Merge branch 'master' into alexey-milovidov-patch-5

This commit is contained in:
Alexey Milovidov 2022-11-08 16:27:54 +01:00 committed by GitHub
commit f2e6e08fb8
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
365 changed files with 6928 additions and 3621 deletions

View File

@ -2,7 +2,7 @@
name: Debug
'on':
[push, pull_request, release, workflow_dispatch]
[push, pull_request, release, workflow_dispatch, workflow_call]
jobs:
DebugInfo:

View File

@ -1056,6 +1056,23 @@ jobs:
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH"
MarkReleaseReady:
needs:
- BuilderBinDarwin
- BuilderBinDarwinAarch64
- BuilderDebRelease
- BuilderDebAarch64
runs-on: [self-hosted, style-checker]
steps:
- name: Clear repository
run: |
sudo rm -fr "$GITHUB_WORKSPACE" && mkdir "$GITHUB_WORKSPACE"
- name: Check out repository code
uses: actions/checkout@v2
- name: Mark Commit Release Ready
run: |
cd "$GITHUB_WORKSPACE/tests/ci"
python3 mark_release_ready.py
##############################################################################################
########################### FUNCTIONAl STATELESS TESTS #######################################
##############################################################################################
@ -2994,10 +3011,83 @@ jobs:
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH"
##############################################################################################
###################################### SQLANCER FUZZERS ######################################
##############################################################################################
SQLancerTestRelease:
needs: [BuilderDebRelease]
runs-on: [self-hosted, fuzzer-unit-tester]
steps:
- name: Set envs
run: |
cat >> "$GITHUB_ENV" << 'EOF'
TEMP_PATH=${{runner.temp}}/sqlancer_release
REPORTS_PATH=${{runner.temp}}/reports_dir
CHECK_NAME=SQLancer (release)
REPO_COPY=${{runner.temp}}/sqlancer_release/ClickHouse
EOF
- name: Download json reports
uses: actions/download-artifact@v2
with:
path: ${{ env.REPORTS_PATH }}
- name: Clear repository
run: |
sudo rm -fr "$GITHUB_WORKSPACE" && mkdir "$GITHUB_WORKSPACE"
- name: Check out repository code
uses: actions/checkout@v2
- name: SQLancer
run: |
sudo rm -fr "$TEMP_PATH"
mkdir -p "$TEMP_PATH"
cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
cd "$REPO_COPY/tests/ci"
python3 sqlancer_check.py "$CHECK_NAME"
- name: Cleanup
if: always()
run: |
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH"
SQLancerTestDebug:
needs: [BuilderDebDebug]
runs-on: [self-hosted, fuzzer-unit-tester]
steps:
- name: Set envs
run: |
cat >> "$GITHUB_ENV" << 'EOF'
TEMP_PATH=${{runner.temp}}/sqlancer_debug
REPORTS_PATH=${{runner.temp}}/reports_dir
CHECK_NAME=SQLancer (debug)
REPO_COPY=${{runner.temp}}/sqlancer_debug/ClickHouse
EOF
- name: Download json reports
uses: actions/download-artifact@v2
with:
path: ${{ env.REPORTS_PATH }}
- name: Clear repository
run: |
sudo rm -fr "$GITHUB_WORKSPACE" && mkdir "$GITHUB_WORKSPACE"
- name: Check out repository code
uses: actions/checkout@v2
- name: SQLancer
run: |
sudo rm -fr "$TEMP_PATH"
mkdir -p "$TEMP_PATH"
cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
cd "$REPO_COPY/tests/ci"
python3 sqlancer_check.py "$CHECK_NAME"
- name: Cleanup
if: always()
run: |
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH"
FinishCheck:
needs:
- DockerHubPush
- BuilderReport
- BuilderSpecialReport
- MarkReleaseReady
- FunctionalStatelessTestDebug0
- FunctionalStatelessTestDebug1
- FunctionalStatelessTestDebug2
@ -3053,6 +3143,8 @@ jobs:
- UnitTestsUBsan
- UnitTestsReleaseClang
- SharedBuildSmokeTest
- SQLancerTestRelease
- SQLancerTestDebug
runs-on: [self-hosted, style-checker]
steps:
- name: Clear repository

View File

@ -10,6 +10,9 @@ env:
workflow_dispatch:
jobs:
Debug:
# The task for having a preserved ENV and event.json for later investigation
uses: ./.github/workflows/debug.yml
DockerHubPushAarch64:
runs-on: [self-hosted, style-checker-aarch64]
steps:

View File

@ -3491,6 +3491,77 @@ jobs:
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH"
##############################################################################################
###################################### SQLANCER FUZZERS ######################################
##############################################################################################
SQLancerTestRelease:
needs: [BuilderDebRelease]
runs-on: [self-hosted, fuzzer-unit-tester]
steps:
- name: Set envs
run: |
cat >> "$GITHUB_ENV" << 'EOF'
TEMP_PATH=${{runner.temp}}/sqlancer_release
REPORTS_PATH=${{runner.temp}}/reports_dir
CHECK_NAME=SQLancer (release)
REPO_COPY=${{runner.temp}}/sqlancer_release/ClickHouse
EOF
- name: Download json reports
uses: actions/download-artifact@v2
with:
path: ${{ env.REPORTS_PATH }}
- name: Clear repository
run: |
sudo rm -fr "$GITHUB_WORKSPACE" && mkdir "$GITHUB_WORKSPACE"
- name: Check out repository code
uses: actions/checkout@v2
- name: SQLancer
run: |
sudo rm -fr "$TEMP_PATH"
mkdir -p "$TEMP_PATH"
cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
cd "$REPO_COPY/tests/ci"
python3 sqlancer_check.py "$CHECK_NAME"
- name: Cleanup
if: always()
run: |
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH"
SQLancerTestDebug:
needs: [BuilderDebDebug]
runs-on: [self-hosted, fuzzer-unit-tester]
steps:
- name: Set envs
run: |
cat >> "$GITHUB_ENV" << 'EOF'
TEMP_PATH=${{runner.temp}}/sqlancer_debug
REPORTS_PATH=${{runner.temp}}/reports_dir
CHECK_NAME=SQLancer (debug)
REPO_COPY=${{runner.temp}}/sqlancer_debug/ClickHouse
EOF
- name: Download json reports
uses: actions/download-artifact@v2
with:
path: ${{ env.REPORTS_PATH }}
- name: Clear repository
run: |
sudo rm -fr "$GITHUB_WORKSPACE" && mkdir "$GITHUB_WORKSPACE"
- name: Check out repository code
uses: actions/checkout@v2
- name: SQLancer
run: |
sudo rm -fr "$TEMP_PATH"
mkdir -p "$TEMP_PATH"
cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
cd "$REPO_COPY/tests/ci"
python3 sqlancer_check.py "$CHECK_NAME"
- name: Cleanup
if: always()
run: |
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH"
#############################################################################################
###################################### JEPSEN TESTS #########################################
#############################################################################################
@ -3501,7 +3572,6 @@ jobs:
if: contains(github.event.pull_request.labels.*.name, 'jepsen-test')
needs: [BuilderBinRelease]
uses: ./.github/workflows/jepsen.yml
FinishCheck:
needs:
- StyleCheck
@ -3509,6 +3579,7 @@ jobs:
- DockerServerImages
- CheckLabels
- BuilderReport
- BuilderSpecialReport
- FastTest
- FunctionalStatelessTestDebug0
- FunctionalStatelessTestDebug1
@ -3576,6 +3647,8 @@ jobs:
- SharedBuildSmokeTest
- CompatibilityCheck
- IntegrationTestsFlakyCheck
- SQLancerTestRelease
- SQLancerTestDebug
runs-on: [self-hosted, style-checker]
steps:
- name: Clear repository

View File

@ -615,6 +615,23 @@ jobs:
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH"
MarkReleaseReady:
needs:
- BuilderBinDarwin
- BuilderBinDarwinAarch64
- BuilderDebRelease
- BuilderDebAarch64
runs-on: [self-hosted, style-checker]
steps:
- name: Clear repository
run: |
sudo rm -fr "$GITHUB_WORKSPACE" && mkdir "$GITHUB_WORKSPACE"
- name: Check out repository code
uses: actions/checkout@v2
- name: Mark Commit Release Ready
run: |
cd "$GITHUB_WORKSPACE/tests/ci"
python3 mark_release_ready.py
##############################################################################################
########################### FUNCTIONAl STATELESS TESTS #######################################
##############################################################################################
@ -1888,6 +1905,7 @@ jobs:
- DockerServerImages
- BuilderReport
- BuilderSpecialReport
- MarkReleaseReady
- FunctionalStatelessTestDebug0
- FunctionalStatelessTestDebug1
- FunctionalStatelessTestDebug2

3
.gitignore vendored
View File

@ -154,3 +154,6 @@ website/package-lock.json
/programs/server/metadata
/programs/server/store
# temporary test files
tests/queries/0_stateless/test_*
tests/queries/0_stateless/*.binary

View File

@ -1,8 +1,10 @@
#if defined(OS_LINUX)
# include <sys/syscall.h>
#endif
#include <cstdlib>
#include <unistd.h>
#include <base/safeExit.h>
#include <base/defines.h> /// for THREAD_SANITIZER
[[noreturn]] void safeExit(int code)
{

View File

@ -8,6 +8,8 @@
#include <link.h> // ElfW
#include <errno.h>
#include "syscall.h"
#define ARRAY_SIZE(a) sizeof((a))/sizeof((a[0]))
/// Suppress TSan since it is possible for this code to be called from multiple threads,
@ -39,7 +41,9 @@ ssize_t __retry_read(int fd, void * buf, size_t count)
{
for (;;)
{
ssize_t ret = read(fd, buf, count);
// We cannot use the read syscall as it will be intercept by sanitizers, which aren't
// initialized yet. Emit syscall directly.
ssize_t ret = __syscall_ret(__syscall(SYS_read, fd, buf, count));
if (ret == -1)
{
if (errno == EINTR)
@ -90,6 +94,11 @@ static unsigned long NO_SANITIZE_THREAD __auxv_init_procfs(unsigned long type)
_Static_assert(sizeof(aux) < 4096, "Unexpected sizeof(aux)");
while (__retry_read(fd, &aux, sizeof(aux)) == sizeof(aux))
{
#if defined(__has_feature)
#if __has_feature(memory_sanitizer)
__msan_unpoison(&aux, sizeof(aux));
#endif
#endif
if (aux.a_type == AT_NULL)
{
break;

View File

@ -3,10 +3,20 @@ option (ENABLE_CLANG_TIDY "Use clang-tidy static analyzer" OFF)
if (ENABLE_CLANG_TIDY)
find_program (CLANG_TIDY_PATH NAMES "clang-tidy" "clang-tidy-15" "clang-tidy-14" "clang-tidy-13" "clang-tidy-12")
find_program (CLANG_TIDY_CACHE_PATH NAMES "clang-tidy-cache")
if (CLANG_TIDY_CACHE_PATH)
find_program (_CLANG_TIDY_PATH NAMES "clang-tidy" "clang-tidy-15" "clang-tidy-14" "clang-tidy-13" "clang-tidy-12")
# Why do we use ';' here?
# It's a cmake black magic: https://cmake.org/cmake/help/latest/prop_tgt/LANG_CLANG_TIDY.html#prop_tgt:%3CLANG%3E_CLANG_TIDY
# The CLANG_TIDY_PATH is passed to CMAKE_CXX_CLANG_TIDY, which follows CXX_CLANG_TIDY syntax.
set (CLANG_TIDY_PATH "${CLANG_TIDY_CACHE_PATH};${_CLANG_TIDY_PATH}" CACHE STRING "A combined command to run clang-tidy with caching wrapper")
else ()
find_program (CLANG_TIDY_PATH NAMES "clang-tidy" "clang-tidy-15" "clang-tidy-14" "clang-tidy-13" "clang-tidy-12")
endif ()
if (CLANG_TIDY_PATH)
message(STATUS
message (STATUS
"Using clang-tidy: ${CLANG_TIDY_PATH}.
The checks will be run during build process.
See the .clang-tidy file at the root directory to configure the checks.")
@ -15,11 +25,15 @@ if (ENABLE_CLANG_TIDY)
# clang-tidy requires assertions to guide the analysis
# Note that NDEBUG is set implicitly by CMake for non-debug builds
set(COMPILER_FLAGS "${COMPILER_FLAGS} -UNDEBUG")
set (COMPILER_FLAGS "${COMPILER_FLAGS} -UNDEBUG")
# The variable CMAKE_CXX_CLANG_TIDY will be set inside src and base directories with non third-party code.
# The variable CMAKE_CXX_CLANG_TIDY will be set inside the following directories with non third-party code.
# - base
# - programs
# - src
# - utils
# set (CMAKE_CXX_CLANG_TIDY "${CLANG_TIDY_PATH}")
else ()
message(${RECONFIGURE_MESSAGE_LEVEL} "clang-tidy is not found")
message (${RECONFIGURE_MESSAGE_LEVEL} "clang-tidy is not found")
endif ()
endif ()

2
contrib/NuRaft vendored

@ -1 +1 @@
Subproject commit 1be805e7cb2494aa8170015493474379b0362dfc
Subproject commit e4e746a24eb56861a86f3672771e3308d8c40722

View File

@ -91,6 +91,9 @@ ENV PATH="$PATH:/usr/local/go/bin"
ENV GOPATH=/workdir/go
ENV GOCACHE=/workdir/
RUN curl https://raw.githubusercontent.com/matus-chochlik/ctcache/7fd516e91c17779cbc6fc18bd119313d9532dd90/clang-tidy-cache -Lo /usr/bin/clang-tidy-cache \
&& chmod +x /usr/bin/clang-tidy-cache
RUN mkdir /workdir && chmod 777 /workdir
WORKDIR /workdir

View File

@ -258,6 +258,10 @@ def parse_env_variables(
if clang_tidy:
# 15G is not enough for tidy build
cache_maxsize = "25G"
# `CTCACHE_DIR` has the same purpose as the `CCACHE_DIR` above.
# It's there to have the clang-tidy cache embedded into our standard `CCACHE_DIR`
result.append("CTCACHE_DIR=/ccache/clang-tidy-cache")
result.append(f"CCACHE_MAXSIZE={cache_maxsize}")
if distcc_hosts:
@ -282,9 +286,7 @@ def parse_env_variables(
cmake_flags.append("-DENABLE_TESTS=1")
if shared_libraries:
cmake_flags.append(
"-DUSE_STATIC_LIBRARIES=0 -DSPLIT_SHARED_LIBRARIES=1"
)
cmake_flags.append("-DUSE_STATIC_LIBRARIES=0 -DSPLIT_SHARED_LIBRARIES=1")
# We can't always build utils because it requires too much space, but
# we have to build them at least in some way in CI. The shared library
# build is probably the least heavy disk-wise.

View File

@ -1,5 +1,5 @@
# docker build -t clickhouse/sqlancer-test .
FROM ubuntu:20.04
FROM ubuntu:22.04
# ARG for quick switch to a given ubuntu mirror
ARG apt_archive="http://archive.ubuntu.com"

View File

@ -11,13 +11,15 @@ def process_result(result_folder):
summary = []
paths = []
tests = [
"TLPWhere",
"TLPAggregate",
"TLPDistinct",
"TLPGroupBy",
"TLPHaving",
"TLPWhere",
"TLPWhereGroupBy",
"TLPDistinct",
"TLPAggregate",
"NoREC",
]
failed_tests = []
for test in tests:
err_path = "{}/{}.err".format(result_folder, test)
@ -33,15 +35,11 @@ def process_result(result_folder):
with open(err_path, "r") as f:
if "AssertionError" in f.read():
summary.append((test, "FAIL"))
failed_tests.append(test)
status = "failure"
else:
summary.append((test, "OK"))
logs_path = "{}/logs.tar.gz".format(result_folder)
if not os.path.exists(logs_path):
logging.info("No logs tar on path %s", logs_path)
else:
paths.append(logs_path)
stdout_path = "{}/stdout.log".format(result_folder)
if not os.path.exists(stdout_path):
logging.info("No stdout log on path %s", stdout_path)
@ -53,18 +51,23 @@ def process_result(result_folder):
else:
paths.append(stderr_path)
description = "SQLancer test run. See report"
description = "SQLancer run successfully"
if status == "failure":
description = f"Failed oracles: {failed_tests}"
return status, description, summary, paths
def write_results(results_file, status_file, results, status):
def write_results(
results_file, status_file, description_file, results, status, description
):
with open(results_file, "w") as f:
out = csv.writer(f, delimiter="\t")
out.writerows(results)
with open(status_file, "w") as f:
out = csv.writer(f, delimiter="\t")
out.writerow(status)
f.write(status + "\n")
with open(description_file, "w") as f:
f.write(description + "\n")
if __name__ == "__main__":
@ -72,13 +75,20 @@ if __name__ == "__main__":
parser = argparse.ArgumentParser(
description="ClickHouse script for parsing results of sqlancer test"
)
parser.add_argument("--in-results-dir", default="/test_output/")
parser.add_argument("--out-results-file", default="/test_output/test_results.tsv")
parser.add_argument("--out-status-file", default="/test_output/check_status.tsv")
parser.add_argument("--in-results-dir", default="/workspace/")
parser.add_argument("--out-results-file", default="/workspace/summary.tsv")
parser.add_argument("--out-description-file", default="/workspace/description.txt")
parser.add_argument("--out-status-file", default="/workspace/status.txt")
args = parser.parse_args()
state, description, test_results, logs = process_result(args.in_results_dir)
status, description, summary, logs = process_result(args.in_results_dir)
logging.info("Result parsed")
status = (state, description)
write_results(args.out_results_file, args.out_status_file, test_results, status)
write_results(
args.out_results_file,
args.out_status_file,
args.out_description_file,
summary,
status,
description,
)
logging.info("Result written")

View File

@ -1,33 +1,62 @@
#!/bin/bash
set -exu
trap "exit" INT TERM
set -e -x
function wget_with_retry
{
for _ in 1 2 3 4; do
if wget -nv -nd -c "$1";then
return 0
else
sleep 0.5
fi
done
return 1
}
dpkg -i package_folder/clickhouse-common-static_*.deb
dpkg -i package_folder/clickhouse-common-static-dbg_*.deb
dpkg -i package_folder/clickhouse-server_*.deb
dpkg -i package_folder/clickhouse-client_*.deb
if [ -z ${BINARY_URL_TO_DOWNLOAD+x} ]
then
echo "No BINARY_URL_TO_DOWNLOAD provided."
else
wget_with_retry "$BINARY_URL_TO_DOWNLOAD"
chmod +x /clickhouse
fi
service clickhouse-server start && sleep 5
if [[ -f "/clickhouse" ]]; then
echo "/clickhouse exists"
else
exit 1
fi
cd /workspace
/clickhouse server -P /workspace/clickhouse-server.pid -L /workspace/clickhouse-server.log -E /workspace/clickhouse-server.log.err --daemon
for _ in $(seq 1 60); do if [[ $(wget -q 'localhost:8123' -O-) == 'Ok.' ]]; then break ; else sleep 1; fi ; done
cd /sqlancer/sqlancer-master
export TIMEOUT=300
export NUM_QUERIES=1000
TIMEOUT=300
NUM_QUERIES=1000
NUM_THREADS=10
TESTS=( "TLPGroupBy" "TLPHaving" "TLPWhere" "TLPDistinct" "TLPAggregate" "NoREC" )
echo "${TESTS[@]}"
( java -jar target/sqlancer-*.jar --num-threads 10 --timeout-seconds $TIMEOUT --num-queries $NUM_QUERIES --username default --password "" clickhouse --oracle TLPWhere | tee /test_output/TLPWhere.out ) 3>&1 1>&2 2>&3 | tee /test_output/TLPWhere.err
( java -jar target/sqlancer-*.jar --num-threads 10 --timeout-seconds $TIMEOUT --num-queries $NUM_QUERIES --username default --password "" clickhouse --oracle TLPGroupBy | tee /test_output/TLPGroupBy.out ) 3>&1 1>&2 2>&3 | tee /test_output/TLPGroupBy.err
( java -jar target/sqlancer-*.jar --num-threads 10 --timeout-seconds $TIMEOUT --num-queries $NUM_QUERIES --username default --password "" clickhouse --oracle TLPHaving | tee /test_output/TLPHaving.out ) 3>&1 1>&2 2>&3 | tee /test_output/TLPHaving.err
( java -jar target/sqlancer-*.jar --num-threads 10 --timeout-seconds $TIMEOUT --num-queries $NUM_QUERIES --username default --password "" clickhouse --oracle TLPWhere --oracle TLPGroupBy | tee /test_output/TLPWhereGroupBy.out ) 3>&1 1>&2 2>&3 | tee /test_output/TLPWhereGroupBy.err
( java -jar target/sqlancer-*.jar --num-threads 10 --timeout-seconds $TIMEOUT --num-queries $NUM_QUERIES --username default --password "" clickhouse --oracle TLPDistinct | tee /test_output/TLPDistinct.out ) 3>&1 1>&2 2>&3 | tee /test_output/TLPDistinct.err
( java -jar target/sqlancer-*.jar --num-threads 10 --timeout-seconds $TIMEOUT --num-queries $NUM_QUERIES --username default --password "" clickhouse --oracle TLPAggregate | tee /test_output/TLPAggregate.out ) 3>&1 1>&2 2>&3 | tee /test_output/TLPAggregate.err
for TEST in "${TESTS[@]}"; do
echo "$TEST"
if [[ $(wget -q 'localhost:8123' -O-) == 'Ok.' ]]
then
echo "Server is OK"
( java -jar target/sqlancer-*.jar --log-each-select true --print-failed false --num-threads "$NUM_THREADS" --timeout-seconds "$TIMEOUT" --num-queries "$NUM_QUERIES" --username default --password "" clickhouse --oracle "$TEST" | tee "/workspace/$TEST.out" ) 3>&1 1>&2 2>&3 | tee "/workspace/$TEST.err"
else
touch "/workspace/$TEST.err" "/workspace/$TEST.out"
echo "Server is not responding" | tee /workspace/server_crashed.log
fi
done
service clickhouse stop
ls /workspace
pkill -F /workspace/clickhouse-server.pid || true
ls /var/log/clickhouse-server/
tar czf /test_output/logs.tar.gz -C /var/log/clickhouse-server/ .
tail -n 1000 /var/log/clickhouse-server/stderr.log > /test_output/stderr.log
tail -n 1000 /var/log/clickhouse-server/stdout.log > /test_output/stdout.log
tail -n 1000 /var/log/clickhouse-server/clickhouse-server.log > /test_output/clickhouse-server.log
for _ in $(seq 1 60); do if [[ $(wget -q 'localhost:8123' -O-) == 'Ok.' ]]; then sleep 1 ; else break; fi ; done
/process_sqlancer_result.py || echo -e "failure\tCannot parse results" > /test_output/check_status.tsv
ls /test_output
/process_sqlancer_result.py || echo -e "failure\tCannot parse results" > /workspace/check_status.tsv
ls /workspace

View File

@ -1,7 +1,7 @@
# docker build -t clickhouse/style-test .
FROM ubuntu:20.04
ARG ACT_VERSION=0.2.25
ARG ACTIONLINT_VERSION=1.6.8
ARG ACT_VERSION=0.2.33
ARG ACTIONLINT_VERSION=1.6.22
# ARG for quick switch to a given ubuntu mirror
ARG apt_archive="http://archive.ubuntu.com"

View File

@ -86,7 +86,7 @@ node1 :) SELECT materialize(hostName()) AS host, groupArray(n) FROM r.d GROUP BY
``` text
┌─hosts─┬─groupArray(n)─┐
│ node1 │ [1,3,5,7,9] │
│ node3 │ [1,3,5,7,9] │
│ node2 │ [0,2,4,6,8] │
└───────┴───────────────┘
```

View File

@ -68,36 +68,57 @@ In the results of `SELECT` query, the values of `AggregateFunction` type have im
## Example of an Aggregated Materialized View {#example-of-an-aggregated-materialized-view}
`AggregatingMergeTree` materialized view that watches the `test.visits` table:
We will create the table `test.visits` that contain the raw data:
``` sql
CREATE MATERIALIZED VIEW test.basic
ENGINE = AggregatingMergeTree() PARTITION BY toYYYYMM(StartDate) ORDER BY (CounterID, StartDate)
CREATE TABLE test.visits
(
StartDate DateTime64 NOT NULL,
CounterID UInt64,
Sign Nullable(Int32),
UserID Nullable(Int32)
) ENGINE = MergeTree ORDER BY (StartDate, CounterID);
```
`AggregatingMergeTree` materialized view that watches the `test.visits` table, and use the `AggregateFunction` type:
``` sql
CREATE MATERIALIZED VIEW test.mv_visits
(
StartDate DateTime64 NOT NULL,
CounterID UInt64,
Visits AggregateFunction(sum, Nullable(Int32)),
Users AggregateFunction(uniq, Nullable(Int32))
)
ENGINE = AggregatingMergeTree() ORDER BY (StartDate, CounterID)
AS SELECT
CounterID,
StartDate,
sumState(Sign) AS Visits,
CounterID,
sumState(Sign) AS Visits,
uniqState(UserID) AS Users
FROM test.visits
GROUP BY CounterID, StartDate;
GROUP BY StartDate, CounterID;
```
Inserting data into the `test.visits` table.
``` sql
INSERT INTO test.visits ...
INSERT INTO test.visits (StartDate, CounterID, Sign, UserID)
VALUES (1667446031, 1, 3, 4)
INSERT INTO test.visits (StartDate, CounterID, Sign, UserID)
VALUES (1667446031, 1, 6, 3)
```
The data are inserted in both the table and view `test.basic` that will perform the aggregation.
The data are inserted in both the table and the materialized view `test.mv_visits`.
To get the aggregated data, we need to execute a query such as `SELECT ... GROUP BY ...` from the view `test.basic`:
To get the aggregated data, we need to execute a query such as `SELECT ... GROUP BY ...` from the materialized view `test.mv_visits`:
``` sql
SELECT
StartDate,
sumMerge(Visits) AS Visits,
uniqMerge(Users) AS Users
FROM test.basic
FROM test.mv_visits
GROUP BY StartDate
ORDER BY StartDate;
```

View File

@ -4,25 +4,39 @@ sidebar_label: Cell Towers
sidebar_position: 3
title: "Cell Towers"
---
import ConnectionDetails from '@site/docs/en/_snippets/_gather_your_details_http.mdx';
import Tabs from '@theme/Tabs';
import TabItem from '@theme/TabItem';
import CodeBlock from '@theme/CodeBlock';
import ActionsMenu from '@site/docs/en/_snippets/_service_actions_menu.md';
import SQLConsoleDetail from '@site/docs/en/_snippets/_launch_sql_console.md';
import SupersetDocker from '@site/docs/en/_snippets/_add_superset_detail.md';
This dataset is from [OpenCellid](https://www.opencellid.org/) - The world's largest Open Database of Cell Towers.
## Goal
In this guide you will learn how to:
- Load the OpenCelliD data in Clickhouse
- Connect Apache Superset to ClickHouse
- Build a dashboard based on data available in the dataset
Here is a preview of the dashboard created in this guide:
![Dashboard of cell towers by radio type in mcc 204](@site/docs/en/getting-started/example-datasets/images/superset-cell-tower-dashboard.png)
## Get the Dataset {#get-the-dataset}
This dataset is from [OpenCelliD](https://www.opencellid.org/) - The world's largest Open Database of Cell Towers.
As of 2021, it contains more than 40 million records about cell towers (GSM, LTE, UMTS, etc.) around the world with their geographical coordinates and metadata (country code, network, etc).
OpenCelliD Project is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License, and we redistribute a snapshot of this dataset under the terms of the same license. The up-to-date version of the dataset is available to download after sign in.
## Get the Dataset {#get-the-dataset}
<Tabs groupId="deployMethod">
<TabItem value="serverless" label="ClickHouse Cloud" default>
### Load the sample data
ClickHouse Cloud provides an easy-button for uploading this dataset from S3. Log in to your ClickHouse Cloud organization, or create a free trial at [ClickHouse.cloud](https://clickhouse.cloud).
<ActionsMenu menu="Load Data" />
@ -30,13 +44,33 @@ Choose the **Cell Towers** dataset from the **Sample data** tab, and **Load data
![Load cell towers dataset](@site/docs/en/_snippets/images/cloud-load-data-sample.png)
Examine the schema of the cell_towers table:
### Examine the schema of the cell_towers table
```sql
DESCRIBE TABLE cell_towers
```
<SQLConsoleDetail />
This is the output of `DESCRIBE`. Down further in this guide the field type choices will be described.
```response
┌─name──────────┬─type──────────────────────────────────────────────────────────────────┬
│ radio │ Enum8('' = 0, 'CDMA' = 1, 'GSM' = 2, 'LTE' = 3, 'NR' = 4, 'UMTS' = 5) │
│ mcc │ UInt16 │
│ net │ UInt16 │
│ area │ UInt16 │
│ cell │ UInt64 │
│ unit │ Int16 │
│ lon │ Float64 │
│ lat │ Float64 │
│ range │ UInt32 │
│ samples │ UInt32 │
│ changeable │ UInt8 │
│ created │ DateTime │
│ updated │ DateTime │
│ averageSignal │ UInt8 │
└───────────────┴───────────────────────────────────────────────────────────────────────┴
```
</TabItem>
<TabItem value="selfmanaged" label="Self-managed">
@ -86,7 +120,7 @@ clickhouse-client --query "INSERT INTO cell_towers FORMAT CSVWithNames" < cell_t
</TabItem>
</Tabs>
## Example queries {#examples}
## Run some example queries {#examples}
1. A number of cell towers by type:
@ -127,13 +161,13 @@ SELECT mcc, count() FROM cell_towers GROUP BY mcc ORDER BY count() DESC LIMIT 10
10 rows in set. Elapsed: 0.019 sec. Processed 43.28 million rows, 86.55 MB (2.33 billion rows/s., 4.65 GB/s.)
```
So, the top countries are: the USA, Germany, and Russia.
Based on the above query and the [MCC list](https://en.wikipedia.org/wiki/Mobile_country_code), the countries with the most cell towers are: the USA, Germany, and Russia.
You may want to create an [External Dictionary](../../sql-reference/dictionaries/external-dictionaries/external-dicts.md) in ClickHouse to decode these values.
## Use case: Incorporate geo data {#use-case}
Using `pointInPolygon` function.
Using the [`pointInPolygon`](/docs/en/sql-reference/functions/geo/coordinates.md/#pointinpolygon) function.
1. Create a table where we will store polygons:
@ -224,6 +258,110 @@ WHERE pointInPolygon((lon, lat), (SELECT * FROM moscow))
1 rows in set. Elapsed: 0.067 sec. Processed 43.28 million rows, 692.42 MB (645.83 million rows/s., 10.33 GB/s.)
```
The data is also available for interactive queries in the [Playground](https://play.clickhouse.com/play?user=play), [example](https://play.clickhouse.com/play?user=play#U0VMRUNUIG1jYywgY291bnQoKSBGUk9NIGNlbGxfdG93ZXJzIEdST1VQIEJZIG1jYyBPUkRFUiBCWSBjb3VudCgpIERFU0M=).
## Review of the schema
Although you cannot create temporary tables there.
Before building visualizations in Superset have a look at the columns that you will use. This dataset primarily provides the location (Longitude and Latitude) and radio types at mobile cellular towers worldwide. The column descriptions can be found in the [community forum](https://community.opencellid.org/t/documenting-the-columns-in-the-downloadable-cells-database-csv/186). The columns used in the visualizations that will be built are described below
Here is a description of the columns taken from the OpenCelliD forum:
| Column | Description |
|--------------|--------------------------------------------------------|
| radio | Technology generation: CDMA, GSM, UMTS, 5G NR |
| mcc | Mobile Country Code: `204` is The Netherlands |
| lon | Longitude: With Latitude, approximate tower location |
| lat | Latitude: With Longitude, approximate tower location |
:::tip mcc
To find your MCC check [Mobile network codes](https://en.wikipedia.org/wiki/Mobile_country_code), and use the three digits in the **Mobile country code** column.
:::
The schema for this table was designed for compact storage on disk and query speed.
- The `radio` data is stored as an `Enum8` (`UInt8`) rather than a string.
- `mcc` or Mobile country code, is stored as a `UInt16` as we know the range is 1 - 999.
- `lon` and `lat` are `Float64`.
None of the other fields are used in the queries or visualizations in this guide, but they are described in the forum linked above if you are interested.
## Build visualizations with Apache Superset
Superset is easy to run from Docker. If you already have Superset running, all you need to do is add ClickHouse Connect with `pip install clickhouse-connect`. If you need to install Superset open the **Launch Apache Superset in Docker** directly below.
<SupersetDocker />
To build a Superset dashboard using the OpenCelliD dataset you should:
- Add your ClickHouse service as a Superset **database**
- Add the table **cell_towers** as a Superset **dataset**
- Create some **charts**
- Add the charts to a **dashboard**
### Add your ClickHouse service as a Superset database
<ConnectionDetails />
In Superset a database can be added by choosing the database type, and then providing the connection details. Open Superset and look for the **+**, it has a menu with **Data** and then **Connect database** options.
![Add a database](@site/docs/en/getting-started/example-datasets/images/superset-add.png)
Choose **ClickHouse Connect** from the list:
![Choose clickhouse connect as database type](@site/docs/en/getting-started/example-datasets/images/superset-choose-a-database.png)
:::note
If **ClickHouse Connect** is not one of your options, then you will need to install it. The comand is `pip install clickhouse-connect`, and more info is [available here](https://pypi.org/project/clickhouse-connect/).
:::
#### Add your connection details:
:::tip
Make sure that you set **SSL** on when connecting to ClickHouse Cloud or other ClickHouse systems that enforce the use of SSL.
:::
![Add ClickHouse as a Superset datasource](@site/docs/en/getting-started/example-datasets/images/superset-connect-a-database.png)
### Add the table **cell_towers** as a Superset **dataset**
In Superset a **dataset** maps to a table within a database. Click on add a dataset and choose your ClickHouse service, the database containing your table (`default`), and choose the `cell_towers` table:
![Add cell_towers table as a dataset](@site/docs/en/getting-started/example-datasets/images/superset-add-dataset.png)
### Create some **charts**
When you choose to add a chart in Superset you have to specify the dataset (`cell_towers`) and the chart type. Since the OpenCelliD dataset provides longitude and latitude coordinates for cell towers we will create a **Map** chart. The **deck.gL Scatterplot** type is suited to this dataset as it works well with dense data points on a map.
![Create a map in Superset](@site/docs/en/getting-started/example-datasets/images/superset-create-map.png)
#### Specify the query used for the map
A deck.gl Scatterplot requires a longitude and latitude, and one or more filters can also be applied to the query. In this example two filters are applied, one for cell towers with UMTS radios, and one for the Mobile country code assigned to The Netherlands.
The fields `lon` and `lat` contain the longitude and latitude:
![Specify longitude and latitude fields](@site/docs/en/getting-started/example-datasets/images/superset-lon-lat.png)
Add a filter with `mcc` = `204` (or substitute any other `mcc` value):
![Filter on MCC 204](@site/docs/en/getting-started/example-datasets/images/superset-mcc-204.png)
Add a filter with `radio` = `'UMTS'` (or substitute any other `radio` value, you can see the choices in the output of `DESCRIBE TABLE cell_towers`):
![Filter on radio = UMTS](@site/docs/en/getting-started/example-datasets/images/superset-radio-umts.png)
This is the full configuration for the chart that filters on `radio = 'UMTS'` and `mcc = 204`:
![Chart for UMTS radios in MCC 204](@site/docs/en/getting-started/example-datasets/images/superset-umts-netherlands.png)
Click on **UPDATE CHART** to render the visualization.
### Add the charts to a **dashboard**
This screenshot shows cell tower locations with LTE, UMTS, and GSM radios. The charts are all created in the same way and they are added to a dashboard.
![Dashboard of cell towers by radio type in mcc 204](@site/docs/en/getting-started/example-datasets/images/superset-cell-tower-dashboard.png)
:::tip
The data is also available for interactive queries in the [Playground](https://play.clickhouse.com/play?user=play).
This [example](https://play.clickhouse.com/play?user=play#U0VMRUNUIG1jYywgY291bnQoKSBGUk9NIGNlbGxfdG93ZXJzIEdST1VQIEJZIG1jYyBPUkRFUiBCWSBjb3VudCgpIERFU0M=) will populate the username and even the query for you.
Although you cannot create tables in the Playground, you can run all of the queries and even use Superset (adjust the hostname and port number).
:::

Binary file not shown.

After

Width:  |  Height:  |  Size: 35 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 35 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 475 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 53 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 73 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 290 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 38 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 12 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 12 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 46 KiB

View File

@ -4,7 +4,7 @@ sidebar_label: Recipes Dataset
title: "Recipes Dataset"
---
RecipeNLG dataset is available for download [here](https://recipenlg.cs.put.poznan.pl/dataset). It contains 2.2 million recipes. The size is slightly less than 1 GB.
The RecipeNLG dataset is available for download [here](https://recipenlg.cs.put.poznan.pl/dataset). It contains 2.2 million recipes. The size is slightly less than 1 GB.
## Download and Unpack the Dataset

View File

@ -1,9 +1,5 @@
---
slug: /en/operations/troubleshooting
sidebar_position: 46
sidebar_label: Troubleshooting
title: Troubleshooting
---
[//]: # (This file is included in FAQ > Troubleshooting)
- [Installation](#troubleshooting-installation-errors)
- [Connecting to the server](#troubleshooting-accepts-no-connections)
@ -28,18 +24,34 @@ sudo apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv 8919F6BD2B48D7
sudo apt-get update
```
### You Get the Unsupported Architecture Warning with Apt-get {#you-get-the-unsupported-architecture-warning-with-apt-get}
### You Get Different Warnings with `apt-get update` {#you-get-different-warnings-with-apt-get-update}
- The completed warning message is as follows:
- The completed warning messages are as one of following:
```
N: Skipping acquire of configured file 'main/binary-i386/Packages' as repository 'https://packages.clickhouse.com/deb stable InRelease' doesn't support architecture 'i386'
```
```
E: Failed to fetch https://packages.clickhouse.com/deb/dists/stable/main/binary-amd64/Packages.gz File has unexpected size (30451 != 28154). Mirror sync in progress?
```
```
E: Repository 'https://packages.clickhouse.com/deb stable InRelease' changed its 'Origin' value from 'Artifactory' to 'ClickHouse'
E: Repository 'https://packages.clickhouse.com/deb stable InRelease' changed its 'Label' value from 'Artifactory' to 'ClickHouse'
N: Repository 'https://packages.clickhouse.com/deb stable InRelease' changed its 'Suite' value from 'stable' to ''
N: This must be accepted explicitly before updates for this repository can be applied. See apt-secure(8) manpage for details.
```
```
Err:11 https://packages.clickhouse.com/deb stable InRelease
400 Bad Request [IP: 172.66.40.249 443]
```
To resolve the above issue, please use the following script:
```bash
sudo rm /var/lib/apt/lists/packages.clickhouse.com_* /var/lib/dpkg/arch
sudo rm /var/lib/apt/lists/packages.clickhouse.com_* /var/lib/dpkg/arch /var/lib/apt/lists/partial/packages.clickhouse.com_*
sudo apt-get clean
sudo apt-get autoclean
```

View File

@ -126,7 +126,7 @@ clickhouse keeper --config /etc/your_path_to_config/config.xml
ClickHouse Keeper also provides 4lw commands which are almost the same with Zookeeper. Each command is composed of four letters such as `mntr`, `stat` etc. There are some more interesting commands: `stat` gives some general information about the server and connected clients, while `srvr` and `cons` give extended details on server and connections respectively.
The 4lw commands has a white list configuration `four_letter_word_white_list` which has default value `conf,cons,crst,envi,ruok,srst,srvr,stat,wchc,wchs,dirs,mntr,isro`.
The 4lw commands has a white list configuration `four_letter_word_white_list` which has default value `conf,cons,crst,envi,ruok,srst,srvr,stat,wchs,dirs,mntr,isro,rcvr,apiv,csnp,lgif`.
You can issue the commands to ClickHouse Keeper via telnet or nc, at the client port.
@ -309,7 +309,26 @@ Sessions with Ephemerals (1):
/clickhouse/task_queue/ddl
```
## [experimental] Migration from ZooKeeper {#migration-from-zookeeper}
- `csnp`: Schedule a snapshot creation task. Return the last committed log index of the scheduled snapshot if success or `Failed to schedule snapshot creation task.` if failed. Note that `lgif` command can help you determine whether the snapshot is done.
```
100
```
- `lgif`: Keeper log information. `first_log_idx` : my first log index in log store; `first_log_term` : my first log term; `last_log_idx` : my last log index in log store; `last_log_term` : my last log term; `last_committed_log_idx` : my last committed log index in state machine; `leader_committed_log_idx` : leader's committed log index from my perspective; `target_committed_log_idx` : target log index should be committed to; `last_snapshot_idx` : the largest committed log index in last snapshot.
```
first_log_idx 1
first_log_term 1
last_log_idx 101
last_log_term 1
last_committed_log_idx 100
leader_committed_log_idx 101
target_committed_log_idx 101
last_snapshot_idx 50
```
## Migration from ZooKeeper {#migration-from-zookeeper}
Seamlessly migration from ZooKeeper to ClickHouse Keeper is impossible you have to stop your ZooKeeper cluster, convert data and start ClickHouse Keeper. `clickhouse-keeper-converter` tool allows converting ZooKeeper logs and snapshots to ClickHouse Keeper snapshot. It works only with ZooKeeper > 3.4. Steps for migration:

View File

@ -2939,7 +2939,7 @@ Possible values:
- 0 — Projection optimization disabled.
- 1 — Projection optimization enabled.
Default value: `0`.
Default value: `1`.
## force_optimize_projection {#force-optimize-projection}

View File

@ -24,6 +24,7 @@ Columns:
- `DOUBLE_SHA1_PASSWORD`
- `LDAP`
- `KERBEROS`
- `SSL_CERTIFICATE`
- `profiles` ([Array](../../sql-reference/data-types/array.md)([LowCardinality(String)](../../sql-reference/data-types/lowcardinality.md))) — The list of profiles set for all roles and/or users.
- `roles` ([Array](../../sql-reference/data-types/array.md)([LowCardinality(String)](../../sql-reference/data-types/lowcardinality.md))) — The list of roles to which the profile is applied.
- `settings` ([Array](../../sql-reference/data-types/array.md)([Tuple](../../sql-reference/data-types/tuple.md)([LowCardinality(String)](../../sql-reference/data-types/lowcardinality.md), [String](../../sql-reference/data-types/string.md)))) — Settings that were changed when the client logged in/out.

View File

@ -12,7 +12,7 @@ Columns:
- `storage` ([String](../../sql-reference/data-types/string.md)) — Path to the storage of users. Configured in the `access_control_path` parameter.
- `auth_type` ([Enum8](../../sql-reference/data-types/enum.md)('no_password' = 0,'plaintext_password' = 1, 'sha256_password' = 2, 'double_sha1_password' = 3)) — Shows the authentication type. There are multiple ways of user identification: with no password, with plain text password, with [SHA256](https://ru.wikipedia.org/wiki/SHA-2)-encoded password or with [double SHA-1](https://ru.wikipedia.org/wiki/SHA-1)-encoded password.
- `auth_type` ([Enum8](../../sql-reference/data-types/enum.md)('no_password' = 0,'plaintext_password' = 1, 'sha256_password' = 2, 'double_sha1_password' = 3, 'ldap' = 4, 'kerberos' = 5, 'ssl_certificate' = 6)) — Shows the authentication type. There are multiple ways of user identification: with no password, with plain text password, with [SHA256](https://ru.wikipedia.org/wiki/SHA-2)-encoded password or with [double SHA-1](https://ru.wikipedia.org/wiki/SHA-1)-encoded password.
- `auth_params` ([String](../../sql-reference/data-types/string.md)) — Authentication parameters in the JSON format depending on the `auth_type`.

View File

@ -109,56 +109,38 @@ In the report you can find:
`clickhouse-benchmark` can compare performances for two running ClickHouse servers.
To use the comparison mode, specify endpoints of both servers by two pairs of `--host`, `--port` keys. Keys matched together by position in arguments list, the first `--host` is matched with the first `--port` and so on. `clickhouse-benchmark` establishes connections to both servers, then sends queries. Each query addressed to a randomly selected server. The results are shown for each server separately.
To use the comparison mode, specify endpoints of both servers by two pairs of `--host`, `--port` keys. Keys matched together by position in arguments list, the first `--host` is matched with the first `--port` and so on. `clickhouse-benchmark` establishes connections to both servers, then sends queries. Each query addressed to a randomly selected server. The results are shown in a table.
## Example {#clickhouse-benchmark-example}
``` bash
$ echo "SELECT * FROM system.numbers LIMIT 10000000 OFFSET 10000000" | clickhouse-benchmark -i 10
$ echo "SELECT * FROM system.numbers LIMIT 10000000 OFFSET 10000000" | clickhouse-benchmark --host=localhost --port=9001 --host=localhost --port=9000 -i 10
```
``` text
Loaded 1 queries.
Queries executed: 6.
Queries executed: 5.
localhost:9000, queries 6, QPS: 6.153, RPS: 123398340.957, MiB/s: 941.455, result RPS: 61532982.200, result MiB/s: 469.459.
localhost:9001, queries 2, QPS: 3.764, RPS: 75446929.370, MiB/s: 575.614, result RPS: 37639659.982, result MiB/s: 287.168.
localhost:9000, queries 3, QPS: 3.815, RPS: 76466659.385, MiB/s: 583.394, result RPS: 38148392.297, result MiB/s: 291.049.
0.000% 0.159 sec.
10.000% 0.159 sec.
20.000% 0.159 sec.
30.000% 0.160 sec.
40.000% 0.160 sec.
50.000% 0.162 sec.
60.000% 0.164 sec.
70.000% 0.165 sec.
80.000% 0.166 sec.
90.000% 0.166 sec.
95.000% 0.167 sec.
99.000% 0.167 sec.
99.900% 0.167 sec.
99.990% 0.167 sec.
0.000% 0.258 sec. 0.250 sec.
10.000% 0.258 sec. 0.250 sec.
20.000% 0.258 sec. 0.250 sec.
30.000% 0.258 sec. 0.267 sec.
40.000% 0.258 sec. 0.267 sec.
50.000% 0.273 sec. 0.267 sec.
60.000% 0.273 sec. 0.267 sec.
70.000% 0.273 sec. 0.267 sec.
80.000% 0.273 sec. 0.269 sec.
90.000% 0.273 sec. 0.269 sec.
95.000% 0.273 sec. 0.269 sec.
99.000% 0.273 sec. 0.269 sec.
99.900% 0.273 sec. 0.269 sec.
99.990% 0.273 sec. 0.269 sec.
Queries executed: 10.
localhost:9000, queries 10, QPS: 6.082, RPS: 121959604.568, MiB/s: 930.478, result RPS: 60815551.642, result MiB/s: 463.986.
0.000% 0.159 sec.
10.000% 0.159 sec.
20.000% 0.160 sec.
30.000% 0.163 sec.
40.000% 0.164 sec.
50.000% 0.165 sec.
60.000% 0.166 sec.
70.000% 0.166 sec.
80.000% 0.167 sec.
90.000% 0.167 sec.
95.000% 0.170 sec.
99.000% 0.172 sec.
99.900% 0.172 sec.
99.990% 0.172 sec.
No difference proven at 99.5% confidence
```
[Original article](https://clickhouse.com/docs/en/operations/utilities/clickhouse-benchmark.md) <!--hide-->

View File

@ -303,17 +303,25 @@ or
CREATE DICTIONARY somedict (
id UInt64,
first Date,
last Date
last Date,
advertiser_id UInt64
)
PRIMARY KEY id
SOURCE(CLICKHOUSE(TABLE 'date_table'))
LIFETIME(MIN 1 MAX 1000)
LAYOUT(RANGE_HASHED())
RANGE(MIN first MAX last)
```
To work with these dictionaries, you need to pass an additional argument to the `dictGetT` function, for which a range is selected:
To work with these dictionaries, you need to pass an additional argument to the `dictGet` function, for which a range is selected:
``` sql
dictGetT('dict_name', 'attr_name', id, date)
dictGet('dict_name', 'attr_name', id, date)
```
Query example:
``` sql
SELECT dictGet('somedict', 'advertiser_id', 1, '2022-10-20 23:20:10.000'::DateTime64::UInt64);
```
This function returns the value for the specified `id`s and the date range that includes the passed date.

View File

@ -8,70 +8,69 @@ title: "Geo Functions"
## Geographical Coordinates Functions
- [greatCircleDistance](./coordinates.md#greatCircleDistance)
- [geoDistance](./coordinates.md#geoDistance)
- [greatCircleAngle](./coordinates.md#greatCircleAngle)
- [pointInEllipses](./coordinates.md#pointInEllipses)
- [pointInPolygon](./coordinates.md#pointInPolygon)
- [greatCircleDistance](./coordinates.md#greatcircledistance)
- [geoDistance](./coordinates.md#geodistance)
- [greatCircleAngle](./coordinates.md#greatcircleangle)
- [pointInEllipses](./coordinates.md#pointinellipses)
- [pointInPolygon](./coordinates.md#pointinpolygon)
## Geohash Functions
- [geohashEncode](./geohash.md#geohashEncode)
- [geohashDecode](./geohash.md#geohashDecode)
- [geohashesInBox](./geohash.md#geohashesInBox)
- [geohashEncode](./geohash.md#geohashencode)
- [geohashDecode](./geohash.md#geohashdecode)
- [geohashesInBox](./geohash.md#geohashesinbox)
## H3 Indexes Functions
- [h3IsValid](./h3.md#h3IsValid)
- [h3GetResolution](./h3.md#h3GetResolution)
- [h3EdgeAngle](./h3.md#h3EdgeAngle)
- [h3EdgeLengthM](./h3.md#h3EdgeLengthM)
- [h3EdgeLengthKm](./h3.md#h3EdgeLengthKm)
- [geoToH3](./h3.md#geoToH3)
- [h3ToGeo](./h3.md#h3ToGeo)
- [h3ToGeoBoundary](./h3.md#h3ToGeoBoundary)
- [h3kRing](./h3.md#h3kRing)
- [h3GetBaseCell](./h3.md#h3GetBaseCell)
- [h3HexAreaM2](./h3.md#h3HexAreaM2)
- [h3HexAreaKm2](./h3.md#h3HexAreaKm2)
- [h3IndexesAreNeighbors](./h3.md#h3IndexesAreNeighbors)
- [h3ToChildren](./h3.md#h3ToChildren)
- [h3ToParent](./h3.md#h3ToParent)
- [h3ToString](./h3.md#h3ToString)
- [stringToH3](./h3.md#stringToH3)
- [h3GetResolution](./h3.md#h3GetResolution)
- [h3IsResClassIII](./h3.md#h3IsResClassIII)
- [h3IsPentagon](./h3.md#h3IsPentagon)
- [h3GetFaces](./h3.md#h3GetFaces)
- [h3CellAreaM2](./h3.md#h3CellAreaM2)
- [h3CellAreaRads2](./h3.md#h3CellAreaRads2)
- [h3ToCenterChild](./h3.md#h3ToCenterChild)
- [h3ExactEdgeLengthM](./h3.md#h3ExactEdgeLengthM)
- [h3ExactEdgeLengthKm](./h3.md#h3ExactEdgeLengthKm)
- [h3ExactEdgeLengthRads](./h3.md#h3ExactEdgeLengthRads)
- [h3NumHexagons](./h3.md#h3NumHexagons)
- [h3Line](./h3.md#h3Line)
- [h3Distance](./h3.md#h3Distance)
- [h3HexRing](./h3.md#h3HexRing)
- [h3GetUnidirectionalEdge](./h3.md#h3GetUnidirectionalEdge)
- [h3UnidirectionalEdgeIsValid](./h3.md#h3UnidirectionalEdgeIsValid)
- [h3GetOriginIndexFromUnidirectionalEdge](./h3.md#h3GetOriginIndexFromUnidirectionalEdge)
- [h3GetDestinationIndexFromUnidirectionalEdge](./h3.md#h3GetDestinationIndexFromUnidirectionalEdge)
- [h3GetIndexesFromUnidirectionalEdge](./h3.md#h3GetIndexesFromUnidirectionalEdge)
- [h3GetUnidirectionalEdgesFromHexagon](./h3.md#h3GetUnidirectionalEdgesFromHexagon)
- [h3GetUnidirectionalEdgeBoundary](./h3.md#h3GetUnidirectionalEdgeBoundary)
- [h3IsValid](./h3.md#h3isvalid)
- [h3GetResolution](./h3.md#h3getresolution)
- [h3EdgeAngle](./h3.md#h3edgeangle)
- [h3EdgeLengthM](./h3.md#h3edgelengthm)
- [h3EdgeLengthKm](./h3.md#h3edgelengthkm)
- [geoToH3](./h3.md#geotoh3)
- [h3ToGeo](./h3.md#h3togeo)
- [h3ToGeoBoundary](./h3.md#h3togeoboundary)
- [h3kRing](./h3.md#h3kring)
- [h3GetBaseCell](./h3.md#h3getbasecell)
- [h3HexAreaM2](./h3.md#h3hexaream2)
- [h3HexAreaKm2](./h3.md#h3hexareakm2)
- [h3IndexesAreNeighbors](./h3.md#h3indexesareneighbors)
- [h3ToChildren](./h3.md#h3tochildren)
- [h3ToParent](./h3.md#h3toparent)
- [h3ToString](./h3.md#h3tostring)
- [stringToH3](./h3.md#stringtoh3)
- [h3GetResolution](./h3.md#h3getresolution)
- [h3IsResClassIII](./h3.md#h3isresclassiii)
- [h3IsPentagon](./h3.md#h3ispentagon)
- [h3GetFaces](./h3.md#h3getfaces)
- [h3CellAreaM2](./h3.md#h3cellaream2)
- [h3CellAreaRads2](./h3.md#h3cellarearads2)
- [h3ToCenterChild](./h3.md#h3tocenterchild)
- [h3ExactEdgeLengthM](./h3.md#h3exactedgelengthm)
- [h3ExactEdgeLengthKm](./h3.md#h3exactedgelengthkm)
- [h3ExactEdgeLengthRads](./h3.md#h3exactedgelengthrads)
- [h3NumHexagons](./h3.md#h3numhexagons)
- [h3Line](./h3.md#h3line)
- [h3Distance](./h3.md#h3distance)
- [h3HexRing](./h3.md#h3hexring)
- [h3GetUnidirectionalEdge](./h3.md#h3getunidirectionaledge)
- [h3UnidirectionalEdgeIsValid](./h3.md#h3unidirectionaledgeisvalid)
- [h3GetOriginIndexFromUnidirectionalEdge](./h3.md#h3getoriginindexfromunidirectionaledge)
- [h3GetDestinationIndexFromUnidirectionalEdge](./h3.md#h3getdestinationindexfromunidirectionaledge)
- [h3GetIndexesFromUnidirectionalEdge](./h3.md#h3getindexesfromunidirectionaledge)
- [h3GetUnidirectionalEdgesFromHexagon](./h3.md#h3getunidirectionaledgesfromhexagon)
- [h3GetUnidirectionalEdgeBoundary](./h3.md#h3getunidirectionaledgeboundary)
## S2 Index Functions
- [geoToS2](./s2.md#geoToS2)
- [s2ToGeo](./s2.md#s2ToGeo)
- [s2GetNeighbors](./s2.md#s2GetNeighbors)
- [s2CellsIntersect](./s2.md#s2CellsIntersect)
- [s2CapContains](./s2.md#s2CapContains)
- [s2CapUnion](./s2.md#s2CapUnion)
- [s2RectAdd](./s2.md#s2RectAdd)
- [s2RectContains](./s2.md#s2RectContains)
- [s2RectUinion](./s2.md#s2RectUinion)
- [s2RectIntersection](./s2.md#s2RectIntersection)
- [geoToS2](./s2.md#geotos2)
- [s2ToGeo](./s2.md#s2togeo)
- [s2GetNeighbors](./s2.md#s2getneighbors)
- [s2CellsIntersect](./s2.md#s2cellsintersect)
- [s2CapContains](./s2.md#s2capcontains)
- [s2CapUnion](./s2.md#s2capunion)
- [s2RectAdd](./s2.md#s2rectadd)
- [s2RectContains](./s2.md#s2rectcontains)
- [s2RectUnion](./s2.md#s2rectunion)
- [s2RectIntersection](./s2.md#s2rectintersection)
[Original article](https://clickhouse.com/docs/en/sql-reference/functions/geo/) <!--hide-->

View File

@ -593,6 +593,27 @@ LIMIT 10
└────────────────┴─────────┘
```
## formatReadableDecimalSize(x)
Accepts the size (number of bytes). Returns a rounded size with a suffix (KB, MB, etc.) as a string.
Example:
``` sql
SELECT
arrayJoin([1, 1024, 1024*1024, 192851925]) AS filesize_bytes,
formatReadableDecimalSize(filesize_bytes) AS filesize
```
``` text
┌─filesize_bytes─┬─filesize───┐
│ 1 │ 1.00 B │
│ 1024 │ 1.02 KB │
│ 1048576 │ 1.05 MB │
│ 192851925 │ 192.85 MB │
└────────────────┴────────────┘
```
## formatReadableSize(x)
Accepts the size (number of bytes). Returns a rounded size with a suffix (KiB, MiB, etc.) as a string.

View File

@ -571,13 +571,13 @@ Similar to base58Decode, but returns an empty string in case of error.
## base64Encode(s)
Encodes s string into base64
Encodes s FixedString or String into base64.
Alias: `TO_BASE64`.
## base64Decode(s)
Decode base64-encoded string s into original string. In case of failure raises an exception.
Decode base64-encoded FixedString or String s into original string. In case of failure raises an exception.
Alias: `FROM_BASE64`.
@ -1150,3 +1150,13 @@ A text with tags .
The content within <b>CDATA</b>
Do Nothing for 2 Minutes 2:00 &nbsp;
```
## ascii(s) {#ascii}
Returns the ASCII code point of the first character of str. The result type is Int32.
If s is empty, the result is 0. If the first character is not an ASCII character or not part of the Latin-1 Supplement range of UTF-16, the result is undefined.

View File

@ -6,28 +6,29 @@ sidebar_label: For Replacing in Strings
# Functions for Searching and Replacing in Strings
:::note
:::note
Functions for [searching](../../sql-reference/functions/string-search-functions.md) and [other manipulations with strings](../../sql-reference/functions/string-functions.md) are described separately.
:::
## replaceOne(haystack, pattern, replacement)
Replaces the first occurrence, if it exists, of the pattern substring in haystack with the replacement substring.
Hereafter, pattern and replacement must be constants.
Replaces the first occurrence of the substring pattern (if it exists) in haystack by the replacement string.
pattern and replacement must be constants.
## replaceAll(haystack, pattern, replacement), replace(haystack, pattern, replacement)
Replaces all occurrences of the pattern substring in haystack with the replacement substring.
Replaces all occurrences of the substring pattern in haystack by the replacement string.
## replaceRegexpOne(haystack, pattern, replacement)
Replacement using the pattern regular expression. A re2 regular expression.
Replaces only the first occurrence, if it exists.
A pattern can be specified as replacement. This pattern can include substitutions `\0-\9`.
The substitution `\0` includes the entire regular expression. Substitutions `\1-\9` correspond to the subpattern numbers.To use the `\` character in a template, escape it using `\`.
Also keep in mind that a string literal requires an extra escape.
Replaces the first occurrence of the substring matching the regular expression pattern in haystack by the replacement string.
pattern must be a constant [re2 regular expression](https://github.com/google/re2/wiki/Syntax).
replacement must be a plain constant string or a constant string containing substitutions `\0-\9`.
Substitutions `\1-\9` correspond to the 1st to 9th capturing group (submatch), substitution `\0` corresponds to the entire match.
To use a verbatim `\` character in the pattern or replacement string, escape it using `\`.
Also keep in mind that string literals require an extra escaping.
Example 1. Converting the date to American format:
Example 1. Converting ISO dates to American format:
``` sql
SELECT DISTINCT
@ -62,7 +63,7 @@ SELECT replaceRegexpOne('Hello, World!', '.*', '\\0\\0\\0\\0\\0\\0\\0\\0\\0\\0')
## replaceRegexpAll(haystack, pattern, replacement)
This does the same thing, but replaces all the occurrences. Example:
Like replaceRegexpOne, but replaces all occurrences of the pattern. Example:
``` sql
SELECT replaceRegexpAll('Hello, World!', '.', '\\0\\0') AS res

View File

@ -7,18 +7,26 @@ title: "Manipulating Projections"
The following operations with [projections](../../../engines/table-engines/mergetree-family/mergetree.md#projections) are available:
- `ALTER TABLE [db].name ADD PROJECTION name ( SELECT <COLUMN LIST EXPR> [GROUP BY] [ORDER BY] )` - Adds projection description to tables metadata.
## ADD PROJECTION
- `ALTER TABLE [db].name DROP PROJECTION name` - Removes projection description from tables metadata and deletes projection files from disk. Implemented as a [mutation](../../../sql-reference/statements/alter/index.md#mutations).
`ALTER TABLE [db].name ADD PROJECTION name ( SELECT <COLUMN LIST EXPR> [GROUP BY] [ORDER BY] )` - Adds projection description to tables metadata.
- `ALTER TABLE [db.]table MATERIALIZE PROJECTION name IN PARTITION partition_name` - The query rebuilds the projection `name` in the partition `partition_name`. Implemented as a [mutation](../../../sql-reference/statements/alter/index.md#mutations).
## DROP PROJECTION
- `ALTER TABLE [db.]table CLEAR PROJECTION name IN PARTITION partition_name` - Deletes projection files from disk without removing description. Implemented as a [mutation](../../../sql-reference/statements/alter/index.md#mutations).
`ALTER TABLE [db].name DROP PROJECTION name` - Removes projection description from tables metadata and deletes projection files from disk. Implemented as a [mutation](../../../sql-reference/statements/alter/index.md#mutations).
## MATERIALIZE PROJECTION
`ALTER TABLE [db.]table MATERIALIZE PROJECTION name IN PARTITION partition_name` - The query rebuilds the projection `name` in the partition `partition_name`. Implemented as a [mutation](../../../sql-reference/statements/alter/index.md#mutations).
## CLEAR PROJECTION
`ALTER TABLE [db.]table CLEAR PROJECTION name IN PARTITION partition_name` - Deletes projection files from disk without removing description. Implemented as a [mutation](../../../sql-reference/statements/alter/index.md#mutations).
The commands `ADD`, `DROP` and `CLEAR` are lightweight in a sense that they only change metadata or remove files.
Also, they are replicated, syncing projections metadata via ZooKeeper.
Also, they are replicated, syncing projections metadata via ClickHouse Keeper or ZooKeeper.
:::note
Projection manipulation is supported only for tables with [`*MergeTree`](../../../engines/table-engines/mergetree-family/mergetree.md) engine (including [replicated](../../../engines/table-engines/mergetree-family/replication.md) variants).

View File

@ -12,7 +12,7 @@ Syntax:
``` sql
ALTER USER [IF EXISTS] name1 [ON CLUSTER cluster_name1] [RENAME TO new_name1]
[, name2 [ON CLUSTER cluster_name2] [RENAME TO new_name2] ...]
[NOT IDENTIFIED | IDENTIFIED {[WITH {no_password | plaintext_password | sha256_password | sha256_hash | double_sha1_password | double_sha1_hash}] BY {'password' | 'hash'}} | {WITH ldap SERVER 'server_name'} | {WITH kerberos [REALM 'realm']}]
[NOT IDENTIFIED | IDENTIFIED {[WITH {no_password | plaintext_password | sha256_password | sha256_hash | double_sha1_password | double_sha1_hash}] BY {'password' | 'hash'}} | {WITH ldap SERVER 'server_name'} | {WITH kerberos [REALM 'realm']} | {WITH ssl_certificate CN 'common_name'}]
[[ADD | DROP] HOST {LOCAL | NAME 'name' | REGEXP 'name_regexp' | IP 'address' | LIKE 'pattern'} [,...] | ANY | NONE]
[DEFAULT ROLE role [,...] | ALL | ALL EXCEPT role [,...] ]
[GRANTEES {user | role | ANY | NONE} [,...] [EXCEPT {user | role} [,...]]]

View File

@ -8,7 +8,7 @@ title: "CHECK TABLE Statement"
Checks if the data in the table is corrupted.
``` sql
CHECK TABLE [db.]name
CHECK TABLE [db.]name [PARTITION partition_expr]
```
The `CHECK TABLE` query compares actual file sizes with the expected values which are stored on the server. If the file sizes do not match the stored values, it means the data is corrupted. This can be caused, for example, by a system crash during query execution.

View File

@ -12,7 +12,7 @@ Syntax:
``` sql
CREATE USER [IF NOT EXISTS | OR REPLACE] name1 [ON CLUSTER cluster_name1]
[, name2 [ON CLUSTER cluster_name2] ...]
[NOT IDENTIFIED | IDENTIFIED {[WITH {no_password | plaintext_password | sha256_password | sha256_hash | double_sha1_password | double_sha1_hash}] BY {'password' | 'hash'}} | {WITH ldap SERVER 'server_name'} | {WITH kerberos [REALM 'realm']}]
[NOT IDENTIFIED | IDENTIFIED {[WITH {no_password | plaintext_password | sha256_password | sha256_hash | double_sha1_password | double_sha1_hash}] BY {'password' | 'hash'}} | {WITH ldap SERVER 'server_name'} | {WITH kerberos [REALM 'realm']} | {WITH ssl_certificate CN 'common_name'}]
[HOST {LOCAL | NAME 'name' | REGEXP 'name_regexp' | IP 'address' | LIKE 'pattern'} [,...] | ANY | NONE]
[DEFAULT ROLE role [,...]]
[DEFAULT DATABASE database | NONE]
@ -34,6 +34,7 @@ There are multiple ways of user identification:
- `IDENTIFIED WITH double_sha1_hash BY 'hash'`
- `IDENTIFIED WITH ldap SERVER 'server_name'`
- `IDENTIFIED WITH kerberos` or `IDENTIFIED WITH kerberos REALM 'realm'`
- `IDENTIFIED WITH ssl_certificate CN 'mysite.com:user'`
For identification with sha256_hash using `SALT` - hash must be calculated from concatination of 'password' and 'salt'.
@ -54,7 +55,7 @@ Another way of specifying host is to use `@` syntax following the username. Exam
- `CREATE USER mira@'localhost'` — Equivalent to the `HOST LOCAL` syntax.
- `CREATE USER mira@'192.168.%.%'` — Equivalent to the `HOST LIKE` syntax.
:::warning
:::warning
ClickHouse treats `user_name@'address'` as a username as a whole. Thus, technically you can create multiple users with the same `user_name` and different constructions after `@`. However, we do not recommend to do so.
:::

View File

@ -281,8 +281,8 @@ After running this statement the `[db.]replicated_merge_tree_family_table_name`
### RESTART REPLICA
Provides possibility to reinitialize Zookeeper sessions state for `ReplicatedMergeTree` table, will compare current state with Zookeeper as source of true and add tasks to Zookeeper queue if needed.
Initialization replication queue based on ZooKeeper date happens in the same way as `ATTACH TABLE` statement. For a short time the table will be unavailable for any operations.
Provides possibility to reinitialize Zookeeper session's state for `ReplicatedMergeTree` table, will compare current state with Zookeeper as source of truth and add tasks to Zookeeper queue if needed.
Initialization of replication queue based on ZooKeeper data happens in the same way as for `ATTACH TABLE` statement. For a short time, the table will be unavailable for any operations.
``` sql
SYSTEM RESTART REPLICA [db.]replicated_merge_tree_family_table_name

View File

@ -39,3 +39,7 @@ You cant use table functions if the [allow_ddl](../../operations/settings/per
| [s3](../../sql-reference/table-functions/s3.md) | Creates a [S3](../../engines/table-engines/integrations/s3.md)-engine table. |
| [sqlite](../../sql-reference/table-functions/sqlite.md) | Creates a [sqlite](../../engines/table-engines/integrations/sqlite.md)-engine table. |
:::note
Only these table functions are enabled in readonly mode :
null, view, viewIfPermitted, numbers, numbers_mt, generateRandom, values, cluster, clusterAllReplicas
:::

View File

@ -37,7 +37,7 @@ deb:
contents:
- src: root/etc/clickhouse-client/config.xml
dst: /etc/clickhouse-client/config.xml
type: config
type: config|noreplace
- src: root/usr/bin/clickhouse-benchmark
dst: /usr/bin/clickhouse-benchmark
- src: root/usr/bin/clickhouse-compressor

View File

@ -29,7 +29,7 @@ deb:
contents:
- src: root/etc/clickhouse-keeper/keeper_config.xml
dst: /etc/clickhouse-keeper/keeper_config.xml
type: config
type: config|noreplace
- src: root/usr/bin/clickhouse-keeper
dst: /usr/bin/clickhouse-keeper
# docs

View File

@ -44,10 +44,10 @@ deb:
contents:
- src: root/etc/clickhouse-server/config.xml
dst: /etc/clickhouse-server/config.xml
type: config
type: config|noreplace
- src: root/etc/clickhouse-server/users.xml
dst: /etc/clickhouse-server/users.xml
type: config
type: config|noreplace
- src: clickhouse-server.init
dst: /etc/init.d/clickhouse-server
- src: clickhouse-server.service

View File

@ -1088,7 +1088,8 @@ void Client::processConfig()
}
else
{
need_render_progress = config().getBool("progress", false);
std::string progress = config().getString("progress", "tty");
need_render_progress = (Poco::icompare(progress, "off") && Poco::icompare(progress, "no") && Poco::icompare(progress, "false") && Poco::icompare(progress, "0"));
echo_queries = config().getBool("echo", false);
ignore_error = config().getBool("ignore-error", false);

View File

@ -1,7 +1,6 @@
#include "ExternalDictionaryLibraryHandler.h"
#include <base/scope_guard.h>
#include <base/bit_cast.h>
#include <base/find_symbols.h>
#include <IO/ReadHelpers.h>
@ -113,7 +112,7 @@ Block ExternalDictionaryLibraryHandler::loadAll()
Block ExternalDictionaryLibraryHandler::loadIds(const std::vector<uint64_t> & ids)
{
const ExternalDictionaryLibraryAPI::VectorUInt64 ids_data{bit_cast<decltype(ExternalDictionaryLibraryAPI::VectorUInt64::data)>(ids.data()), ids.size()};
const ExternalDictionaryLibraryAPI::VectorUInt64 ids_data{std::bit_cast<decltype(ExternalDictionaryLibraryAPI::VectorUInt64::data)>(ids.data()), ids.size()};
auto columns_holder = std::make_unique<ExternalDictionaryLibraryAPI::CString[]>(attributes_names.size());
ExternalDictionaryLibraryAPI::CStrings columns_pass{static_cast<decltype(ExternalDictionaryLibraryAPI::CStrings::data)>(columns_holder.get()), attributes_names.size()};

View File

@ -2,7 +2,6 @@
#include <Common/StringUtils/StringUtils.h>
#include <Core/Block.h>
#include <base/bit_cast.h>
#include <base/range.h>
#include "ExternalDictionaryLibraryAPI.h"

View File

@ -489,7 +489,8 @@ void LocalServer::processConfig()
}
else
{
need_render_progress = config().getBool("progress", false);
std::string progress = config().getString("progress", "tty");
need_render_progress = (Poco::icompare(progress, "off") && Poco::icompare(progress, "no") && Poco::icompare(progress, "false") && Poco::icompare(progress, "0"));
echo_queries = config().hasOption("echo") || config().hasOption("verbose");
ignore_error = config().getBool("ignore-error", false);
is_multiquery = true;

View File

@ -32,7 +32,6 @@
#include <Core/Block.h>
#include <base/StringRef.h>
#include <Common/DateLUT.h>
#include <base/bit_cast.h>
#include <IO/ReadBufferFromFileDescriptor.h>
#include <IO/WriteBufferFromFileDescriptor.h>
#include <IO/ReadBufferFromFile.h>
@ -278,9 +277,9 @@ Float transformFloatMantissa(Float x, UInt64 seed)
using UInt = std::conditional_t<std::is_same_v<Float, Float32>, UInt32, UInt64>;
constexpr size_t mantissa_num_bits = std::is_same_v<Float, Float32> ? 23 : 52;
UInt x_uint = bit_cast<UInt>(x);
UInt x_uint = std::bit_cast<UInt>(x);
x_uint = static_cast<UInt>(feistelNetwork(x_uint, mantissa_num_bits, seed));
return bit_cast<Float>(x_uint);
return std::bit_cast<Float>(x_uint);
}

View File

@ -1336,17 +1336,13 @@
name - name for the rule (optional)
regexp - RE2 compatible regular expression (mandatory)
replace - substitution string for sensitive data (optional, by default - six asterisks)
-->
<query_masking_rules>
<rule>
<name>hide encrypt/decrypt arguments</name>
<regexp>((?:aes_)?(?:encrypt|decrypt)(?:_mysql)?)\s*\(\s*(?:'(?:\\'|.)+'|.*?)\s*\)</regexp>
<!-- or more secure, but also more invasive:
(aes_\w+)\s*\(.*\)
-->
<replace>\1(???)</replace>
</rule>
</query_masking_rules>
</query_masking_rules> -->
<!-- Uncomment to use custom http handlers.
rules are checked from top to bottom, first match runs the handler

View File

@ -18,8 +18,10 @@ AggregateFunctionPtr createAggregateFunctionAnalysisOfVariance(const std::string
assertNoParameters(name, parameters);
assertBinary(name, arguments);
if (!isNumber(arguments[0]) || !isNumber(arguments[1]))
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Aggregate function {} only supports numerical types", name);
if (!isNumber(arguments[0]))
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Aggregate function {} only supports numerical argument types", name);
if (!WhichDataType(arguments[1]).isNativeUInt())
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Second argument of aggregate function {} should be a native unsigned integer", name);
return std::make_shared<AggregateFunctionAnalysisOfVariance>(arguments, parameters);
}

View File

@ -77,7 +77,7 @@ public:
void insertResultInto(AggregateDataPtr __restrict place, IColumn & to, Arena *) const override
{
auto f_stat = data(place).getFStatistic();
if (std::isinf(f_stat) || isNaN(f_stat))
if (std::isinf(f_stat) || isNaN(f_stat) || f_stat < 0)
throw Exception("F statistic is not defined or infinite for these arguments", ErrorCodes::BAD_ARGUMENTS);
auto p_value = data(place).getPValue(f_stat);

View File

@ -482,6 +482,8 @@ struct ZTestMoments
template <typename T>
struct AnalysisOfVarianceMoments
{
constexpr static size_t MAX_GROUPS_NUMBER = 1024 * 1024;
/// Sums of values within a group
std::vector<T> xs1{};
/// Sums of squared values within a group
@ -494,6 +496,10 @@ struct AnalysisOfVarianceMoments
if (xs1.size() >= possible_size)
return;
if (possible_size > MAX_GROUPS_NUMBER)
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Too many groups for analysis of variance (should be no more than {}, got {})",
MAX_GROUPS_NUMBER, possible_size);
xs1.resize(possible_size, 0.0);
xs2.resize(possible_size, 0.0);
ns.resize(possible_size, 0);

View File

@ -1,7 +1,6 @@
#pragma once
#include <base/types.h>
#include <base/bit_cast.h>
#include <base/sort.h>
#include <Common/HashTable/HashMap.h>
@ -104,13 +103,13 @@ private:
/// Take the most significant 16 bits of the floating point number.
BFloat16 toBFloat16(const Value & x) const
{
return bit_cast<UInt32>(static_cast<Float32>(x)) >> 16;
return std::bit_cast<UInt32>(static_cast<Float32>(x)) >> 16;
}
/// Put the bits into most significant 16 bits of the floating point number and fill other bits with zeros.
Float32 toFloat32(const BFloat16 & x) const
{
return bit_cast<Float32>(x << 16);
return std::bit_cast<Float32>(x << 16);
}
using Pair = PairNoInit<Float32, Weight>;

View File

@ -16,7 +16,7 @@ namespace DB
* Dependencies between passes must be avoided.
*/
class IQueryTreePass;
using QueryTreePassPtr = std::shared_ptr<IQueryTreePass>;
using QueryTreePassPtr = std::unique_ptr<IQueryTreePass>;
using QueryTreePasses = std::vector<QueryTreePassPtr>;
class IQueryTreePass

View File

@ -5527,9 +5527,15 @@ void QueryAnalyzer::resolveQuery(const QueryTreeNodePtr & query_node, Identifier
* 3. Check that there are no columns that are not specified in GROUP BY keys.
* 4. Validate GROUP BY modifiers.
*/
assertNoAggregateFunctionNodes(query_node_typed.getJoinTree(), "in JOIN TREE");
assertNoGroupingFunction(query_node_typed.getJoinTree(), "in JOIN TREE");
assertNoWindowFunctionNodes(query_node_typed.getJoinTree(), "in JOIN TREE");
auto join_tree_node_type = query_node_typed.getJoinTree()->getNodeType();
bool join_tree_is_subquery = join_tree_node_type == QueryTreeNodeType::QUERY || join_tree_node_type == QueryTreeNodeType::UNION;
if (!join_tree_is_subquery)
{
assertNoAggregateFunctionNodes(query_node_typed.getJoinTree(), "in JOIN TREE");
assertNoGroupingFunction(query_node_typed.getJoinTree(), "in JOIN TREE");
assertNoWindowFunctionNodes(query_node_typed.getJoinTree(), "in JOIN TREE");
}
if (query_node_typed.hasWhere())
{

View File

@ -18,6 +18,9 @@
#include <IO/Operators.h>
#include <Interpreters/Context.h>
#include <Analyzer/ColumnNode.h>
#include <Analyzer/InDepthQueryTreeVisitor.h>
#include <Common/Exception.h>
namespace DB
{
@ -25,6 +28,38 @@ namespace DB
namespace ErrorCodes
{
extern const int BAD_ARGUMENTS;
extern const int LOGICAL_ERROR;
}
namespace
{
#ifndef NDEBUG
/** This visitor checks if Query Tree structure is valid after each pass
* in debug build.
*/
class ValidationChecker : public InDepthQueryTreeVisitor<ValidationChecker>
{
String pass_name;
public:
explicit ValidationChecker(String pass_name_)
: pass_name(std::move(pass_name_))
{}
void visitImpl(QueryTreeNodePtr & node) const
{
auto * column = node->as<ColumnNode>();
if (!column)
return;
if (column->getColumnSourceOrNull() == nullptr)
throw Exception(ErrorCodes::LOGICAL_ERROR,
"Column {} {} query tree node does not have valid source node after running {} pass",
column->getColumnName(), column->getColumnType(), pass_name);
}
};
#endif
}
/** ClickHouse query tree pass manager.
@ -61,7 +96,12 @@ void QueryTreePassManager::run(QueryTreeNodePtr query_tree_node)
size_t passes_size = passes.size();
for (size_t i = 0; i < passes_size; ++i)
{
passes[i]->run(query_tree_node, current_context);
#ifndef NDEBUG
ValidationChecker(passes[i]->getName()).visit(query_tree_node);
#endif
}
}
void QueryTreePassManager::run(QueryTreeNodePtr query_tree_node, size_t up_to_pass_index)
@ -75,7 +115,12 @@ void QueryTreePassManager::run(QueryTreeNodePtr query_tree_node, size_t up_to_pa
auto current_context = getContext();
for (size_t i = 0; i < up_to_pass_index; ++i)
{
passes[i]->run(query_tree_node, current_context);
#ifndef NDEBUG
ValidationChecker(passes[i]->getName()).visit(query_tree_node);
#endif
}
}
void QueryTreePassManager::dump(WriteBuffer & buffer)
@ -114,38 +159,38 @@ void addQueryTreePasses(QueryTreePassManager & manager)
auto context = manager.getContext();
const auto & settings = context->getSettingsRef();
manager.addPass(std::make_shared<QueryAnalysisPass>());
manager.addPass(std::make_unique<QueryAnalysisPass>());
if (settings.optimize_functions_to_subcolumns)
manager.addPass(std::make_shared<FunctionToSubcolumnsPass>());
manager.addPass(std::make_unique<FunctionToSubcolumnsPass>());
if (settings.count_distinct_optimization)
manager.addPass(std::make_shared<CountDistinctPass>());
manager.addPass(std::make_unique<CountDistinctPass>());
if (settings.optimize_rewrite_sum_if_to_count_if)
manager.addPass(std::make_shared<SumIfToCountIfPass>());
manager.addPass(std::make_unique<SumIfToCountIfPass>());
if (settings.optimize_normalize_count_variants)
manager.addPass(std::make_shared<NormalizeCountVariantsPass>());
manager.addPass(std::make_unique<NormalizeCountVariantsPass>());
manager.addPass(std::make_shared<CustomizeFunctionsPass>());
manager.addPass(std::make_unique<CustomizeFunctionsPass>());
if (settings.optimize_arithmetic_operations_in_aggregate_functions)
manager.addPass(std::make_shared<AggregateFunctionsArithmericOperationsPass>());
manager.addPass(std::make_unique<AggregateFunctionsArithmericOperationsPass>());
if (settings.optimize_injective_functions_inside_uniq)
manager.addPass(std::make_shared<UniqInjectiveFunctionsEliminationPass>());
manager.addPass(std::make_unique<UniqInjectiveFunctionsEliminationPass>());
if (settings.optimize_multiif_to_if)
manager.addPass(std::make_shared<MultiIfToIfPass>());
manager.addPass(std::make_unique<MultiIfToIfPass>());
manager.addPass(std::make_shared<IfConstantConditionPass>());
manager.addPass(std::make_unique<IfConstantConditionPass>());
if (settings.optimize_if_chain_to_multiif)
manager.addPass(std::make_shared<IfChainToMultiIfPass>());
manager.addPass(std::make_unique<IfChainToMultiIfPass>());
manager.addPass(std::make_shared<OrderByTupleEliminationPass>());
manager.addPass(std::make_shared<OrderByLimitByDuplicateEliminationPass>());
manager.addPass(std::make_unique<OrderByTupleEliminationPass>());
manager.addPass(std::make_unique<OrderByLimitByDuplicateEliminationPass>());
}
}

View File

@ -126,6 +126,7 @@ BackupWriterS3::BackupWriterS3(
, max_single_read_retries(context_->getSettingsRef().s3_max_single_read_retries)
, read_settings(context_->getReadSettings())
, rw_settings(context_->getStorageS3Settings().getSettings(s3_uri.uri.toString()).rw_settings)
, log(&Poco::Logger::get("BackupWriterS3"))
{
rw_settings.updateFromSettingsIfEmpty(context_->getSettingsRef());
}
@ -146,9 +147,12 @@ void BackupWriterS3::copyObjectImpl(
const String & src_key,
const String & dst_bucket,
const String & dst_key,
std::optional<Aws::S3::Model::HeadObjectResult> head,
std::optional<ObjectAttributes> metadata) const
const Aws::S3::Model::HeadObjectResult & head,
const std::optional<ObjectAttributes> & metadata) const
{
size_t size = head.GetContentLength();
LOG_TRACE(log, "Copying {} bytes using single-operation copy", size);
Aws::S3::Model::CopyObjectRequest request;
request.SetCopySource(src_bucket + "/" + src_key);
request.SetBucket(dst_bucket);
@ -186,13 +190,11 @@ void BackupWriterS3::copyObjectMultipartImpl(
const String & src_key,
const String & dst_bucket,
const String & dst_key,
std::optional<Aws::S3::Model::HeadObjectResult> head,
std::optional<ObjectAttributes> metadata) const
const Aws::S3::Model::HeadObjectResult & head,
const std::optional<ObjectAttributes> & metadata) const
{
if (!head)
head = requestObjectHeadData(src_bucket, src_key).GetResult();
size_t size = head->GetContentLength();
size_t size = head.GetContentLength();
LOG_TRACE(log, "Copying {} bytes using multipart upload copy", size);
String multipart_upload_id;
@ -213,16 +215,20 @@ void BackupWriterS3::copyObjectMultipartImpl(
std::vector<String> part_tags;
size_t position = 0;
size_t upload_part_size = rw_settings.min_upload_part_size;
for (size_t position = 0, part_number = 1; position < size; ++part_number, position += upload_part_size)
for (size_t part_number = 1; position < size; ++part_number)
{
size_t next_position = std::min(position + upload_part_size, size);
Aws::S3::Model::UploadPartCopyRequest part_request;
part_request.SetCopySource(src_bucket + "/" + src_key);
part_request.SetBucket(dst_bucket);
part_request.SetKey(dst_key);
part_request.SetUploadId(multipart_upload_id);
part_request.SetPartNumber(static_cast<int>(part_number));
part_request.SetCopySourceRange(fmt::format("bytes={}-{}", position, std::min(size, position + upload_part_size) - 1));
part_request.SetCopySourceRange(fmt::format("bytes={}-{}", position, next_position - 1));
auto outcome = client->UploadPartCopy(part_request);
if (!outcome.IsSuccess())
@ -239,6 +245,14 @@ void BackupWriterS3::copyObjectMultipartImpl(
auto etag = outcome.GetResult().GetCopyPartResult().GetETag();
part_tags.push_back(etag);
position = next_position;
if (part_number % rw_settings.upload_part_size_multiply_parts_count_threshold == 0)
{
upload_part_size *= rw_settings.upload_part_size_multiply_factor;
upload_part_size = std::min(upload_part_size, rw_settings.max_upload_part_size);
}
}
{
@ -280,15 +294,14 @@ void BackupWriterS3::copyFileNative(DiskPtr from_disk, const String & file_name_
auto file_path = fs::path(s3_uri.key) / file_name_to;
auto head = requestObjectHeadData(source_bucket, objects[0].absolute_path).GetResult();
static constexpr int64_t multipart_upload_threashold = 5UL * 1024 * 1024 * 1024;
if (head.GetContentLength() >= multipart_upload_threashold)
if (static_cast<size_t>(head.GetContentLength()) < rw_settings.max_single_operation_copy_size)
{
copyObjectMultipartImpl(
copyObjectImpl(
source_bucket, objects[0].absolute_path, s3_uri.bucket, file_path, head);
}
else
{
copyObjectImpl(
copyObjectMultipartImpl(
source_bucket, objects[0].absolute_path, s3_uri.bucket, file_path, head);
}
}

View File

@ -61,7 +61,6 @@ public:
void copyFileNative(DiskPtr from_disk, const String & file_name_from, const String & file_name_to) override;
private:
Aws::S3::Model::HeadObjectOutcome requestObjectHeadData(const std::string & bucket_from, const std::string & key) const;
void copyObjectImpl(
@ -69,22 +68,23 @@ private:
const String & src_key,
const String & dst_bucket,
const String & dst_key,
std::optional<Aws::S3::Model::HeadObjectResult> head = std::nullopt,
std::optional<ObjectAttributes> metadata = std::nullopt) const;
const Aws::S3::Model::HeadObjectResult & head,
const std::optional<ObjectAttributes> & metadata = std::nullopt) const;
void copyObjectMultipartImpl(
const String & src_bucket,
const String & src_key,
const String & dst_bucket,
const String & dst_key,
std::optional<Aws::S3::Model::HeadObjectResult> head = std::nullopt,
std::optional<ObjectAttributes> metadata = std::nullopt) const;
const Aws::S3::Model::HeadObjectResult & head,
const std::optional<ObjectAttributes> & metadata = std::nullopt) const;
S3::URI s3_uri;
std::shared_ptr<Aws::S3::S3Client> client;
UInt64 max_single_read_retries;
ReadSettings read_settings;
S3Settings::ReadWriteSettings rw_settings;
Poco::Logger * log;
};
}

View File

@ -138,12 +138,12 @@ private:
BackupImpl::BackupImpl(
const String & backup_name_,
const String & backup_name_for_logging_,
const ArchiveParams & archive_params_,
const std::optional<BackupInfo> & base_backup_info_,
std::shared_ptr<IBackupReader> reader_,
const ContextPtr & context_)
: backup_name(backup_name_)
: backup_name_for_logging(backup_name_for_logging_)
, archive_params(archive_params_)
, use_archives(!archive_params.archive_name.empty())
, open_mode(OpenMode::READ)
@ -158,7 +158,7 @@ BackupImpl::BackupImpl(
BackupImpl::BackupImpl(
const String & backup_name_,
const String & backup_name_for_logging_,
const ArchiveParams & archive_params_,
const std::optional<BackupInfo> & base_backup_info_,
std::shared_ptr<IBackupWriter> writer_,
@ -166,7 +166,7 @@ BackupImpl::BackupImpl(
bool is_internal_backup_,
const std::shared_ptr<IBackupCoordination> & coordination_,
const std::optional<UUID> & backup_uuid_)
: backup_name(backup_name_)
: backup_name_for_logging(backup_name_for_logging_)
, archive_params(archive_params_)
, use_archives(!archive_params.archive_name.empty())
, open_mode(OpenMode::WRITE)
@ -225,10 +225,19 @@ void BackupImpl::open(const ContextPtr & context)
base_backup = BackupFactory::instance().createBackup(params);
if (open_mode == OpenMode::WRITE)
{
base_backup_uuid = base_backup->getUUID();
}
else if (base_backup_uuid != base_backup->getUUID())
throw Exception(ErrorCodes::WRONG_BASE_BACKUP, "Backup {}: The base backup {} has different UUID ({} != {})",
backup_name, base_backup->getName(), toString(base_backup->getUUID()), (base_backup_uuid ? toString(*base_backup_uuid) : ""));
{
throw Exception(
ErrorCodes::WRONG_BASE_BACKUP,
"Backup {}: The base backup {} has different UUID ({} != {})",
backup_name_for_logging,
base_backup->getNameForLogging(),
toString(base_backup->getUUID()),
(base_backup_uuid ? toString(*base_backup_uuid) : ""));
}
}
}
@ -349,14 +358,14 @@ void BackupImpl::readBackupMetadata()
if (use_archives)
{
if (!reader->fileExists(archive_params.archive_name))
throw Exception(ErrorCodes::BACKUP_NOT_FOUND, "Backup {} not found", backup_name);
throw Exception(ErrorCodes::BACKUP_NOT_FOUND, "Backup {} not found", backup_name_for_logging);
setCompressedSize();
in = getArchiveReader("")->readFile(".backup");
}
else
{
if (!reader->fileExists(".backup"))
throw Exception(ErrorCodes::BACKUP_NOT_FOUND, "Backup {} not found", backup_name);
throw Exception(ErrorCodes::BACKUP_NOT_FOUND, "Backup {} not found", backup_name_for_logging);
in = reader->readFile(".backup");
}
@ -369,7 +378,8 @@ void BackupImpl::readBackupMetadata()
version = config->getInt("version");
if ((version < INITIAL_BACKUP_VERSION) || (version > CURRENT_BACKUP_VERSION))
throw Exception(ErrorCodes::BACKUP_VERSION_NOT_SUPPORTED, "Backup {}: Version {} is not supported", backup_name, version);
throw Exception(
ErrorCodes::BACKUP_VERSION_NOT_SUPPORTED, "Backup {}: Version {} is not supported", backup_name_for_logging, version);
timestamp = parse<LocalDateTime>(config->getString("timestamp")).to_time_t();
uuid = parse<UUID>(config->getString("uuid"));
@ -400,7 +410,13 @@ void BackupImpl::readBackupMetadata()
use_base = true;
if (info.base_size > info.size)
throw Exception(ErrorCodes::BACKUP_DAMAGED, "Backup {}: Base size must not be greater than the size of entry {}", backup_name, quoteString(info.file_name));
{
throw Exception(
ErrorCodes::BACKUP_DAMAGED,
"Backup {}: Base size must not be greater than the size of entry {}",
backup_name_for_logging,
quoteString(info.file_name));
}
if (use_base)
{
@ -436,14 +452,14 @@ void BackupImpl::checkBackupDoesntExist() const
file_name_to_check_existence = ".backup";
if (writer->fileExists(file_name_to_check_existence))
throw Exception(ErrorCodes::BACKUP_ALREADY_EXISTS, "Backup {} already exists", backup_name);
throw Exception(ErrorCodes::BACKUP_ALREADY_EXISTS, "Backup {} already exists", backup_name_for_logging);
/// Check that no other backup (excluding internal backups) is writing to the same destination.
if (!is_internal_backup)
{
assert(!lock_file_name.empty());
if (writer->fileExists(lock_file_name))
throw Exception(ErrorCodes::BACKUP_ALREADY_EXISTS, "Backup {} is being written already", backup_name);
throw Exception(ErrorCodes::BACKUP_ALREADY_EXISTS, "Backup {} is being written already", backup_name_for_logging);
}
}
@ -466,8 +482,16 @@ bool BackupImpl::checkLockFile(bool throw_if_failed) const
if (throw_if_failed)
{
if (!writer->fileExists(lock_file_name))
throw Exception(ErrorCodes::FAILED_TO_SYNC_BACKUP_OR_RESTORE, "Lock file {} suddenly disappeared while writing backup {}", lock_file_name, backup_name);
throw Exception(ErrorCodes::BACKUP_ALREADY_EXISTS, "A concurrent backup writing to the same destination {} detected", backup_name);
{
throw Exception(
ErrorCodes::FAILED_TO_SYNC_BACKUP_OR_RESTORE,
"Lock file {} suddenly disappeared while writing backup {}",
lock_file_name,
backup_name_for_logging);
}
throw Exception(
ErrorCodes::BACKUP_ALREADY_EXISTS, "A concurrent backup writing to the same destination {} detected", backup_name_for_logging);
}
return false;
}
@ -514,8 +538,13 @@ UInt64 BackupImpl::getFileSize(const String & file_name) const
auto adjusted_path = removeLeadingSlash(file_name);
auto info = coordination->getFileInfo(adjusted_path);
if (!info)
{
throw Exception(
ErrorCodes::BACKUP_ENTRY_NOT_FOUND, "Backup {}: Entry {} not found in the backup", backup_name, quoteString(file_name));
ErrorCodes::BACKUP_ENTRY_NOT_FOUND,
"Backup {}: Entry {} not found in the backup",
backup_name_for_logging,
quoteString(file_name));
}
return info->size;
}
@ -525,8 +554,13 @@ UInt128 BackupImpl::getFileChecksum(const String & file_name) const
auto adjusted_path = removeLeadingSlash(file_name);
auto info = coordination->getFileInfo(adjusted_path);
if (!info)
{
throw Exception(
ErrorCodes::BACKUP_ENTRY_NOT_FOUND, "Backup {}: Entry {} not found in the backup", backup_name, quoteString(file_name));
ErrorCodes::BACKUP_ENTRY_NOT_FOUND,
"Backup {}: Entry {} not found in the backup",
backup_name_for_logging,
quoteString(file_name));
}
return info->checksum;
}
@ -536,8 +570,13 @@ SizeAndChecksum BackupImpl::getFileSizeAndChecksum(const String & file_name) con
auto adjusted_path = removeLeadingSlash(file_name);
auto info = coordination->getFileInfo(adjusted_path);
if (!info)
{
throw Exception(
ErrorCodes::BACKUP_ENTRY_NOT_FOUND, "Backup {}: Entry {} not found in the backup", backup_name, quoteString(file_name));
ErrorCodes::BACKUP_ENTRY_NOT_FOUND,
"Backup {}: Entry {} not found in the backup",
backup_name_for_logging,
quoteString(file_name));
}
return {info->size, info->checksum};
}
@ -560,8 +599,13 @@ BackupEntryPtr BackupImpl::readFile(const SizeAndChecksum & size_and_checksum) c
auto info_opt = coordination->getFileInfo(size_and_checksum);
if (!info_opt)
{
throw Exception(
ErrorCodes::BACKUP_ENTRY_NOT_FOUND, "Backup {}: Entry {} not found in the backup", backup_name, formatSizeAndChecksum(size_and_checksum));
ErrorCodes::BACKUP_ENTRY_NOT_FOUND,
"Backup {}: Entry {} not found in the backup",
backup_name_for_logging,
formatSizeAndChecksum(size_and_checksum));
}
const auto & info = *info_opt;
@ -577,7 +621,7 @@ BackupEntryPtr BackupImpl::readFile(const SizeAndChecksum & size_and_checksum) c
throw Exception(
ErrorCodes::NO_BASE_BACKUP,
"Backup {}: Entry {} is marked to be read from a base backup, but there is no base backup specified",
backup_name, formatSizeAndChecksum(size_and_checksum));
backup_name_for_logging, formatSizeAndChecksum(size_and_checksum));
}
if (!base_backup->fileExists(std::pair(info.base_size, info.base_checksum)))
@ -585,7 +629,7 @@ BackupEntryPtr BackupImpl::readFile(const SizeAndChecksum & size_and_checksum) c
throw Exception(
ErrorCodes::WRONG_BASE_BACKUP,
"Backup {}: Entry {} is marked to be read from a base backup, but doesn't exist there",
backup_name, formatSizeAndChecksum(size_and_checksum));
backup_name_for_logging, formatSizeAndChecksum(size_and_checksum));
}
auto base_entry = base_backup->readFile(std::pair{info.base_size, info.base_checksum});
@ -695,9 +739,12 @@ void BackupImpl::writeFile(const String & file_name, BackupEntryPtr entry)
LOG_TRACE(log, "Writing backup for file {} from {}", file_name, from_file_name);
auto adjusted_path = removeLeadingSlash(file_name);
if (coordination->getFileInfo(adjusted_path))
{
throw Exception(
ErrorCodes::BACKUP_ENTRY_ALREADY_EXISTS, "Backup {}: Entry {} already exists", backup_name, quoteString(file_name));
ErrorCodes::BACKUP_ENTRY_ALREADY_EXISTS, "Backup {}: Entry {} already exists", backup_name_for_logging, quoteString(file_name));
}
FileInfo info
{
@ -893,12 +940,12 @@ void BackupImpl::finalizeWriting()
if (!is_internal_backup)
{
LOG_TRACE(log, "Finalizing backup {}", backup_name);
LOG_TRACE(log, "Finalizing backup {}", backup_name_for_logging);
writeBackupMetadata();
closeArchives();
setCompressedSize();
removeLockFile();
LOG_TRACE(log, "Finalized backup {}", backup_name);
LOG_TRACE(log, "Finalized backup {}", backup_name_for_logging);
}
writing_finalized = true;
@ -971,7 +1018,7 @@ void BackupImpl::removeAllFilesAfterFailure()
try
{
LOG_INFO(log, "Removing all files of backup {} after failure", backup_name);
LOG_INFO(log, "Removing all files of backup {} after failure", backup_name_for_logging);
Strings files_to_remove;
if (use_archives)

View File

@ -35,14 +35,14 @@ public:
};
BackupImpl(
const String & backup_name_,
const String & backup_name_for_logging_,
const ArchiveParams & archive_params_,
const std::optional<BackupInfo> & base_backup_info_,
std::shared_ptr<IBackupReader> reader_,
const ContextPtr & context_);
BackupImpl(
const String & backup_name_,
const String & backup_name_for_logging_,
const ArchiveParams & archive_params_,
const std::optional<BackupInfo> & base_backup_info_,
std::shared_ptr<IBackupWriter> writer_,
@ -53,7 +53,7 @@ public:
~BackupImpl() override;
const String & getName() const override { return backup_name; }
const String & getNameForLogging() const override { return backup_name_for_logging; }
OpenMode getOpenMode() const override { return open_mode; }
time_t getTimestamp() const override { return timestamp; }
UUID getUUID() const override { return *uuid; }
@ -107,7 +107,7 @@ private:
/// Calculates and sets `compressed_size`.
void setCompressedSize();
const String backup_name;
const String backup_name_for_logging;
const ArchiveParams archive_params;
const bool use_archives;
const OpenMode open_mode;

View File

@ -6,6 +6,7 @@
#include <Parsers/ExpressionElementParsers.h>
#include <Parsers/formatAST.h>
#include <Parsers/parseQuery.h>
#include <Interpreters/maskSensitiveInfoInQueryForLogging.h>
namespace DB
@ -92,4 +93,10 @@ BackupInfo BackupInfo::fromAST(const IAST & ast)
}
String BackupInfo::toStringForLogging(const ContextPtr & context) const
{
ASTPtr ast = toAST();
return maskSensitiveInfoInBackupNameForLogging(serializeAST(*ast), ast, context);
}
}

View File

@ -1,6 +1,7 @@
#pragma once
#include <Core/Field.h>
#include <Interpreters/Context_fwd.h>
namespace DB
@ -20,6 +21,8 @@ struct BackupInfo
ASTPtr toAST() const;
static BackupInfo fromAST(const IAST & ast);
String toStringForLogging(const ContextPtr & context) const;
};
}

View File

@ -166,9 +166,10 @@ OperationID BackupsWorker::startMakingBackup(const ASTPtr & query, const Context
}
auto backup_info = BackupInfo::fromAST(*backup_query->backup_name);
String backup_name_for_logging = backup_info.toStringForLogging(context);
try
{
addInfo(backup_id, backup_info.toString(), backup_settings.internal, BackupStatus::CREATING_BACKUP);
addInfo(backup_id, backup_name_for_logging, backup_settings.internal, BackupStatus::CREATING_BACKUP);
/// Prepare context to use.
ContextPtr context_in_use = context;
@ -184,13 +185,14 @@ OperationID BackupsWorker::startMakingBackup(const ASTPtr & query, const Context
if (backup_settings.async)
{
backups_thread_pool.scheduleOrThrowOnError(
[this, backup_query, backup_id, backup_settings, backup_info, backup_coordination, context_in_use, mutable_context]
[this, backup_query, backup_id, backup_name_for_logging, backup_info, backup_settings, backup_coordination, context_in_use, mutable_context]
{
doBackup(
backup_query,
backup_id,
backup_settings,
backup_name_for_logging,
backup_info,
backup_settings,
backup_coordination,
context_in_use,
mutable_context,
@ -202,8 +204,9 @@ OperationID BackupsWorker::startMakingBackup(const ASTPtr & query, const Context
doBackup(
backup_query,
backup_id,
backup_settings,
backup_name_for_logging,
backup_info,
backup_settings,
backup_coordination,
context_in_use,
mutable_context,
@ -214,7 +217,7 @@ OperationID BackupsWorker::startMakingBackup(const ASTPtr & query, const Context
}
catch (...)
{
tryLogCurrentException(log, fmt::format("Failed to start {} {}", (backup_settings.internal ? "internal backup" : "backup"), backup_info.toString()));
tryLogCurrentException(log, fmt::format("Failed to start {} {}", (backup_settings.internal ? "internal backup" : "backup"), backup_name_for_logging));
/// Something bad happened, the backup has not built.
setStatusSafe(backup_id, BackupStatus::BACKUP_FAILED);
sendCurrentExceptionToCoordination(backup_coordination, backup_settings.host_id);
@ -226,8 +229,9 @@ OperationID BackupsWorker::startMakingBackup(const ASTPtr & query, const Context
void BackupsWorker::doBackup(
const std::shared_ptr<ASTBackupQuery> & backup_query,
const OperationID & backup_id,
BackupSettings backup_settings,
const String & backup_name_for_logging,
const BackupInfo & backup_info,
BackupSettings backup_settings,
std::shared_ptr<IBackupCoordination> backup_coordination,
const ContextPtr & context,
ContextMutablePtr mutable_context,
@ -336,7 +340,7 @@ void BackupsWorker::doBackup(
/// Close the backup.
backup.reset();
LOG_INFO(log, "{} {} was created successfully", (backup_settings.internal ? "Internal backup" : "Backup"), backup_info.toString());
LOG_INFO(log, "{} {} was created successfully", (backup_settings.internal ? "Internal backup" : "Backup"), backup_name_for_logging);
setStatus(backup_id, BackupStatus::BACKUP_CREATED);
setNumFilesAndSize(backup_id, num_files, uncompressed_size, compressed_size);
}
@ -345,7 +349,7 @@ void BackupsWorker::doBackup(
/// Something bad happened, the backup has not built.
if (called_async)
{
tryLogCurrentException(log, fmt::format("Failed to make {} {}", (backup_settings.internal ? "internal backup" : "backup"), backup_info.toString()));
tryLogCurrentException(log, fmt::format("Failed to make {} {}", (backup_settings.internal ? "internal backup" : "backup"), backup_name_for_logging));
setStatusSafe(backup_id, BackupStatus::BACKUP_FAILED);
sendCurrentExceptionToCoordination(backup_coordination, backup_settings.host_id);
}
@ -384,7 +388,8 @@ OperationID BackupsWorker::startRestoring(const ASTPtr & query, ContextMutablePt
try
{
auto backup_info = BackupInfo::fromAST(*restore_query->backup_name);
addInfo(restore_id, backup_info.toString(), restore_settings.internal, BackupStatus::RESTORING);
String backup_name_for_logging = backup_info.toStringForLogging(context);
addInfo(restore_id, backup_name_for_logging, restore_settings.internal, BackupStatus::RESTORING);
/// Prepare context to use.
ContextMutablePtr context_in_use = context;
@ -399,12 +404,14 @@ OperationID BackupsWorker::startRestoring(const ASTPtr & query, ContextMutablePt
if (restore_settings.async)
{
backups_thread_pool.scheduleOrThrowOnError(
[this, restore_query, restore_id, restore_settings, backup_info, restore_coordination, context_in_use] {
[this, restore_query, restore_id, backup_name_for_logging, backup_info, restore_settings, restore_coordination, context_in_use]
{
doRestore(
restore_query,
restore_id,
restore_settings,
backup_name_for_logging,
backup_info,
restore_settings,
restore_coordination,
context_in_use,
/* called_async= */ true);
@ -415,8 +422,9 @@ OperationID BackupsWorker::startRestoring(const ASTPtr & query, ContextMutablePt
doRestore(
restore_query,
restore_id,
restore_settings,
backup_name_for_logging,
backup_info,
restore_settings,
restore_coordination,
context_in_use,
/* called_async= */ false);
@ -437,8 +445,9 @@ OperationID BackupsWorker::startRestoring(const ASTPtr & query, ContextMutablePt
void BackupsWorker::doRestore(
const std::shared_ptr<ASTBackupQuery> & restore_query,
const OperationID & restore_id,
RestoreSettings restore_settings,
const String & backup_name_for_logging,
const BackupInfo & backup_info,
RestoreSettings restore_settings,
std::shared_ptr<IRestoreCoordination> restore_coordination,
ContextMutablePtr context,
bool called_async)
@ -541,7 +550,7 @@ void BackupsWorker::doRestore(
restore_coordination->setStage(restore_settings.host_id, Stage::COMPLETED, "");
}
LOG_INFO(log, "Restored from {} {} successfully", (restore_settings.internal ? "internal backup" : "backup"), backup_info.toString());
LOG_INFO(log, "Restored from {} {} successfully", (restore_settings.internal ? "internal backup" : "backup"), backup_name_for_logging);
setStatus(restore_id, BackupStatus::RESTORED);
}
catch (...)
@ -549,7 +558,7 @@ void BackupsWorker::doRestore(
/// Something bad happened, the backup has not built.
if (called_async)
{
tryLogCurrentException(log, fmt::format("Failed to restore from {} {}", (restore_settings.internal ? "internal backup" : "backup"), backup_info.toString()));
tryLogCurrentException(log, fmt::format("Failed to restore from {} {}", (restore_settings.internal ? "internal backup" : "backup"), backup_name_for_logging));
setStatusSafe(restore_id, BackupStatus::RESTORE_FAILED);
sendCurrentExceptionToCoordination(restore_coordination, restore_settings.host_id);
}

View File

@ -76,14 +76,28 @@ public:
private:
OperationID startMakingBackup(const ASTPtr & query, const ContextPtr & context);
void doBackup(const std::shared_ptr<ASTBackupQuery> & backup_query, const OperationID & backup_id, BackupSettings backup_settings,
const BackupInfo & backup_info, std::shared_ptr<IBackupCoordination> backup_coordination, const ContextPtr & context,
ContextMutablePtr mutable_context, bool called_async);
void doBackup(
const std::shared_ptr<ASTBackupQuery> & backup_query,
const OperationID & backup_id,
const String & backup_name_for_logging,
const BackupInfo & backup_info,
BackupSettings backup_settings,
std::shared_ptr<IBackupCoordination> backup_coordination,
const ContextPtr & context,
ContextMutablePtr mutable_context,
bool called_async);
OperationID startRestoring(const ASTPtr & query, ContextMutablePtr context);
void doRestore(const std::shared_ptr<ASTBackupQuery> & restore_query, const OperationID & restore_id, RestoreSettings restore_settings, const BackupInfo & backup_info,
std::shared_ptr<IRestoreCoordination> restore_coordination, ContextMutablePtr context, bool called_async);
void doRestore(
const std::shared_ptr<ASTBackupQuery> & restore_query,
const OperationID & restore_id,
const String & backup_name_for_logging,
const BackupInfo & backup_info,
RestoreSettings restore_settings,
std::shared_ptr<IRestoreCoordination> restore_coordination,
ContextMutablePtr context,
bool called_async);
void addInfo(const OperationID & id, const String & name, bool internal, BackupStatus status);
void setStatus(const OperationID & id, BackupStatus status, bool throw_if_error = true);

View File

@ -19,7 +19,8 @@ public:
virtual ~IBackup() = default;
/// Name of the backup.
virtual const String & getName() const = 0;
//virtual const String & getName() const = 0;
virtual const String & getNameForLogging() const = 0;
enum class OpenMode
{

View File

@ -47,7 +47,7 @@ void registerBackupEngineS3(BackupFactory & factory)
auto creator_fn = []([[maybe_unused]] const BackupFactory::CreateParams & params) -> std::unique_ptr<IBackup>
{
#if USE_AWS_S3
String backup_name = params.backup_info.toString();
String backup_name_for_logging = params.backup_info.toStringForLogging(params.context);
const String & id_arg = params.backup_info.id_arg;
const auto & args = params.backup_info.args;
@ -111,12 +111,12 @@ void registerBackupEngineS3(BackupFactory & factory)
if (params.open_mode == IBackup::OpenMode::READ)
{
auto reader = std::make_shared<BackupReaderS3>(S3::URI{Poco::URI{s3_uri}}, access_key_id, secret_access_key, params.context);
return std::make_unique<BackupImpl>(backup_name, archive_params, params.base_backup_info, reader, params.context);
return std::make_unique<BackupImpl>(backup_name_for_logging, archive_params, params.base_backup_info, reader, params.context);
}
else
{
auto writer = std::make_shared<BackupWriterS3>(S3::URI{Poco::URI{s3_uri}}, access_key_id, secret_access_key, params.context);
return std::make_unique<BackupImpl>(backup_name, archive_params, params.base_backup_info, writer, params.context, params.is_internal_backup, params.backup_coordination, params.backup_uuid);
return std::make_unique<BackupImpl>(backup_name_for_logging, archive_params, params.base_backup_info, writer, params.context, params.is_internal_backup, params.backup_coordination, params.backup_uuid);
}
#else
throw Exception("S3 support is disabled", ErrorCodes::SUPPORT_IS_DISABLED);

View File

@ -99,7 +99,7 @@ void registerBackupEnginesFileAndDisk(BackupFactory & factory)
{
auto creator_fn = [](const BackupFactory::CreateParams & params) -> std::unique_ptr<IBackup>
{
String backup_name = params.backup_info.toString();
String backup_name_for_logging = params.backup_info.toStringForLogging(params.context);
const String & engine_name = params.backup_info.backup_engine_name;
if (!params.backup_info.id_arg.empty())
@ -172,7 +172,7 @@ void registerBackupEnginesFileAndDisk(BackupFactory & factory)
reader = std::make_shared<BackupReaderFile>(path);
else
reader = std::make_shared<BackupReaderDisk>(disk, path);
return std::make_unique<BackupImpl>(backup_name, archive_params, params.base_backup_info, reader, params.context);
return std::make_unique<BackupImpl>(backup_name_for_logging, archive_params, params.base_backup_info, reader, params.context);
}
else
{
@ -181,7 +181,7 @@ void registerBackupEnginesFileAndDisk(BackupFactory & factory)
writer = std::make_shared<BackupWriterFile>(path);
else
writer = std::make_shared<BackupWriterDisk>(disk, path);
return std::make_unique<BackupImpl>(backup_name, archive_params, params.base_backup_info, writer, params.context, params.is_internal_backup, params.backup_coordination, params.backup_uuid);
return std::make_unique<BackupImpl>(backup_name_for_logging, archive_params, params.base_backup_info, writer, params.context, params.is_internal_backup, params.backup_coordination, params.backup_uuid);
}
};

View File

@ -65,10 +65,12 @@
#include <Interpreters/ReplaceQueryParameterVisitor.h>
#include <Interpreters/ProfileEventsExt.h>
#include <IO/WriteBufferFromOStream.h>
#include <IO/WriteBufferFromFileDescriptor.h>
#include <IO/CompressionMethod.h>
#include <Client/InternalTextLogs.h>
#include <IO/ForkWriteBuffer.h>
#include <Parsers/Kusto/ParserKQLStatement.h>
#include <boost/algorithm/string/case_conv.hpp>
namespace fs = std::filesystem;
@ -103,6 +105,7 @@ namespace ErrorCodes
extern const int CANNOT_SET_SIGNAL_HANDLER;
extern const int UNRECOGNIZED_ARGUMENTS;
extern const int LOGICAL_ERROR;
extern const int CANNOT_OPEN_FILE;
}
}
@ -116,6 +119,25 @@ namespace ProfileEvents
namespace DB
{
std::istream& operator>> (std::istream & in, ProgressOption & progress)
{
std::string token;
in >> token;
boost::to_upper(token);
if (token == "OFF" || token == "FALSE" || token == "0" || token == "NO")
progress = ProgressOption::OFF;
else if (token == "TTY" || token == "ON" || token == "TRUE" || token == "1" || token == "YES")
progress = ProgressOption::TTY;
else if (token == "ERR")
progress = ProgressOption::ERR;
else
throw boost::program_options::validation_error(boost::program_options::validation_error::invalid_option_value);
return in;
}
static ClientInfo::QueryKind parseQueryKind(const String & query_kind)
{
if (query_kind == "initial_query")
@ -413,8 +435,8 @@ void ClientBase::onData(Block & block, ASTPtr parsed_query)
return;
/// If results are written INTO OUTFILE, we can avoid clearing progress to avoid flicker.
if (need_render_progress && (stdout_is_a_tty || is_interactive) && (!select_into_file || select_into_file_and_stdout))
progress_indication.clearProgressOutput();
if (need_render_progress && tty_buf && (!select_into_file || select_into_file_and_stdout))
progress_indication.clearProgressOutput(*tty_buf);
try
{
@ -431,11 +453,11 @@ void ClientBase::onData(Block & block, ASTPtr parsed_query)
output_format->flush();
/// Restore progress bar after data block.
if (need_render_progress && (stdout_is_a_tty || is_interactive))
if (need_render_progress && tty_buf)
{
if (select_into_file && !select_into_file_and_stdout)
std::cerr << "\r";
progress_indication.writeProgress();
progress_indication.writeProgress(*tty_buf);
}
}
@ -443,7 +465,8 @@ void ClientBase::onData(Block & block, ASTPtr parsed_query)
void ClientBase::onLogData(Block & block)
{
initLogsOutputStream();
progress_indication.clearProgressOutput();
if (need_render_progress && tty_buf)
progress_indication.clearProgressOutput(*tty_buf);
logs_out_stream->writeLogs(block);
logs_out_stream->flush();
}
@ -639,6 +662,58 @@ void ClientBase::initLogsOutputStream()
}
}
void ClientBase::initTtyBuffer(bool to_err)
{
if (!tty_buf)
{
static constexpr auto tty_file_name = "/dev/tty";
/// Output all progress bar commands to terminal at once to avoid flicker.
/// This size is usually greater than the window size.
static constexpr size_t buf_size = 1024;
if (!to_err)
{
std::error_code ec;
std::filesystem::file_status tty = std::filesystem::status(tty_file_name, ec);
if (!ec && exists(tty) && is_character_file(tty)
&& (tty.permissions() & std::filesystem::perms::others_write) != std::filesystem::perms::none)
{
try
{
tty_buf = std::make_unique<WriteBufferFromFile>(tty_file_name, buf_size);
/// It is possible that the terminal file has writeable permissions
/// but we cannot write anything there. Check it with invisible character.
tty_buf->write('\0');
tty_buf->next();
return;
}
catch (const Exception & e)
{
if (tty_buf)
tty_buf.reset();
if (e.code() != ErrorCodes::CANNOT_OPEN_FILE)
throw;
/// It is normal if file exists, indicated as writeable but still cannot be opened.
/// Fallback to other options.
}
}
}
if (stderr_is_a_tty)
{
tty_buf = std::make_unique<WriteBufferFromFileDescriptor>(STDERR_FILENO, buf_size);
}
else
need_render_progress = false;
}
}
void ClientBase::updateSuggest(const ASTPtr & ast)
{
std::vector<std::string> new_words;
@ -937,14 +1012,15 @@ void ClientBase::onProgress(const Progress & value)
if (output_format)
output_format->onProgress(value);
if (need_render_progress)
progress_indication.writeProgress();
if (need_render_progress && tty_buf)
progress_indication.writeProgress(*tty_buf);
}
void ClientBase::onEndOfStream()
{
progress_indication.clearProgressOutput();
if (need_render_progress && tty_buf)
progress_indication.clearProgressOutput(*tty_buf);
if (output_format)
output_format->finalize();
@ -952,10 +1028,7 @@ void ClientBase::onEndOfStream()
resetOutput();
if (is_interactive && !written_first_block)
{
progress_indication.clearProgressOutput();
std::cout << "Ok." << std::endl;
}
}
@ -998,15 +1071,16 @@ void ClientBase::onProfileEvents(Block & block)
}
progress_indication.updateThreadEventData(thread_times);
if (need_render_progress)
progress_indication.writeProgress();
if (need_render_progress && tty_buf)
progress_indication.writeProgress(*tty_buf);
if (profile_events.print)
{
if (profile_events.watch.elapsedMilliseconds() >= profile_events.delay_ms)
{
initLogsOutputStream();
progress_indication.clearProgressOutput();
if (need_render_progress && tty_buf)
progress_indication.clearProgressOutput(*tty_buf);
logs_out_stream->writeProfileEvents(block);
logs_out_stream->flush();
@ -1173,14 +1247,15 @@ void ClientBase::sendData(Block & sample, const ColumnsDescription & columns_des
bool have_data_in_stdin = !is_interactive && !stdin_is_a_tty && !std_in.eof();
if (need_render_progress && have_data_in_stdin)
if (need_render_progress)
{
/// Set total_bytes_to_read for current fd.
FileProgress file_progress(0, std_in.getFileSize());
progress_indication.updateProgress(Progress(file_progress));
/// Set callback to be called on file progress.
progress_indication.setFileProgressCallback(global_context, true);
if (tty_buf)
progress_indication.setFileProgressCallback(global_context, *tty_buf);
}
/// If data fetched from file (maybe compressed file)
@ -1432,12 +1507,12 @@ bool ClientBase::receiveEndOfQuery()
void ClientBase::cancelQuery()
{
connection->sendCancel();
if (need_render_progress && tty_buf)
progress_indication.clearProgressOutput(*tty_buf);
if (is_interactive)
{
progress_indication.clearProgressOutput();
std::cout << "Cancelling query." << std::endl;
}
cancelled = true;
}
@ -1557,7 +1632,8 @@ void ClientBase::processParsedSingleQuery(const String & full_query, const Strin
if (profile_events.last_block)
{
initLogsOutputStream();
progress_indication.clearProgressOutput();
if (need_render_progress && tty_buf)
progress_indication.clearProgressOutput(*tty_buf);
logs_out_stream->writeProfileEvents(profile_events.last_block);
logs_out_stream->flush();
@ -2248,7 +2324,7 @@ void ClientBase::init(int argc, char ** argv)
("stage", po::value<std::string>()->default_value("complete"), "Request query processing up to specified stage: complete,fetch_columns,with_mergeable_state,with_mergeable_state_after_aggregation,with_mergeable_state_after_aggregation_and_limit")
("query_kind", po::value<std::string>()->default_value("initial_query"), "One of initial_query/secondary_query/no_query")
("query_id", po::value<std::string>(), "query_id")
("progress", "print progress of queries execution")
("progress", po::value<ProgressOption>()->implicit_value(ProgressOption::TTY, "tty")->default_value(ProgressOption::TTY, "tty"), "Print progress of queries execution - to TTY (default): tty|on|1|true|yes; to STDERR: err; OFF: off|0|false|no")
("disable_suggestion,A", "Disable loading suggestion data. Note that suggestion data is loaded asynchronously through a second connection to ClickHouse server. Also it is reasonable to disable suggestion if you want to paste a query with TAB characters. Shorthand option -A is for those who get used to mysql client.")
("time,t", "print query execution time to stderr in non-interactive mode (for benchmarks)")
@ -2303,6 +2379,11 @@ void ClientBase::init(int argc, char ** argv)
parseAndCheckOptions(options_description, options, common_arguments);
po::notify(options);
if (options["progress"].as<ProgressOption>() == ProgressOption::OFF)
need_render_progress = false;
else
initTtyBuffer(options["progress"].as<ProgressOption>() == ProgressOption::ERR);
if (options.count("version") || options.count("V"))
{
showClientVersion();
@ -2353,7 +2434,20 @@ void ClientBase::init(int argc, char ** argv)
if (options.count("profile-events-delay-ms"))
config().setUInt64("profile-events-delay-ms", options["profile-events-delay-ms"].as<UInt64>());
if (options.count("progress"))
config().setBool("progress", true);
{
switch (options["progress"].as<ProgressOption>())
{
case OFF:
config().setString("progress", "off");
break;
case TTY:
config().setString("progress", "tty");
break;
case ERR:
config().setString("progress", "err");
break;
}
}
if (options.count("echo"))
config().setBool("echo", true);
if (options.count("disable_suggestion"))

View File

@ -15,6 +15,7 @@
#include <Storages/StorageFile.h>
#include <Storages/SelectQueryInfo.h>
namespace po = boost::program_options;
@ -35,9 +36,18 @@ enum MultiQueryProcessingStage
PARSING_FAILED,
};
enum ProgressOption
{
OFF,
TTY,
ERR,
};
std::istream& operator>> (std::istream & in, ProgressOption & progress);
void interruptSignalHandler(int signum);
class InternalTextLogs;
class WriteBufferFromFileDescriptor;
class ClientBase : public Poco::Util::Application, public IHints<2, ClientBase>
{
@ -143,6 +153,7 @@ private:
void initOutputFormat(const Block & block, ASTPtr parsed_query);
void initLogsOutputStream();
void initTtyBuffer(bool to_err = false);
String prompt() const;
@ -218,6 +229,10 @@ protected:
String server_logs_file;
std::unique_ptr<InternalTextLogs> logs_out_stream;
/// /dev/tty if accessible or std::cerr - for progress bar.
/// We prefer to output progress bar directly to tty to allow user to redirect stdout and stderr and still get the progress indication.
std::unique_ptr<WriteBufferFromFileDescriptor> tty_buf;
String home_path;
String history_file; /// Path to a file containing command history.

View File

@ -1331,7 +1331,7 @@ public:
}
template <typename DateOrTime>
inline auto addQuarters(DateOrTime d, Int64 delta) const
inline auto NO_SANITIZE_UNDEFINED addQuarters(DateOrTime d, Int64 delta) const
{
return addMonths(d, delta * 3);
}

View File

@ -12,6 +12,7 @@
M(FailedQuery, "Number of failed queries.") \
M(FailedSelectQuery, "Same as FailedQuery, but only for SELECT queries.") \
M(FailedInsertQuery, "Same as FailedQuery, but only for INSERT queries.") \
M(FailedAsyncInsertQuery, "Number of failed ASYNC INSERT queries.") \
M(QueryTimeMicroseconds, "Total time of all queries.") \
M(SelectQueryTimeMicroseconds, "Total time of SELECT queries.") \
M(InsertQueryTimeMicroseconds, "Total time of INSERT queries.") \

View File

@ -2,6 +2,7 @@
#include <algorithm>
#include <cstddef>
#include <numeric>
#include <filesystem>
#include <cmath>
#include <IO/WriteBufferFromFileDescriptor.h>
#include <base/types.h>
@ -11,6 +12,9 @@
#include "IO/WriteBufferFromString.h"
#include <Databases/DatabaseMemory.h>
/// http://en.wikipedia.org/wiki/ANSI_escape_code
#define CLEAR_TO_END_OF_LINE "\033[K"
namespace
{
@ -44,15 +48,6 @@ bool ProgressIndication::updateProgress(const Progress & value)
return progress.incrementPiecewiseAtomically(value);
}
void ProgressIndication::clearProgressOutput()
{
if (written_progress_chars)
{
written_progress_chars = 0;
std::cerr << "\r" CLEAR_TO_END_OF_LINE;
}
}
void ProgressIndication::resetProgress()
{
watch.restart();
@ -67,15 +62,12 @@ void ProgressIndication::resetProgress()
}
}
void ProgressIndication::setFileProgressCallback(ContextMutablePtr context, bool write_progress_on_update_)
void ProgressIndication::setFileProgressCallback(ContextMutablePtr context, WriteBufferFromFileDescriptor & message)
{
write_progress_on_update = write_progress_on_update_;
context->setFileProgressCallback([&](const FileProgress & file_progress)
{
progress.incrementPiecewiseAtomically(Progress(file_progress));
if (write_progress_on_update)
writeProgress();
writeProgress(message);
});
}
@ -142,13 +134,10 @@ void ProgressIndication::writeFinalProgress()
std::cout << ". ";
}
void ProgressIndication::writeProgress()
void ProgressIndication::writeProgress(WriteBufferFromFileDescriptor & message)
{
std::lock_guard lock(progress_mutex);
/// Output all progress bar commands to stderr at once to avoid flicker.
WriteBufferFromFileDescriptor message(STDERR_FILENO, 1024);
static size_t increment = 0;
static const char * indicators[8] = {
"\033[1;30m→\033[0m",
@ -307,4 +296,14 @@ void ProgressIndication::writeProgress()
message.next();
}
void ProgressIndication::clearProgressOutput(WriteBufferFromFileDescriptor & message)
{
if (written_progress_chars)
{
written_progress_chars = 0;
message << "\r" CLEAR_TO_END_OF_LINE;
message.next();
}
}
}

View File

@ -9,12 +9,12 @@
#include <Common/Stopwatch.h>
#include <Common/EventRateMeter.h>
/// http://en.wikipedia.org/wiki/ANSI_escape_code
#define CLEAR_TO_END_OF_LINE "\033[K"
namespace DB
{
class WriteBufferFromFileDescriptor;
struct ThreadEventData
{
UInt64 time() const noexcept { return user_ms + system_ms; }
@ -30,14 +30,13 @@ using HostToThreadTimesMap = std::unordered_map<String, ThreadIdToTimeMap>;
class ProgressIndication
{
public:
/// Write progress to stderr.
void writeProgress();
/// Write progress bar.
void writeProgress(WriteBufferFromFileDescriptor & message);
void clearProgressOutput(WriteBufferFromFileDescriptor & message);
/// Write summary.
void writeFinalProgress();
/// Clear stderr output.
void clearProgressOutput();
/// Reset progress values.
void resetProgress();
@ -52,7 +51,7 @@ public:
/// In some cases there is a need to update progress value, when there is no access to progress_inidcation object.
/// In this case it is added via context.
/// `write_progress_on_update` is needed to write progress for loading files data via pipe in non-interactive mode.
void setFileProgressCallback(ContextMutablePtr context, bool write_progress_on_update = false);
void setFileProgressCallback(ContextMutablePtr context, WriteBufferFromFileDescriptor & message);
/// How much seconds passed since query execution start.
double elapsedSeconds() const { return getElapsedNanoseconds() / 1e9; }

View File

@ -3,10 +3,10 @@
#include <cstring>
#include <iostream>
#include <Core/Defines.h>
#include <Common/Stopwatch.h>
#include <Common/TargetSpecific.h>
#include <base/types.h>
#include <base/unaligned.h>
#include <Common/Stopwatch.h>
#include <Common/TargetSpecific.h>
#ifdef __SSE2__
#include <emmintrin.h>
@ -599,6 +599,9 @@ bool NO_INLINE decompressImpl(const char * const source, char * const dest, size
copy_end = op + length;
if (unlikely(copy_end > output_end))
return false;
/** Here we can write up to copy_amount - 1 - 4 * 2 bytes after buffer.
* The worst case when offset = 1 and length = 4
*/

View File

@ -1,8 +1,5 @@
#include <Compression/CompressionFactory.h>
#include <Common/PODArray.h>
#include <Common/Stopwatch.h>
#include <base/types.h>
#include <DataTypes/DataTypesNumber.h>
#include <DataTypes/IDataType.h>
#include <IO/ReadBufferFromMemory.h>
@ -10,6 +7,12 @@
#include <Parsers/ExpressionElementParsers.h>
#include <Parsers/IParser.h>
#include <Parsers/TokenIterator.h>
#include <base/types.h>
#include <Common/PODArray.h>
#include <Common/Stopwatch.h>
#include <Compression/LZ4_decompress_faster.h>
#include <IO/BufferWithOwnMemory.h>
#include <random>
#include <bitset>
@ -1319,4 +1322,34 @@ INSTANTIATE_TEST_SUITE_P(Gorilla,
// ),
//);
TEST(LZ4Test, DecompressMalformedInput)
{
/// This malformed input was initially found by lz4_decompress_fuzzer and causes failure under UBSAN.
constexpr unsigned char data[]
= {0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x07, 0x00,
0x00, 0x20, 0x00, 0x00, 0x66, 0x66, 0x66, 0x66, 0xff, 0xff, 0xff, 0x17, 0xff, 0xff, 0x0f, 0x00, 0x00, 0xff, 0xff, 0xff, 0xff,
0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
0xfe, 0x1f, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00};
const char * const source = reinterpret_cast<const char * const>(data);
const uint32_t source_size = std::size(data);
constexpr uint32_t uncompressed_size = 80;
DB::Memory<> memory;
memory.resize(ICompressionCodec::getHeaderSize() + uncompressed_size + LZ4::ADDITIONAL_BYTES_AT_END_OF_BUFFER);
unalignedStoreLE<uint8_t>(memory.data(), static_cast<uint8_t>(CompressionMethodByte::LZ4));
unalignedStoreLE<uint32_t>(&memory[1], source_size);
unalignedStoreLE<uint32_t>(&memory[5], uncompressed_size);
auto codec = CompressionCodecFactory::instance().get("LZ4", {});
ASSERT_THROW(codec->decompress(source, source_size, memory.data()), Exception);
}
}

View File

@ -36,7 +36,7 @@ void CoordinationSettings::loadFromConfig(const String & config_elem, const Poco
}
const String KeeperConfigurationAndSettings::DEFAULT_FOUR_LETTER_WORD_CMD = "conf,cons,crst,envi,ruok,srst,srvr,stat,wchs,dirs,mntr,isro,rcvr,apiv";
const String KeeperConfigurationAndSettings::DEFAULT_FOUR_LETTER_WORD_CMD = "conf,cons,crst,envi,ruok,srst,srvr,stat,wchs,dirs,mntr,isro,rcvr,apiv,csnp,lgif";
KeeperConfigurationAndSettings::KeeperConfigurationAndSettings()
: server_id(NOT_EXIST)

View File

@ -136,6 +136,12 @@ void FourLetterCommandFactory::registerCommands(KeeperDispatcher & keeper_dispat
FourLetterCommandPtr api_version_command = std::make_shared<ApiVersionCommand>(keeper_dispatcher);
factory.registerCommand(api_version_command);
FourLetterCommandPtr create_snapshot_command = std::make_shared<CreateSnapshotCommand>(keeper_dispatcher);
factory.registerCommand(create_snapshot_command);
FourLetterCommandPtr log_info_command = std::make_shared<LogInfoCommand>(keeper_dispatcher);
factory.registerCommand(log_info_command);
factory.initializeAllowList(keeper_dispatcher);
factory.setInitialize(true);
}
@ -472,4 +478,33 @@ String ApiVersionCommand::run()
return toString(static_cast<uint8_t>(Coordination::current_keeper_api_version));
}
String CreateSnapshotCommand::run()
{
auto log_index = keeper_dispatcher.createSnapshot();
return log_index > 0 ? std::to_string(log_index) : "Failed to schedule snapshot creation task.";
}
String LogInfoCommand::run()
{
KeeperLogInfo log_info = keeper_dispatcher.getKeeperLogInfo();
StringBuffer ret;
auto append = [&ret] (String key, uint64_t value) -> void
{
writeText(key, ret);
writeText('\t', ret);
writeText(std::to_string(value), ret);
writeText('\n', ret);
};
append("first_log_idx", log_info.first_log_idx);
append("first_log_term", log_info.first_log_idx);
append("last_log_idx", log_info.last_log_idx);
append("last_log_term", log_info.last_log_term);
append("last_committed_log_idx", log_info.last_committed_log_idx);
append("leader_committed_log_idx", log_info.leader_committed_log_idx);
append("target_committed_log_idx", log_info.target_committed_log_idx);
append("last_snapshot_idx", log_info.last_snapshot_idx);
return ret.str();
}
}

View File

@ -17,6 +17,7 @@ using FourLetterCommandPtr = std::shared_ptr<DB::IFourLetterCommand>;
/// Just like zookeeper Four Letter Words commands, CH Keeper responds to a small set of commands.
/// Each command is composed of four letters, these commands are useful to monitor and issue system problems.
/// The feature is based on Zookeeper 3.5.9, details is in https://zookeeper.apache.org/doc/r3.5.9/zookeeperAdmin.html#sc_zkCommands.
/// Also we add some additional commands such as csnp, lgif etc.
struct IFourLetterCommand
{
public:
@ -327,4 +328,40 @@ struct ApiVersionCommand : public IFourLetterCommand
String run() override;
~ApiVersionCommand() override = default;
};
/// Create snapshot manually
struct CreateSnapshotCommand : public IFourLetterCommand
{
explicit CreateSnapshotCommand(KeeperDispatcher & keeper_dispatcher_)
: IFourLetterCommand(keeper_dispatcher_)
{
}
String name() override { return "csnp"; }
String run() override;
~CreateSnapshotCommand() override = default;
};
/** Raft log information:
* first_log_idx 1
* first_log_term 1
* last_log_idx 101
* last_log_term 1
* last_committed_idx 100
* leader_committed_log_idx 101
* target_committed_log_idx 101
* last_snapshot_idx 50
*/
struct LogInfoCommand : public IFourLetterCommand
{
explicit LogInfoCommand(KeeperDispatcher & keeper_dispatcher_)
: IFourLetterCommand(keeper_dispatcher_)
{
}
String name() override { return "lgif"; }
String run() override;
~LogInfoCommand() override = default;
};
}

View File

@ -47,4 +47,32 @@ struct Keeper4LWInfo
}
};
/// Keeper log information for 4lw commands
struct KeeperLogInfo
{
/// My first log index in log store.
uint64_t first_log_idx;
/// My first log term.
uint64_t first_log_term;
/// My last log index in log store.
uint64_t last_log_idx;
/// My last log term.
uint64_t last_log_term;
/// My last committed log index in state machine.
uint64_t last_committed_log_idx;
/// Leader's committed log index from my perspective.
uint64_t leader_committed_log_idx;
/// Target log index should be committed to.
uint64_t target_committed_log_idx;
/// The largest committed log index in last snapshot.
uint64_t last_snapshot_idx;
};
}

View File

@ -203,6 +203,18 @@ public:
{
keeper_stats.reset();
}
/// Create snapshot manually, return the last committed log index in the snapshot
uint64_t createSnapshot()
{
return server->createSnapshot();
}
/// Get Raft information
KeeperLogInfo getKeeperLogInfo()
{
return server->getKeeperLogInfo();
}
};
}

View File

@ -907,4 +907,29 @@ Keeper4LWInfo KeeperServer::getPartiallyFilled4LWInfo() const
return result;
}
uint64_t KeeperServer::createSnapshot()
{
uint64_t log_idx = raft_instance->create_snapshot();
if (log_idx != 0)
LOG_INFO(log, "Snapshot creation scheduled with last committed log index {}.", log_idx);
else
LOG_WARNING(log, "Failed to schedule snapshot creation task.");
return log_idx;
}
KeeperLogInfo KeeperServer::getKeeperLogInfo()
{
KeeperLogInfo log_info;
auto log_store = state_manager->load_log_store();
log_info.first_log_idx = log_store->start_index();
log_info.first_log_term = log_store->term_at(log_info.first_log_idx);
log_info.last_log_idx = raft_instance->get_last_log_idx();
log_info.last_log_term = raft_instance->get_last_log_term();
log_info.last_committed_log_idx = raft_instance->get_committed_log_idx();
log_info.leader_committed_log_idx = raft_instance->get_leader_committed_log_idx();
log_info.target_committed_log_idx = raft_instance->get_target_committed_log_idx();
log_info.last_snapshot_idx = raft_instance->get_last_snapshot_idx();
return log_info;
}
}

View File

@ -131,6 +131,10 @@ public:
/// Wait configuration update for action. Used by followers.
/// Return true if update was successfully received.
bool waitConfigurationUpdate(const ConfigUpdateAction & task);
uint64_t createSnapshot();
KeeperLogInfo getKeeperLogInfo();
};
}

View File

@ -377,6 +377,9 @@ void KeeperStorage::UncommittedState::commit(int64_t commit_zxid)
{
assert(deltas.empty() || deltas.front().zxid >= commit_zxid);
// collect nodes that have no further modification in the current transaction
std::unordered_set<std::string> modified_nodes;
while (!deltas.empty() && deltas.front().zxid == commit_zxid)
{
if (std::holds_alternative<SubDeltaEnd>(deltas.front().operation))
@ -393,7 +396,17 @@ void KeeperStorage::UncommittedState::commit(int64_t commit_zxid)
assert(path_deltas.front() == &front_delta);
path_deltas.pop_front();
if (path_deltas.empty())
{
deltas_for_path.erase(front_delta.path);
// no more deltas for path -> no modification
modified_nodes.insert(std::move(front_delta.path));
}
else if (path_deltas.front()->zxid > commit_zxid)
{
// next delta has a zxid from a different transaction -> no modification in this transaction
modified_nodes.insert(std::move(front_delta.path));
}
}
else if (auto * add_auth = std::get_if<AddAuthDelta>(&front_delta.operation))
{
@ -409,9 +422,12 @@ void KeeperStorage::UncommittedState::commit(int64_t commit_zxid)
}
// delete all cached nodes that were not modified after the commit_zxid
// the commit can end on SubDeltaEnd so we don't want to clear cached nodes too soon
if (deltas.empty() || deltas.front().zxid > commit_zxid)
std::erase_if(nodes, [commit_zxid](const auto & node) { return node.second.zxid == commit_zxid; });
// we only need to check the nodes that were modified in this transaction
for (const auto & node : modified_nodes)
{
if (nodes[node].zxid == commit_zxid)
nodes.erase(node);
}
}
void KeeperStorage::UncommittedState::rollback(int64_t rollback_zxid)

View File

@ -411,6 +411,7 @@ inline bool isDecimal(const DataTypePtr & data_type) { return WhichDataType(data
inline bool isTuple(const DataTypePtr & data_type) { return WhichDataType(data_type).isTuple(); }
inline bool isArray(const DataTypePtr & data_type) { return WhichDataType(data_type).isArray(); }
inline bool isMap(const DataTypePtr & data_type) {return WhichDataType(data_type).isMap(); }
inline bool isInterval(const DataTypePtr & data_type) {return WhichDataType(data_type).isInterval(); }
inline bool isNothing(const DataTypePtr & data_type) { return WhichDataType(data_type).isNothing(); }
inline bool isUUID(const DataTypePtr & data_type) { return WhichDataType(data_type).isUUID(); }

View File

@ -84,11 +84,12 @@ void SerializationString::deserializeBinary(IColumn & column, ReadBuffer & istr)
void SerializationString::serializeBinaryBulk(const IColumn & column, WriteBuffer & ostr, size_t offset, size_t limit) const
{
const ColumnString & column_string = typeid_cast<const ColumnString &>(column);
const auto & full_column = column.convertToFullColumnIfLowCardinality();
const ColumnString & column_string = typeid_cast<const ColumnString &>(*full_column);
const ColumnString::Chars & data = column_string.getChars();
const ColumnString::Offsets & offsets = column_string.getOffsets();
size_t size = column.size();
size_t size = column_string.size();
if (!size)
return;

View File

@ -62,11 +62,12 @@ std::pair<String, StoragePtr> createTableFromAST(
if (ast_create_query.as_table_function)
{
const auto & factory = TableFunctionFactory::instance();
auto table_function = factory.get(ast_create_query.as_table_function, context);
auto table_function_ast = ast_create_query.as_table_function->ptr();
auto table_function = factory.get(table_function_ast, context);
ColumnsDescription columns;
if (ast_create_query.columns_list && ast_create_query.columns_list->columns)
columns = InterpreterCreateQuery::getColumnsDescription(*ast_create_query.columns_list->columns, context, true);
StoragePtr storage = table_function->execute(ast_create_query.as_table_function, context, ast_create_query.getTable(), std::move(columns));
StoragePtr storage = table_function->execute(table_function_ast, context, ast_create_query.getTable(), std::move(columns));
storage->renameInMemory(ast_create_query);
return {ast_create_query.getTable(), storage};
}

View File

@ -24,7 +24,6 @@
#include <Common/quoteString.h>
#include <Common/setThreadName.h>
#include <base/sleep.h>
#include <base/bit_cast.h>
#include <boost/algorithm/string/split.hpp>
#include <boost/algorithm/string/trim.hpp>
#include <Parsers/CommonParsers.h>

View File

@ -443,6 +443,11 @@ ASTPtr DatabasePostgreSQL::getColumnDeclaration(const DataTypePtr & data_type) c
if (which.isArray())
return makeASTFunction("Array", getColumnDeclaration(typeid_cast<const DataTypeArray *>(data_type.get())->getNestedType()));
if (which.isDateTime64())
{
return makeASTFunction("DateTime64", std::make_shared<ASTLiteral>(static_cast<UInt32>(6)));
}
return std::make_shared<ASTIdentifier>(data_type->getName());
}

View File

@ -11,6 +11,7 @@ enum class DataSourceType
Local,
RAM,
S3,
S3_Plain,
HDFS,
WebServer,
AzureBlobStorage,
@ -26,6 +27,8 @@ inline String toString(DataSourceType data_source_type)
return "memory";
case DataSourceType::S3:
return "s3";
case DataSourceType::S3_Plain:
return "s3_plain";
case DataSourceType::HDFS:
return "hdfs";
case DataSourceType::WebServer:

View File

@ -213,7 +213,9 @@ public:
template <class ...Args>
S3PlainObjectStorage(Args && ...args)
: S3ObjectStorage("S3PlainObjectStorage", std::forward<Args>(args)...)
{}
{
data_source_description.type = DataSourceType::S3_Plain;
}
};
}

View File

@ -2,21 +2,19 @@
#include "config.h"
#if USE_BASE64
# include <Columns/ColumnConst.h>
# include <Common/MemorySanitizer.h>
# include <Columns/ColumnFixedString.h>
# include <Columns/ColumnString.h>
# include <DataTypes/DataTypeString.h>
# include <Functions/FunctionFactory.h>
# include <Functions/FunctionHelpers.h>
# include <Functions/GatherUtils/Algorithms.h>
# include <IO/WriteHelpers.h>
# include <Functions/IFunction.h>
# include <Interpreters/Context_fwd.h>
# include <turbob64.h>
# include <Common/MemorySanitizer.h>
# include <span>
namespace DB
{
using namespace GatherUtils;
namespace ErrorCodes
{
extern const int BAD_ARGUMENTS;
@ -25,33 +23,86 @@ namespace ErrorCodes
extern const int INCORRECT_DATA;
}
namespace Detail
{
inline size_t base64Decode(const std::span<const UInt8> src, UInt8 * dst)
{
# if defined(__aarch64__)
return tb64sdec(reinterpret_cast<const uint8_t *>(src.data()), src.size(), reinterpret_cast<uint8_t *>(dst));
# else
return _tb64d(reinterpret_cast<const uint8_t *>(src.data()), src.size(), reinterpret_cast<uint8_t *>(dst));
# endif
}
}
struct Base64Encode
{
static constexpr auto name = "base64Encode";
static size_t getBufferSize(size_t string_length, size_t string_count)
static size_t getBufferSize(const size_t string_length, const size_t string_count)
{
return ((string_length - string_count) / 3 + string_count) * 4 + string_count;
}
static size_t performCoding(const std::span<const UInt8> src, UInt8 * dst)
{
/*
* Some bug in sse arm64 implementation?
* `base64Encode(repeat('a', 46))` returns wrong padding character
*/
# if defined(__aarch64__)
return tb64senc(reinterpret_cast<const uint8_t *>(src.data()), src.size(), reinterpret_cast<uint8_t *>(dst));
# else
return _tb64e(reinterpret_cast<const uint8_t *>(src.data()), src.size(), reinterpret_cast<uint8_t *>(dst));
# endif
}
};
struct Base64Decode
{
static constexpr auto name = "base64Decode";
static size_t getBufferSize(size_t string_length, size_t string_count)
static size_t getBufferSize(const size_t string_length, const size_t string_count)
{
return ((string_length - string_count) / 4 + string_count) * 3 + string_count;
}
static size_t performCoding(const std::span<const UInt8> src, UInt8 * dst)
{
const auto outlen = Detail::base64Decode(src, dst);
if (src.size() > 0 && !outlen)
throw Exception(
ErrorCodes::INCORRECT_DATA,
"Failed to {} input '{}'",
name,
String(reinterpret_cast<const char *>(src.data()), src.size()));
return outlen;
}
};
struct TryBase64Decode
{
static constexpr auto name = "tryBase64Decode";
static size_t getBufferSize(size_t string_length, size_t string_count)
static size_t getBufferSize(const size_t string_length, const size_t string_count)
{
return Base64Decode::getBufferSize(string_length, string_count);
}
static size_t performCoding(const std::span<const UInt8> src, UInt8 * dst)
{
if (src.empty())
return 0;
const auto outlen = Detail::base64Decode(src, dst);
// during decoding character array can be partially polluted
// if fail, revert back and clean
if (!outlen)
*dst = 0;
return outlen;
}
};
template <typename Func>
@ -71,99 +122,60 @@ public:
if (arguments.size() != 1)
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Wrong number of arguments for function {}: 1 expected.", getName());
if (!WhichDataType(arguments[0].type).isString())
if (!WhichDataType(arguments[0].type).isStringOrFixedString())
throw Exception(
ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT,
"Illegal type {} of 1st argument of function {}. Must be String.",
arguments[0].type->getName(), getName());
"Illegal type {} of 1st argument of function {}. Must be FixedString or String.",
arguments[0].type->getName(),
getName());
return std::make_shared<DataTypeString>();
}
ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override
ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, const size_t input_rows_count) const override
{
const ColumnPtr column_string = arguments[0].column;
const ColumnString * input = checkAndGetColumn<ColumnString>(column_string.get());
const auto & input_column = arguments[0].column;
if (const auto * src_column_as_fixed_string = checkAndGetColumn<ColumnFixedString>(*input_column))
return execute(*src_column_as_fixed_string, input_rows_count);
else if (const auto * src_column_as_string = checkAndGetColumn<ColumnString>(*input_column))
return execute(*src_column_as_string, input_rows_count);
if (!input)
throw Exception(
ErrorCodes::ILLEGAL_COLUMN,
"Illegal column {} of first argument of function {}, must be of type String",
arguments[0].column->getName(), getName());
throw Exception(
ErrorCodes::ILLEGAL_COLUMN,
"Illegal column {} of first argument of function {}, must be of type FixedString or String.",
input_column->getName(),
getName());
}
private:
static ColumnPtr execute(const ColumnString & src_column, const size_t src_row_count)
{
auto dst_column = ColumnString::create();
auto & dst_data = dst_column->getChars();
auto & dst_chars = dst_column->getChars();
auto & dst_offsets = dst_column->getOffsets();
size_t reserve = Func::getBufferSize(input->getChars().size(), input->size());
dst_data.resize(reserve);
dst_offsets.resize(input_rows_count);
const auto reserve = Func::getBufferSize(src_column.byteSize(), src_column.size());
dst_chars.resize(reserve);
dst_offsets.resize(src_row_count);
const ColumnString::Offsets & src_offsets = input->getOffsets();
const auto & src_chars = src_column.getChars();
const auto & src_offsets = src_column.getOffsets();
const auto * source = input->getChars().data();
auto * dst = dst_data.data();
auto * dst = dst_chars.data();
auto * dst_pos = dst;
const auto * src = src_chars.data();
size_t src_offset_prev = 0;
for (size_t row = 0; row < input_rows_count; ++row)
for (size_t row = 0; row < src_row_count; ++row)
{
size_t srclen = src_offsets[row] - src_offset_prev - 1;
size_t outlen = 0;
if constexpr (std::is_same_v<Func, Base64Encode>)
{
/*
* Some bug in sse arm64 implementation?
* `base64Encode(repeat('a', 46))` returns wrong padding character
*/
#if defined(__aarch64__)
outlen = tb64senc(reinterpret_cast<const uint8_t *>(source), srclen, reinterpret_cast<uint8_t *>(dst_pos));
#else
outlen = _tb64e(reinterpret_cast<const uint8_t *>(source), srclen, reinterpret_cast<uint8_t *>(dst_pos));
#endif
}
else if constexpr (std::is_same_v<Func, Base64Decode>)
{
if (srclen > 0)
{
#if defined(__aarch64__)
outlen = tb64sdec(reinterpret_cast<const uint8_t *>(source), srclen, reinterpret_cast<uint8_t *>(dst_pos));
#else
outlen = _tb64d(reinterpret_cast<const uint8_t *>(source), srclen, reinterpret_cast<uint8_t *>(dst_pos));
#endif
if (!outlen)
throw Exception(
ErrorCodes::INCORRECT_DATA,
"Failed to {} input '{}'",
getName(), String(reinterpret_cast<const char *>(source), srclen));
}
}
else
{
if (srclen > 0)
{
// during decoding character array can be partially polluted
// if fail, revert back and clean
auto * savepoint = dst_pos;
outlen = _tb64d(reinterpret_cast<const uint8_t *>(source), srclen, reinterpret_cast<uint8_t *>(dst_pos));
if (!outlen)
{
outlen = 0;
dst_pos = savepoint; //-V1048
// clean the symbol
dst_pos[0] = 0;
}
}
}
const size_t src_length = src_offsets[row] - src_offset_prev - 1;
const auto outlen = Func::performCoding({src, src_length}, dst_pos);
/// Base64 library is using AVX-512 with some shuffle operations.
/// Memory sanitizer don't understand if there was uninitialized memory in SIMD register but it was not used in the result of shuffle.
__msan_unpoison(dst_pos, outlen);
source += srclen + 1;
src += src_length + 1;
dst_pos += outlen;
*dst_pos = '\0';
dst_pos += 1;
@ -172,8 +184,44 @@ public:
src_offset_prev = src_offsets[row];
}
dst_data.resize(dst_pos - dst);
dst_chars.resize(dst_pos - dst);
return dst_column;
}
static ColumnPtr execute(const ColumnFixedString & src_column, const size_t src_row_count)
{
auto dst_column = ColumnString::create();
auto & dst_chars = dst_column->getChars();
auto & dst_offsets = dst_column->getOffsets();
const auto reserve = Func::getBufferSize(src_column.byteSize(), src_column.size());
dst_chars.resize(reserve);
dst_offsets.resize(src_row_count);
const auto & src_chars = src_column.getChars();
const auto & src_n = src_column.getN();
auto * dst = dst_chars.data();
auto * dst_pos = dst;
const auto * src = src_chars.data();
for (size_t row = 0; row < src_row_count; ++row)
{
const auto outlen = Func::performCoding({src, src_n}, dst_pos);
/// Base64 library is using AVX-512 with some shuffle operations.
/// Memory sanitizer don't understand if there was uninitialized memory in SIMD register but it was not used in the result of shuffle.
__msan_unpoison(dst_pos, outlen);
src += src_n;
dst_pos += outlen;
*dst_pos = '\0';
dst_pos += 1;
dst_offsets[row] = dst_pos - dst;
}
dst_chars.resize(dst_pos - dst);
return dst_column;
}
};

View File

@ -22,6 +22,7 @@
#include <DataTypes/DataTypeFactory.h>
#include <DataTypes/DataTypeFixedString.h>
#include <DataTypes/DataTypeInterval.h>
#include <DataTypes/DataTypeTuple.h>
#include <DataTypes/DataTypeString.h>
#include <DataTypes/DataTypesDecimal.h>
#include <DataTypes/DataTypesNumber.h>
@ -642,7 +643,8 @@ class FunctionBinaryArithmetic : public IFunction
DataTypeInt8, DataTypeInt16, DataTypeInt32, DataTypeInt64, DataTypeInt128, DataTypeInt256,
DataTypeDecimal32, DataTypeDecimal64, DataTypeDecimal128, DataTypeDecimal256,
DataTypeDate, DataTypeDateTime,
DataTypeFixedString, DataTypeString>;
DataTypeFixedString, DataTypeString,
DataTypeInterval>;
using Floats = TypeList<DataTypeFloat32, DataTypeFloat64>;
@ -717,6 +719,82 @@ class FunctionBinaryArithmetic : public IFunction
return FunctionFactory::instance().get(function_name, context);
}
static FunctionOverloadResolverPtr
getFunctionForDateTupleOfIntervalsArithmetic(const DataTypePtr & type0, const DataTypePtr & type1, ContextPtr context)
{
bool first_is_date_or_datetime = isDateOrDate32(type0) || isDateTime(type0) || isDateTime64(type0);
bool second_is_date_or_datetime = isDateOrDate32(type1) || isDateTime(type1) || isDateTime64(type1);
/// Exactly one argument must be Date or DateTime
if (first_is_date_or_datetime == second_is_date_or_datetime)
return {};
if (!isTuple(type0) && !isTuple(type1))
return {};
/// Special case when the function is plus or minus, one of arguments is Date/DateTime and another is Tuple.
/// We construct another function and call it.
if constexpr (!is_plus && !is_minus)
return {};
if (isTuple(type0) && second_is_date_or_datetime && is_minus)
throw Exception("Wrong order of arguments for function " + String(name) + ": argument of Tuple type cannot be first",
ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT);
std::string function_name;
if (is_plus)
{
function_name = "addTupleOfIntervals";
}
else
{
function_name = "subtractTupleOfIntervals";
}
return FunctionFactory::instance().get(function_name, context);
}
static FunctionOverloadResolverPtr
getFunctionForMergeIntervalsArithmetic(const DataTypePtr & type0, const DataTypePtr & type1, ContextPtr context)
{
/// Special case when the function is plus or minus, first argument is Interval or Tuple of Intervals
/// and the second argument is the Interval of a different kind.
/// We construct another function (example: addIntervals) and call it
if constexpr (!is_plus && !is_minus)
return {};
const auto * tuple_data_type_0 = checkAndGetDataType<DataTypeTuple>(type0.get());
const auto * interval_data_type_0 = checkAndGetDataType<DataTypeInterval>(type0.get());
const auto * interval_data_type_1 = checkAndGetDataType<DataTypeInterval>(type1.get());
if ((!tuple_data_type_0 && !interval_data_type_0) || !interval_data_type_1)
return {};
if (interval_data_type_0 && interval_data_type_0->equals(*interval_data_type_1))
return {};
if (tuple_data_type_0)
{
auto & tuple_types = tuple_data_type_0->getElements();
for (auto & type : tuple_types)
if (!isInterval(type))
return {};
}
std::string function_name;
if (is_plus)
{
function_name = "addInterval";
}
else
{
function_name = "subtractInterval";
}
return FunctionFactory::instance().get(function_name, context);
}
static FunctionOverloadResolverPtr
getFunctionForTupleArithmetic(const DataTypePtr & type0, const DataTypePtr & type1, ContextPtr context)
{
@ -915,6 +993,30 @@ class FunctionBinaryArithmetic : public IFunction
return function->execute(new_arguments, result_type, input_rows_count);
}
ColumnPtr executeDateTimeTupleOfIntervalsPlusMinus(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type,
size_t input_rows_count, const FunctionOverloadResolverPtr & function_builder) const
{
ColumnsWithTypeAndName new_arguments = arguments;
/// Tuple argument must be second.
if (isTuple(arguments[0].type))
std::swap(new_arguments[0], new_arguments[1]);
auto function = function_builder->build(new_arguments);
return function->execute(new_arguments, result_type, input_rows_count);
}
ColumnPtr executeIntervalTupleOfIntervalsPlusMinus(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type,
size_t input_rows_count, const FunctionOverloadResolverPtr & function_builder) const
{
ColumnsWithTypeAndName new_arguments = arguments;
auto function = function_builder->build(new_arguments);
return function->execute(new_arguments, result_type, input_rows_count);
}
ColumnPtr executeTupleNumberOperator(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type,
size_t input_rows_count, const FunctionOverloadResolverPtr & function_builder) const
{
@ -1134,6 +1236,34 @@ public:
return function->getResultType();
}
/// Special case when the function is plus or minus, one of arguments is Date/DateTime and another is Tuple.
if (auto function_builder = getFunctionForDateTupleOfIntervalsArithmetic(arguments[0], arguments[1], context))
{
ColumnsWithTypeAndName new_arguments(2);
for (size_t i = 0; i < 2; ++i)
new_arguments[i].type = arguments[i];
/// Tuple argument must be second.
if (isTuple(new_arguments[0].type))
std::swap(new_arguments[0], new_arguments[1]);
auto function = function_builder->build(new_arguments);
return function->getResultType();
}
/// Special case when the function is plus or minus, one of arguments is Interval/Tuple of Intervals and another is Interval.
if (auto function_builder = getFunctionForMergeIntervalsArithmetic(arguments[0], arguments[1], context))
{
ColumnsWithTypeAndName new_arguments(2);
for (size_t i = 0; i < 2; ++i)
new_arguments[i].type = arguments[i];
auto function = function_builder->build(new_arguments);
return function->getResultType();
}
/// Special case when the function is multiply or divide, one of arguments is Tuple and another is Number.
if (auto function_builder = getFunctionForTupleAndNumberArithmetic(arguments[0], arguments[1], context))
{
@ -1185,6 +1315,21 @@ public:
type_res = std::make_shared<DataTypeString>();
return true;
}
else if constexpr (std::is_same_v<LeftDataType, DataTypeInterval> || std::is_same_v<RightDataType, DataTypeInterval>)
{
if constexpr (std::is_same_v<LeftDataType, DataTypeInterval> &&
std::is_same_v<RightDataType, DataTypeInterval>)
{
if constexpr (is_plus || is_minus)
{
if (left.getKind() == right.getKind())
{
type_res = std::make_shared<LeftDataType>(left.getKind());
return true;
}
}
}
}
else
{
using ResultDataType = typename BinaryOperationTraits<Op, LeftDataType, RightDataType>::ResultDataType;
@ -1566,6 +1711,18 @@ public:
return executeDateTimeIntervalPlusMinus(arguments, result_type, input_rows_count, function_builder);
}
/// Special case when the function is plus or minus, one of arguments is Date/DateTime and another is Tuple.
if (auto function_builder = getFunctionForDateTupleOfIntervalsArithmetic(arguments[0].type, arguments[1].type, context))
{
return executeDateTimeTupleOfIntervalsPlusMinus(arguments, result_type, input_rows_count, function_builder);
}
/// Special case when the function is plus or minus, one of arguments is Interval/Tuple of Intervals and another is Interval.
if (auto function_builder = getFunctionForMergeIntervalsArithmetic(arguments[0].type, arguments[1].type, context))
{
return executeIntervalTupleOfIntervalsPlusMinus(arguments, result_type, input_rows_count, function_builder);
}
/// Special case when the function is plus, minus or multiply, both arguments are tuples.
if (auto function_builder = getFunctionForTupleArithmetic(arguments[0].type, arguments[1].type, context))
{

View File

@ -38,18 +38,21 @@ public:
{
if (!isStringOrFixedString(arguments[0]))
throw Exception(
"Illegal type " + arguments[0]->getName() + " of first argument of function " + getName(),
ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT);
ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT,
"Illegal type {} of first argument of function {}",
arguments[0]->getName(), getName());
if (!isStringOrFixedString(arguments[1]))
throw Exception(
"Illegal type " + arguments[1]->getName() + " of second argument of function " + getName(),
ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT);
ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT,
"Illegal type {} of second argument of function {}",
arguments[1]->getName(), getName());
if (!isStringOrFixedString(arguments[2]))
throw Exception(
"Illegal type " + arguments[2]->getName() + " of third argument of function " + getName(),
ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT);
ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT,
"Illegal type {} of third argument of function {}",
arguments[2]->getName(), getName());
return std::make_shared<DataTypeString>();
}
@ -61,7 +64,10 @@ public:
const ColumnPtr column_replacement = arguments[2].column;
if (!isColumnConst(*column_needle) || !isColumnConst(*column_replacement))
throw Exception("2nd and 3rd arguments of function " + getName() + " must be constants.", ErrorCodes::ILLEGAL_COLUMN);
throw Exception(
ErrorCodes::ILLEGAL_COLUMN,
"2nd and 3rd arguments of function {} must be constants.",
getName());
const IColumn * c1 = arguments[1].column.get();
const IColumn * c2 = arguments[2].column.get();
@ -71,7 +77,9 @@ public:
String replacement = c2_const->getValue<String>();
if (needle.empty())
throw Exception("Length of the second argument of function replace must be greater than 0.", ErrorCodes::ARGUMENT_OUT_OF_BOUND);
throw Exception(
ErrorCodes::ARGUMENT_OUT_OF_BOUND,
"Length of the second argument of function replace must be greater than 0.");
if (const ColumnString * col = checkAndGetColumn<ColumnString>(column_src.get()))
{
@ -87,8 +95,9 @@ public:
}
else
throw Exception(
"Illegal column " + arguments[0].column->getName() + " of first argument of function " + getName(),
ErrorCodes::ILLEGAL_COLUMN);
ErrorCodes::ILLEGAL_COLUMN,
"Illegal column {} of first argument of function {}",
arguments[0].column->getName(), getName());
}
};

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