/test/run_from_same_connection.c
View File

@ -190,6 +190,9 @@ run_commands_on_session_level_connection_to_node(PG_FUNCTION_ARGS)
/*
* override_backend_data_gpid is a wrapper around SetBackendDataGpid().
* Also sets distributedCommandOriginator to true since the only caller of
* this method calls this function actually wants this backend to
* be treated as a distributed command originator with the given global pid.
*/
Datum
override_backend_data_gpid(PG_FUNCTION_ARGS)
@ -199,6 +202,7 @@ override_backend_data_gpid(PG_FUNCTION_ARGS)
uint64 gpid = PG_GETARG_INT64(0);
SetBackendDataGlobalPID(gpid);
SetBackendDataDistributedCommandOriginator(true);
PG_RETURN_VOID();
}

5
src/backend/distributed/test/shared_connection_counters.c
View File

@ -49,13 +49,8 @@ makeIntConst(int val, int location)
{
A_Const *n = makeNode(A_Const);
#if PG_VERSION_NUM >= PG_VERSION_15
n->val.ival.type = T_Integer;
n->val.ival.ival = val;
#else
n->val.type = T_Integer;
n->val.val.ival = val;
#endif
n->location = location;
return (Node *) n;

42
src/backend/distributed/transaction/backend_data.c
View File

@ -395,7 +395,7 @@ StoreAllActiveTransactions(Tuplestorestate *tupleStore, TupleDesc tupleDescripto
bool showCurrentBackendDetails = showAllBackends;
BackendData *currentBackend =
&backendManagementShmemData->backends[backendIndex];
PGPROC *currentProc = &ProcGlobal->allProcs[backendIndex];
PGPROC *currentProc = GetPGProcByNumber(backendIndex);
/* to work on data after releasing g spinlock to protect against errors */
uint64 transactionNumber = 0;
@ -420,7 +420,7 @@ StoreAllActiveTransactions(Tuplestorestate *tupleStore, TupleDesc tupleDescripto
}
Oid databaseId = currentBackend->databaseId;
int backendPid = ProcGlobal->allProcs[backendIndex].pid;
int backendPid = GetPGProcByNumber(backendIndex)->pid;
/*
* We prefer to use worker_query instead of distributedCommandOriginator in
@ -519,15 +519,6 @@ UserHasPermissionToViewStatsOf(Oid currentUserId, Oid backendOwnedId)
void
InitializeBackendManagement(void)
{
/* on PG 15, we use shmem_request_hook_type */
#if PG_VERSION_NUM < PG_VERSION_15
/* allocate shared memory */
if (!IsUnderPostmaster)
{
RequestAddinShmemSpace(BackendManagementShmemSize());
}
#endif
prev_shmem_startup_hook = shmem_startup_hook;
shmem_startup_hook = BackendManagementShmemInit;
}
@ -855,6 +846,16 @@ GetCurrentDistributedTransactionId(void)
void
AssignDistributedTransactionId(void)
{
/*
* MyBackendData should always be available. However, we observed some
* crashes where certain hooks were not executed.
* Bug 3697586: Server crashes when assigning distributed transaction
*/
if (!MyBackendData)
{
ereport(ERROR, (errmsg("backend is not ready for distributed transactions")));
}
pg_atomic_uint64 *transactionNumberSequence =
&backendManagementShmemData->nextTransactionNumber;
@ -964,6 +965,23 @@ SetBackendDataGlobalPID(uint64 gpid)
}
/*
* SetBackendDataDistributedCommandOriginator sets the distributedCommandOriginator
* field on MyBackendData.
*/
void
SetBackendDataDistributedCommandOriginator(bool distributedCommandOriginator)
{
if (!MyBackendData)
{
return;
}
SpinLockAcquire(&MyBackendData->mutex);
MyBackendData->distributedCommandOriginator = distributedCommandOriginator;
SpinLockRelease(&MyBackendData->mutex);
}
/*
* GetGlobalPID returns the global process id of the current backend.
*/
@ -1280,7 +1298,7 @@ ActiveDistributedTransactionNumbers(void)
/* build list of starting procs */
for (int curBackend = 0; curBackend < MaxBackends; curBackend++)
{
PGPROC *currentProc = &ProcGlobal->allProcs[curBackend];
PGPROC *currentProc = GetPGProcByNumber(curBackend);
BackendData currentBackendData;
if (currentProc->pid == 0)

2
src/backend/distributed/transaction/distributed_deadlock_detection.c
View File

@ -375,7 +375,7 @@ AssociateDistributedTransactionWithBackendProc(TransactionNode *transactionNode)
for (int backendIndex = 0; backendIndex < MaxBackends; ++backendIndex)
{
PGPROC *currentProc = &ProcGlobal->allProcs[backendIndex];
PGPROC *currentProc = GetPGProcByNumber(backendIndex);
BackendData currentBackendData;
/* we're not interested in processes that are not active or waiting on a lock */

2
src/backend/distributed/transaction/lock_graph.c
View File

@ -561,7 +561,7 @@ BuildLocalWaitGraph(bool onlyDistributedTx)
/* build list of starting procs */
for (int curBackend = 0; curBackend < totalProcs; curBackend++)
{
PGPROC *currentProc = &ProcGlobal->allProcs[curBackend];
PGPROC *currentProc = GetPGProcByNumber(curBackend);
BackendData currentBackendData;
if (currentProc->pid == 0)

51
src/backend/distributed/transaction/transaction_recovery.c
View File

@ -53,7 +53,8 @@ PG_FUNCTION_INFO_V1(recover_prepared_transactions);
/* Local functions forward declarations */
static int RecoverWorkerTransactions(WorkerNode *workerNode);
static int RecoverWorkerTransactions(WorkerNode *workerNode,
MultiConnection *connection);
static List * PendingWorkerTransactionList(MultiConnection *connection);
static bool IsTransactionInProgress(HTAB *activeTransactionNumberSet,
char *preparedTransactionName);
@ -123,10 +124,51 @@ RecoverTwoPhaseCommits(void)
LockTransactionRecovery(ShareUpdateExclusiveLock);
List *workerList = ActivePrimaryNodeList(NoLock);
List *workerConnections = NIL;
WorkerNode *workerNode = NULL;
MultiConnection *connection = NULL;
/*
* Pre-establish all connections to worker nodes.
*
* We do this to enforce a consistent lock acquisition order and prevent deadlocks.
* Currently, during extension updates, we take strong locks on the Citus
* catalog tables in a specific order: first on pg_dist_authinfo, then on
* pg_dist_transaction. It's critical that any operation locking these two
* tables adheres to this order, or a deadlock could occur.
*
* Note that RecoverWorkerTransactions() retains its lock until the end
* of the transaction, while GetNodeConnection() releases its lock after
* the catalog lookup. So when there are multiple workers in the active primary
* node list, the lock acquisition order may reverse in subsequent iterations
* of the loop calling RecoverWorkerTransactions(), increasing the risk
* of deadlock.
*
* By establishing all worker connections upfront, we ensure that
* RecoverWorkerTransactions() deals with a single distributed catalog table,
* thereby preventing deadlocks regardless of the lock acquisition sequence
* used in the upgrade extension script.
*/
foreach_declared_ptr(workerNode, workerList)
{
recoveredTransactionCount += RecoverWorkerTransactions(workerNode);
int connectionFlags = 0;
char *nodeName = workerNode->workerName;
int nodePort = workerNode->workerPort;
connection = GetNodeConnection(connectionFlags, nodeName, nodePort);
Assert(connection != NULL);
/*
* We don't verify connection validity here.
* Instead, RecoverWorkerTransactions() performs the necessary
* sanity checks on the connection state.
*/
workerConnections = lappend(workerConnections, connection);
}
forboth_ptr(workerNode, workerList, connection, workerConnections)
{
recoveredTransactionCount += RecoverWorkerTransactions(workerNode, connection);
}
return recoveredTransactionCount;
@ -138,7 +180,7 @@ RecoverTwoPhaseCommits(void)
* started by this node on the specified worker.
*/
static int
RecoverWorkerTransactions(WorkerNode *workerNode)
RecoverWorkerTransactions(WorkerNode *workerNode, MultiConnection *connection)
{
int recoveredTransactionCount = 0;
@ -156,8 +198,7 @@ RecoverWorkerTransactions(WorkerNode *workerNode)
bool recoveryFailed = false;
int connectionFlags = 0;
MultiConnection *connection = GetNodeConnection(connectionFlags, nodeName, nodePort);
Assert(connection != NULL);
if (connection->pgConn == NULL || PQstatus(connection->pgConn) != CONNECTION_OK)
{
ereport(WARNING, (errmsg("transaction recovery cannot connect to %s:%d",

81
src/backend/distributed/utils/background_jobs.c
View File

@ -1393,87 +1393,6 @@ CalculateBackoffDelay(int retryCount)
}
#if PG_VERSION_NUM < PG_VERSION_15
static const char *
error_severity(int elevel)
{
const char *prefix;
switch (elevel)
{
case DEBUG1:
case DEBUG2:
case DEBUG3:
case DEBUG4:
case DEBUG5:
{
prefix = gettext_noop("DEBUG");
break;
}
case LOG:
case LOG_SERVER_ONLY:
{
prefix = gettext_noop("LOG");
break;
}
case INFO:
{
prefix = gettext_noop("INFO");
break;
}
case NOTICE:
{
prefix = gettext_noop("NOTICE");
break;
}
case WARNING:
{
prefix = gettext_noop("WARNING");
break;
}
case WARNING_CLIENT_ONLY:
{
prefix = gettext_noop("WARNING");
break;
}
case ERROR:
{
prefix = gettext_noop("ERROR");
break;
}
case FATAL:
{
prefix = gettext_noop("FATAL");
break;
}
case PANIC:
{
prefix = gettext_noop("PANIC");
break;
}
default:
{
prefix = "???";
break;
}
}
return prefix;
}
#endif
/*
* bgw_generate_returned_message -
* generates the message to be inserted into the job_run_details table

11
src/backend/distributed/utils/citus_stat_tenants.c
View File

@ -15,6 +15,7 @@
#include "unistd.h"
#include "access/hash.h"
#include "common/pg_prng.h"
#include "executor/execdesc.h"
#include "storage/ipc.h"
#include "storage/lwlock.h"
@ -38,10 +39,6 @@
#include "distributed/tuplestore.h"
#include "distributed/utils/citus_stat_tenants.h"
#if (PG_VERSION_NUM >= PG_VERSION_15)
#include "common/pg_prng.h"
#endif
static void AttributeMetricsIfApplicable(void);
ExecutorEnd_hook_type prev_ExecutorEnd = NULL;
@ -298,13 +295,7 @@ AttributeTask(char *tenantId, int colocationId, CmdType commandType)
/* If the tenant is not found in the hash table, we will track the query with a probability of StatTenantsSampleRateForNewTenants. */
if (!found)
{
#if (PG_VERSION_NUM >= PG_VERSION_15)
double randomValue = pg_prng_double(&pg_global_prng_state);
#else
/* Generate a random double between 0 and 1 */
double randomValue = (double) random() / MAX_RANDOM_VALUE;
#endif
bool shouldTrackQuery = randomValue <= StatTenantsSampleRateForNewTenants;
if (!shouldTrackQuery)
{

6
src/backend/distributed/utils/colocation_utils.c
View File

@ -362,10 +362,8 @@ ErrorIfShardPlacementsNotColocated(Oid leftRelationId, Oid rightRelationId)
leftRelationName, rightRelationName)));
}
List *leftPlacementList = ShardPlacementListSortedByWorker(
leftShardId);
List *rightPlacementList = ShardPlacementListSortedByWorker(
rightShardId);
List *leftPlacementList = ShardPlacementList(leftShardId);
List *rightPlacementList = ShardPlacementList(rightShardId);
if (list_length(leftPlacementList) != list_length(rightPlacementList))
{

5
src/backend/distributed/utils/shardinterval_utils.c
View File

@ -470,12 +470,11 @@ SingleReplicatedTable(Oid relationId)
return false;
}
List *shardIntervalList = LoadShardList(relationId);
uint64 *shardIdPointer = NULL;
foreach_declared_ptr(shardIdPointer, shardIntervalList)
foreach_declared_ptr(shardIdPointer, shardList)
{
uint64 shardId = *shardIdPointer;
shardPlacementList = ShardPlacementListSortedByWorker(shardId);
shardPlacementList = ShardPlacementList(shardId);
if (list_length(shardPlacementList) != 1)
{

6
src/backend/distributed/worker/worker_drop_protocol.c
View File

@ -170,14 +170,10 @@ WorkerDropDistributedTable(Oid relationId)
*/
if (!IsAnyObjectAddressOwnedByExtension(list_make1(distributedTableObject), NULL))
{
char *relName = get_rel_name(relationId);
Oid schemaId = get_rel_namespace(relationId);
char *schemaName = get_namespace_name(schemaId);
StringInfo dropCommand = makeStringInfo();
appendStringInfo(dropCommand, "DROP%sTABLE %s CASCADE",
IsForeignTable(relationId) ? " FOREIGN " : " ",
quote_qualified_identifier(schemaName, relName));
generate_qualified_relation_name(relationId));
Node *dropCommandNode = ParseTreeNode(dropCommand->data);

33
src/backend/distributed/worker/worker_shard_visibility.c
View File

@ -441,7 +441,7 @@ FilterShardsFromPgclass(Node *node, void *context)
/*
* We process the whole rtable rather than visiting individual RangeTblEntry's
* in the walker, since we need to know the varno to generate the right
* fiter.
* filter.
*/
int varno = 0;
RangeTblEntry *rangeTableEntry = NULL;
@ -471,20 +471,39 @@ FilterShardsFromPgclass(Node *node, void *context)
/* make sure the expression is in the right memory context */
MemoryContext originalContext = MemoryContextSwitchTo(queryContext);
/* add relation_is_a_known_shard(oid) IS NOT TRUE to the quals of the query */
Node *newQual = CreateRelationIsAKnownShardFilter(varno);
Node *oldQuals = query->jointree->quals;
if (oldQuals)
#if PG_VERSION_NUM >= PG_VERSION_17
/*
* In PG17, MERGE queries introduce a new struct `mergeJoinCondition`.
* We need to handle this condition safely.
*/
if (query->mergeJoinCondition != NULL)
{
query->jointree->quals = (Node *) makeBoolExpr(
/* Add the filter to mergeJoinCondition */
query->mergeJoinCondition = (Node *) makeBoolExpr(
AND_EXPR,
list_make2(oldQuals, newQual),
list_make2(query->mergeJoinCondition, newQual),
-1);
}
else
#endif
{
query->jointree->quals = newQual;
/* Handle older versions or queries without mergeJoinCondition */
Node *oldQuals = query->jointree->quals;
if (oldQuals)
{
query->jointree->quals = (Node *) makeBoolExpr(
AND_EXPR,
list_make2(oldQuals, newQual),
-1);
}
else
{
query->jointree->quals = newQual;
}
}
MemoryContextSwitchTo(originalContext);

8
src/include/columnar/columnar_version_compat.h
View File

@ -14,14 +14,6 @@
#include "pg_version_constants.h"
#if PG_VERSION_NUM >= PG_VERSION_15
#define ExecARDeleteTriggers_compat(a, b, c, d, e, f) \
ExecARDeleteTriggers(a, b, c, d, e, f)
#else
#define ExecARDeleteTriggers_compat(a, b, c, d, e, f) \
ExecARDeleteTriggers(a, b, c, d, e)
#endif
#define ACLCHECK_OBJECT_TABLE OBJECT_TABLE
#define ExplainPropertyLong(qlabel, value, es) \

1
src/include/distributed/backend_data.h
View File

@ -61,6 +61,7 @@ extern void AssignGlobalPID(const char *applicationName);
extern uint64 GetGlobalPID(void);
extern void SetBackendDataDatabaseId(void);
extern void SetBackendDataGlobalPID(uint64 gpid);
extern void SetBackendDataDistributedCommandOriginator(bool distributedCommandOriginator);
extern uint64 ExtractGlobalPID(const char *applicationName);
extern int ExtractNodeIdFromGlobalPID(uint64 globalPID, bool missingOk);
extern int ExtractProcessIdFromGlobalPID(uint64 globalPID);

11
src/include/distributed/commands.h
View File

@ -510,6 +510,11 @@ extern List * AlterSchemaOwnerStmtObjectAddress(Node *node, bool missing_ok,
extern List * AlterSchemaRenameStmtObjectAddress(Node *node, bool missing_ok, bool
isPostprocess);
/* seclabel.c - forward declarations*/
extern List * PostprocessSecLabelStmt(Node *node, const char *queryString);
extern List * SecLabelStmtObjectAddress(Node *node, bool missing_ok, bool isPostprocess);
extern void citus_test_object_relabel(const ObjectAddress *object, const char *seclabel);
/* sequence.c - forward declarations */
extern List * PreprocessAlterSequenceStmt(Node *node, const char *queryString,
ProcessUtilityContext processUtilityContext);
@ -520,13 +525,11 @@ extern List * PostprocessAlterSequenceSchemaStmt(Node *node, const char *querySt
extern List * PreprocessAlterSequenceOwnerStmt(Node *node, const char *queryString,
ProcessUtilityContext processUtilityContext);
extern List * PostprocessAlterSequenceOwnerStmt(Node *node, const char *queryString);
#if (PG_VERSION_NUM >= PG_VERSION_15)
extern List * PreprocessAlterSequencePersistenceStmt(Node *node, const char *queryString,
ProcessUtilityContext
processUtilityContext);
extern List * PreprocessSequenceAlterTableStmt(Node *node, const char *queryString,
ProcessUtilityContext processUtilityContext);
#endif
extern List * PreprocessDropSequenceStmt(Node *node, const char *queryString,
ProcessUtilityContext processUtilityContext);
extern List * SequenceDropStmtObjectAddress(Node *stmt, bool missing_ok, bool
@ -542,10 +545,8 @@ extern List * AlterSequenceSchemaStmtObjectAddress(Node *node, bool missing_ok,
isPostprocess);
extern List * AlterSequenceOwnerStmtObjectAddress(Node *node, bool missing_ok, bool
isPostprocess);
#if (PG_VERSION_NUM >= PG_VERSION_15)
extern List * AlterSequencePersistenceStmtObjectAddress(Node *node, bool missing_ok, bool
isPostprocess);
#endif
extern List * RenameSequenceStmtObjectAddress(Node *node, bool missing_ok, bool
isPostprocess);
extern void ErrorIfUnsupportedSeqStmt(CreateSeqStmt *createSeqStmt);
@ -749,8 +750,6 @@ extern List * CreateTriggerStmtObjectAddress(Node *node, bool missingOk, bool
isPostprocess);
extern void CreateTriggerEventExtendNames(CreateTrigStmt *createTriggerStmt,
char *schemaName, uint64 shardId);
extern List * PreprocessAlterTriggerRenameStmt(Node *node, const char *queryString,
ProcessUtilityContext processUtilityContext);
extern List * PostprocessAlterTriggerRenameStmt(Node *node, const char *queryString);
extern void AlterTriggerRenameEventExtendNames(RenameStmt *renameTriggerStmt,
char *schemaName, uint64 shardId);

7
src/include/distributed/deparser.h
View File

@ -259,14 +259,15 @@ extern void QualifyRenameTextSearchDictionaryStmt(Node *node);
extern void QualifyTextSearchConfigurationCommentStmt(Node *node);
extern void QualifyTextSearchDictionaryCommentStmt(Node *node);
/* forward declarations for deparse_seclabel_stmts.c */
extern char * DeparseSecLabelStmt(Node *node);
/* forward declarations for deparse_sequence_stmts.c */
extern char * DeparseDropSequenceStmt(Node *node);
extern char * DeparseRenameSequenceStmt(Node *node);
extern char * DeparseAlterSequenceSchemaStmt(Node *node);
extern char * DeparseAlterSequenceOwnerStmt(Node *node);
#if (PG_VERSION_NUM >= PG_VERSION_15)
extern char * DeparseAlterSequencePersistenceStmt(Node *node);
#endif
extern char * DeparseGrantOnSequenceStmt(Node *node);
/* forward declarations for qualify_sequence_stmt.c */
@ -274,9 +275,7 @@ extern void QualifyRenameSequenceStmt(Node *node);
extern void QualifyDropSequenceStmt(Node *node);
extern void QualifyAlterSequenceSchemaStmt(Node *node);
extern void QualifyAlterSequenceOwnerStmt(Node *node);
#if (PG_VERSION_NUM >= PG_VERSION_15)
extern void QualifyAlterSequencePersistenceStmt(Node *node);
#endif
extern void QualifyGrantOnSequenceStmt(Node *node);
#endif /* CITUS_DEPARSER_H */

5
src/include/distributed/distributed_planner.h
View File

@ -28,11 +28,6 @@
#define CURSOR_OPT_FORCE_DISTRIBUTED 0x080000
/* Hack to compile Citus on pre-MERGE Postgres versions */
#if PG_VERSION_NUM < PG_VERSION_15
#define CMD_MERGE CMD_UNKNOWN
#endif
/* level of planner calls */
extern int PlannerLevel;

3
src/include/distributed/metadata_sync.h
View File

@ -128,9 +128,6 @@ extern List * IdentitySequenceDependencyCommandList(Oid targetRelationId);
extern List * DDLCommandsForSequence(Oid sequenceOid, char *ownerName);
extern List * GetSequencesFromAttrDef(Oid attrdefOid);
#if PG_VERSION_NUM < PG_VERSION_15
ObjectAddress GetAttrDefaultColumnAddress(Oid attrdefoid);
#endif
extern List * GetAttrDefsFromSequence(Oid seqOid);
extern void GetDependentSequencesWithRelation(Oid relationId, List **seqInfoList,
AttrNumber attnum, char depType);

2
src/include/distributed/multi_router_planner.h
View File

@ -118,7 +118,7 @@ extern bool HasDangerousJoinUsing(List *rtableList, Node *jtnode);
extern Job * RouterJob(Query *originalQuery,
PlannerRestrictionContext *plannerRestrictionContext,
DeferredErrorMessage **planningError);
extern bool ContainsOnlyLocalTables(RTEListProperties *rteProperties);
extern bool ContainsOnlyLocalOrReferenceTables(RTEListProperties *rteProperties);
extern RangeTblEntry * ExtractSourceResultRangeTableEntry(Query *query);
#endif /* MULTI_ROUTER_PLANNER_H */

2
src/include/distributed/query_colocation_checker.h
View File

@ -39,5 +39,7 @@ extern Query * WrapRteRelationIntoSubquery(RangeTblEntry *rteRelation,
List *requiredAttributes,
RTEPermissionInfo *perminfo);
extern List * CreateAllTargetListForRelation(Oid relationId, List *requiredAttributes);
extern List * CreateFilteredTargetListForRelation(Oid relationId,
List *requiredAttributes);
#endif /* QUERY_COLOCATION_CHECKER_H */

1
src/include/distributed/query_pushdown_planning.h
View File

@ -46,6 +46,7 @@ extern DeferredErrorMessage * DeferErrorIfCannotPushdownSubquery(Query *subquery
bool
outerMostQueryHasLimit);
extern DeferredErrorMessage * DeferErrorIfUnsupportedUnionQuery(Query *queryTree);
extern bool IsJsonTableRTE(RangeTblEntry *rte);
#endif /* QUERY_PUSHDOWN_PLANNING_H */

1
src/include/distributed/recursive_planning.h
View File

@ -40,6 +40,7 @@ extern Query * BuildReadIntermediateResultsArrayQuery(List *targetEntryList,
List *columnAliasList,
List *resultIdList,
bool useBinaryCopyFormat);
extern Query * BuildEmptyResultQuery(List *targetEntryList, char *resultId);
extern bool GeneratingSubplans(void);
extern bool ContainsLocalTableDistributedTableJoin(List *rangeTableList);
extern void ReplaceRTERelationWithRteSubquery(RangeTblEntry *rangeTableEntry,

5
src/include/distributed/repartition_executor.h
View File

@ -28,5 +28,10 @@ extern List * GenerateTaskListWithRedistributedResults(
bool useBinaryFormat);
extern bool IsSupportedRedistributionTarget(Oid targetRelationId);
extern bool IsRedistributablePlan(Plan *selectPlan);
extern bool HasMergeNotMatchedBySource(Query *query);
extern void AdjustTaskQueryForEmptySource(Oid targetRelationId,
Query *mergeQuery,
List *emptySourceTaskList,
char *resultIdPrefix);
#endif /* REPARTITION_EXECUTOR_H */

Some files were not shown because too many files have changed in this diff Show More