thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-09-24 02:30:51 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'master' of https://github.com/ClickHouse/ClickHouse into HEAD

Browse Source
This commit is contained in:
achimbab 2021-02-27 18:33:13 +09:00
commit ee5d76a82b
535 changed files with 15267 additions and 4852 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
.github/codecov.yml vendored
View File

@ -1,5 +1,5 @@
codecov:
max_report_age: off
max_report_age: "off"
strict_yaml_branch: "master"
ignore:
@ -14,4 +14,4 @@ ignore:
comment: false
github_checks:
annotations: false
annotations: false

38
.github/workflows/anchore-analysis.yml vendored
View File

@ -8,9 +8,9 @@
name: Docker Container Scan (clickhouse-server)
on:
"on":
pull_request:
paths:
paths:
- docker/server/Dockerfile
- .github/workflows/anchore-analysis.yml
schedule:
@ -20,20 +20,20 @@ jobs:
Anchore-Build-Scan:
runs-on: ubuntu-latest
steps:
- name: Checkout the code
uses: actions/checkout@v2
- name: Build the Docker image
run: |
cd docker/server
perl -pi -e 's|=\$version||g' Dockerfile
docker build . --file Dockerfile --tag localbuild/testimage:latest
- name: Run the local Anchore scan action itself with GitHub Advanced Security code scanning integration enabled
uses: anchore/scan-action@v2
id: scan
with:
image: "localbuild/testimage:latest"
acs-report-enable: true
- name: Upload Anchore Scan Report
uses: github/codeql-action/upload-sarif@v1
with:
sarif_file: ${{ steps.scan.outputs.sarif }}
- name: Checkout the code
uses: actions/checkout@v2
- name: Build the Docker image
run: |
cd docker/server
perl -pi -e 's|=\$version||g' Dockerfile
docker build . --file Dockerfile --tag localbuild/testimage:latest
- name: Run the local Anchore scan action itself with GitHub Advanced Security code scanning integration enabled
uses: anchore/scan-action@v2
id: scan
with:
image: "localbuild/testimage:latest"
acs-report-enable: true
- name: Upload Anchore Scan Report
uses: github/codeql-action/upload-sarif@v1
with:
sarif_file: ${{ steps.scan.outputs.sarif }}

16
.potato.yml
View File

@ -14,14 +14,14 @@ handlers:
# The trigger for creating the Yandex.Tracker issue. When the specified event occurs, it transfers PR data to Yandex.Tracker.
github:pullRequest:labeled:
data:
# The Yandex.Tracker queue to create the issue in. Each issue in Tracker belongs to one of the project queues.
queue: CLICKHOUSEDOCS
# The issue title.
summary: '[Potato] Pull Request #{{pullRequest.number}}'
# The issue description.
description: >
# The Yandex.Tracker queue to create the issue in. Each issue in Tracker belongs to one of the project queues.
queue: CLICKHOUSEDOCS
# The issue title.
summary: '[Potato] Pull Request #{{pullRequest.number}}'
# The issue description.
description: >
{{pullRequest.description}}
Ссылка на Pull Request: {{pullRequest.webUrl}}
# The condition for creating the Yandex.Tracker issue.
condition: eventPayload.labels.filter(label => ['pr-feature'].includes(label.name)).length
# The condition for creating the Yandex.Tracker issue.
condition: eventPayload.labels.filter(label => ['pr-feature'].includes(label.name)).length

15
.yamllint Normal file
View File

@ -0,0 +1,15 @@
# vi: ft=yaml
extends: default
rules:
indentation:
level: warning
indent-sequences: consistent
line-length:
# there are some bash -c "", so this is OK
max: 300
level: warning
comments:
min-spaces-from-content: 1
document-start:
present: false

36
base/common/DateLUTImpl.h
View File

@ -7,6 +7,7 @@
#include <ctime>
#include <string>
#define DATE_LUT_MAX (0xFFFFFFFFU - 86400)
#define DATE_LUT_MAX_DAY_NUM (0xFFFFFFFFU / 86400)
/// Table size is bigger than DATE_LUT_MAX_DAY_NUM to fill all indices within UInt16 range: this allows to remove extra check.
@ -249,7 +250,7 @@ public:
{
DayNum index = findIndex(t);
if (unlikely(index == 0))
if (unlikely(index == 0 || index > DATE_LUT_MAX_DAY_NUM))
return t + offset_at_start_of_epoch;
time_t res = t - lut[index].date;
@ -264,18 +265,18 @@ public:
{
DayNum index = findIndex(t);
/// If it is not 1970 year (findIndex found nothing appropriate),
/// than limit number of hours to avoid insane results like 1970-01-01 89:28:15
if (unlikely(index == 0))
/// If it is overflow case,
/// then limit number of hours to avoid insane results like 1970-01-01 89:28:15
if (unlikely(index == 0 || index > DATE_LUT_MAX_DAY_NUM))
return static_cast<unsigned>((t + offset_at_start_of_epoch) / 3600) % 24;
time_t res = t - lut[index].date;
time_t time = t - lut[index].date;
/// Data is cleaned to avoid possibility of underflow.
if (res >= lut[index].time_at_offset_change)
res += lut[index].amount_of_offset_change;
if (time >= lut[index].time_at_offset_change)
time += lut[index].amount_of_offset_change;
return res / 3600;
unsigned res = time / 3600;
return res <= 23 ? res : 0;
}
/** Calculating offset from UTC in seconds.
@ -314,12 +315,12 @@ public:
* each minute, with added or subtracted leap second, spans exactly 60 unix timestamps.
*/
inline unsigned toSecond(time_t t) const { return t % 60; }
inline unsigned toSecond(time_t t) const { return UInt32(t) % 60; }
inline unsigned toMinute(time_t t) const
{
if (offset_is_whole_number_of_hours_everytime)
return (t / 60) % 60;
return (UInt32(t) / 60) % 60;
UInt32 date = find(t).date;
return (UInt32(t) - date) / 60 % 60;
@ -555,9 +556,7 @@ public:
}
}
/*
* check and change mode to effective
*/
/// Check and change mode to effective.
inline UInt8 check_week_mode(UInt8 mode) const
{
UInt8 week_format = (mode & 7);
@ -566,10 +565,9 @@ public:
return week_format;
}
/*
* Calc weekday from d
* Returns 0 for monday, 1 for tuesday ...
*/
/** Calculate weekday from d.
* Returns 0 for monday, 1 for tuesday...
*/
inline unsigned calc_weekday(DayNum d, bool sunday_first_day_of_week) const
{
if (!sunday_first_day_of_week)
@ -578,7 +576,7 @@ public:
return toDayOfWeek(DayNum(d + 1)) - 1;
}
/* Calc days in one year. */
/// Calculate days in one year.
inline unsigned calc_days_in_year(UInt16 year) const
{
return ((year & 3) == 0 && (year % 100 || (year % 400 == 0 && year)) ? 366 : 365);

50
base/common/arithmeticOverflow.h
View File

@ -6,6 +6,25 @@
namespace common
{
/// Multiply and ignore overflow.
template <typename T1, typename T2>
inline auto NO_SANITIZE_UNDEFINED mulIgnoreOverflow(T1 x, T2 y)
{
return x * y;
}
template <typename T1, typename T2>
inline auto NO_SANITIZE_UNDEFINED addIgnoreOverflow(T1 x, T2 y)
{
return x + y;
}
template <typename T1, typename T2>
inline auto NO_SANITIZE_UNDEFINED subIgnoreOverflow(T1 x, T2 y)
{
return x - y;
}
template <typename T>
inline bool addOverflow(T x, T y, T & res)
{
@ -35,14 +54,14 @@ namespace common
{
static constexpr __int128 min_int128 = minInt128();
static constexpr __int128 max_int128 = maxInt128();
res = x + y;
res = addIgnoreOverflow(x, y);
return (y > 0 && x > max_int128 - y) || (y < 0 && x < min_int128 - y);
}
template <>
inline bool addOverflow(wInt256 x, wInt256 y, wInt256 & res)
{
res = x + y;
res = addIgnoreOverflow(x, y);
return (y > 0 && x > std::numeric_limits<wInt256>::max() - y) ||
(y < 0 && x < std::numeric_limits<wInt256>::min() - y);
}
@ -50,7 +69,7 @@ namespace common
template <>
inline bool addOverflow(wUInt256 x, wUInt256 y, wUInt256 & res)
{
res = x + y;
res = addIgnoreOverflow(x, y);
return x > std::numeric_limits<wUInt256>::max() - y;
}
@ -83,14 +102,14 @@ namespace common
{
static constexpr __int128 min_int128 = minInt128();
static constexpr __int128 max_int128 = maxInt128();
res = x - y;
res = subIgnoreOverflow(x, y);
return (y < 0 && x > max_int128 + y) || (y > 0 && x < min_int128 + y);
}
template <>
inline bool subOverflow(wInt256 x, wInt256 y, wInt256 & res)
{
res = x - y;
res = subIgnoreOverflow(x, y);
return (y < 0 && x > std::numeric_limits<wInt256>::max() + y) ||
(y > 0 && x < std::numeric_limits<wInt256>::min() + y);
}
@ -98,7 +117,7 @@ namespace common
template <>
inline bool subOverflow(wUInt256 x, wUInt256 y, wUInt256 & res)
{
res = x - y;
res = subIgnoreOverflow(x, y);
return x < y;
}
@ -129,40 +148,33 @@ namespace common
template <>
inline bool mulOverflow(__int128 x, __int128 y, __int128 & res)
{
res = static_cast<unsigned __int128>(x) * static_cast<unsigned __int128>(y); /// Avoid signed integer overflow.
res = mulIgnoreOverflow(x, y);
if (!x || !y)
return false;
unsigned __int128 a = (x > 0) ? x : -x;
unsigned __int128 b = (y > 0) ? y : -y;
return (a * b) / b != a;
return mulIgnoreOverflow(a, b) / b != a;
}
template <>
inline bool mulOverflow(wInt256 x, wInt256 y, wInt256 & res)
{
res = x * y;
res = mulIgnoreOverflow(x, y);
if (!x || !y)
return false;
wInt256 a = (x > 0) ? x : -x;
wInt256 b = (y > 0) ? y : -y;
return (a * b) / b != a;
return mulIgnoreOverflow(a, b) / b != a;
}
template <>
inline bool mulOverflow(wUInt256 x, wUInt256 y, wUInt256 & res)
{
res = x * y;
res = mulIgnoreOverflow(x, y);
if (!x || !y)
return false;
return (x * y) / y != x;
}
/// Multiply and ignore overflow.
template <typename T1, typename T2>
inline auto NO_SANITIZE_UNDEFINED mulIgnoreOverflow(T1 x, T2 y)
{
return x * y;
return res / y != x;
}
}

27
base/common/defines.h
View File

@ -1,5 +1,20 @@
#pragma once
/// __has_feature supported only by clang.
///
/// But libcxx/libcxxabi overrides it to 0,
/// thus the checks for __has_feature will be wrong.
///
/// NOTE:
/// - __has_feature cannot be simply undefined,
/// since this will be broken if some C++ header will be included after
/// including <common/defines.h>
/// - it should not have fallback to 0,
/// since this may create false-positive detection (common problem)
#if defined(__clang__) && defined(__has_feature)
# define ch_has_feature __has_feature
#endif
#if defined(_MSC_VER)
# if !defined(likely)
# define likely(x) (x)
@ -32,8 +47,8 @@
/// Check for presence of address sanitizer
#if !defined(ADDRESS_SANITIZER)
# if defined(__has_feature)
# if __has_feature(address_sanitizer)
# if defined(ch_has_feature)
# if ch_has_feature(address_sanitizer)
# define ADDRESS_SANITIZER 1
# endif
# elif defined(__SANITIZE_ADDRESS__)
@ -42,8 +57,8 @@
#endif
#if !defined(THREAD_SANITIZER)
# if defined(__has_feature)
# if __has_feature(thread_sanitizer)
# if defined(ch_has_feature)
# if ch_has_feature(thread_sanitizer)
# define THREAD_SANITIZER 1
# endif
# elif defined(__SANITIZE_THREAD__)
@ -52,8 +67,8 @@
#endif
#if !defined(MEMORY_SANITIZER)
# if defined(__has_feature)
# if __has_feature(memory_sanitizer)
# if defined(ch_has_feature)
# if ch_has_feature(memory_sanitizer)
# define MEMORY_SANITIZER 1
# endif
# elif defined(__MEMORY_SANITIZER__)

10
base/common/phdr_cache.cpp
View File

@ -15,11 +15,11 @@
#endif
#define __msan_unpoison(X, Y) // NOLINT
#if defined(__has_feature)
# if __has_feature(memory_sanitizer)
# undef __msan_unpoison
# include <sanitizer/msan_interface.h>
# endif
#if defined(ch_has_feature)
# if ch_has_feature(memory_sanitizer)
# undef __msan_unpoison
# include <sanitizer/msan_interface.h>
# endif
#endif
#include <link.h>

4
base/daemon/BaseDaemon.cpp
View File

@ -416,7 +416,9 @@ static void sanitizerDeathCallback()
else
log_message = "Terminate called without an active exception";
static const size_t buf_size = 1024;
/// POSIX.1 says that write(2)s of less than PIPE_BUF bytes must be atomic - man 7 pipe
/// And the buffer should not be too small because our exception messages can be large.
static constexpr size_t buf_size = PIPE_BUF;
if (log_message.size() > buf_size - 16)
log_message.resize(buf_size - 16);

6
base/daemon/BaseDaemon.h
View File

@ -83,7 +83,7 @@ public:
template <class T>
void writeToGraphite(const std::string & key, const T & value, const std::string & config_name = DEFAULT_GRAPHITE_CONFIG_NAME, time_t timestamp = 0, const std::string & custom_root_path = "")
{
auto writer = getGraphiteWriter(config_name);
auto *writer = getGraphiteWriter(config_name);
if (writer)
writer->write(key, value, timestamp, custom_root_path);
}
@ -91,7 +91,7 @@ public:
template <class T>
void writeToGraphite(const GraphiteWriter::KeyValueVector<T> & key_vals, const std::string & config_name = DEFAULT_GRAPHITE_CONFIG_NAME, time_t timestamp = 0, const std::string & custom_root_path = "")
{
auto writer = getGraphiteWriter(config_name);
auto *writer = getGraphiteWriter(config_name);
if (writer)
writer->write(key_vals, timestamp, custom_root_path);
}
@ -99,7 +99,7 @@ public:
template <class T>
void writeToGraphite(const GraphiteWriter::KeyValueVector<T> & key_vals, const std::chrono::system_clock::time_point & current_time, const std::string & custom_root_path)
{
auto writer = getGraphiteWriter();
auto *writer = getGraphiteWriter();
if (writer)
writer->write(key_vals, std::chrono::system_clock::to_time_t(current_time), custom_root_path);
}

13
base/mysqlxx/Connection.cpp
View File

@ -51,10 +51,11 @@ Connection::Connection(
const char* ssl_key,
unsigned timeout,
unsigned rw_timeout,
bool enable_local_infile)
bool enable_local_infile,
bool opt_reconnect)
: Connection()
{
connect(db, server, user, password, port, socket, ssl_ca, ssl_cert, ssl_key, timeout, rw_timeout, enable_local_infile);
connect(db, server, user, password, port, socket, ssl_ca, ssl_cert, ssl_key, timeout, rw_timeout, enable_local_infile, opt_reconnect);
}
Connection::Connection(const std::string & config_name)
@ -80,7 +81,8 @@ void Connection::connect(const char* db,
const char * ssl_key,
unsigned timeout,
unsigned rw_timeout,
bool enable_local_infile)
bool enable_local_infile,
bool opt_reconnect)
{
if (is_connected)
disconnect();
@ -104,9 +106,8 @@ void Connection::connect(const char* db,
if (mysql_options(driver.get(), MYSQL_OPT_LOCAL_INFILE, &enable_local_infile_arg))
throw ConnectionFailed(errorMessage(driver.get()), mysql_errno(driver.get()));
/// Enables auto-reconnect.
bool reconnect = true;
if (mysql_options(driver.get(), MYSQL_OPT_RECONNECT, reinterpret_cast<const char *>(&reconnect)))
/// See C API Developer Guide: Automatic Reconnection Control
if (mysql_options(driver.get(), MYSQL_OPT_RECONNECT, reinterpret_cast<const char *>(&opt_reconnect)))
throw ConnectionFailed(errorMessage(driver.get()), mysql_errno(driver.get()));
/// Specifies particular ssl key and certificate if it needs

12
base/mysqlxx/Connection.h
View File

@ -14,6 +14,8 @@
/// Disable LOAD DATA LOCAL INFILE because it is insecure
#define MYSQLXX_DEFAULT_ENABLE_LOCAL_INFILE false
/// See https://dev.mysql.com/doc/c-api/5.7/en/c-api-auto-reconnect.html
#define MYSQLXX_DEFAULT_MYSQL_OPT_RECONNECT true
namespace mysqlxx
@ -76,7 +78,8 @@ public:
const char * ssl_key = "",
unsigned timeout = MYSQLXX_DEFAULT_TIMEOUT,
unsigned rw_timeout = MYSQLXX_DEFAULT_RW_TIMEOUT,
bool enable_local_infile = MYSQLXX_DEFAULT_ENABLE_LOCAL_INFILE);
bool enable_local_infile = MYSQLXX_DEFAULT_ENABLE_LOCAL_INFILE,
bool opt_reconnect = MYSQLXX_DEFAULT_MYSQL_OPT_RECONNECT);
/// Creates connection. Can be used if Poco::Util::Application is using.
/// All settings will be got from config_name section of configuration.
@ -96,7 +99,8 @@ public:
const char* ssl_key,
unsigned timeout = MYSQLXX_DEFAULT_TIMEOUT,
unsigned rw_timeout = MYSQLXX_DEFAULT_RW_TIMEOUT,
bool enable_local_infile = MYSQLXX_DEFAULT_ENABLE_LOCAL_INFILE);
bool enable_local_infile = MYSQLXX_DEFAULT_ENABLE_LOCAL_INFILE,
bool opt_reconnect = MYSQLXX_DEFAULT_MYSQL_OPT_RECONNECT);
void connect(const std::string & config_name)
{
@ -112,6 +116,7 @@ public:
std::string ssl_cert = cfg.getString(config_name + ".ssl_cert", "");
std::string ssl_key = cfg.getString(config_name + ".ssl_key", "");
bool enable_local_infile = cfg.getBool(config_name + ".enable_local_infile", MYSQLXX_DEFAULT_ENABLE_LOCAL_INFILE);
bool opt_reconnect = cfg.getBool(config_name + ".opt_reconnect", MYSQLXX_DEFAULT_MYSQL_OPT_RECONNECT);
unsigned timeout =
cfg.getInt(config_name + ".connect_timeout",
@ -135,7 +140,8 @@ public:
ssl_key.c_str(),
timeout,
rw_timeout,
enable_local_infile);
enable_local_infile,
opt_reconnect);
}
/// If MySQL connection was established.

9
base/mysqlxx/Exception.h
View File

@ -26,6 +26,15 @@ struct ConnectionFailed : public Exception
};
/// Connection to MySQL server was lost
struct ConnectionLost : public Exception
{
ConnectionLost(const std::string & msg, int code = 0) : Exception(msg, code) {}
const char * name() const throw() override { return "mysqlxx::ConnectionLost"; }
const char * className() const throw() override { return "mysqlxx::ConnectionLost"; }
};
/// Erroneous query.
struct BadQuery : public Exception
{

62
base/mysqlxx/Pool.cpp
View File

@ -10,7 +10,6 @@
#include <common/sleep.h>
#include <Poco/Util/Application.h>
#include <Poco/Util/LayeredConfiguration.h>
@ -41,7 +40,9 @@ void Pool::Entry::decrementRefCount()
Pool::Pool(const Poco::Util::AbstractConfiguration & cfg, const std::string & config_name,
unsigned default_connections_, unsigned max_connections_,
const char * parent_config_name_)
: default_connections(default_connections_), max_connections(max_connections_)
: logger(Poco::Logger::get("mysqlxx::Pool"))
, default_connections(default_connections_)
, max_connections(max_connections_)
{
server = cfg.getString(config_name + ".host");
@ -78,6 +79,9 @@ Pool::Pool(const Poco::Util::AbstractConfiguration & cfg, const std::string & co
enable_local_infile = cfg.getBool(config_name + ".enable_local_infile",
cfg.getBool(parent_config_name + ".enable_local_infile", MYSQLXX_DEFAULT_ENABLE_LOCAL_INFILE));
opt_reconnect = cfg.getBool(config_name + ".opt_reconnect",
cfg.getBool(parent_config_name + ".opt_reconnect", MYSQLXX_DEFAULT_MYSQL_OPT_RECONNECT));
}
else
{
@ -96,6 +100,8 @@ Pool::Pool(const Poco::Util::AbstractConfiguration & cfg, const std::string & co
enable_local_infile = cfg.getBool(
config_name + ".enable_local_infile", MYSQLXX_DEFAULT_ENABLE_LOCAL_INFILE);
opt_reconnect = cfg.getBool(config_name + ".opt_reconnect", MYSQLXX_DEFAULT_MYSQL_OPT_RECONNECT);
}
connect_timeout = cfg.getInt(config_name + ".connect_timeout",
@ -125,20 +131,30 @@ Pool::Entry Pool::get()
initialize();
for (;;)
{
logger.trace("(%s): Iterating through existing MySQL connections", getDescription());
for (auto & connection : connections)
{
if (connection->ref_count == 0)
return Entry(connection, this);
}
logger.trace("(%s): Trying to allocate a new connection.", getDescription());
if (connections.size() < static_cast<size_t>(max_connections))
{
Connection * conn = allocConnection();
if (conn)
return Entry(conn, this);
logger.trace("(%s): Unable to create a new connection: Allocation failed.", getDescription());
}
else
{
logger.trace("(%s): Unable to create a new connection: Max number of connections has been reached.", getDescription());
}
lock.unlock();
logger.trace("(%s): Sleeping for %d seconds.", getDescription(), MYSQLXX_POOL_SLEEP_ON_CONNECT_FAIL);
sleepForSeconds(MYSQLXX_POOL_SLEEP_ON_CONNECT_FAIL);
lock.lock();
}
@ -162,8 +178,7 @@ Pool::Entry Pool::tryGet()
if (res.tryForceConnected()) /// Tries to reestablish connection as well
return res;
auto & logger = Poco::Util::Application::instance().logger();
logger.information("Idle connection to mysql server cannot be recovered, dropping it.");
logger.debug("(%s): Idle connection to MySQL server cannot be recovered, dropping it.", getDescription());
/// This one is disconnected, cannot be reestablished and so needs to be disposed of.
connection_it = connections.erase(connection_it);
@ -186,6 +201,8 @@ Pool::Entry Pool::tryGet()
void Pool::removeConnection(Connection* connection)
{
logger.trace("(%s): Removing connection.", getDescription());
std::lock_guard<std::mutex> lock(mutex);
if (connection)
{
@ -210,8 +227,6 @@ void Pool::Entry::forceConnected() const
if (data == nullptr)
throw Poco::RuntimeException("Tried to access NULL database connection.");
Poco::Util::Application & app = Poco::Util::Application::instance();
bool first = true;
while (!tryForceConnected())
{
@ -220,7 +235,7 @@ void Pool::Entry::forceConnected() const
else
sleepForSeconds(MYSQLXX_POOL_SLEEP_ON_CONNECT_FAIL);
app.logger().information("MYSQL: Reconnecting to " + pool->description);
pool->logger.debug("Entry: Reconnecting to MySQL server %s", pool->description);
data->conn.connect(
pool->db.c_str(),
pool->server.c_str(),
@ -233,7 +248,8 @@ void Pool::Entry::forceConnected() const
pool->ssl_key.c_str(),
pool->connect_timeout,
pool->rw_timeout,
pool->enable_local_infile);
pool->enable_local_infile,
pool->opt_reconnect);
}
}
@ -242,18 +258,22 @@ bool Pool::Entry::tryForceConnected() const
{
auto * const mysql_driver = data->conn.getDriver();
const auto prev_connection_id = mysql_thread_id(mysql_driver);
pool->logger.trace("Entry(connection %lu): sending PING to check if it is alive.", prev_connection_id);
if (data->conn.ping()) /// Attempts to reestablish lost connection
{
const auto current_connection_id = mysql_thread_id(mysql_driver);
if (prev_connection_id != current_connection_id)
{
auto & logger = Poco::Util::Application::instance().logger();
logger.information("Connection to mysql server has been reestablished. Connection id changed: %lu -> %lu",
prev_connection_id, current_connection_id);
pool->logger.debug("Entry(connection %lu): Reconnected to MySQL server. Connection id changed: %lu -> %lu",
current_connection_id, prev_connection_id, current_connection_id);
}
pool->logger.trace("Entry(connection %lu): PING ok.", current_connection_id);
return true;
}
pool->logger.trace("Entry(connection %lu): PING failed.", prev_connection_id);
return false;
}
@ -274,15 +294,13 @@ void Pool::initialize()
Pool::Connection * Pool::allocConnection(bool dont_throw_if_failed_first_time)
{
Poco::Util::Application & app = Poco::Util::Application::instance();
std::unique_ptr<Connection> conn(new Connection);
std::unique_ptr<Connection> conn_ptr{new Connection};
try
{
app.logger().information("MYSQL: Connecting to " + description);
logger.debug("Connecting to %s", description);
conn->conn.connect(
conn_ptr->conn.connect(
db.c_str(),
server.c_str(),
user.c_str(),
@ -294,29 +312,29 @@ Pool::Connection * Pool::allocConnection(bool dont_throw_if_failed_first_time)
ssl_key.c_str(),
connect_timeout,
rw_timeout,
enable_local_infile);
enable_local_infile,
opt_reconnect);
}
catch (mysqlxx::ConnectionFailed & e)
{
logger.error(e.what());
if ((!was_successful && !dont_throw_if_failed_first_time)
|| e.errnum() == ER_ACCESS_DENIED_ERROR
|| e.errnum() == ER_DBACCESS_DENIED_ERROR
|| e.errnum() == ER_BAD_DB_ERROR)
{
app.logger().error(e.what());
throw;
}
else
{
app.logger().error(e.what());
return nullptr;
}
}
connections.push_back(conn_ptr.get());
was_successful = true;
auto * connection = conn.release();
connections.push_back(connection);
return connection;
return conn_ptr.release();
}
}

19
base/mysqlxx/Pool.h
View File

@ -6,6 +6,8 @@
#include <atomic>
#include <Poco/Exception.h>
#include <Poco/Logger.h>
#include <mysqlxx/Connection.h>
@ -165,19 +167,21 @@ public:
unsigned rw_timeout_ = MYSQLXX_DEFAULT_RW_TIMEOUT,
unsigned default_connections_ = MYSQLXX_POOL_DEFAULT_START_CONNECTIONS,
unsigned max_connections_ = MYSQLXX_POOL_DEFAULT_MAX_CONNECTIONS,
unsigned enable_local_infile_ = MYSQLXX_DEFAULT_ENABLE_LOCAL_INFILE)
: default_connections(default_connections_), max_connections(max_connections_),
db(db_), server(server_), user(user_), password(password_), port(port_), socket(socket_),
connect_timeout(connect_timeout_), rw_timeout(rw_timeout_), enable_local_infile(enable_local_infile_) {}
unsigned enable_local_infile_ = MYSQLXX_DEFAULT_ENABLE_LOCAL_INFILE,
bool opt_reconnect_ = MYSQLXX_DEFAULT_MYSQL_OPT_RECONNECT)
: logger(Poco::Logger::get("mysqlxx::Pool")), default_connections(default_connections_),
max_connections(max_connections_), db(db_), server(server_), user(user_), password(password_), port(port_), socket(socket_),
connect_timeout(connect_timeout_), rw_timeout(rw_timeout_), enable_local_infile(enable_local_infile_),
opt_reconnect(opt_reconnect_) {}
Pool(const Pool & other)
: default_connections{other.default_connections},
: logger(other.logger), default_connections{other.default_connections},
max_connections{other.max_connections},
db{other.db}, server{other.server},
user{other.user}, password{other.password},
port{other.port}, socket{other.socket},
connect_timeout{other.connect_timeout}, rw_timeout{other.rw_timeout},
enable_local_infile{other.enable_local_infile}
enable_local_infile{other.enable_local_infile}, opt_reconnect(other.opt_reconnect)
{}
Pool & operator=(const Pool &) = delete;
@ -201,6 +205,8 @@ public:
void removeConnection(Connection * connection);
protected:
Poco::Logger & logger;
/// Number of MySQL connections which are created at launch.
unsigned default_connections;
/// Maximum possible number of connections
@ -231,6 +237,7 @@ private:
std::string ssl_cert;
std::string ssl_key;
bool enable_local_infile;
bool opt_reconnect;
/// True if connection was established at least once.
bool was_successful{false};

18
base/mysqlxx/PoolWithFailover.cpp
View File

@ -1,3 +1,8 @@
#include <algorithm>
#include <ctime>
#include <random>
#include <thread>
#include <mysqlxx/PoolWithFailover.h>
@ -33,6 +38,19 @@ PoolWithFailover::PoolWithFailover(const Poco::Util::AbstractConfiguration & con
std::make_shared<Pool>(config_, replica_name, default_connections_, max_connections_, config_name_.c_str()));
}
}
/// PoolWithFailover objects are stored in a cache inside PoolFactory.
/// This cache is reset by ExternalDictionariesLoader after every SYSTEM RELOAD DICTIONAR{Y|IES}
/// which triggers massive re-constructing of connection pools.
/// The state of PRNGs like std::mt19937 is considered to be quite heavy
/// thus here we attempt to optimize its construction.
static thread_local std::mt19937 rnd_generator(
std::hash<std::thread::id>{}(std::this_thread::get_id()) + std::clock());
for (auto & [_, replicas] : replicas_by_priority)
{
if (replicas.size() > 1)
std::shuffle(replicas.begin(), replicas.end(), rnd_generator);
}
}
else
{

25
base/mysqlxx/Query.cpp
View File

@ -1,11 +1,16 @@
#if __has_include(<mysql.h>)
#include <errmsg.h>
#include <mysql.h>
#else
#include <mysql/errmsg.h>
#include <mysql/mysql.h>
#endif
#include <Poco/Logger.h>
#include <mysqlxx/Connection.h>
#include <mysqlxx/Query.h>
#include <mysqlxx/Types.h>
namespace mysqlxx
@ -57,8 +62,24 @@ void Query::reset()
void Query::executeImpl()
{
std::string query_string = query_buf.str();
if (mysql_real_query(conn->getDriver(), query_string.data(), query_string.size()))
throw BadQuery(errorMessage(conn->getDriver()), mysql_errno(conn->getDriver()));
MYSQL* mysql_driver = conn->getDriver();
auto & logger = Poco::Logger::get("mysqlxx::Query");
logger.trace("Running MySQL query using connection %lu", mysql_thread_id(mysql_driver));
if (mysql_real_query(mysql_driver, query_string.data(), query_string.size()))
{
const auto err_no = mysql_errno(mysql_driver);
switch (err_no)
{
case CR_SERVER_GONE_ERROR:
[[fallthrough]];
case CR_SERVER_LOST:
throw ConnectionLost(errorMessage(mysql_driver), err_no);
default:
throw BadQuery(errorMessage(mysql_driver), err_no);
}
}
}
UseQueryResult Query::use()

16
cmake/find/ccache.cmake
View File

@ -32,27 +32,25 @@ if (CCACHE_FOUND AND NOT COMPILER_MATCHES_CCACHE)
if (CCACHE_VERSION VERSION_GREATER "3.2.0" OR NOT CMAKE_CXX_COMPILER_ID STREQUAL "Clang")
message(STATUS "Using ${CCACHE_FOUND} ${CCACHE_VERSION}")
# debian (debhlpers) set SOURCE_DATE_EPOCH environment variable, that is
set_property (GLOBAL PROPERTY RULE_LAUNCH_COMPILE ${CCACHE_FOUND})
set_property (GLOBAL PROPERTY RULE_LAUNCH_LINK ${CCACHE_FOUND})
# debian (debhelpers) set SOURCE_DATE_EPOCH environment variable, that is
# filled from the debian/changelog or current time.
#
# - 4.0+ ccache always includes this environment variable into the hash
# of the manifest, which do not allow to use previous cache,
# - 4.2+ ccache ignores SOURCE_DATE_EPOCH under time_macros sloppiness.
# - 4.2+ ccache ignores SOURCE_DATE_EPOCH for every file w/o __DATE__/__TIME__
#
# So for:
# - 4.2+ time_macros sloppiness is used,
# - 4.2+ does not require any sloppiness
# - 4.0+ will ignore SOURCE_DATE_EPOCH environment variable.
if (CCACHE_VERSION VERSION_GREATER_EQUAL "4.2")
message(STATUS "Use time_macros sloppiness for ccache")
set_property (GLOBAL PROPERTY RULE_LAUNCH_COMPILE "${CCACHE_FOUND} --set-config=sloppiness=time_macros")
set_property (GLOBAL PROPERTY RULE_LAUNCH_LINK "${CCACHE_FOUND} --set-config=sloppiness=time_macros")
message(STATUS "ccache is 4.2+ no quirks for SOURCE_DATE_EPOCH required")
elseif (CCACHE_VERSION VERSION_GREATER_EQUAL "4.0")
message(STATUS "Ignore SOURCE_DATE_EPOCH for ccache")
set_property (GLOBAL PROPERTY RULE_LAUNCH_COMPILE "env -u SOURCE_DATE_EPOCH ${CCACHE_FOUND}")
set_property (GLOBAL PROPERTY RULE_LAUNCH_LINK "env -u SOURCE_DATE_EPOCH ${CCACHE_FOUND}")
else()
set_property (GLOBAL PROPERTY RULE_LAUNCH_COMPILE ${CCACHE_FOUND})
set_property (GLOBAL PROPERTY RULE_LAUNCH_LINK ${CCACHE_FOUND})
endif()
else ()
message(${RECONFIGURE_MESSAGE_LEVEL} "Not using ${CCACHE_FOUND} ${CCACHE_VERSION} bug: https://bugzilla.samba.org/show_bug.cgi?id=8118")

2
contrib/NuRaft vendored

@ -1 +1 @@
Subproject commit 7adf7ae33e7d5c307342431b577c8ab1025ee793
Subproject commit 9a0d78de4b90546368d954b6434f0e9a823e8d80

1
docker/test/fasttest/Dockerfile
View File

@ -47,6 +47,7 @@ RUN apt-get update \
expect \
fakeroot \
git \
gdb \
gperf \
lld-${LLVM_VERSION} \
llvm-${LLVM_VERSION} \

18
docker/test/fasttest/run.sh
View File

@ -70,6 +70,7 @@ function start_server
--path "$FASTTEST_DATA"
--user_files_path "$FASTTEST_DATA/user_files"
--top_level_domains_path "$FASTTEST_DATA/top_level_domains"
--test_keeper_server.log_storage_path "$FASTTEST_DATA/coordination"
)
clickhouse-server "${opts[@]}" &>> "$FASTTEST_OUTPUT/server.log" &
server_pid=$!
@ -107,6 +108,18 @@ function start_server
fi
echo "ClickHouse server pid '$server_pid' started and responded"
echo "
handle all noprint
handle SIGSEGV stop print
handle SIGBUS stop print
handle SIGABRT stop print
continue
thread apply all backtrace
continue
" > script.gdb
gdb -batch -command script.gdb -p "$server_pid" &
}
function clone_root
@ -259,6 +272,7 @@ function run_tests
00929_multi_match_edit_distance
01681_hyperscan_debug_assertion
01176_mysql_client_interactive # requires mysql client
01031_mutations_interpreter_and_context
01053_ssd_dictionary # this test mistakenly requires acces to /var/lib/clickhouse -- can't run this locally, disabled
01083_expressions_in_engine_arguments
@ -326,7 +340,7 @@ function run_tests
# Look at DistributedFilesToInsert, so cannot run in parallel.
01460_DistributedFilesToInsert
01541_max_memory_usage_for_user
01541_max_memory_usage_for_user_long
# Require python libraries like scipy, pandas and numpy
01322_ttest_scipy
@ -362,7 +376,7 @@ function run_tests
stop_server ||:
# Clean the data so that there is no interference from the previous test run.
rm -rf "$FASTTEST_DATA"/{{meta,}data,user_files} ||:
rm -rf "$FASTTEST_DATA"/{{meta,}data,user_files,coordination} ||:
start_server

2
docker/test/integration/runner/Dockerfile
View File

@ -58,7 +58,7 @@ RUN dockerd --version; docker --version
RUN python3 -m pip install \
PyMySQL \
aerospike \
aerospike==4.0.0 \
avro \
cassandra-driver \
confluent-kafka==1.5.0 \

2
docker/test/integration/runner/compose/docker_compose_cassandra.yml
View File

@ -4,4 +4,4 @@ services:
image: cassandra
restart: always
ports:
- 9043:9042
- 9043:9042

4
docker/test/integration/runner/compose/docker_compose_hdfs.yml
View File

@ -5,6 +5,6 @@ services:
hostname: hdfs1
restart: always
ports:
- 50075:50075
- 50070:50070
- 50075:50075
- 50070:50070
entrypoint: /etc/bootstrap.sh -d

44
docker/test/integration/runner/compose/docker_compose_kafka.yml
View File

@ -5,42 +5,42 @@ services:
image: zookeeper:3.4.9
hostname: kafka_zookeeper
environment:
ZOO_MY_ID: 1
ZOO_PORT: 2181
ZOO_SERVERS: server.1=kafka_zookeeper:2888:3888
ZOO_MY_ID: 1
ZOO_PORT: 2181
ZOO_SERVERS: server.1=kafka_zookeeper:2888:3888
security_opt:
- label:disable
- label:disable
kafka1:
image: confluentinc/cp-kafka:5.2.0
hostname: kafka1
ports:
- "9092:9092"
- "9092:9092"
environment:
KAFKA_ADVERTISED_LISTENERS: INSIDE://localhost:9092,OUTSIDE://kafka1:19092
KAFKA_LISTENERS: INSIDE://:9092,OUTSIDE://:19092
KAFKA_LISTENER_SECURITY_PROTOCOL_MAP: INSIDE:PLAINTEXT,OUTSIDE:PLAINTEXT
KAFKA_INTER_BROKER_LISTENER_NAME: INSIDE
KAFKA_BROKER_ID: 1
KAFKA_ZOOKEEPER_CONNECT: "kafka_zookeeper:2181"
KAFKA_LOG4J_LOGGERS: "kafka.controller=INFO,kafka.producer.async.DefaultEventHandler=INFO,state.change.logger=INFO"
KAFKA_OFFSETS_TOPIC_REPLICATION_FACTOR: 1
KAFKA_ADVERTISED_LISTENERS: INSIDE://localhost:9092,OUTSIDE://kafka1:19092
KAFKA_LISTENERS: INSIDE://:9092,OUTSIDE://:19092
KAFKA_LISTENER_SECURITY_PROTOCOL_MAP: INSIDE:PLAINTEXT,OUTSIDE:PLAINTEXT
KAFKA_INTER_BROKER_LISTENER_NAME: INSIDE
KAFKA_BROKER_ID: 1
KAFKA_ZOOKEEPER_CONNECT: "kafka_zookeeper:2181"
KAFKA_LOG4J_LOGGERS: "kafka.controller=INFO,kafka.producer.async.DefaultEventHandler=INFO,state.change.logger=INFO"
KAFKA_OFFSETS_TOPIC_REPLICATION_FACTOR: 1
depends_on:
- kafka_zookeeper
- kafka_zookeeper
security_opt:
- label:disable
- label:disable
schema-registry:
image: confluentinc/cp-schema-registry:5.2.0
hostname: schema-registry
ports:
- "8081:8081"
- "8081:8081"
environment:
SCHEMA_REGISTRY_HOST_NAME: schema-registry
SCHEMA_REGISTRY_KAFKASTORE_SECURITY_PROTOCOL: PLAINTEXT
SCHEMA_REGISTRY_KAFKASTORE_BOOTSTRAP_SERVERS: PLAINTEXT://kafka1:19092
SCHEMA_REGISTRY_HOST_NAME: schema-registry
SCHEMA_REGISTRY_KAFKASTORE_SECURITY_PROTOCOL: PLAINTEXT
SCHEMA_REGISTRY_KAFKASTORE_BOOTSTRAP_SERVERS: PLAINTEXT://kafka1:19092
depends_on:
- kafka_zookeeper
- kafka1
- kafka_zookeeper
- kafka1
security_opt:
- label:disable
- label:disable

14
docker/test/integration/runner/compose/docker_compose_kerberized_hdfs.yml
View File

@ -8,22 +8,22 @@ services:
hostname: kerberizedhdfs1
restart: always
volumes:
- ${KERBERIZED_HDFS_DIR}/../../hdfs_configs/bootstrap.sh:/etc/bootstrap.sh:ro
- ${KERBERIZED_HDFS_DIR}/secrets:/usr/local/hadoop/etc/hadoop/conf
- ${KERBERIZED_HDFS_DIR}/secrets/krb_long.conf:/etc/krb5.conf:ro
- ${KERBERIZED_HDFS_DIR}/../../hdfs_configs/bootstrap.sh:/etc/bootstrap.sh:ro
- ${KERBERIZED_HDFS_DIR}/secrets:/usr/local/hadoop/etc/hadoop/conf
- ${KERBERIZED_HDFS_DIR}/secrets/krb_long.conf:/etc/krb5.conf:ro
ports:
- 1006:1006
- 50070:50070
- 9010:9010
depends_on:
- hdfskerberos
- hdfskerberos
entrypoint: /etc/bootstrap.sh -d
hdfskerberos:
image: yandex/clickhouse-kerberos-kdc:${DOCKER_KERBEROS_KDC_TAG}
hostname: hdfskerberos
volumes:
- ${KERBERIZED_HDFS_DIR}/secrets:/tmp/keytab
- ${KERBERIZED_HDFS_DIR}/../../kerberos_image_config.sh:/config.sh
- /dev/urandom:/dev/random
- ${KERBERIZED_HDFS_DIR}/secrets:/tmp/keytab
- ${KERBERIZED_HDFS_DIR}/../../kerberos_image_config.sh:/config.sh
- /dev/urandom:/dev/random
ports: [88, 749]

64
docker/test/integration/runner/compose/docker_compose_kerberized_kafka.yml
View File

@ -6,54 +6,54 @@ services:
# restart: always
hostname: kafka_kerberized_zookeeper
environment:
ZOOKEEPER_SERVER_ID: 1
ZOOKEEPER_CLIENT_PORT: 2181
ZOOKEEPER_SERVERS: "kafka_kerberized_zookeeper:2888:3888"
KAFKA_OPTS: "-Djava.security.auth.login.config=/etc/kafka/secrets/zookeeper_jaas.conf -Djava.security.krb5.conf=/etc/kafka/secrets/krb.conf -Dzookeeper.authProvider.1=org.apache.zookeeper.server.auth.SASLAuthenticationProvider -Dsun.security.krb5.debug=true"
ZOOKEEPER_SERVER_ID: 1
ZOOKEEPER_CLIENT_PORT: 2181
ZOOKEEPER_SERVERS: "kafka_kerberized_zookeeper:2888:3888"
KAFKA_OPTS: "-Djava.security.auth.login.config=/etc/kafka/secrets/zookeeper_jaas.conf -Djava.security.krb5.conf=/etc/kafka/secrets/krb.conf -Dzookeeper.authProvider.1=org.apache.zookeeper.server.auth.SASLAuthenticationProvider -Dsun.security.krb5.debug=true"
volumes:
- ${KERBERIZED_KAFKA_DIR}/secrets:/etc/kafka/secrets
- /dev/urandom:/dev/random
- ${KERBERIZED_KAFKA_DIR}/secrets:/etc/kafka/secrets
- /dev/urandom:/dev/random
depends_on:
- kafka_kerberos
- kafka_kerberos
security_opt:
- label:disable
- label:disable
kerberized_kafka1:
image: confluentinc/cp-kafka:5.2.0
# restart: always
hostname: kerberized_kafka1
ports:
- "9092:9092"
- "9093:9093"
- "9092:9092"
- "9093:9093"
environment:
KAFKA_LISTENERS: OUTSIDE://:19092,UNSECURED_OUTSIDE://:19093,UNSECURED_INSIDE://:9093
KAFKA_ADVERTISED_LISTENERS: OUTSIDE://kerberized_kafka1:19092,UNSECURED_OUTSIDE://kerberized_kafka1:19093,UNSECURED_INSIDE://localhost:9093
# KAFKA_LISTENERS: INSIDE://kerberized_kafka1:9092,OUTSIDE://kerberized_kafka1:19092
# KAFKA_ADVERTISED_LISTENERS: INSIDE://localhost:9092,OUTSIDE://kerberized_kafka1:19092
KAFKA_SASL_MECHANISM_INTER_BROKER_PROTOCOL: GSSAPI
KAFKA_SASL_ENABLED_MECHANISMS: GSSAPI
KAFKA_SASL_KERBEROS_SERVICE_NAME: kafka
KAFKA_LISTENER_SECURITY_PROTOCOL_MAP: OUTSIDE:SASL_PLAINTEXT,UNSECURED_OUTSIDE:PLAINTEXT,UNSECURED_INSIDE:PLAINTEXT,
KAFKA_INTER_BROKER_LISTENER_NAME: OUTSIDE
KAFKA_BROKER_ID: 1
KAFKA_ZOOKEEPER_CONNECT: "kafka_kerberized_zookeeper:2181"
KAFKA_LOG4J_LOGGERS: "kafka.controller=INFO,kafka.producer.async.DefaultEventHandler=INFO,state.change.logger=INFO"
KAFKA_OFFSETS_TOPIC_REPLICATION_FACTOR: 1
KAFKA_OPTS: "-Djava.security.auth.login.config=/etc/kafka/secrets/broker_jaas.conf -Djava.security.krb5.conf=/etc/kafka/secrets/krb.conf -Dsun.security.krb5.debug=true"
KAFKA_LISTENERS: OUTSIDE://:19092,UNSECURED_OUTSIDE://:19093,UNSECURED_INSIDE://:9093
KAFKA_ADVERTISED_LISTENERS: OUTSIDE://kerberized_kafka1:19092,UNSECURED_OUTSIDE://kerberized_kafka1:19093,UNSECURED_INSIDE://localhost:9093
# KAFKA_LISTENERS: INSIDE://kerberized_kafka1:9092,OUTSIDE://kerberized_kafka1:19092
# KAFKA_ADVERTISED_LISTENERS: INSIDE://localhost:9092,OUTSIDE://kerberized_kafka1:19092
KAFKA_SASL_MECHANISM_INTER_BROKER_PROTOCOL: GSSAPI
KAFKA_SASL_ENABLED_MECHANISMS: GSSAPI
KAFKA_SASL_KERBEROS_SERVICE_NAME: kafka
KAFKA_LISTENER_SECURITY_PROTOCOL_MAP: OUTSIDE:SASL_PLAINTEXT,UNSECURED_OUTSIDE:PLAINTEXT,UNSECURED_INSIDE:PLAINTEXT,
KAFKA_INTER_BROKER_LISTENER_NAME: OUTSIDE
KAFKA_BROKER_ID: 1
KAFKA_ZOOKEEPER_CONNECT: "kafka_kerberized_zookeeper:2181"
KAFKA_LOG4J_LOGGERS: "kafka.controller=INFO,kafka.producer.async.DefaultEventHandler=INFO,state.change.logger=INFO"
KAFKA_OFFSETS_TOPIC_REPLICATION_FACTOR: 1
KAFKA_OPTS: "-Djava.security.auth.login.config=/etc/kafka/secrets/broker_jaas.conf -Djava.security.krb5.conf=/etc/kafka/secrets/krb.conf -Dsun.security.krb5.debug=true"
volumes:
- ${KERBERIZED_KAFKA_DIR}/secrets:/etc/kafka/secrets
- /dev/urandom:/dev/random
- ${KERBERIZED_KAFKA_DIR}/secrets:/etc/kafka/secrets
- /dev/urandom:/dev/random
depends_on:
- kafka_kerberized_zookeeper
- kafka_kerberos
- kafka_kerberized_zookeeper
- kafka_kerberos
security_opt:
- label:disable
- label:disable
kafka_kerberos:
image: yandex/clickhouse-kerberos-kdc:${DOCKER_KERBEROS_KDC_TAG:-latest}
hostname: kafka_kerberos
volumes:
- ${KERBERIZED_KAFKA_DIR}/secrets:/tmp/keytab
- ${KERBERIZED_KAFKA_DIR}/../../kerberos_image_config.sh:/config.sh
- /dev/urandom:/dev/random
- ${KERBERIZED_KAFKA_DIR}/secrets:/tmp/keytab
- ${KERBERIZED_KAFKA_DIR}/../../kerberos_image_config.sh:/config.sh
- /dev/urandom:/dev/random
ports: [88, 749]

2
docker/test/integration/runner/compose/docker_compose_mongo.yml
View File

@ -7,5 +7,5 @@ services:
MONGO_INITDB_ROOT_USERNAME: root
MONGO_INITDB_ROOT_PASSWORD: clickhouse
ports:
- 27018:27017
- 27018:27017
command: --profile=2 --verbose

2
docker/test/integration/runner/compose/docker_compose_mysql.yml
View File

@ -6,5 +6,5 @@ services:
environment:
MYSQL_ROOT_PASSWORD: clickhouse
ports:
- 3308:3306
- 3308:3306
command: --server_id=100 --log-bin='mysql-bin-1.log' --default-time-zone='+3:00' --gtid-mode="ON" --enforce-gtid-consistency

8
docker/test/integration/runner/compose/docker_compose_mysql_5_7_for_materialize_mysql.yml
View File

@ -6,5 +6,9 @@ services:
environment:
MYSQL_ROOT_PASSWORD: clickhouse
ports:
- 3308:3306
command: --server_id=100 --log-bin='mysql-bin-1.log' --default-time-zone='+3:00' --gtid-mode="ON" --enforce-gtid-consistency
- 3308:3306
command: --server_id=100 --log-bin='mysql-bin-1.log'
--default-time-zone='+3:00'
--gtid-mode="ON"
--enforce-gtid-consistency
--log-error-verbosity=3

9
docker/test/integration/runner/compose/docker_compose_mysql_8_0_for_materialize_mysql.yml
View File

@ -6,5 +6,10 @@ services:
environment:
MYSQL_ROOT_PASSWORD: clickhouse
ports:
- 33308:3306
command: --server_id=100 --log-bin='mysql-bin-1.log' --default_authentication_plugin='mysql_native_password' --default-time-zone='+3:00' --gtid-mode="ON" --enforce-gtid-consistency
- 33308:3306
command: --server_id=100 --log-bin='mysql-bin-1.log'
--default_authentication_plugin='mysql_native_password'
--default-time-zone='+3:00'
--gtid-mode="ON"
--enforce-gtid-consistency
--log-error-verbosity=3

2
docker/test/integration/runner/compose/docker_compose_mysql_client.yml
View File

@ -7,7 +7,7 @@ services:
MYSQL_ALLOW_EMPTY_PASSWORD: 1
command: --federated --socket /var/run/mysqld/mysqld.sock
healthcheck:
test: ["CMD", "mysqladmin" ,"ping", "-h", "localhost"]
test: ["CMD", "mysqladmin", "ping", "-h", "localhost"]
interval: 1s
timeout: 2s
retries: 100

2
docker/test/integration/runner/compose/docker_compose_postgesql.yml
View File

@ -11,4 +11,4 @@ services:
ports:
- "5433:5433"
environment:
POSTGRES_HOST_AUTH_METHOD: "trust"
POSTGRES_HOST_AUTH_METHOD: "trust"

8
docker/test/integration/runner/compose/docker_compose_postgres.yml
View File

@ -6,8 +6,8 @@ services:
environment:
POSTGRES_PASSWORD: mysecretpassword
ports:
- 5432:5432
- 5432:5432
networks:
default:
aliases:
- postgre-sql.local
default:
aliases:
- postgre-sql.local

2
docker/test/integration/runner/compose/docker_compose_redis.yml
View File

@ -4,5 +4,5 @@ services:
image: redis
restart: always
ports:
- 6380:6379
- 6380:6379
command: redis-server --requirepass "clickhouse" --databases 32

1
docker/test/performance-comparison/compare.sh
View File

@ -97,6 +97,7 @@ function configure
rm -r right/db ||:
rm -r db0/preprocessed_configs ||:
rm -r db0/{data,metadata}/system ||:
rm db0/status ||:
cp -al db0/ left/db/
cp -al db0/ right/db/
}

4
docker/test/stateful/run.sh
View File

@ -60,4 +60,8 @@ fi
# more idiologically correct.
read -ra ADDITIONAL_OPTIONS <<< "${ADDITIONAL_OPTIONS:-}"
if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then
ADDITIONAL_OPTIONS+=('--replicated-database')
fi
clickhouse-test --testname --shard --zookeeper --no-stateless --hung-check --print-time "$SKIP_LIST_OPT" "${ADDITIONAL_OPTIONS[@]}" "$SKIP_TESTS_OPTION" 2>&1 | ts '%Y-%m-%d %H:%M:%S' | tee test_output/test_result.txt

6
docker/test/stateless/Dockerfile
View File

@ -3,6 +3,9 @@ FROM yandex/clickhouse-test-base
ARG odbc_driver_url="https://github.com/ClickHouse/clickhouse-odbc/releases/download/v1.1.4.20200302/clickhouse-odbc-1.1.4-Linux.tar.gz"
RUN echo "deb [trusted=yes] http://repo.mysql.com/apt/ubuntu/ bionic mysql-5.7" >> /etc/apt/sources.list \
&& apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 8C718D3B5072E1F5
RUN apt-get update -y \
&& env DEBIAN_FRONTEND=noninteractive \
apt-get install --yes --no-install-recommends \
@ -24,7 +27,8 @@ RUN apt-get update -y \
telnet \
tree \
unixodbc \
wget
wget \
mysql-client=5.7*
RUN pip3 install numpy scipy pandas

4
docker/test/stateless/run.sh
View File

@ -57,6 +57,10 @@ function run_tests()
ADDITIONAL_OPTIONS+=('4')
fi
if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then
ADDITIONAL_OPTIONS+=('--replicated-database')
fi
clickhouse-test --testname --shard --zookeeper --hung-check --print-time \
--test-runs "$NUM_TRIES" \
"$SKIP_LIST_OPT" "${ADDITIONAL_OPTIONS[@]}" 2>&1 \

48
docker/test/stress/run.sh
View File

@ -8,16 +8,23 @@ dpkg -i package_folder/clickhouse-server_*.deb
dpkg -i package_folder/clickhouse-client_*.deb
dpkg -i package_folder/clickhouse-test_*.deb
function configure()
{
# install test configs
/usr/share/clickhouse-test/config/install.sh
# for clickhouse-server (via service)
echo "ASAN_OPTIONS='malloc_context_size=10 verbosity=1 allocator_release_to_os_interval_ms=10000'" >> /etc/environment
# for clickhouse-client
export ASAN_OPTIONS='malloc_context_size=10 allocator_release_to_os_interval_ms=10000'
# since we run clickhouse from root
sudo chown root: /var/lib/clickhouse
}
function stop()
{
timeout 120 service clickhouse-server stop
# Wait for process to disappear from processlist and also try to kill zombies.
while kill -9 "$(pidof clickhouse-server)"
do
echo "Killed clickhouse-server"
sleep 0.5
done
clickhouse stop
}
function start()
@ -33,19 +40,26 @@ function start()
tail -n1000 /var/log/clickhouse-server/clickhouse-server.log
break
fi
timeout 120 service clickhouse-server start
# use root to match with current uid
clickhouse start --user root >/var/log/clickhouse-server/stdout.log 2>/var/log/clickhouse-server/stderr.log
sleep 0.5
counter=$((counter + 1))
done
echo "
handle all noprint
handle SIGSEGV stop print
handle SIGBUS stop print
handle SIGABRT stop print
continue
thread apply all backtrace
continue
" > script.gdb
gdb -batch -command script.gdb -p "$(cat /var/run/clickhouse-server/clickhouse-server.pid)" &
}
# install test configs
/usr/share/clickhouse-test/config/install.sh
# for clickhouse-server (via service)
echo "ASAN_OPTIONS='malloc_context_size=10 verbosity=1 allocator_release_to_os_interval_ms=10000'" >> /etc/environment
# for clickhouse-client
export ASAN_OPTIONS='malloc_context_size=10 allocator_release_to_os_interval_ms=10000'
configure
start
@ -67,6 +81,8 @@ clickhouse-client --query "SHOW TABLES FROM test"
./stress --hung-check --output-folder test_output --skip-func-tests "$SKIP_TESTS_OPTION" && echo "OK" > /test_output/script_exit_code.txt || echo "FAIL" > /test_output/script_exit_code.txt
stop
# TODO remove me when persistent snapshots will be ready
rm -fr /var/lib/clickhouse/coordination ||:
start
clickhouse-client --query "SELECT 'Server successfuly started'" > /test_output/alive_check.txt || echo 'Server failed to start' > /test_output/alive_check.txt

7
docker/test/stress/stress
View File

@ -23,12 +23,15 @@ def get_options(i):
if 0 < i:
options += " --order=random"
if i % 2 == 1:
if i % 3 == 1:
options += " --db-engine=Ordinary"
if i % 3 == 2:
options += ''' --db-engine="Replicated('/test/db/test_{}', 's1', 'r1')"'''.format(i)
# If database name is not specified, new database is created for each functional test.
# Run some threads with one database for all tests.
if i % 3 == 1:
if i % 2 == 1:
options += " --database=test_{}".format(i)
if i == 13:

9
docker/test/style/Dockerfile
View File

@ -1,7 +1,14 @@
# docker build -t yandex/clickhouse-style-test .
FROM ubuntu:20.04
RUN apt-get update && env DEBIAN_FRONTEND=noninteractive apt-get install --yes shellcheck libxml2-utils git python3-pip pylint && pip3 install codespell
RUN apt-get update && env DEBIAN_FRONTEND=noninteractive apt-get install --yes \
shellcheck \
libxml2-utils \
git \
python3-pip \
pylint \
yamllint \
&& pip3 install codespell
# For |& syntax

28
docs/en/engines/table-engines/integrations/rabbitmq.md
View File

@ -59,6 +59,21 @@ Optional parameters:
- `rabbitmq_max_block_size`
- `rabbitmq_flush_interval_ms`
Also format settings can be added along with rabbitmq-related settings.
Example:
``` sql
CREATE TABLE queue (
key UInt64,
value UInt64,
date DateTime
) ENGINE = RabbitMQ SETTINGS rabbitmq_host_port = 'localhost:5672',
rabbitmq_exchange_name = 'exchange1',
rabbitmq_format = 'JSONEachRow',
rabbitmq_num_consumers = 5,
date_time_input_format = 'best_effort';
```
The RabbitMQ server configuration should be added using the ClickHouse config file.
@ -79,18 +94,6 @@ Additional configuration:
</rabbitmq>
```
Example:
``` sql
CREATE TABLE queue (
key UInt64,
value UInt64
) ENGINE = RabbitMQ SETTINGS rabbitmq_host_port = 'localhost:5672',
rabbitmq_exchange_name = 'exchange1',
rabbitmq_format = 'JSONEachRow',
rabbitmq_num_consumers = 5;
```
## Description {#description}
`SELECT` is not particularly useful for reading messages (except for debugging), because each message can be read only once. It is more practical to create real-time threads using [materialized views](../../../sql-reference/statements/create/view.md). To do this:
@ -114,6 +117,7 @@ Exchange type options:
- `consistent_hash` - Data is evenly distributed between all bound tables (where the exchange name is the same). Note that this exchange type must be enabled with RabbitMQ plugin: `rabbitmq-plugins enable rabbitmq_consistent_hash_exchange`.
Setting `rabbitmq_queue_base` may be used for the following cases:
- to let different tables share queues, so that multiple consumers could be registered for the same queues, which makes a better performance. If using `rabbitmq_num_consumers` and/or `rabbitmq_num_queues` settings, the exact match of queues is achieved in case these parameters are the same.
- to be able to restore reading from certain durable queues when not all messages were successfully consumed. To resume consumption from one specific queue - set its name in `rabbitmq_queue_base` setting and do not specify `rabbitmq_num_consumers` and `rabbitmq_num_queues` (defaults to 1). To resume consumption from all queues, which were declared for a specific table - just specify the same settings: `rabbitmq_queue_base`, `rabbitmq_num_consumers`, `rabbitmq_num_queues`. By default, queue names will be unique to tables.
- to reuse queues as they are declared durable and not auto-deleted. (Can be deleted via any of RabbitMQ CLI tools.)

3
docs/en/engines/table-engines/special/file.md
View File

@ -66,7 +66,8 @@ SELECT * FROM file_engine_table
## Usage in ClickHouse-local {#usage-in-clickhouse-local}
In [clickhouse-local](../../../operations/utilities/clickhouse-local.md) File engine accepts file path in addition to `Format`. Default input/output streams can be specified using numeric or human-readable names like `0` or `stdin`, `1` or `stdout`.
In [clickhouse-local](../../../operations/utilities/clickhouse-local.md) File engine accepts file path in addition to `Format`. Default input/output streams can be specified using numeric or human-readable names like `0` or `stdin`, `1` or `stdout`. It is possible to read and write compressed files based on an additional engine parameter or file extension (`gz`, `br` or `xz`).
**Example:**
``` bash

8
docs/en/getting-started/example-datasets/brown-benchmark.md
View File

@ -5,7 +5,7 @@ toc_title: Brown University Benchmark
# Brown University Benchmark
MgBench - A new analytical benchmark for machine-generated log data, [Andrew Crotty](http://cs.brown.edu/people/acrotty/).
`MgBench` is a new analytical benchmark for machine-generated log data, [Andrew Crotty](http://cs.brown.edu/people/acrotty/).
Download the data:
```
@ -153,7 +153,7 @@ ORDER BY dt,
hr;
-- Q1.4: Over a 1-month period, how often was each server blocked on disk I/O?
-- Q1.4: Over 1 month, how often was each server blocked on disk I/O?
SELECT machine_name,
COUNT(*) AS spikes
@ -301,7 +301,7 @@ WHERE event_type = 'temperature'
AND log_time >= '2019-11-29 17:00:00.000';
-- Q3.4: Over the past 6 months, how frequently was each door opened?
-- Q3.4: Over the past 6 months, how frequently were each door opened?
SELECT device_name,
device_floor,
@ -412,3 +412,5 @@ ORDER BY yr,
```
The data is also available for interactive queries in the [Playground](https://gh-api.clickhouse.tech/play?user=play), [example](https://gh-api.clickhouse.tech/play?user=play#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).
[Original article](https://clickhouse.tech/docs/en/getting_started/example_datasets/brown-benchmark/) <!--hide-->

44
docs/en/interfaces/http.md
View File

@ -148,28 +148,48 @@ $ echo 'DROP TABLE t' | curl 'http://localhost:8123/' --data-binary @-
For successful requests that don’t return a data table, an empty response body is returned.
You can use the internal ClickHouse compression format when transmitting data. The compressed data has a non-standard format, and you will need to use the special `clickhouse-compressor` program to work with it (it is installed with the `clickhouse-client` package). To increase the efficiency of data insertion, you can disable server-side checksum verification by using the [http_native_compression_disable_checksumming_on_decompress](../operations/settings/settings.md#settings-http_native_compression_disable_checksumming_on_decompress) setting.
If you specified `compress=1` in the URL, the server compresses the data it sends you.
If you specified `decompress=1` in the URL, the server decompresses the same data that you pass in the `POST` method.
## Compression {#compression}
You can also choose to use [HTTP compression](https://en.wikipedia.org/wiki/HTTP_compression). To send a compressed `POST` request, append the request header `Content-Encoding: compression_method`. In order for ClickHouse to compress the response, you must append `Accept-Encoding: compression_method`. ClickHouse supports `gzip`, `br`, and `deflate` [compression methods](https://en.wikipedia.org/wiki/HTTP_compression#Content-Encoding_tokens). To enable HTTP compression, you must use the ClickHouse [enable_http_compression](../operations/settings/settings.md#settings-enable_http_compression) setting. You can configure the data compression level in the [http_zlib_compression_level](#settings-http_zlib_compression_level) setting for all the compression methods.
You can use compression to reduce network traffic when transmitting a large amount of data or for creating dumps that are immediately compressed.
You can use this to reduce network traffic when transmitting a large amount of data, or for creating dumps that are immediately compressed.
You can use the internal ClickHouse compression format when transmitting data. The compressed data has a non-standard format, and you need `clickhouse-compressor` program to work with it. It is installed with the `clickhouse-client` package. To increase the efficiency of data insertion, you can disable server-side checksum verification by using the [http_native_compression_disable_checksumming_on_decompress](../operations/settings/settings.md#settings-http_native_compression_disable_checksumming_on_decompress) setting.
Examples of sending data with compression:
If you specify `compress=1` in the URL, the server will compress the data it sends to you. If you specify `decompress=1` in the URL, the server will decompress the data which you pass in the `POST` method.
``` bash
#Sending data to the server:
$ curl -vsS "http://localhost:8123/?enable_http_compression=1" -d 'SELECT number FROM system.numbers LIMIT 10' -H 'Accept-Encoding: gzip'
You can also choose to use [HTTP compression](https://en.wikipedia.org/wiki/HTTP_compression). ClickHouse supports the following [compression methods](https://en.wikipedia.org/wiki/HTTP_compression#Content-Encoding_tokens):
#Sending data to the client:
$ echo "SELECT 1" | gzip -c | curl -sS --data-binary @- -H 'Content-Encoding: gzip' 'http://localhost:8123/'
```
- `gzip`
- `br`
- `deflate`
- `xz`
To send a compressed `POST` request, append the request header `Content-Encoding: compression_method`.
In order for ClickHouse to compress the response, enable compression with [enable_http_compression](../operations/settings/settings.md#settings-enable_http_compression) setting and append `Accept-Encoding: compression_method` header to the request. You can configure the data compression level in the [http_zlib_compression_level](../operations/settings/settings.md#settings-http_zlib_compression_level) setting for all compression methods.
!!! note "Note"
Some HTTP clients might decompress data from the server by default (with `gzip` and `deflate`) and you might get decompressed data even if you use the compression settings correctly.
**Examples**
``` bash
# Sending compressed data to the server
$ echo "SELECT 1" | gzip -c | \
curl -sS --data-binary @- -H 'Content-Encoding: gzip' 'http://localhost:8123/'
```
``` bash
# Receiving compressed data from the server
$ curl -vsS "http://localhost:8123/?enable_http_compression=1" \
-H 'Accept-Encoding: gzip' --output result.gz -d 'SELECT number FROM system.numbers LIMIT 3'
$ zcat result.gz
0
1
2
```
## Default Database {#default-database}
You can use the ‘database’ URL parameter or the ‘X-ClickHouse-Database’ header to specify the default database.
``` bash

15
docs/en/operations/caches.md
View File

@ -8,18 +8,21 @@ toc_title: Caches
When performing queries, ClichHouse uses different caches.
Main cache types:
- `mark_cache` — Cache of marks used by table engines of the [MergeTree](../engines/table-engines/mergetree-family/mergetree.md) family.
- `uncompressed_cache` — Cache of uncompressed data used by table engines of the [MergeTree](../engines/table-engines/mergetree-family/mergetree.md) family.
Additional cache types:
- DNS cache
- [regexp](../interfaces/formats.md#data-format-regexp) cache
- compiled expressions cache
- [Avro format](../interfaces/formats.md#data-format-avro) schemas cache
- [dictionaries data cache](../sql-reference/dictionaries/index.md)
- DNS cache.
- [Regexp](../interfaces/formats.md#data-format-regexp) cache.
- Compiled expressions cache.
- [Avro format](../interfaces/formats.md#data-format-avro) schemas cache.
- [Dictionaries](../sql-reference/dictionaries/index.md) data cache.
Indirectly used:
- OS page cache
- OS page cache.
To drop cache, use [SYSTEM DROP ... CACHE](../sql-reference/statements/system.md) statements.

13
docs/en/operations/external-authenticators/index.md Normal file
View File

@ -0,0 +1,13 @@
---
toc_folder_title: External User Authenticators and Directories
toc_priority: 48
toc_title: Introduction
---
# External User Authenticators and Directories {#external-authenticators}
ClickHouse supports authenticating and managing users using external services.
The following external authenticators and directories are supported:
- [LDAP](./ldap.md#external-authenticators-ldap) [Authenticator](./ldap.md#ldap-external-authenticator) and [Directory](./ldap.md#ldap-external-user-directory)

156
docs/en/operations/external-authenticators/ldap.md Normal file
View File

@ -0,0 +1,156 @@
# LDAP {#external-authenticators-ldap}
LDAP server can be used to authenticate ClickHouse users. There are two different approaches for doing this:
- use LDAP as an external authenticator for existing users, which are defined in `users.xml` or in local access control paths
- use LDAP as an external user directory and allow locally undefined users to be authenticated if they exist on the LDAP server
For both of these approaches, an internally named LDAP server must be defined in the ClickHouse config so that other parts of config are able to refer to it.
## LDAP Server Definition {#ldap-server-definition}
To define LDAP server you must add `ldap_servers` section to the `config.xml`. For example,
```xml
<yandex>
<!- ... -->
<ldap_servers>
<my_ldap_server>
<host>localhost</host>
<port>636</port>
<bind_dn>uid={user_name},ou=users,dc=example,dc=com</bind_dn>
<verification_cooldown>300</verification_cooldown>
<enable_tls>yes</enable_tls>
<tls_minimum_protocol_version>tls1.2</tls_minimum_protocol_version>
<tls_require_cert>demand</tls_require_cert>
<tls_cert_file>/path/to/tls_cert_file</tls_cert_file>
<tls_key_file>/path/to/tls_key_file</tls_key_file>
<tls_ca_cert_file>/path/to/tls_ca_cert_file</tls_ca_cert_file>
<tls_ca_cert_dir>/path/to/tls_ca_cert_dir</tls_ca_cert_dir>
<tls_cipher_suite>ECDHE-ECDSA-AES256-GCM-SHA384:ECDHE-RSA-AES256-GCM-SHA384:AES256-GCM-SHA384</tls_cipher_suite>
</my_ldap_server>
</ldap_servers>
</yandex>
```
Note, that you can define multiple LDAP servers inside the `ldap_servers` section using distinct names.
Parameters:
- `host` - LDAP server hostname or IP, this parameter is mandatory and cannot be empty.
- `port` - LDAP server port, default is `636` if `enable_tls` is set to `true`, `389` otherwise.
- `bind_dn` - template used to construct the DN to bind to.
- The resulting DN will be constructed by replacing all `{user_name}` substrings of the
template with the actual user name during each authentication attempt.
- `verification_cooldown` - a period of time, in seconds, after a successful bind attempt,
during which the user will be assumed to be successfully authenticated for all consecutive
requests without contacting the LDAP server.
- Specify `0` (the default) to disable caching and force contacting the LDAP server for each authentication request.
- `enable_tls` - flag to trigger use of secure connection to the LDAP server.
- Specify `no` for plain text `ldap://` protocol (not recommended).
- Specify `yes` for LDAP over SSL/TLS `ldaps://` protocol (recommended, the default).
- Specify `starttls` for legacy StartTLS protocol (plain text `ldap://` protocol, upgraded to TLS).
- `tls_minimum_protocol_version` - the minimum protocol version of SSL/TLS.
- Accepted values are: `ssl2`, `ssl3`, `tls1.0`, `tls1.1`, `tls1.2` (the default).
- `tls_require_cert` - SSL/TLS peer certificate verification behavior.
- Accepted values are: `never`, `allow`, `try`, `demand` (the default).
- `tls_cert_file` - path to certificate file.
- `tls_key_file` - path to certificate key file.
- `tls_ca_cert_file` - path to CA certificate file.
- `tls_ca_cert_dir` - path to the directory containing CA certificates.
- `tls_cipher_suite` - allowed cipher suite (in OpenSSL notation).
## LDAP External Authenticator {#ldap-external-authenticator}
A remote LDAP server can be used as a method for verifying passwords for locally defined users (users defined in `users.xml` or in local access control paths). In order to achieve this, specify previously defined LDAP server name instead of `password` or similar sections in the user definition.
At each login attempt, ClickHouse will try to "bind" to the specified DN defined by the `bind_dn` parameter in the [LDAP server definition](#ldap-server-definition) using the provided credentials, and if successful, the user will be considered authenticated. This is often called a "simple bind" method.
For example,
```xml
<yandex>
<!- ... -->
<users>
<!- ... -->
<my_user>
<!- ... -->
<ldap>
<server>my_ldap_server</server>
</ldap>
</my_user>
</users>
</yandex>
```
Note, that user `my_user` refers to `my_ldap_server`. This LDAP server must be configured in the main `config.xml` file as described previously.
When SQL-driven [Access Control and Account Management](../access-rights.md#access-control) is enabled in ClickHouse, users that are authenticated by LDAP servers can also be created using the [CRATE USER](../../sql-reference/statements/create/user.md#create-user-statement) statement.
```sql
CREATE USER my_user IDENTIFIED WITH ldap_server BY 'my_ldap_server'
```
## LDAP Exernal User Directory {#ldap-external-user-directory}
In addition to the locally defined users, a remote LDAP server can be used as a source of user definitions. In order to achieve this, specify previously defined LDAP server name (see [LDAP Server Definition](#ldap-server-definition)) in the `ldap` section inside the `users_directories` section of the `config.xml` file.
At each login attempt, ClickHouse will try to find the user definition locally and authenticate it as usual, but if the user is not defined, ClickHouse will assume it exists in the external LDAP directory, and will try to "bind" to the specified DN at the LDAP server using the provided credentials. If successful, the user will be considered existing and authenticated. The user will be assigned roles from the list specified in the `roles` section. Additionally, LDAP "search" can be performed and results can be transformed and treated as role names and then be assigned to the user if the `role_mapping` section is also configured. All this implies that the SQL-driven [Access Control and Account Management](../access-rights.md#access-control) is enabled and roles are created using the [CREATE ROLE](../../sql-reference/statements/create/role.md#create-role-statement) statement.
Example (goes into `config.xml`):
```xml
<yandex>
<!- ... -->
<user_directories>
<!- ... -->
<ldap>
<server>my_ldap_server</server>
<roles>
<my_local_role1 />
<my_local_role2 />
</roles>
<role_mapping>
<base_dn>ou=groups,dc=example,dc=com</base_dn>
<scope>subtree</scope>
<search_filter>(&amp;(objectClass=groupOfNames)(member={bind_dn}))</search_filter>
<attribute>cn</attribute>
<prefix>clickhouse_</prefix>
</role_mapping>
</ldap>
</user_directories>
</yandex>
```
Note that `my_ldap_server` referred in the `ldap` section inside the `user_directories` section must be a previously
defined LDAP server that is configured in the `config.xml` (see [LDAP Server Definition](#ldap-server-definition)).
Parameters:
- `server` - one of LDAP server names defined in the `ldap_servers` config section above.
This parameter is mandatory and cannot be empty.
- `roles` - section with a list of locally defined roles that will be assigned to each user retrieved from the LDAP server.
- If no roles are specified here or assigned during role mapping (below), user will not be able
to perform any actions after authentication.
- `role_mapping` - section with LDAP search parameters and mapping rules.
- When a user authenticates, while still bound to LDAP, an LDAP search is performed using `search_filter`
and the name of the logged in user. For each entry found during that search, the value of the specified
attribute is extracted. For each attribute value that has the specified prefix, the prefix is removed,
and the rest of the value becomes the name of a local role defined in ClickHouse,
which is expected to be created beforehand by the [CREATE ROLE](../../sql-reference/statements/create/role.md#create-role-statement) statement.
- There can be multiple `role_mapping` sections defined inside the same `ldap` section. All of them will be applied.
- `base_dn` - template used to construct the base DN for the LDAP search.
- The resulting DN will be constructed by replacing all `{user_name}` and `{bind_dn}`
substrings of the template with the actual user name and bind DN during each LDAP search.
- `scope` - scope of the LDAP search.
- Accepted values are: `base`, `one_level`, `children`, `subtree` (the default).
- `search_filter` - template used to construct the search filter for the LDAP search.
- The resulting filter will be constructed by replacing all `{user_name}`, `{bind_dn}`, and `{base_dn}`
substrings of the template with the actual user name, bind DN, and base DN during each LDAP search.
- Note, that the special characters must be escaped properly in XML.
- `attribute` - attribute name whose values will be returned by the LDAP search.
- `prefix` - prefix, that will be expected to be in front of each string in the original
list of strings returned by the LDAP search. Prefix will be removed from the original
strings and resulting strings will be treated as local role names. Empty, by default.

2
docs/en/operations/settings/settings-users.md
View File

@ -139,7 +139,7 @@ You can assign a quotas set for the user. For a detailed description of quotas c
### user_name/databases {#user-namedatabases}
In this section, you can you can limit rows that are returned by ClickHouse for `SELECT` queries made by the current user, thus implementing basic row-level security.
In this section, you can limit rows that are returned by ClickHouse for `SELECT` queries made by the current user, thus implementing basic row-level security.
**Example**

22
docs/en/operations/settings/settings.md
View File

@ -1104,7 +1104,7 @@ The maximum number of replicas for each shard when executing a query. In limited
- the sampling key is an expression that is expensive to calculate
- the cluster's latency distribution has a long tail, so that querying more servers increases the query's overall latency
In addition, this setting will produce incorrect results when joins or subqueries are involved, and all tables don't meet certain conditions. See [Distributed Subqueries and max_parallel_replicas](../../sql-reference/operators/in.md/#max_parallel_replica-subqueries) for more details.
In addition, this setting will produce incorrect results when joins or subqueries are involved, and all tables don't meet certain conditions. See [Distributed Subqueries and max_parallel_replicas](../../sql-reference/operators/in.md#max_parallel_replica-subqueries) for more details.
## compile {#compile}
@ -2659,3 +2659,23 @@ Result:
Note that this setting influences [Materialized view](../../sql-reference/statements/create/view.md#materialized) and [MaterializeMySQL](../../engines/database-engines/materialize-mysql.md) behaviour.
[Original article](https://clickhouse.tech/docs/en/operations/settings/settings/) <!-- hide -->
## engine_file_empty_if_not_exists {#engine-file-empty_if-not-exists}
Allows to select data from a file engine table without file.
Possible values:
- 0 — `SELECT` throws exception.
- 1 — `SELECT` returns empty result.
Default value: `0`.
## engine_file_truncate_on_insert {#engine-file-truncate-on-insert}
Enables or disables truncate before insert in file engine tables.
Possible values:
- 0 — Disabled.
- 1 — Enabled.
Default value: `0`.

2
docs/en/operations/system-tables/distributed_ddl_queue.md
View File

@ -14,7 +14,7 @@ Columns:
- `initiator` ([String](../../sql-reference/data-types/string.md)) — Node that executed the query.
- `query_start_time` ([DateTime](../../sql-reference/data-types/datetime.md)) — Query start time.
- `query_finish_time` ([DateTime](../../sql-reference/data-types/datetime.md)) — Query finish time.
- `query_duration_ms` ([UInt64](../../sql-reference/data-types/datetime64.md)) — Duration of query execution (in milliseconds).
- `query_duration_ms` ([UInt64](../../sql-reference/data-types/int-uint.md)) — Duration of query execution (in milliseconds).
- `exception_code` ([Enum8](../../sql-reference/data-types/enum.md)) — Exception code from [ZooKeeper](../../operations/tips.md#zookeeper).
**Example**

8
docs/en/sql-reference/aggregate-functions/reference/argmax.md
View File

@ -52,15 +52,15 @@ Input table:
Query:
``` sql
SELECT argMax(user, salary), argMax(tuple(user, salary)) FROM salary;
SELECT argMax(user, salary), argMax(tuple(user, salary), salary), argMax(tuple(user, salary)) FROM salary;
```
Result:
``` text
┌─argMax(user, salary)─┬─argMax(tuple(user, salary))─┐
│ director │ ('director',5000) │
└──────────────────────┴─────────────────────────────┘
┌─argMax(user, salary)─┬─argMax(tuple(user, salary), salary)─┬─argMax(tuple(user, salary))─┐
│ director │ ('director',5000) │ ('director',5000) │
└──────────────────────┴─────────────────────────────────────┴─────────────────────────────┘
```
[Original article](https://clickhouse.tech/docs/en/sql-reference/aggregate-functions/reference/argmax/) <!--hide-->

2
docs/en/sql-reference/aggregate-functions/reference/avg.md
View File

@ -9,7 +9,7 @@ Calculates the arithmetic mean.
**Syntax**
``` sql
avgWeighted(x)
avg(x)
```
**Arguments**

1
docs/en/sql-reference/aggregate-functions/reference/mannwhitneyutest.md
View File

@ -32,6 +32,7 @@ The null hypothesis is that two populations are stochastically equal. Also one-s
**Returned values**
[Tuple](../../../sql-reference/data-types/tuple.md) with two elements:
- calculated U-statistic. [Float64](../../../sql-reference/data-types/float.md).
- calculated p-value. [Float64](../../../sql-reference/data-types/float.md).

1
docs/en/sql-reference/aggregate-functions/reference/studentttest.md
View File

@ -24,6 +24,7 @@ The null hypothesis is that means of populations are equal. Normal distribution
**Returned values**
[Tuple](../../../sql-reference/data-types/tuple.md) with two elements:
- calculated t-statistic. [Float64](../../../sql-reference/data-types/float.md).
- calculated p-value. [Float64](../../../sql-reference/data-types/float.md).

1
docs/en/sql-reference/aggregate-functions/reference/welchttest.md
View File

@ -24,6 +24,7 @@ The null hypothesis is that means of populations are equal. Normal distribution
**Returned values**
[Tuple](../../../sql-reference/data-types/tuple.md) with two elements:
- calculated t-statistic. [Float64](../../../sql-reference/data-types/float.md).
- calculated p-value. [Float64](../../../sql-reference/data-types/float.md).

18
docs/en/sql-reference/functions/date-time-functions.md
View File

@ -61,40 +61,58 @@ int32samoa: 1546300800
Converts a date or date with time to a UInt16 number containing the year number (AD).
Alias: `YEAR`.
## toQuarter {#toquarter}
Converts a date or date with time to a UInt8 number containing the quarter number.
Alias: `QUARTER`.
## toMonth {#tomonth}
Converts a date or date with time to a UInt8 number containing the month number (1-12).
Alias: `MONTH`.
## toDayOfYear {#todayofyear}
Converts a date or date with time to a UInt16 number containing the number of the day of the year (1-366).
Alias: `DAYOFYEAR`.
## toDayOfMonth {#todayofmonth}
Converts a date or date with time to a UInt8 number containing the number of the day of the month (1-31).
Aliases: `DAYOFMONTH`, `DAY`.
## toDayOfWeek {#todayofweek}
Converts a date or date with time to a UInt8 number containing the number of the day of the week (Monday is 1, and Sunday is 7).
Alias: `DAYOFWEEK`.
## toHour {#tohour}
Converts a date with time to a UInt8 number containing the number of the hour in 24-hour time (0-23).
This function assumes that if clocks are moved ahead, it is by one hour and occurs at 2 a.m., and if clocks are moved back, it is by one hour and occurs at 3 a.m. (which is not always true – even in Moscow the clocks were twice changed at a different time).
Alias: `HOUR`.
## toMinute {#tominute}
Converts a date with time to a UInt8 number containing the number of the minute of the hour (0-59).
Alias: `MINUTE`.
## toSecond {#tosecond}
Converts a date with time to a UInt8 number containing the number of the second in the minute (0-59).
Leap seconds are not accounted for.
Alias: `SECOND`.
## toUnixTimestamp {#to-unix-timestamp}
For DateTime argument: converts value to the number with type UInt32 -- Unix Timestamp (https://en.wikipedia.org/wiki/Unix_time).

2
docs/en/sql-reference/functions/encoding-functions.md
View File

@ -75,6 +75,8 @@ Result:
Returns a string containing the argument’s hexadecimal representation.
Alias: `HEX`.
**Syntax**
``` sql

2
docs/en/sql-reference/functions/encryption-functions.md
View File

@ -132,7 +132,7 @@ aes_encrypt_mysql('mode', 'plaintext', 'key' [, iv])
- `mode` — Encryption mode. [String](../../sql-reference/data-types/string.md#string).
- `plaintext` — Text that needs to be encrypted. [String](../../sql-reference/data-types/string.md#string).
- `key` — Encryption key. If key is longer than required by mode, MySQL-specific key folding is performed. [String](../../sql-reference/data-types/string.md#string).
- `iv` — Initialization vector. Optinal, only first 16 bytes are taken into account [String](../../sql-reference/data-types/string.md#string).
- `iv` — Initialization vector. Optional, only first 16 bytes are taken into account [String](../../sql-reference/data-types/string.md#string).
**Returned value**

2
docs/en/sql-reference/functions/functions-for-nulls.md
View File

@ -13,6 +13,8 @@ Checks whether the argument is [NULL](../../sql-reference/syntax.md#null-literal
isNull(x)
```
Alias: `ISNULL`.
**Arguments**
- `x` — A value with a non-compound data type.

4
docs/en/sql-reference/functions/hash-functions.md
View File

@ -9,7 +9,7 @@ Hash functions can be used for the deterministic pseudo-random shuffling of elem
## halfMD5 {#hash-functions-halfmd5}
[Interprets](../../sql-reference/functions/type-conversion-functions.md#type_conversion_functions-reinterpretAsString) all the input parameters as strings and calculates the [MD5](https://en.wikipedia.org/wiki/MD5) hash value for each of them. Then combines hashes, takes the first 8 bytes of the hash of the resulting string, and interprets them as `UInt64` in big-endian byte order.
[Interprets](../../sql-reference/functions/type-conversion-functions.md#type_conversion_function-reinterpretAsString) all the input parameters as strings and calculates the [MD5](https://en.wikipedia.org/wiki/MD5) hash value for each of them. Then combines hashes, takes the first 8 bytes of the hash of the resulting string, and interprets them as `UInt64` in big-endian byte order.
``` sql
halfMD5(par1, ...)
@ -54,7 +54,7 @@ sipHash64(par1,...)
This is a cryptographic hash function. It works at least three times faster than the [MD5](#hash_functions-md5) function.
Function [interprets](../../sql-reference/functions/type-conversion-functions.md#type_conversion_functions-reinterpretAsString) all the input parameters as strings and calculates the hash value for each of them. Then combines hashes by the following algorithm:
Function [interprets](../../sql-reference/functions/type-conversion-functions.md#type_conversion_function-reinterpretAsString) all the input parameters as strings and calculates the hash value for each of them. Then combines hashes by the following algorithm:
1. After hashing all the input parameters, the function gets the array of hashes.
2. Function takes the first and the second elements and calculates a hash for the array of them.

12
docs/en/sql-reference/functions/ip-address-functions.md
View File

@ -9,10 +9,14 @@ toc_title: IP Addresses
Takes a UInt32 number. Interprets it as an IPv4 address in big endian. Returns a string containing the corresponding IPv4 address in the format A.B.C.d (dot-separated numbers in decimal form).
Alias: `INET_NTOA`.
## IPv4StringToNum(s) {#ipv4stringtonums}
The reverse function of IPv4NumToString. If the IPv4 address has an invalid format, it returns 0.
Alias: `INET_ATON`.
## IPv4NumToStringClassC(num) {#ipv4numtostringclasscnum}
Similar to IPv4NumToString, but using xxx instead of the last octet.
@ -49,7 +53,11 @@ Since using ‘xxx’ is highly unusual, this may be changed in the future. We r
### IPv6NumToString(x) {#ipv6numtostringx}
Accepts a FixedString(16) value containing the IPv6 address in binary format. Returns a string containing this address in text format.
IPv6-mapped IPv4 addresses are output in the format ::ffff:111.222.33.44. Examples:
IPv6-mapped IPv4 addresses are output in the format ::ffff:111.222.33.44.
Alias: `INET6_NTOA`.
Examples:
``` sql
SELECT IPv6NumToString(toFixedString(unhex('2A0206B8000000000000000000000011'), 16)) AS addr
@ -119,6 +127,8 @@ The reverse function of IPv6NumToString. If the IPv6 address has an invalid form
If the IP address is a valid IPv4 address then the IPv6 equivalent of the IPv4 address is returned.
HEX can be uppercase or lowercase.
Alias: `INET6_ATON`.
``` sql
SELECT cutIPv6(IPv6StringToNum('127.0.0.1'), 0, 0);
```

48
docs/en/sql-reference/functions/string-functions.md
View File

@ -98,6 +98,8 @@ SELECT toValidUTF8('\x61\xF0\x80\x80\x80b')
Repeats a string as many times as specified and concatenates the replicated values as a single string.
Alias: `REPEAT`.
**Syntax**
``` sql
@ -276,10 +278,14 @@ Returns the string ‘s’ that was converted from the encoding in ‘from’ to
Encodes ‘s’ string into base64
Alias: `TO_BASE64`.
## base64Decode(s) {#base64decode}
Decode base64-encoded string ‘s’ into original string. In case of failure raises an exception.
Alias: `FROM_BASE64`.
## tryBase64Decode(s) {#trybase64decode}
Similar to base64Decode, but in case of error an empty string would be returned.
@ -600,4 +606,46 @@ Hello, &quot;world&quot;!
&apos;foo&apos;
```
## decodeXMLComponent {#decode-xml-component}
Replaces XML predefined entities with characters. Predefined entities are `&quot;` `&amp;` `&apos;` `&gt;` `&lt;`
This function also replaces numeric character references with Unicode characters. Both decimal (like `&#10003;`) and hexadecimal (`&#x2713;`) forms are supported.
**Syntax**
``` sql
decodeXMLComponent(x)
```
**Parameters**
- `x` — A sequence of characters. [String](../../sql-reference/data-types/string.md).
**Returned value**
- The sequence of characters after replacement.
Type: [String](../../sql-reference/data-types/string.md).
**Example**
Query:
``` sql
SELECT decodeXMLComponent('&apos;foo&apos;');
SELECT decodeXMLComponent('&lt; &#x3A3; &gt;');
```
Result:
``` text
'foo'
< Σ >
```
**See Also**
- [List of XML and HTML character entity references](https://en.wikipedia.org/wiki/List_of_XML_and_HTML_character_entity_references)
[Original article](https://clickhouse.tech/docs/en/query_language/functions/string_functions/) <!--hide-->

480
docs/en/sql-reference/functions/type-conversion-functions.md
View File

@ -36,10 +36,14 @@ The behavior of functions for the [NaN and Inf](../../sql-reference/data-types/f
**Example**
Query:
``` sql
SELECT toInt64(nan), toInt32(32), toInt16('16'), toInt8(8.8)
SELECT toInt64(nan), toInt32(32), toInt16('16'), toInt8(8.8);
```
Result:
``` text
┌─────────toInt64(nan)─┬─toInt32(32)─┬─toInt16('16')─┬─toInt8(8.8)─┐
│ -9223372036854775808 │ 32 │ 16 │ 8 │
@ -52,10 +56,14 @@ It takes an argument of type String and tries to parse it into Int (8 \| 16 \| 3
**Example**
Query:
``` sql
select toInt64OrZero('123123'), toInt8OrZero('123qwe123')
SELECT toInt64OrZero('123123'), toInt8OrZero('123qwe123');
```
Result:
``` text
┌─toInt64OrZero('123123')─┬─toInt8OrZero('123qwe123')─┐
│ 123123 │ 0 │
@ -68,10 +76,14 @@ It takes an argument of type String and tries to parse it into Int (8 \| 16 \| 3
**Example**
Query:
``` sql
select toInt64OrNull('123123'), toInt8OrNull('123qwe123')
SELECT toInt64OrNull('123123'), toInt8OrNull('123qwe123');
```
Result:
``` text
┌─toInt64OrNull('123123')─┬─toInt8OrNull('123qwe123')─┐
│ 123123 │ ᴺᵁᴸᴸ │
@ -102,10 +114,14 @@ The behavior of functions for negative agruments and for the [NaN and Inf](../..
**Example**
Query:
``` sql
SELECT toUInt64(nan), toUInt32(-32), toUInt16('16'), toUInt8(8.8)
SELECT toUInt64(nan), toUInt32(-32), toUInt16('16'), toUInt8(8.8);
```
Result:
``` text
┌───────toUInt64(nan)─┬─toUInt32(-32)─┬─toUInt16('16')─┬─toUInt8(8.8)─┐
│ 9223372036854775808 │ 4294967264 │ 16 │ 8 │
@ -124,6 +140,8 @@ SELECT toUInt64(nan), toUInt32(-32), toUInt16('16'), toUInt8(8.8)
## toDate {#todate}
Alias: `DATE`.
## toDateOrZero {#todateorzero}
## toDateOrNull {#todateornull}
@ -168,20 +186,28 @@ A value in the `Nullable(Decimal(P,S))` data type. The value contains:
**Examples**
Query:
``` sql
SELECT toDecimal32OrNull(toString(-1.111), 5) AS val, toTypeName(val)
SELECT toDecimal32OrNull(toString(-1.111), 5) AS val, toTypeName(val);
```
Result:
``` text
┌──────val─┬─toTypeName(toDecimal32OrNull(toString(-1.111), 5))─┐
│ -1.11100 │ Nullable(Decimal(9, 5)) │
└──────────┴────────────────────────────────────────────────────┘
```
Query:
``` sql
SELECT toDecimal32OrNull(toString(-1.111), 2) AS val, toTypeName(val)
SELECT toDecimal32OrNull(toString(-1.111), 2) AS val, toTypeName(val);
```
Result:
``` text
┌──val─┬─toTypeName(toDecimal32OrNull(toString(-1.111), 2))─┐
│ ᴺᵁᴸᴸ │ Nullable(Decimal(9, 2)) │
@ -213,20 +239,28 @@ A value in the `Nullable(Decimal(P,S))` data type. The value contains:
**Example**
Query:
``` sql
SELECT toDecimal32OrZero(toString(-1.111), 5) AS val, toTypeName(val)
SELECT toDecimal32OrZero(toString(-1.111), 5) AS val, toTypeName(val);
```
Result:
``` text
┌──────val─┬─toTypeName(toDecimal32OrZero(toString(-1.111), 5))─┐
│ -1.11100 │ Decimal(9, 5) │
└──────────┴────────────────────────────────────────────────────┘
```
Query:
``` sql
SELECT toDecimal32OrZero(toString(-1.111), 2) AS val, toTypeName(val)
SELECT toDecimal32OrZero(toString(-1.111), 2) AS val, toTypeName(val);
```
Result:
``` text
┌──val─┬─toTypeName(toDecimal32OrZero(toString(-1.111), 2))─┐
│ 0.00 │ Decimal(9, 2) │
@ -258,12 +292,18 @@ Conversion between numeric types uses the same rules as assignments between diff
Additionally, the toString function of the DateTime argument can take a second String argument containing the name of the time zone. Example: `Asia/Yekaterinburg` In this case, the time is formatted according to the specified time zone.
**Example**
Query:
``` sql
SELECT
now() AS now_local,
toString(now(), 'Asia/Yekaterinburg') AS now_yekat
toString(now(), 'Asia/Yekaterinburg') AS now_yekat;
```
Result:
``` text
┌───────────now_local─┬─now_yekat───────────┐
│ 2016-06-15 00:11:21 │ 2016-06-15 02:11:21 │
@ -281,52 +321,39 @@ If the string has fewer bytes than N, it is padded with null bytes to the right.
Accepts a String or FixedString argument. Returns the String with the content truncated at the first zero byte found.
Example:
**Example**
Query:
``` sql
SELECT toFixedString('foo', 8) AS s, toStringCutToZero(s) AS s_cut
SELECT toFixedString('foo', 8) AS s, toStringCutToZero(s) AS s_cut;
```
Result:
``` text
┌─s─────────────┬─s_cut─┐
│ foo\0\0\0\0\0 │ foo │
└───────────────┴───────┘
```
Query:
``` sql
SELECT toFixedString('foo\0bar', 8) AS s, toStringCutToZero(s) AS s_cut
SELECT toFixedString('foo\0bar', 8) AS s, toStringCutToZero(s) AS s_cut;
```
Result:
``` text
┌─s──────────┬─s_cut─┐
│ foo\0bar\0 │ foo │
└────────────┴───────┘
```
## reinterpretAs(x, T) {#type_conversion_function-cast}
## reinterpretAsUInt(8\|16\|32\|64) {#reinterpretasuint8163264}
Performs byte reinterpretation of ‘x’ as ‘t’ data type.
Following reinterpretations are allowed:
1. Any type that has fixed size and value of that type can be represented continuously into FixedString.
2. Any type that if value of that type can be represented continuously into String. Null bytes are dropped from the end. For example, a UInt32 type value of 255 is a string that is one byte long.
3. FixedString, String, types that can be interpreted as numeric (Integers, Float, Date, DateTime, UUID) into types that can be interpreted as numeric (Integers, Float, Date, DateTime, UUID) into FixedString,
``` sql
SELECT reinterpretAs(toInt8(-1), 'UInt8') as int_to_uint,
reinterpretAs(toInt8(1), 'Float32') as int_to_float,
reinterpretAs('1', 'UInt32') as string_to_int;
```
``` text
┌─int_to_uint─┬─int_to_float─┬─string_to_int─┐
│ 255 │ 1e-45 │ 49 │
└─────────────┴──────────────┴───────────────┘
```
## reinterpretAsUInt(8\|16\|32\|64\|256) {#reinterpretasuint8163264256}
## reinterpretAsInt(8\|16\|32\|64\|128\|256) {#reinterpretasint8163264128256}
## reinterpretAsInt(8\|16\|32\|64) {#reinterpretasint8163264}
## reinterpretAsFloat(32\|64) {#reinterpretasfloat3264}
@ -334,19 +361,87 @@ SELECT reinterpretAs(toInt8(-1), 'UInt8') as int_to_uint,
## reinterpretAsDateTime {#reinterpretasdatetime}
These functions accept a string and interpret the bytes placed at the beginning of the string as a number in host order (little endian). If the string isn’t long enough, the functions work as if the string is padded with the necessary number of null bytes. If the string is longer than needed, the extra bytes are ignored. A date is interpreted as the number of days since the beginning of the Unix Epoch, and a date with time is interpreted as the number of seconds since the beginning of the Unix Epoch.
## reinterpretAsString {#type_conversion_functions-reinterpretAsString}
This function accepts a number or date or date with time, and returns a string containing bytes representing the corresponding value in host order (little endian). Null bytes are dropped from the end. For example, a UInt32 type value of 255 is a string that is one byte long.
## reinterpretAsFixedString {#reinterpretasfixedstring}
This function accepts a number or date or date with time, and returns a FixedString containing bytes representing the corresponding value in host order (little endian). Null bytes are dropped from the end. For example, a UInt32 type value of 255 is a FixedString that is one byte long.
## reinterpretAsUUID {#reinterpretasuuid}
These functions are aliases for `reinterpretAs`function.
This function accepts 16 bytes string, and returns UUID containing bytes representing the corresponding value in network byte order (big-endian). If the string isn't long enough, the functions work as if the string is padded with the necessary number of null bytes to the end. If the string longer than 16 bytes, the extra bytes at the end are ignored.
**Syntax**
``` sql
reinterpretAsUUID(fixed_string)
```
**Parameters**
- `fixed_string` — Big-endian byte string. [FixedString](../../sql-reference/data-types/fixedstring.md#fixedstring).
## reinterpret(x, T) {#type_conversion_function-reinterpret}
**Returned value**
- The UUID type value. [UUID](../../sql-reference/data-types/uuid.md#uuid-data-type).
**Examples**
String to UUID.
Query:
``` sql
SELECT reinterpret(toInt8(-1), 'UInt8') as int_to_uint,
reinterpret(toInt8(1), 'Float32') as int_to_float,
reinterpret('1', 'UInt32') as string_to_int;
```
Result:
``` text
┌─reinterpretAsUUID(reverse(unhex('000102030405060708090a0b0c0d0e0f')))─┐
│ 08090a0b-0c0d-0e0f-0001-020304050607 │
└───────────────────────────────────────────────────────────────────────┘
```
Going back and forth from String to UUID.
Query:
``` sql
WITH
generateUUIDv4() AS uuid,
identity(lower(hex(reverse(reinterpretAsString(uuid))))) AS str,
reinterpretAsUUID(reverse(unhex(str))) AS uuid2
SELECT uuid = uuid2;
```
Result:
``` text
┌─equals(uuid, uuid2)─┐
│ 1 │
└─────────────────────┘
```
## CAST(x, T) {#type_conversion_function-cast}
Converts ‘x’ to the ‘t’ data type. The syntax CAST(x AS t) is also supported.
Converts input value `x` to the `T` data type.
Example:
The syntax `CAST(x AS t)` is also supported.
Note, that if value `x` does not fit the bounds of type T, the function overflows. For example, CAST(-1, 'UInt8') returns 255.
**Example**
Query:
``` sql
SELECT
@ -354,9 +449,11 @@ SELECT
CAST(timestamp AS DateTime) AS datetime,
CAST(timestamp AS Date) AS date,
CAST(timestamp, 'String') AS string,
CAST(timestamp, 'FixedString(22)') AS fixed_string
CAST(timestamp, 'FixedString(22)') AS fixed_string;
```
Result:
``` text
┌─timestamp───────────┬────────────datetime─┬───────date─┬─string──────────────┬─fixed_string──────────────┐
│ 2016-06-15 23:00:00 │ 2016-06-15 23:00:00 │ 2016-06-15 │ 2016-06-15 23:00:00 │ 2016-06-15 23:00:00\0\0\0 │
@ -365,12 +462,18 @@ SELECT
Conversion to FixedString(N) only works for arguments of type String or FixedString(N).
Type conversion to [Nullable](../../sql-reference/data-types/nullable.md) and back is supported. Example:
Type conversion to [Nullable](../../sql-reference/data-types/nullable.md) and back is supported.
**Example**
Query:
``` sql
SELECT toTypeName(x) FROM t_null
SELECT toTypeName(x) FROM t_null;
```
Result:
``` text
┌─toTypeName(x)─┐
│ Int8 │
@ -378,10 +481,14 @@ SELECT toTypeName(x) FROM t_null
└───────────────┘
```
Query:
``` sql
SELECT toTypeName(CAST(x, 'Nullable(UInt16)')) FROM t_null
SELECT toTypeName(CAST(x, 'Nullable(UInt16)')) FROM t_null;
```
Result:
``` text
┌─toTypeName(CAST(x, 'Nullable(UInt16)'))─┐
│ Nullable(UInt16) │
@ -395,15 +502,19 @@ SELECT toTypeName(CAST(x, 'Nullable(UInt16)')) FROM t_null
## accurateCast(x, T) {#type_conversion_function-accurate-cast}
Converts ‘x’ to the ‘t’ data type. The differente from cast(x, T) is that accurateCast
does not allow overflow of numeric types during cast if type value x does not fit
bounds of type T.
Converts `x` to the `T` data type.
The difference from [cast(x, T)](#type_conversion_function-cast) is that `accurateCast` does not allow overflow of numeric types during cast if type value `x` does not fit the bounds of type `T`. For example, `accurateCast(-1, 'UInt8')` throws an exception.
**Example**
Query:
Example
``` sql
SELECT cast(-1, 'UInt8') as uint8;
SELECT cast(-1, 'UInt8') as uint8;
```
Result:
``` text
┌─uint8─┐
@ -411,38 +522,46 @@ SELECT cast(-1, 'UInt8') as uint8;
└───────┘
```
Query:
```sql
SELECT accurateCast(-1, 'UInt8') as uint8;
```
Result:
``` text
Code: 70. DB::Exception: Received from localhost:9000. DB::Exception: Value in column Int8 cannot be safely converted into type UInt8: While processing accurateCast(-1, 'UInt8') AS uint8.
```
## accurateCastOrNull(x, T) {#type_conversion_function-accurate-cast_or_null}
Converts ‘x’ to the ‘t’ data type. Always returns nullable type and returns NULL
if the casted value is not representable in the target type.
Converts input value `x` to the specified data type `T`. Always returns [Nullable](../../sql-reference/data-types/nullable.md) type and returns [NULL](../../sql-reference/syntax.md#null-literal) if the casted value is not representable in the target type.
Example:
**Syntax**
```sql
accurateCastOrNull(x, T)
```
**Parameters**
- `x` — Input value.
- `T` — The name of the returned data type.
**Returned value**
- The value, converted to the specified data type `T`.
**Example**
Query:
``` sql
SELECT
accurateCastOrNull(-1, 'UInt8') as uint8,
accurateCastOrNull(128, 'Int8') as int8,
accurateCastOrNull('Test', 'FixedString(2)') as fixed_string
SELECT toTypeName(accurateCastOrNull(5, 'UInt8'));
```
``` text
┌─uint8─┬─int8─┬─fixed_string─┐
│ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │
└───────┴──────┴──────────────┘┘
```
``` sql
SELECT toTypeName(accurateCastOrNull(5, 'UInt8'))
```
Result:
``` text
┌─toTypeName(accurateCastOrNull(5, 'UInt8'))─┐
@ -450,6 +569,23 @@ SELECT toTypeName(accurateCastOrNull(5, 'UInt8'))
└────────────────────────────────────────────┘
```
Query:
``` sql
SELECT
accurateCastOrNull(-1, 'UInt8') as uint8,
accurateCastOrNull(128, 'Int8') as int8,
accurateCastOrNull('Test', 'FixedString(2)') as fixed_string;
```
Result:
``` text
┌─uint8─┬─int8─┬─fixed_string─┐
│ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │
└───────┴──────┴──────────────┘
```
## toInterval(Year\|Quarter\|Month\|Week\|Day\|Hour\|Minute\|Second) {#function-tointerval}
Converts a Number type argument to an [Interval](../../sql-reference/data-types/special-data-types/interval.md) data type.
@ -477,6 +613,8 @@ toIntervalYear(number)
**Example**
Query:
``` sql
WITH
toDate('2019-01-01') AS date,
@ -484,9 +622,11 @@ WITH
toIntervalWeek(1) AS interval_to_week
SELECT
date + interval_week,
date + interval_to_week
date + interval_to_week;
```
Result:
``` text
┌─plus(date, interval_week)─┬─plus(date, interval_to_week)─┐
│ 2019-01-08 │ 2019-01-08 │
@ -502,7 +642,7 @@ The function parses [ISO 8601](https://en.wikipedia.org/wiki/ISO_8601), [RFC 112
**Syntax**
``` sql
parseDateTimeBestEffort(time_string [, time_zone]);
parseDateTimeBestEffort(time_string [, time_zone])
```
**Arguments**
@ -545,7 +685,7 @@ Query:
``` sql
SELECT parseDateTimeBestEffort('Sat, 18 Aug 2018 07:22:16 GMT', 'Europe/Moscow')
AS parseDateTimeBestEffort
AS parseDateTimeBestEffort;
```
Result:
@ -560,7 +700,7 @@ Query:
``` sql
SELECT parseDateTimeBestEffort('1284101485')
AS parseDateTimeBestEffort
AS parseDateTimeBestEffort;
```
Result:
@ -575,7 +715,7 @@ Query:
``` sql
SELECT parseDateTimeBestEffort('2018-12-12 10:12:12')
AS parseDateTimeBestEffort
AS parseDateTimeBestEffort;
```
Result:
@ -589,7 +729,7 @@ Result:
Query:
``` sql
SELECT parseDateTimeBestEffort('10 20:19')
SELECT parseDateTimeBestEffort('10 20:19');
```
Result:
@ -609,12 +749,12 @@ Result:
## parseDateTimeBestEffortUS {#parsedatetimebesteffortUS}
This function is similar to [‘parseDateTimeBestEffort’](#parsedatetimebesteffort), the only difference is that this function prefers US date format (`MM/DD/YYYY` etc.) in case of ambiguity.
This function is similar to [parseDateTimeBestEffort](#parsedatetimebesteffort), the only difference is that this function prefers US date format (`MM/DD/YYYY` etc.) in case of ambiguity.
**Syntax**
``` sql
parseDateTimeBestEffortUS(time_string [, time_zone]);
parseDateTimeBestEffortUS(time_string [, time_zone])
```
**Arguments**
@ -689,6 +829,178 @@ Same as for [parseDateTimeBestEffort](#parsedatetimebesteffort) except that it r
Same as for [parseDateTimeBestEffort](#parsedatetimebesteffort) except that it returns zero date or zero date time when it encounters a date format that cannot be processed.
## parseDateTimeBestEffortUSOrNull {#parsedatetimebesteffortusornull}
Same as [parseDateTimeBestEffortUS](#parsedatetimebesteffortUS) function except that it returns `NULL` when it encounters a date format that cannot be processed.
**Syntax**
``` sql
parseDateTimeBestEffortUSOrNull(time_string[, time_zone])
```
**Parameters**
- `time_string` — String containing a date or date with time to convert. The date must be in the US date format (`MM/DD/YYYY`, etc). [String](../../sql-reference/data-types/string.md).
- `time_zone` — [Timezone](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone). The function parses `time_string` according to the timezone. Optional. [String](../../sql-reference/data-types/string.md).
**Supported non-standard formats**
- A string containing 9..10 digit [unix timestamp](https://en.wikipedia.org/wiki/Unix_time).
- A string with a date and a time components: `YYYYMMDDhhmmss`, `MM/DD/YYYY hh:mm:ss`, `MM-DD-YY hh:mm`, `YYYY-MM-DD hh:mm:ss`, etc.
- A string with a date, but no time component: `YYYY`, `YYYYMM`, `YYYY*MM`, `MM/DD/YYYY`, `MM-DD-YY`, etc.
- A string with a day and time: `DD`, `DD hh`, `DD hh:mm`. In this case, `YYYY-MM` are substituted with `2000-01`.
- A string that includes date and time along with timezone offset information: `YYYY-MM-DD hh:mm:ss ±h:mm`, etc. For example, `2020-12-12 17:36:00 -5:00`.
**Returned values**
- `time_string` converted to the [DateTime](../../sql-reference/data-types/datetime.md) data type.
- `NULL` if the input string cannot be converted to the `DateTime` data type.
**Examples**
Query:
``` sql
SELECT parseDateTimeBestEffortUSOrNull('02/10/2021 21:12:57') AS parseDateTimeBestEffortUSOrNull;
```
Result:
``` text
┌─parseDateTimeBestEffortUSOrNull─┐
│ 2021-02-10 21:12:57 │
└─────────────────────────────────┘
```
Query:
``` sql
SELECT parseDateTimeBestEffortUSOrNull('02-10-2021 21:12:57 GMT', 'Europe/Moscow') AS parseDateTimeBestEffortUSOrNull;
```
Result:
``` text
┌─parseDateTimeBestEffortUSOrNull─┐
│ 2021-02-11 00:12:57 │
└─────────────────────────────────┘
```
Query:
``` sql
SELECT parseDateTimeBestEffortUSOrNull('02.10.2021') AS parseDateTimeBestEffortUSOrNull;
```
Result:
``` text
┌─parseDateTimeBestEffortUSOrNull─┐
│ 2021-02-10 00:00:00 │
└─────────────────────────────────┘
```
Query:
``` sql
SELECT parseDateTimeBestEffortUSOrNull('10.2021') AS parseDateTimeBestEffortUSOrNull;
```
Result:
``` text
┌─parseDateTimeBestEffortUSOrNull─┐
│ ᴺᵁᴸᴸ │
└─────────────────────────────────┘
```
## parseDateTimeBestEffortUSOrZero {#parsedatetimebesteffortusorzero}
Same as [parseDateTimeBestEffortUS](#parsedatetimebesteffortUS) function except that it returns zero date (`1970-01-01`) or zero date with time (`1970-01-01 00:00:00`) when it encounters a date format that cannot be processed.
**Syntax**
``` sql
parseDateTimeBestEffortUSOrZero(time_string[, time_zone])
```
**Parameters**
- `time_string` — String containing a date or date with time to convert. The date must be in the US date format (`MM/DD/YYYY`, etc). [String](../../sql-reference/data-types/string.md).
- `time_zone` — [Timezone](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone). The function parses `time_string` according to the timezone. Optional. [String](../../sql-reference/data-types/string.md).
**Supported non-standard formats**
- A string containing 9..10 digit [unix timestamp](https://en.wikipedia.org/wiki/Unix_time).
- A string with a date and a time components: `YYYYMMDDhhmmss`, `MM/DD/YYYY hh:mm:ss`, `MM-DD-YY hh:mm`, `YYYY-MM-DD hh:mm:ss`, etc.
- A string with a date, but no time component: `YYYY`, `YYYYMM`, `YYYY*MM`, `MM/DD/YYYY`, `MM-DD-YY`, etc.
- A string with a day and time: `DD`, `DD hh`, `DD hh:mm`. In this case, `YYYY-MM` are substituted with `2000-01`.
- A string that includes date and time along with timezone offset information: `YYYY-MM-DD hh:mm:ss ±h:mm`, etc. For example, `2020-12-12 17:36:00 -5:00`.
**Returned values**
- `time_string` converted to the [DateTime](../../sql-reference/data-types/datetime.md) data type.
- Zero date or zero date with time if the input string cannot be converted to the `DateTime` data type.
**Examples**
Query:
``` sql
SELECT parseDateTimeBestEffortUSOrZero('02/10/2021 21:12:57') AS parseDateTimeBestEffortUSOrZero;
```
Result:
``` text
┌─parseDateTimeBestEffortUSOrZero─┐
│ 2021-02-10 21:12:57 │
└─────────────────────────────────┘
```
Query:
``` sql
SELECT parseDateTimeBestEffortUSOrZero('02-10-2021 21:12:57 GMT', 'Europe/Moscow') AS parseDateTimeBestEffortUSOrZero;
```
Result:
``` text
┌─parseDateTimeBestEffortUSOrZero─┐
│ 2021-02-11 00:12:57 │
└─────────────────────────────────┘
```
Query:
``` sql
SELECT parseDateTimeBestEffortUSOrZero('02.10.2021') AS parseDateTimeBestEffortUSOrZero;
```
Result:
``` text
┌─parseDateTimeBestEffortUSOrZero─┐
│ 2021-02-10 00:00:00 │
└─────────────────────────────────┘
```
Query:
``` sql
SELECT parseDateTimeBestEffortUSOrZero('02.2021') AS parseDateTimeBestEffortUSOrZero;
```
Result:
``` text
┌─parseDateTimeBestEffortUSOrZero─┐
│ 1970-01-01 00:00:00 │
└─────────────────────────────────┘
```
## toLowCardinality {#tolowcardinality}
Converts input parameter to the [LowCardianlity](../../sql-reference/data-types/lowcardinality.md) version of same data type.
@ -716,7 +1028,7 @@ Type: `LowCardinality(expr_result_type)`
Query:
``` sql
SELECT toLowCardinality('1')
SELECT toLowCardinality('1');
```
Result:
@ -755,7 +1067,7 @@ Query:
``` sql
WITH toDateTime64('2019-09-16 19:20:12.345678910', 6) AS dt64
SELECT toUnixTimestamp64Milli(dt64)
SELECT toUnixTimestamp64Milli(dt64);
```
Result:
@ -768,7 +1080,7 @@ Result:
``` sql
WITH toDateTime64('2019-09-16 19:20:12.345678910', 6) AS dt64
SELECT toUnixTimestamp64Nano(dt64)
SELECT toUnixTimestamp64Nano(dt64);
```
Result:
@ -802,13 +1114,17 @@ fromUnixTimestamp64Milli(value [, ti])
- `value` converted to the `DateTime64` data type.
**Examples**
**Example**
Query:
``` sql
WITH CAST(1234567891011, 'Int64') AS i64
SELECT fromUnixTimestamp64Milli(i64, 'UTC')
SELECT fromUnixTimestamp64Milli(i64, 'UTC');
```
Result:
``` text
┌─fromUnixTimestamp64Milli(i64, 'UTC')─┐
│ 2009-02-13 23:31:31.011 │
@ -817,9 +1133,9 @@ SELECT fromUnixTimestamp64Milli(i64, 'UTC')
## formatRow {#formatrow}
Converts arbitrary expressions into a string via given format.
Converts arbitrary expressions into a string via given format.
**Syntax**
**Syntax**
``` sql
formatRow(format, x, y, ...)
@ -840,7 +1156,7 @@ Query:
``` sql
SELECT formatRow('CSV', number, 'good')
FROM numbers(3)
FROM numbers(3);
```
Result:
@ -860,7 +1176,7 @@ Result:
Converts arbitrary expressions into a string via given format. The function trims the last `\n` if any.
**Syntax**
**Syntax**
``` sql
formatRowNoNewline(format, x, y, ...)
@ -881,7 +1197,7 @@ Query:
``` sql
SELECT formatRowNoNewline('CSV', number, 'good')
FROM numbers(3)
FROM numbers(3);
```
Result:

20
docs/en/sql-reference/operators/in.md
View File

@ -13,10 +13,28 @@ SELECT (CounterID, UserID) IN ((34, 123), (101500, 456)) FROM ...
If the left side is a single column that is in the index, and the right side is a set of constants, the system uses the index for processing the query.
Don’t list too many values explicitly (i.e. millions). If a data set is large, put it in a temporary table (for example, see the section “External data for query processing”), then use a subquery.
Don’t list too many values explicitly (i.e. millions). If a data set is large, put it in a temporary table (for example, see the section [External data for query processing](../../engines/table-engines/special/external-data.md)), then use a subquery.
The right side of the operator can be a set of constant expressions, a set of tuples with constant expressions (shown in the examples above), or the name of a database table or SELECT subquery in brackets.
ClickHouse allows types to differ in the left and the right parts of `IN` subquery. In this case it converts the left side value to the type of the right side, as if the [accurateCastOrNull](../functions/type-conversion-functions.md#type_conversion_function-accurate-cast_or_null) function is applied. That means, that the data type becomes [Nullable](../../sql-reference/data-types/nullable.md), and if the conversion cannot be performed, it returns [NULL](../../sql-reference/syntax.md#null-literal).
**Example**
Query:
``` sql
SELECT '1' IN (SELECT 1);
```
Result:
``` text
┌─in('1', _subquery49)─┐
│ 1 │
└──────────────────────┘
```
If the right side of the operator is the name of a table (for example, `UserID IN users`), this is equivalent to the subquery `UserID IN (SELECT * FROM users)`. Use this when working with external data that is sent along with the query. For example, the query can be sent together with a set of user IDs loaded to the ‘users’ temporary table, which should be filtered.
If the right side of the operator is a table name that has the Set engine (a prepared data set that is always in RAM), the data set will not be created over again for each query.

17
docs/en/sql-reference/statements/alter/column.md
View File

@ -25,6 +25,7 @@ The following actions are supported:
- [COMMENT COLUMN](#alter_comment-column) — Adds a text comment to the column.
- [MODIFY COLUMN](#alter_modify-column) — Changes column’s type, default expression and TTL.
- [MODIFY COLUMN REMOVE](#modify-remove) — Removes one of the column properties.
- [RENAME COLUMN](#alter_rename-column) — Renames an existing column.
These actions are described in detail below.
@ -183,6 +184,22 @@ ALTER TABLE table_with_ttl MODIFY COLUMN column_ttl REMOVE TTL;
- [REMOVE TTL](ttl.md).
## RENAME COLUMN {#alter_rename-column}
Renames an existing column.
Syntax:
```sql
ALTER TABLE table_name RENAME COLUMN column_name TO new_column_name;
```
**Example**
```sql
ALTER TABLE table_with_ttl RENAME COLUMN column_ttl TO column_ttl_new;
```
## Limitations {#alter-query-limitations}
The `ALTER` query lets you create and delete separate elements (columns) in nested data structures, but not whole nested data structures. To add a nested data structure, you can add columns with a name like `name.nested_name` and the type `Array(T)`. A nested data structure is equivalent to multiple array columns with a name that has the same prefix before the dot.

3
docs/en/sql-reference/statements/create/user.md
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] ...]
[IDENTIFIED [WITH {NO_PASSWORD|PLAINTEXT_PASSWORD|SHA256_PASSWORD|SHA256_HASH|DOUBLE_SHA1_PASSWORD|DOUBLE_SHA1_HASH}] BY {'password'|'hash'}]
[IDENTIFIED [WITH {NO_PASSWORD|PLAINTEXT_PASSWORD|SHA256_PASSWORD|SHA256_HASH|DOUBLE_SHA1_PASSWORD|DOUBLE_SHA1_HASH|LDAP_SERVER}] BY {'password'|'hash'}]
[HOST {LOCAL | NAME 'name' | REGEXP 'name_regexp' | IP 'address' | LIKE 'pattern'} [,...] | ANY | NONE]
[DEFAULT ROLE role [,...]]
[SETTINGS variable [= value] [MIN [=] min_value] [MAX [=] max_value] [READONLY|WRITABLE] | PROFILE 'profile_name'] [,...]
@ -30,6 +30,7 @@ There are multiple ways of user identification:
- `IDENTIFIED WITH sha256_hash BY 'hash'`
- `IDENTIFIED WITH double_sha1_password BY 'qwerty'`
- `IDENTIFIED WITH double_sha1_hash BY 'hash'`
- `IDENTIFIED WITH ldap_server BY 'server'`
## User Host {#user-host}

188
docs/en/sql-reference/statements/create/view.md
View File

@ -41,7 +41,6 @@ SELECT a, b, c FROM (SELECT ...)
CREATE MATERIALIZED VIEW [IF NOT EXISTS] [db.]table_name [ON CLUSTER] [TO[db.]name] [ENGINE = engine] [POPULATE] AS SELECT ...
```
Materialized views store data transformed by the corresponding [SELECT](../../../sql-reference/statements/select/index.md) query.
When creating a materialized view without `TO [db].[table]`, you must specify `ENGINE` – the table engine for storing data.
@ -65,4 +64,191 @@ Views look the same as normal tables. For example, they are listed in the result
There isn’t a separate query for deleting views. To delete a view, use [DROP TABLE](../../../sql-reference/statements/drop.md).
## Live View (Experimental) {#live-view}
!!! important "Important"
This is an experimental feature that may change in backwards-incompatible ways in the future releases.
Enable usage of live views and `WATCH` query using `set allow_experimental_live_view = 1`.
```sql
CREATE LIVE VIEW [IF NOT EXISTS] [db.]table_name [WITH [TIMEOUT [value_in_sec] [AND]] [REFRESH [value_in_sec]]] AS SELECT ...
```
Live views store result of the corresponding [SELECT](../../../sql-reference/statements/select/index.md) query and are updated any time the result of the query changes. Query result as well as partial result needed to combine with new data are stored in memory providing increased performance for repeated queries. Live views can provide push notifications when query result changes using the [WATCH](../../../sql-reference/statements/watch.md) query.
Live views are triggered by insert into the innermost table specified in the query.
Live views work similarly to how a query in a distributed table works. But instead of combining partial results from different servers they combine partial result from current data with partial result from the new data. When a live view query includes a subquery then the cached partial result is only stored for the innermost subquery.
!!! info "Limitations"
- [Table function](../../../sql-reference/table-functions/index.md) is not supported as the innermost table.
- Tables that do not have inserts such as a [dictionary](../../../sql-reference/dictionaries/index.md), [system table](../../../operations/system-tables/index.md), a [normal view](#normal), or a [materialized view](#materialized) will not trigger a live view.
- Only queries where one can combine partial result from the old data plus partial result from the new data will work. Live view will not work for queries that require the complete data set to compute the final result or aggregations where the state of the aggregation must be preserved.
- Does not work with replicated or distributed tables where inserts are performed on different nodes.
- Can't be triggered by multiple tables.
See [WITH REFRESH](#live-view-with-refresh) to force periodic updates of a live view that in some cases can be used as a workaround.
You can watch for changes in the live view query result using the [WATCH](../../../sql-reference/statements/watch.md) query
```sql
WATCH [db.]live_view
```
**Example:**
```sql
CREATE TABLE mt (x Int8) Engine = MergeTree ORDER BY x;
CREATE LIVE VIEW lv AS SELECT sum(x) FROM mt;
```
Watch a live view while doing a parallel insert into the source table.
```sql
WATCH lv
```
```bash
┌─sum(x)─┬─_version─┐
│ 1 │ 1 │
└────────┴──────────┘
┌─sum(x)─┬─_version─┐
│ 2 │ 2 │
└────────┴──────────┘
┌─sum(x)─┬─_version─┐
│ 6 │ 3 │
└────────┴──────────┘
...
```
```sql
INSERT INTO mt VALUES (1);
INSERT INTO mt VALUES (2);
INSERT INTO mt VALUES (3);
```
or add [EVENTS](../../../sql-reference/statements/watch.md#events-clause) clause to just get change events.
```sql
WATCH [db.]live_view EVENTS
```
**Example:**
```sql
WATCH lv EVENTS
```
```bash
┌─version─┐
│ 1 │
└─────────┘
┌─version─┐
│ 2 │
└─────────┘
┌─version─┐
│ 3 │
└─────────┘
...
```
You can execute [SELECT](../../../sql-reference/statements/select/index.md) query on a live view in the same way as for any regular view or a table. If the query result is cached it will return the result immediately without running the stored query on the underlying tables.
```sql
SELECT * FROM [db.]live_view WHERE ...
```
### Force Refresh {#live-view-alter-refresh}
You can force live view refresh using the `ALTER LIVE VIEW [db.]table_name REFRESH` statement.
### With Timeout {#live-view-with-timeout}
When a live view is create with a `WITH TIMEOUT` clause then the live view will be dropped automatically after the specified number of seconds elapse since the end of the last [WATCH](../../../sql-reference/statements/watch.md) query that was watching the live view.
```sql
CREATE LIVE VIEW [db.]table_name WITH TIMEOUT [value_in_sec] AS SELECT ...
```
If the timeout value is not specified then the value specified by the `temporary_live_view_timeout` setting is used.
**Example:**
```sql
CREATE TABLE mt (x Int8) Engine = MergeTree ORDER BY x;
CREATE LIVE VIEW lv WITH TIMEOUT 15 AS SELECT sum(x) FROM mt;
```
### With Refresh {#live-view-with-refresh}
When a live view is created with a `WITH REFRESH` clause then it will be automatically refreshed after the specified number of seconds elapse since the last refresh or trigger.
```sql
CREATE LIVE VIEW [db.]table_name WITH REFRESH [value_in_sec] AS SELECT ...
```
If the refresh value is not specified then the value specified by the `periodic_live_view_refresh` setting is used.
**Example:**
```sql
CREATE LIVE VIEW lv WITH REFRESH 5 AS SELECT now();
WATCH lv
```
```bash
┌───────────────now()─┬─_version─┐
│ 2021-02-21 08:47:05 │ 1 │
└─────────────────────┴──────────┘
┌───────────────now()─┬─_version─┐
│ 2021-02-21 08:47:10 │ 2 │
└─────────────────────┴──────────┘
┌───────────────now()─┬─_version─┐
│ 2021-02-21 08:47:15 │ 3 │
└─────────────────────┴──────────┘
```
You can combine `WITH TIMEOUT` and `WITH REFRESH` clauses using an `AND` clause.
```sql
CREATE LIVE VIEW [db.]table_name WITH TIMEOUT [value_in_sec] AND REFRESH [value_in_sec] AS SELECT ...
```
**Example:**
```sql
CREATE LIVE VIEW lv WITH TIMEOUT 15 AND REFRESH 5 AS SELECT now();
```
After 15 sec the live view will be automatically dropped if there are no active `WATCH` queries.
```sql
WATCH lv
```
```
Code: 60. DB::Exception: Received from localhost:9000. DB::Exception: Table default.lv doesn't exist..
```
### Usage
Most common uses of live view tables include:
- Providing push notifications for query result changes to avoid polling.
- Caching results of most frequent queries to provide immediate query results.
- Watching for table changes and triggering a follow-up select queries.
- Watching metrics from system tables using periodic refresh.
### Settings {#live-view-settings}
You can use the following settings to control the behaviour of live views.
- `allow_experimental_live_view` - enable live views. Default is `0`.
- `live_view_heartbeat_interval` - the heartbeat interval in seconds to indicate live query is alive. Default is `15` seconds.
- `max_live_view_insert_blocks_before_refresh` - maximum number of inserted blocks after which
mergeable blocks are dropped and query is re-executed. Default is `64` inserts.
- `temporary_live_view_timeout` - interval after which live view with timeout is deleted. Default is `5` seconds.
- `periodic_live_view_refresh` - interval after which periodically refreshed live view is forced to refresh. Default is `60` seconds.
[Original article](https://clickhouse.tech/docs/en/sql-reference/statements/create/view/) <!--hide-->

106
docs/en/sql-reference/statements/watch.md Normal file
View File

@ -0,0 +1,106 @@
---
toc_priority: 53
toc_title: WATCH
---
# WATCH Statement (Experimental) {#watch}
!!! important "Important"
This is an experimental feature that may change in backwards-incompatible ways in the future releases.
Enable live views and `WATCH` query using `set allow_experimental_live_view = 1`.
``` sql
WATCH [db.]live_view
[EVENTS]
[LIMIT n]
[FORMAT format]
```
The `WATCH` query performs continuous data retrieval from a [live view](./create/view.md#live-view) table. Unless the `LIMIT` clause is specified it provides an infinite stream of query results from a [live view](./create/view.md#live-view).
```sql
WATCH [db.]live_view
```
The virtual `_version` column in the query result indicates the current result version.
**Example:**
```sql
CREATE LIVE VIEW lv WITH REFRESH 5 AS SELECT now();
WATCH lv
```
```bash
┌───────────────now()─┬─_version─┐
│ 2021-02-21 09:17:21 │ 1 │
└─────────────────────┴──────────┘
┌───────────────now()─┬─_version─┐
│ 2021-02-21 09:17:26 │ 2 │
└─────────────────────┴──────────┘
┌───────────────now()─┬─_version─┐
│ 2021-02-21 09:17:31 │ 3 │
└─────────────────────┴──────────┘
...
```
By default, the requested data is returned to the client, while in conjunction with [INSERT INTO](../../sql-reference/statements/insert-into.md) it can be forwarded to a different table.
```sql
INSERT INTO [db.]table WATCH [db.]live_view ...
```
## EVENTS Clause {#events-clause}
The `EVENTS` clause can be used to obtain a short form of the `WATCH` query where instead of the query result you will just get the latest query result version.
```sql
WATCH [db.]live_view EVENTS
```
**Example:**
```sql
CREATE LIVE VIEW lv WITH REFRESH 5 AS SELECT now();
WATCH lv EVENTS
```
```bash
┌─version─┐
│ 1 │
└─────────┘
┌─version─┐
│ 2 │
└─────────┘
...
```
## LIMIT Clause {#limit-clause}
The `LIMIT n` clause species the number of updates the `WATCH` query should wait for before terminating. By default there is no limit on the number of updates and therefore the query will not terminate. The value of `0` indicates that the `WATCH` query should not wait for any new query results and therefore will return immediately once query is evaluated.
```sql
WATCH [db.]live_view LIMIT 1
```
**Example:**
```sql
CREATE LIVE VIEW lv WITH REFRESH 5 AS SELECT now();
WATCH lv EVENTS LIMIT 1
```
```bash
┌─version─┐
│ 1 │
└─────────┘
```
## FORMAT Clause {#format-clause}
The `FORMAT` clause works the same way as for the [SELECT](../../sql-reference/statements/select/format.md#format-clause).
!!! info "Note"
The [JSONEachRowWithProgress](../../../interfaces/formats/#jsoneachrowwithprogress) format should be used when watching [live view](./create/view.md#live-view) tables over the HTTP interface. The progress messages will be added to the output to keep the long-lived HTTP connection alive until the query result changes. The interval between progress messages is controlled using the [live_view_heartbeat_interval](./create/view.md#live-view-settings) setting.

22
docs/en/sql-reference/table-functions/file.md
View File

@ -5,7 +5,7 @@ toc_title: file
# file {#file}
Creates a table from a file. This table function is similar to [url](../../sql-reference/table-functions/url.md) and [hdfs](../../sql-reference/table-functions/hdfs.md) ones.
Creates a table from a file. This table function is similar to [url](../../sql-reference/table-functions/url.md) and [hdfs](../../sql-reference/table-functions/hdfs.md) ones.
`file` function can be used in `SELECT` and `INSERT` queries on data in [File](../../engines/table-engines/special/file.md) tables.
@ -15,9 +15,9 @@ Creates a table from a file. This table function is similar to [url](../../sql-r
file(path, format, structure)
```
**Input parameters**
**Parameters**
- `path` — The relative path to the file from [user_files_path](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-user_files_path). Path to file support following globs in readonly mode: `*`, `?`, `{abc,def}` and `{N..M}` where `N`, `M` — numbers, `'abc', 'def'` — strings.
- `path` — The relative path to the file from [user_files_path](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-user_files_path). Path to file support following globs in read-only mode: `*`, `?`, `{abc,def}` and `{N..M}` where `N`, `M` — numbers, `'abc', 'def'` — strings.
- `format` — The [format](../../interfaces/formats.md#formats) of the file.
- `structure` — Structure of the table. Format: `'column1_name column1_type, column2_name column2_type, ...'`.
@ -39,7 +39,7 @@ $ cat /var/lib/clickhouse/user_files/test.csv
78,43,45
```
Getting data from a table in `test.csv` and selecting first two rows from it:
Getting data from a table in `test.csv` and selecting the first two rows from it:
``` sql
SELECT * FROM file('test.csv', 'CSV', 'column1 UInt32, column2 UInt32, column3 UInt32') LIMIT 2;
@ -51,7 +51,8 @@ SELECT * FROM file('test.csv', 'CSV', 'column1 UInt32, column2 UInt32, column3 U
│ 3 │ 2 │ 1 │
└─────────┴─────────┴─────────┘
```
Getting the first 10 lines of a table that contains 3 columns of UInt32 type from a CSV file:
Getting the first 10 lines of a table that contains 3 columns of [UInt32](../../sql-reference/data-types/int-uint.md) type from a CSV file:
``` sql
SELECT * FROM file('test.csv', 'CSV', 'column1 UInt32, column2 UInt32, column3 UInt32') LIMIT 10;
@ -71,17 +72,16 @@ SELECT * FROM file('test.csv', 'CSV', 'column1 UInt32, column2 UInt32, column3 U
└─────────┴─────────┴─────────┘
```
## Globs in Path {#globs-in-path}
Multiple path components can have globs. For being processed file should exists and matches to the whole path pattern (not only suffix or prefix).
Multiple path components can have globs. For being processed file must exist and match to the whole path pattern (not only suffix or prefix).
- `*` — Substitutes any number of any characters except `/` including empty string.
- `?` — Substitutes any single character.
- `{some_string,another_string,yet_another_one}` — Substitutes any of strings `'some_string', 'another_string', 'yet_another_one'`.
- `{N..M}` — Substitutes any number in range from N to M including both borders.
Constructions with `{}` are similar to the [remote table function](../../sql-reference/table-functions/remote.md)).
Constructions with `{}` are similar to the [remote](remote.md) table function.
**Example**
@ -94,13 +94,13 @@ Suppose we have several files with the following relative paths:
- 'another_dir/some_file_2'
- 'another_dir/some_file_3'
Query the amount of rows in these files:
Query the number of rows in these files:
``` sql
SELECT count(*) FROM file('{some,another}_dir/some_file_{1..3}', 'TSV', 'name String, value UInt32');
```
Query the amount of rows in all files of these two directories:
Query the number of rows in all files of these two directories:
``` sql
SELECT count(*) FROM file('{some,another}_dir/*', 'TSV', 'name String, value UInt32');
@ -124,6 +124,6 @@ SELECT count(*) FROM file('big_dir/file{0..9}{0..9}{0..9}', 'CSV', 'name String,
**See Also**
- [Virtual columns](https://clickhouse.tech/docs/en/operations/table_engines/#table_engines-virtual_columns)
- [Virtual columns](index.md#table_engines-virtual_columns)
[Original article](https://clickhouse.tech/docs/en/sql-reference/table-functions/file/) <!--hide-->

2
docs/en/sql-reference/table-functions/mysql.md
View File

@ -44,7 +44,7 @@ The rest of the conditions and the `LIMIT` sampling constraint are executed in C
A table object with the same columns as the original MySQL table.
!!! info "Note"
In the `INSERT` query to distinguish table function `mysql(...)` from table name with column names list you must use keywords `FUNCTION` or `TABLE FUNCTION`. See examples below.
In the `INSERT` query to distinguish table function `mysql(...)` from table name with column names list, you must use keywords `FUNCTION` or `TABLE FUNCTION`. See examples below.
**Examples**

26
docs/en/sql-reference/table-functions/remote.md
View File

@ -5,7 +5,7 @@ toc_title: remote
# remote, remoteSecure {#remote-remotesecure}
Allows to access remote servers without creating a [Distributed](../../engines/table-engines/special/distributed.md) table. `remoteSecure` - same as `remote` but with secured connection.
Allows to access remote servers without creating a [Distributed](../../engines/table-engines/special/distributed.md) table. `remoteSecure` - same as `remote` but with a secured connection.
Both functions can be used in `SELECT` and `INSERT` queries.
@ -18,31 +18,31 @@ remoteSecure('addresses_expr', db, table[, 'user'[, 'password'], sharding_key])
remoteSecure('addresses_expr', db.table[, 'user'[, 'password'], sharding_key])
```
**Input parameters**
**Parameters**
- `addresses_expr` – An expression that generates addresses of remote servers. This may be just one server address. The server address is `host:port`, or just `host`.
- `addresses_expr` — An expression that generates addresses of remote servers. This may be just one server address. The server address is `host:port`, or just `host`.
The host can be specified as the server name, or as the IPv4 or IPv6 address. An IPv6 address is specified in square brackets.
The port is the TCP port on the remote server. If the port is omitted, it uses [tcp_port](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-tcp_port) from the server’s config file in `remote` (by default, 9000) and [tcp_port_secure](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-tcp_port_secure) in `remoteSecure` (by default, 9440).
The port is the TCP port on the remote server. If the port is omitted, it uses [tcp_port](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-tcp_port) from the server’s config file in `remote` (by default, 9000) and [tcp_port_secure](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-tcp_port_secure) in `remoteSecure` (by default, 9440).
The port is required for an IPv6 address.
Type: [String](../../sql-reference/data-types/string.md).
- `db` - Database name. Type: [String](../../sql-reference/data-types/string.md).
- `table` - Table name. Type: [String](../../sql-reference/data-types/string.md).
- `user` - User name. If the user is not specified, `default` is used. Type: [String](../../sql-reference/data-types/string.md).
- `password` - User password. If the password is not specified, an empty password is used. Type: [String](../../sql-reference/data-types/string.md).
- `sharding_key` - Sharding key to support distributing data across nodes. For example: `insert into remote('127.0.0.1:9000,127.0.0.2', db, table, 'default', rand())`. Type: [UInt32](../../sql-reference/data-types/int-uint.md).
- `db` — Database name. Type: [String](../../sql-reference/data-types/string.md).
- `table` — Table name. Type: [String](../../sql-reference/data-types/string.md).
- `user` — User name. If the user is not specified, `default` is used. Type: [String](../../sql-reference/data-types/string.md).
- `password` — User password. If the password is not specified, an empty password is used. Type: [String](../../sql-reference/data-types/string.md).
- `sharding_key` — Sharding key to support distributing data across nodes. For example: `insert into remote('127.0.0.1:9000,127.0.0.2', db, table, 'default', rand())`. Type: [UInt32](../../sql-reference/data-types/int-uint.md).
**Returned value**
Dataset from remote servers.
The dataset from remote servers.
**Usage**
Using the `remote` table function is less optimal than creating a `Distributed` table, because in this case the server connection is re-established for every request. In addition, if host names are set, the names are resolved, and errors are not counted when working with various replicas. When processing a large number of queries, always create the `Distributed` table ahead of time, and don’t use the `remote` table function.
Using the `remote` table function is less optimal than creating a `Distributed` table because in this case the server connection is re-established for every request. Also, if hostnames are set, the names are resolved, and errors are not counted when working with various replicas. When processing a large number of queries, always create the `Distributed` table ahead of time, and don’t use the `remote` table function.
The `remote` table function can be useful in the following cases:
@ -62,7 +62,7 @@ localhost
[2a02:6b8:0:1111::11]:9000
```
Multiple addresses can be comma-separated. In this case, ClickHouse will use distributed processing, so it will send the query to all specified addresses (like to shards with different data). Example:
Multiple addresses can be comma-separated. In this case, ClickHouse will use distributed processing, so it will send the query to all specified addresses (like shards with different data). Example:
``` text
example01-01-1,example01-02-1
@ -82,7 +82,7 @@ example01-{01..02}-1
If you have multiple pairs of curly brackets, it generates the direct product of the corresponding sets.
Addresses and parts of addresses in curly brackets can be separated by the pipe symbol (\|). In this case, the corresponding sets of addresses are interpreted as replicas, and the query will be sent to the first healthy replica. However, the replicas are iterated in the order currently set in the [load_balancing](../../operations/settings/settings.md) setting. This example specifies two shards that each have two replicas:
Addresses and parts of addresses in curly brackets can be separated by the pipe symbol (\|). In this case, the corresponding sets of addresses are interpreted as replicas, and the query will be sent to the first healthy replica. However, the replicas are iterated in the order currently set in the [load_balancing](../../operations/settings/settings.md#settings-load_balancing) setting. This example specifies two shards that each have two replicas:
``` text
example01-{01..02}-{1|2}

14
docs/en/sql-reference/table-functions/url.md
View File

@ -15,25 +15,25 @@ toc_title: url
url(URL, format, structure)
```
**Input parameters**
**Parameters**
- `URL` - HTTP or HTTPS server address, which can accept `GET` (for `SELECT`) or `POST` (for `INSERT`) requests. Type: [String](../../sql-reference/data-types/string.md).
- `format` - [Format](../../interfaces/formats.md#formats) of the data. Type: [String](../../sql-reference/data-types/string.md).
- `structure` - Table structure in `'UserID UInt64, Name String'` format. Determines column names and types. Type: [String](../../sql-reference/data-types/string.md).
- `URL` — HTTP or HTTPS server address, which can accept `GET` or `POST` requests (for `SELECT` or `INSERT` queries correspondingly). Type: [String](../../sql-reference/data-types/string.md).
- `format` — [Format](../../interfaces/formats.md#formats) of the data. Type: [String](../../sql-reference/data-types/string.md).
- `structure` — Table structure in `'UserID UInt64, Name String'` format. Determines column names and types. Type: [String](../../sql-reference/data-types/string.md).
**Returned value**
A table with the specified format and structure and with data from the defined URL.
A table with the specified format and structure and with data from the defined `URL`.
**Examples**
Getting the first 3 lines of a table that contains columns of `String` and `UInt32` type from HTTP-server which answers in `CSV` format.
Getting the first 3 lines of a table that contains columns of `String` and [UInt32](../../sql-reference/data-types/int-uint.md) type from HTTP-server which answers in [CSV](../../interfaces/formats.md/#csv) format.
``` sql
SELECT * FROM url('http://127.0.0.1:12345/', CSV, 'column1 String, column2 UInt32') LIMIT 3;
```
Inserting data from a URL into a table:
Inserting data from a `URL` into a table:
``` sql
CREATE TABLE test_table (column1 String, column2 UInt32) ENGINE=Memory;

28
docs/ru/engines/table-engines/integrations/rabbitmq.md
View File

@ -52,6 +52,21 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
- `rabbitmq_max_block_size`
- `rabbitmq_flush_interval_ms`
Настройки форматов данных также могут быть добавлены в списке RabbitMQ настроек.
Example:
``` sql
CREATE TABLE queue (
key UInt64,
value UInt64,
date DateTime
) ENGINE = RabbitMQ SETTINGS rabbitmq_host_port = 'localhost:5672',
rabbitmq_exchange_name = 'exchange1',
rabbitmq_format = 'JSONEachRow',
rabbitmq_num_consumers = 5,
date_time_input_format = 'best_effort';
```
Конфигурация сервера RabbitMQ добавляется с помощью конфигурационного файла ClickHouse.
@ -72,18 +87,6 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
</rabbitmq>
```
Example:
``` sql
CREATE TABLE queue (
key UInt64,
value UInt64
) ENGINE = RabbitMQ SETTINGS rabbitmq_host_port = 'localhost:5672',
rabbitmq_exchange_name = 'exchange1',
rabbitmq_format = 'JSONEachRow',
rabbitmq_num_consumers = 5;
```
## Описание {#description}
Запрос `SELECT` не очень полезен для чтения сообщений (за исключением отладки), поскольку каждое сообщение может быть прочитано только один раз. Практичнее создавать потоки реального времени с помощью [материализованных преставлений](../../../sql-reference/statements/create/view.md). Для этого:
@ -107,6 +110,7 @@ Example:
- `consistent_hash` - данные равномерно распределяются между всеми связанными таблицами, где имя точки обмена совпадает. Обратите внимание, что этот тип обмена должен быть включен с помощью плагина RabbitMQ: `rabbitmq-plugins enable rabbitmq_consistent_hash_exchange`.
Настройка `rabbitmq_queue_base` может быть использована в следующих случаях:
1. чтобы восстановить чтение из ранее созданных очередей, если оно прекратилось по какой-либо причине, но очереди остались непустыми. Для восстановления чтения из одной конкретной очереди, нужно написать ее имя в `rabbitmq_queue_base` настройку и не указывать настройки `rabbitmq_num_consumers` и `rabbitmq_num_queues`. Чтобы восстановить чтение из всех очередей, которые были созданы для конкретной таблицы, необходимо совпадение следующих настроек: `rabbitmq_queue_base`, `rabbitmq_num_consumers`, `rabbitmq_num_queues`. По умолчанию, если настройка `rabbitmq_queue_base` не указана, будут использованы уникальные для каждой таблицы имена очередей.
2. чтобы объявить одни и те же очереди для разных таблиц, что позволяет создавать несколько параллельных подписчиков на каждую из очередей. То есть обеспечивается лучшая производительность. В данном случае, для таких таблиц также необходимо совпадение настроек: `rabbitmq_num_consumers`, `rabbitmq_num_queues`.
3. чтобы повторно использовать созданные c `durable` настройкой очереди, так как они не удаляются автоматически (но могут быть удалены с помощью любого RabbitMQ CLI).

2
docs/ru/engines/table-engines/special/file.md
View File

@ -63,7 +63,7 @@ SELECT * FROM file_engine_table
## Использование движка в Clickhouse-local {#ispolzovanie-dvizhka-v-clickhouse-local}
В [clickhouse-local](../../../engines/table-engines/special/file.md) движок в качестве параметра принимает не только формат, но и путь к файлу. В том числе можно указать стандартные потоки ввода/вывода цифровым или буквенным обозначением `0` или `stdin`, `1` или `stdout`.
В [clickhouse-local](../../../engines/table-engines/special/file.md) движок в качестве параметра принимает не только формат, но и путь к файлу. В том числе можно указать стандартные потоки ввода/вывода цифровым или буквенным обозначением `0` или `stdin`, `1` или `stdout`. Можно записывать и читать сжатые файлы. Для этого нужно задать дополнительный параметр движка или расширение файла (`gz`, `br` или `xz`).
**Пример:**

416
docs/ru/getting-started/example-datasets/brown-benchmark.md Normal file
View File

@ -0,0 +1,416 @@
---
toc_priority: 20
toc_title: Brown University Benchmark
---
# Brown University Benchmark
`MgBench` — это аналитический тест производительности для данных журнала событий, сгенерированных машиной. Бенчмарк разработан [Andrew Crotty](http://cs.brown.edu/people/acrotty/).
Скачать данные:
```
wget https://datasets.clickhouse.tech/mgbench{1..3}.csv.xz
```
Распаковать данные:
```
xz -v -d mgbench{1..3}.csv.xz
```
Создание таблиц:
```
CREATE DATABASE mgbench;
CREATE TABLE mgbench.logs1 (
log_time DateTime,
machine_name LowCardinality(String),
machine_group LowCardinality(String),
cpu_idle Nullable(Float32),
cpu_nice Nullable(Float32),
cpu_system Nullable(Float32),
cpu_user Nullable(Float32),
cpu_wio Nullable(Float32),
disk_free Nullable(Float32),
disk_total Nullable(Float32),
part_max_used Nullable(Float32),
load_fifteen Nullable(Float32),
load_five Nullable(Float32),
load_one Nullable(Float32),
mem_buffers Nullable(Float32),
mem_cached Nullable(Float32),
mem_free Nullable(Float32),
mem_shared Nullable(Float32),
swap_free Nullable(Float32),
bytes_in Nullable(Float32),
bytes_out Nullable(Float32)
)
ENGINE = MergeTree()
ORDER BY (machine_group, machine_name, log_time);
CREATE TABLE mgbench.logs2 (
log_time DateTime,
client_ip IPv4,
request String,
status_code UInt16,
object_size UInt64
)
ENGINE = MergeTree()
ORDER BY log_time;
CREATE TABLE mgbench.logs3 (
log_time DateTime64,
device_id FixedString(15),
device_name LowCardinality(String),
device_type LowCardinality(String),
device_floor UInt8,
event_type LowCardinality(String),
event_unit FixedString(1),
event_value Nullable(Float32)
)
ENGINE = MergeTree()
ORDER BY (event_type, log_time);
```
Вставка данных:
```
clickhouse-client --query "INSERT INTO mgbench.logs1 FORMAT CSVWithNames" < mgbench1.csv
clickhouse-client --query "INSERT INTO mgbench.logs2 FORMAT CSVWithNames" < mgbench2.csv
clickhouse-client --query "INSERT INTO mgbench.logs3 FORMAT CSVWithNames" < mgbench3.csv
```
Запуск тестов производительности:
```
-- Q1.1: What is the CPU/network utilization for each web server since midnight?
SELECT machine_name,
MIN(cpu) AS cpu_min,
MAX(cpu) AS cpu_max,
AVG(cpu) AS cpu_avg,
MIN(net_in) AS net_in_min,
MAX(net_in) AS net_in_max,
AVG(net_in) AS net_in_avg,
MIN(net_out) AS net_out_min,
MAX(net_out) AS net_out_max,
AVG(net_out) AS net_out_avg
FROM (
SELECT machine_name,
COALESCE(cpu_user, 0.0) AS cpu,
COALESCE(bytes_in, 0.0) AS net_in,
COALESCE(bytes_out, 0.0) AS net_out
FROM logs1
WHERE machine_name IN ('anansi','aragog','urd')
AND log_time >= TIMESTAMP '2017-01-11 00:00:00'
) AS r
GROUP BY machine_name;
-- Q1.2: Which computer lab machines have been offline in the past day?
SELECT machine_name,
log_time
FROM logs1
WHERE (machine_name LIKE 'cslab%' OR
machine_name LIKE 'mslab%')
AND load_one IS NULL
AND log_time >= TIMESTAMP '2017-01-10 00:00:00'
ORDER BY machine_name,
log_time;
-- Q1.3: What are the hourly average metrics during the past 10 days for a specific workstation?
SELECT dt,
hr,
AVG(load_fifteen) AS load_fifteen_avg,
AVG(load_five) AS load_five_avg,
AVG(load_one) AS load_one_avg,
AVG(mem_free) AS mem_free_avg,
AVG(swap_free) AS swap_free_avg
FROM (
SELECT CAST(log_time AS DATE) AS dt,
EXTRACT(HOUR FROM log_time) AS hr,
load_fifteen,
load_five,
load_one,
mem_free,
swap_free
FROM logs1
WHERE machine_name = 'babbage'
AND load_fifteen IS NOT NULL
AND load_five IS NOT NULL
AND load_one IS NOT NULL
AND mem_free IS NOT NULL
AND swap_free IS NOT NULL
AND log_time >= TIMESTAMP '2017-01-01 00:00:00'
) AS r
GROUP BY dt,
hr
ORDER BY dt,
hr;
-- Q1.4: Over 1 month, how often was each server blocked on disk I/O?
SELECT machine_name,
COUNT(*) AS spikes
FROM logs1
WHERE machine_group = 'Servers'
AND cpu_wio > 0.99
AND log_time >= TIMESTAMP '2016-12-01 00:00:00'
AND log_time < TIMESTAMP '2017-01-01 00:00:00'
GROUP BY machine_name
ORDER BY spikes DESC
LIMIT 10;
-- Q1.5: Which externally reachable VMs have run low on memory?
SELECT machine_name,
dt,
MIN(mem_free) AS mem_free_min
FROM (
SELECT machine_name,
CAST(log_time AS DATE) AS dt,
mem_free
FROM logs1
WHERE machine_group = 'DMZ'
AND mem_free IS NOT NULL
) AS r
GROUP BY machine_name,
dt
HAVING MIN(mem_free) < 10000
ORDER BY machine_name,
dt;
-- Q1.6: What is the total hourly network traffic across all file servers?
SELECT dt,
hr,
SUM(net_in) AS net_in_sum,
SUM(net_out) AS net_out_sum,
SUM(net_in) + SUM(net_out) AS both_sum
FROM (
SELECT CAST(log_time AS DATE) AS dt,
EXTRACT(HOUR FROM log_time) AS hr,
COALESCE(bytes_in, 0.0) / 1000000000.0 AS net_in,
COALESCE(bytes_out, 0.0) / 1000000000.0 AS net_out
FROM logs1
WHERE machine_name IN ('allsorts','andes','bigred','blackjack','bonbon',
'cadbury','chiclets','cotton','crows','dove','fireball','hearts','huey',
'lindt','milkduds','milkyway','mnm','necco','nerds','orbit','peeps',
'poprocks','razzles','runts','smarties','smuggler','spree','stride',
'tootsie','trident','wrigley','york')
) AS r
GROUP BY dt,
hr
ORDER BY both_sum DESC
LIMIT 10;
-- Q2.1: Which requests have caused server errors within the past 2 weeks?
SELECT *
FROM logs2
WHERE status_code >= 500
AND log_time >= TIMESTAMP '2012-12-18 00:00:00'
ORDER BY log_time;
-- Q2.2: During a specific 2-week period, was the user password file leaked?
SELECT *
FROM logs2
WHERE status_code >= 200
AND status_code < 300
AND request LIKE '%/etc/passwd%'
AND log_time >= TIMESTAMP '2012-05-06 00:00:00'
AND log_time < TIMESTAMP '2012-05-20 00:00:00';
-- Q2.3: What was the average path depth for top-level requests in the past month?
SELECT top_level,
AVG(LENGTH(request) - LENGTH(REPLACE(request, '/', ''))) AS depth_avg
FROM (
SELECT SUBSTRING(request FROM 1 FOR len) AS top_level,
request
FROM (
SELECT POSITION(SUBSTRING(request FROM 2), '/') AS len,
request
FROM logs2
WHERE status_code >= 200
AND status_code < 300
AND log_time >= TIMESTAMP '2012-12-01 00:00:00'
) AS r
WHERE len > 0
) AS s
WHERE top_level IN ('/about','/courses','/degrees','/events',
'/grad','/industry','/news','/people',
'/publications','/research','/teaching','/ugrad')
GROUP BY top_level
ORDER BY top_level;
-- Q2.4: During the last 3 months, which clients have made an excessive number of requests?
SELECT client_ip,
COUNT(*) AS num_requests
FROM logs2
WHERE log_time >= TIMESTAMP '2012-10-01 00:00:00'
GROUP BY client_ip
HAVING COUNT(*) >= 100000
ORDER BY num_requests DESC;
-- Q2.5: What are the daily unique visitors?
SELECT dt,
COUNT(DISTINCT client_ip)
FROM (
SELECT CAST(log_time AS DATE) AS dt,
client_ip
FROM logs2
) AS r
GROUP BY dt
ORDER BY dt;
-- Q2.6: What are the average and maximum data transfer rates (Gbps)?
SELECT AVG(transfer) / 125000000.0 AS transfer_avg,
MAX(transfer) / 125000000.0 AS transfer_max
FROM (
SELECT log_time,
SUM(object_size) AS transfer
FROM logs2
GROUP BY log_time
) AS r;
-- Q3.1: Did the indoor temperature reach freezing over the weekend?
SELECT *
FROM logs3
WHERE event_type = 'temperature'
AND event_value <= 32.0
AND log_time >= '2019-11-29 17:00:00.000';
-- Q3.4: Over the past 6 months, how frequently were each door opened?
SELECT device_name,
device_floor,
COUNT(*) AS ct
FROM logs3
WHERE event_type = 'door_open'
AND log_time >= '2019-06-01 00:00:00.000'
GROUP BY device_name,
device_floor
ORDER BY ct DESC;
-- Q3.5: Where in the building do large temperature variations occur in winter and summer?
WITH temperature AS (
SELECT dt,
device_name,
device_type,
device_floor
FROM (
SELECT dt,
hr,
device_name,
device_type,
device_floor,
AVG(event_value) AS temperature_hourly_avg
FROM (
SELECT CAST(log_time AS DATE) AS dt,
EXTRACT(HOUR FROM log_time) AS hr,
device_name,
device_type,
device_floor,
event_value
FROM logs3
WHERE event_type = 'temperature'
) AS r
GROUP BY dt,
hr,
device_name,
device_type,
device_floor
) AS s
GROUP BY dt,
device_name,
device_type,
device_floor
HAVING MAX(temperature_hourly_avg) - MIN(temperature_hourly_avg) >= 25.0
)
SELECT DISTINCT device_name,
device_type,
device_floor,
'WINTER'
FROM temperature
WHERE dt >= DATE '2018-12-01'
AND dt < DATE '2019-03-01'
UNION
SELECT DISTINCT device_name,
device_type,
device_floor,
'SUMMER'
FROM temperature
WHERE dt >= DATE '2019-06-01'
AND dt < DATE '2019-09-01';
-- Q3.6: For each device category, what are the monthly power consumption metrics?
SELECT yr,
mo,
SUM(coffee_hourly_avg) AS coffee_monthly_sum,
AVG(coffee_hourly_avg) AS coffee_monthly_avg,
SUM(printer_hourly_avg) AS printer_monthly_sum,
AVG(printer_hourly_avg) AS printer_monthly_avg,
SUM(projector_hourly_avg) AS projector_monthly_sum,
AVG(projector_hourly_avg) AS projector_monthly_avg,
SUM(vending_hourly_avg) AS vending_monthly_sum,
AVG(vending_hourly_avg) AS vending_monthly_avg
FROM (
SELECT dt,
yr,
mo,
hr,
AVG(coffee) AS coffee_hourly_avg,
AVG(printer) AS printer_hourly_avg,
AVG(projector) AS projector_hourly_avg,
AVG(vending) AS vending_hourly_avg
FROM (
SELECT CAST(log_time AS DATE) AS dt,
EXTRACT(YEAR FROM log_time) AS yr,
EXTRACT(MONTH FROM log_time) AS mo,
EXTRACT(HOUR FROM log_time) AS hr,
CASE WHEN device_name LIKE 'coffee%' THEN event_value END AS coffee,
CASE WHEN device_name LIKE 'printer%' THEN event_value END AS printer,
CASE WHEN device_name LIKE 'projector%' THEN event_value END AS projector,
CASE WHEN device_name LIKE 'vending%' THEN event_value END AS vending
FROM logs3
WHERE device_type = 'meter'
) AS r
GROUP BY dt,
yr,
mo,
hr
) AS s
GROUP BY yr,
mo
ORDER BY yr,
mo;
```
Данные также доступны для работы с интерактивными запросами через [Playground](https://gh-api.clickhouse.tech/play?user=play), [пример](https://gh-api.clickhouse.tech/play?user=play#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).
[Оригинальная статья](https://clickhouse.tech/docs/ru/getting_started/example_datasets/brown-benchmark/) <!--hide-->

44
docs/ru/interfaces/http.md
View File

@ -149,28 +149,48 @@ $ echo 'DROP TABLE t' | curl 'http://localhost:8123/' --data-binary @-
Для запросов, которые не возвращают таблицу с данными, в случае успеха, выдаётся пустое тело ответа.
Вы можете использовать внутренний формат сжатия Clickhouse при передаче данных. Формат сжатых данных нестандартный, и вам придётся использовать для работы с ним специальную программу `clickhouse-compressor` (устанавливается вместе с пакетом `clickhouse-client`). Для повышения эффективности вставки данных можно отключить проверку контрольной суммы на стороне сервера с помощью настройки[http_native_compression_disable_checksumming_on_decompress](../operations/settings/settings.md#settings-http_native_compression_disable_checksumming_on_decompress).
Если вы указали `compress = 1` в URL, то сервер сжимает данные, которые он отправляет.
Если вы указали `decompress = 1` в URL, сервер распаковывает те данные, которые вы передаёте методом `POST`.
## Сжатие {#compression}
Также, можно использовать [HTTP compression](https://en.wikipedia.org/wiki/HTTP_compression). Для отправки сжатого запроса `POST`, добавьте заголовок `Content-Encoding: compression_method`. Чтобы ClickHouse сжимал ответ, добавьте заголовок `Accept-Encoding: compression_method`. ClickHouse поддерживает следующие [методы сжатия](https://en.wikipedia.org/wiki/HTTP_compression#Content-Encoding_tokens): `gzip`, `br`, and `deflate`. Чтобы включить HTTP compression, используйте настройку ClickHouse [enable_http_compression](../operations/settings/settings.md#settings-enable_http_compression). Уровень сжатия данных для всех методов сжатия можно настроить с помощью настройки [http_zlib_compression_level](#settings-http_zlib_compression_level).
Сжатие можно использовать для уменьшения трафика по сети при передаче большого количества данных, а также для создания сразу сжатых дампов.
Это может быть использовано для уменьшения трафика по сети при передаче большого количества данных, а также для создания сразу сжатых дампов.
Вы можете использовать внутренний формат сжатия Clickhouse при передаче данных. Формат сжатых данных нестандартный, и вам придётся использовать для работы с ним специальную программу `clickhouse-compressor`. Она устанавливается вместе с пакетом `clickhouse-client`. Для повышения эффективности вставки данных можно отключить проверку контрольной суммы на стороне сервера с помощью настройки [http_native_compression_disable_checksumming_on_decompress](../operations/settings/settings.md#settings-http_native_compression_disable_checksumming_on_decompress).
Примеры отправки данных со сжатием:
Если вы указали `compress=1` в URL, то сервер сжимает данные, которые он отправляет. Если вы указали `decompress=1` в URL, сервер распаковывает те данные, которые вы передаёте методом `POST`.
``` bash
$ #Отправка данных на сервер:
$ curl -vsS "http://localhost:8123/?enable_http_compression=1" -d 'SELECT number FROM system.numbers LIMIT 10' -H 'Accept-Encoding: gzip'
Также можно использовать [сжатие HTTP](https://en.wikipedia.org/wiki/HTTP_compression). ClickHouse поддерживает следующие [методы сжатия](https://en.wikipedia.org/wiki/HTTP_compression#Content-Encoding_tokens):
$ #Отправка данных клиенту:
$ echo "SELECT 1" | gzip -c | curl -sS --data-binary @- -H 'Content-Encoding: gzip' 'http://localhost:8123/'
```
- `gzip`
- `br`
- `deflate`
- `xz`
Для отправки сжатого запроса `POST`, добавьте заголовок `Content-Encoding: compression_method`.
Чтобы ClickHouse сжимал ответ, разрешите сжатие настройкой [enable_http_compression](../operations/settings/settings.md#settings-enable_http_compression) и добавьте заголовок `Accept-Encoding: compression_method`. Уровень сжатия данных для всех методов сжатия можно задать с помощью настройки [http_zlib_compression_level](../operations/settings/settings.md#settings-http_zlib_compression_level).
!!! note "Примечание"
Некоторые HTTP-клиенты могут по умолчанию распаковывать данные (`gzip` и `deflate`) с сервера в фоновом режиме и вы можете получить распакованные данные, даже если правильно используете настройки сжатия.
**Примеры**
``` bash
# Отправка сжатых данных на сервер
$ echo "SELECT 1" | gzip -c | \
curl -sS --data-binary @- -H 'Content-Encoding: gzip' 'http://localhost:8123/'
```
``` bash
# Получение сжатых данных с сервера
$ curl -vsS "http://localhost:8123/?enable_http_compression=1" \
-H 'Accept-Encoding: gzip' --output result.gz -d 'SELECT number FROM system.numbers LIMIT 3'
$ zcat result.gz
0
1
2
```
## База данных по умолчанию {#default-database}
Вы можете использовать параметр URL `database` или заголовок `X-ClickHouse-Database`, чтобы указать БД по умолчанию.
``` bash

29
docs/ru/operations/caches.md Normal file
View File

@ -0,0 +1,29 @@
---
toc_priority: 65
toc_title: Кеши
---
# Типы кеша {#cache-types}
При выполнении запросов ClickHouse использует различные типы кеша.
Основные типы кеша:
- `mark_cache` — кеш засечек, используемых движками таблиц семейства [MergeTree](../engines/table-engines/mergetree-family/mergetree.md).
- `uncompressed_cache` — кеш несжатых данных, используемых движками таблиц семейства [MergeTree](../engines/table-engines/mergetree-family/mergetree.md).
Дополнительные типы кеша:
- DNS-кеш.
- Кеш данных формата [regexp](../interfaces/formats.md#data-format-regexp).
- Кеш скомпилированных выражений.
- Кеш схем формата [Avro](../interfaces/formats.md#data-format-avro).
- Кеш данных в [словарях](../sql-reference/dictionaries/index.md).
Непрямое использование:
- Кеш страницы ОС.
Чтобы очистить кеш, используйте выражение [SYSTEM DROP ... CACHE](../sql-reference/statements/system.md).
[Оригинальная статья](https://clickhouse.tech/docs/ru/operations/caches/) <!--hide-->

2
docs/ru/operations/system-tables/distributed_ddl_queue.md
View File

@ -14,7 +14,7 @@
- `initiator` ([String](../../sql-reference/data-types/string.md)) — узел, выполнивший запрос.
- `query_start_time` ([DateTime](../../sql-reference/data-types/datetime.md)) — время начала запроса.
- `query_finish_time` ([DateTime](../../sql-reference/data-types/datetime.md)) — время окончания запроса.
- `query_duration_ms` ([UInt64](../../sql-reference/data-types/datetime64.md)) — продолжительность выполнения запроса (в миллисекундах).
- `query_duration_ms` ([UInt64](../../sql-reference/data-types/int-uint.md)) — продолжительность выполнения запроса (в миллисекундах).
- `exception_code` ([Enum8](../../sql-reference/data-types/enum.md)) — код исключения из [ZooKeeper](../../operations/tips.md#zookeeper).
**Пример**

1
docs/ru/sql-reference/aggregate-functions/reference/mannwhitneyutest.md
View File

@ -31,6 +31,7 @@ mannWhitneyUTest[(alternative[, continuity_correction])](sample_data, sample_ind
**Возвращаемые значения**
[Кортеж](../../../sql-reference/data-types/tuple.md) с двумя элементами:
- вычисленное значение критерия Манна — Уитни. [Float64](../../../sql-reference/data-types/float.md).
- вычисленное p-значение. [Float64](../../../sql-reference/data-types/float.md).

1
docs/ru/sql-reference/aggregate-functions/reference/studentttest.md
View File

@ -24,6 +24,7 @@ studentTTest(sample_data, sample_index)
**Возвращаемые значения**
[Кортеж](../../../sql-reference/data-types/tuple.md) с двумя элементами:
- вычисленное значение критерия Стьюдента. [Float64](../../../sql-reference/data-types/float.md).
- вычисленное p-значение. [Float64](../../../sql-reference/data-types/float.md).

1
docs/ru/sql-reference/aggregate-functions/reference/welchttest.md
View File

@ -24,6 +24,7 @@ welchTTest(sample_data, sample_index)
**Возвращаемые значения**
[Кортеж](../../../sql-reference/data-types/tuple.md) с двумя элементами:
- вычисленное значение критерия Уэлча. [Float64](../../../sql-reference/data-types/float.md).
- вычисленное p-значение. [Float64](../../../sql-reference/data-types/float.md).

2
docs/ru/sql-reference/data-types/lowcardinality.md
View File

@ -23,7 +23,7 @@ LowCardinality(data_type)
Эффективность использования типа данных `LowCarditality` зависит от разнообразия данных. Если словарь содержит менее 10 000 различных значений, ClickHouse в основном показывает более высокую эффективность чтения и хранения данных. Если же словарь содержит более 100 000 различных значений, ClickHouse может работать хуже, чем при использовании обычных типов данных.
При работе со строками, использование `LowCardinality` вместо [Enum](enum.md). `LowCardinality` обеспечивает большую гибкость в использовании и часто показывает такую же или более высокую эффективность.
При работе со строками, использование `LowCardinality` вместо [Enum](enum.md) обеспечивает большую гибкость в использовании и часто показывает такую же или более высокую эффективность.
## Пример

46
docs/ru/sql-reference/functions/array-functions.md
View File

@ -1355,6 +1355,52 @@ SELECT arrayAvg(x -> (x * x), [2, 4]) AS res;
└─────┘
```
**Синтаксис**
``` sql
arraySum(arr)
```
**Возвращаемое значение**
- Число.
Тип: [Int](../../sql-reference/data-types/int-uint.md) или [Float](../../sql-reference/data-types/float.md).
**Параметры**
- `arr` — [Массив](../../sql-reference/data-types/array.md).
**Примеры**
Запрос:
```sql
SELECT arraySum([2,3]) AS res;
```
Результат:
``` text
┌─res─┐
│ 5 │
└─────┘
```
Запрос:
``` sql
SELECT arraySum(x -> x*x, [2, 3]) AS res;
```
Результат:
``` text
┌─res─┐
│ 13 │
└─────┘
```
## arrayCumSum(\[func,\] arr1, …) {#arraycumsumfunc-arr1}
Возвращает массив из частичных сумм элементов исходного массива (сумма с накоплением). Если указана функция `func`, то значения элементов массива преобразуются этой функцией перед суммированием.

18
docs/ru/sql-reference/functions/date-time-functions.md
View File

@ -63,40 +63,58 @@ int32samoa: 1546300800
Переводит дату или дату-с-временем в число типа UInt16, содержащее номер года (AD).
Синоним: `YEAR`.
## toQuarter {#toquarter}
Переводит дату или дату-с-временем в число типа UInt8, содержащее номер квартала.
Синоним: `QUARTER`.
## toMonth {#tomonth}
Переводит дату или дату-с-временем в число типа UInt8, содержащее номер месяца (1-12).
Синоним: `MONTH`.
## toDayOfYear {#todayofyear}
Переводит дату или дату-с-временем в число типа UInt16, содержащее номер дня года (1-366).
Синоним: `DAYOFYEAR`.
## toDayOfMonth {#todayofmonth}
Переводит дату или дату-с-временем в число типа UInt8, содержащее номер дня в месяце (1-31).
Синонимы: `DAYOFMONTH`, `DAY`.
## toDayOfWeek {#todayofweek}
Переводит дату или дату-с-временем в число типа UInt8, содержащее номер дня в неделе (понедельник - 1, воскресенье - 7).
Синоним: `DAYOFWEEK`.
## toHour {#tohour}
Переводит дату-с-временем в число типа UInt8, содержащее номер часа в сутках (0-23).
Функция исходит из допущения, что перевод стрелок вперёд, если осуществляется, то на час, в два часа ночи, а перевод стрелок назад, если осуществляется, то на час, в три часа ночи (что, в общем, не верно - даже в Москве два раза перевод стрелок был осуществлён в другое время).
Синоним: `HOUR`.
## toMinute {#tominute}
Переводит дату-с-временем в число типа UInt8, содержащее номер минуты в часе (0-59).
Синоним: `MINUTE`.
## toSecond {#tosecond}
Переводит дату-с-временем в число типа UInt8, содержащее номер секунды в минуте (0-59).
Секунды координации не учитываются.
Синоним: `SECOND`.
## toUnixTimestamp {#to-unix-timestamp}
Переводит дату-с-временем в число типа UInt32 -- Unix Timestamp (https://en.wikipedia.org/wiki/Unix_time).

2
docs/ru/sql-reference/functions/encoding-functions.md
View File

@ -75,6 +75,8 @@ SELECT char(0xE4, 0xBD, 0xA0, 0xE5, 0xA5, 0xBD) AS hello;
Returns a string containing the argument’s hexadecimal representation.
Синоним: `HEX`.
**Syntax**
``` sql

2
docs/ru/sql-reference/functions/functions-for-nulls.md
View File

@ -13,6 +13,8 @@ toc_title: "\u0424\u0443\u043d\u043a\u0446\u0438\u0438\u0020\u0434\u043b\u044f\u
isNull(x)
```
Синоним: `ISNULL`.
**Параметры**
- `x` — значение с не составным типом данных.

12
docs/ru/sql-reference/functions/ip-address-functions.md
View File

@ -9,10 +9,14 @@ toc_title: "\u0424\u0443\u043d\u043a\u0446\u0438\u0438\u0020\u0434\u043b\u044f\u
Принимает число типа UInt32. Интерпретирует его, как IPv4-адрес в big endian. Возвращает строку, содержащую соответствующий IPv4-адрес в формате A.B.C.D (числа в десятичной форме через точки).
Синоним: `INET_NTOA`.
## IPv4StringToNum(s) {#ipv4stringtonums}
Функция, обратная к IPv4NumToString. Если IPv4 адрес в неправильном формате, то возвращает 0.
Синоним: `INET_ATON`.
## IPv4NumToStringClassC(num) {#ipv4numtostringclasscnum}
Похоже на IPv4NumToString, но вместо последнего октета используется xxx.
@ -49,7 +53,11 @@ LIMIT 10
### IPv6NumToString(x) {#ipv6numtostringx}
Принимает значение типа FixedString(16), содержащее IPv6-адрес в бинарном виде. Возвращает строку, содержащую этот адрес в текстовом виде.
IPv6-mapped IPv4 адреса выводится в формате ::ffff:111.222.33.44. Примеры:
IPv6-mapped IPv4 адреса выводится в формате ::ffff:111.222.33.44.
Примеры: `INET6_NTOA`.
Примеры:
``` sql
SELECT IPv6NumToString(toFixedString(unhex('2A0206B8000000000000000000000011'), 16)) AS addr
@ -118,6 +126,8 @@ LIMIT 10
Функция, обратная к IPv6NumToString. Если IPv6 адрес в неправильном формате, то возвращает строку из нулевых байт.
HEX может быть в любом регистре.
Alias: `INET6_ATON`.
## IPv4ToIPv6(x) {#ipv4toipv6x}
Принимает число типа `UInt32`. Интерпретирует его, как IPv4-адрес в [big endian](https://en.wikipedia.org/wiki/Endianness). Возвращает значение `FixedString(16)`, содержащее адрес IPv6 в двоичном формате. Примеры:

48
docs/ru/sql-reference/functions/string-functions.md
View File

@ -95,6 +95,8 @@ SELECT toValidUTF8('\x61\xF0\x80\x80\x80b')
Повторяет строку определенное количество раз и объединяет повторяемые значения в одну строку.
Синоним: `REPEAT`.
**Синтаксис**
``` sql
@ -273,10 +275,14 @@ SELECT concat(key1, key2), sum(value) FROM key_val GROUP BY (key1, key2)
Производит кодирование строки s в base64-представление.
Синоним: `TO_BASE64`.
## base64Decode(s) {#base64decode}
Декодирует base64-представление s в исходную строку. При невозможности декодирования выбрасывает исключение
Синоним: `FROM_BASE64`.
## tryBase64Decode(s) {#trybase64decode}
Функционал аналогичен base64Decode, но при невозможности декодирования возвращает пустую строку.
@ -597,4 +603,46 @@ Hello, &quot;world&quot;!
&apos;foo&apos;
```
## decodeXMLComponent {#decode-xml-component}
Заменяет символами предопределенные мнемоники XML: `&quot;` `&amp;` `&apos;` `&gt;` `&lt;`
Также эта функция заменяет числовые ссылки соответствующими символами юникод. Поддерживаются десятичная (например, `&#10003;`) и шестнадцатеричная (`&#x2713;`) формы.
**Синтаксис**
``` sql
decodeXMLComponent(x)
```
**Параметры**
- `x` — последовательность символов. [String](../../sql-reference/data-types/string.md).
**Возвращаемое значение**
- Строка с произведенными заменами.
Тип: [String](../../sql-reference/data-types/string.md).
**Пример**
Запрос:
``` sql
SELECT decodeXMLComponent('&apos;foo&apos;');
SELECT decodeXMLComponent('&lt; &#x3A3; &gt;');
```
Результат:
``` text
'foo'
< Σ >
```
**Смотрите также**
- [Мнемоники в HTML](https://ru.wikipedia.org/wiki/%D0%9C%D0%BD%D0%B5%D0%BC%D0%BE%D0%BD%D0%B8%D0%BA%D0%B8_%D0%B2_HTML)
[Оригинальная статья](https://clickhouse.tech/docs/ru/query_language/functions/string_functions/) <!--hide-->

412
docs/ru/sql-reference/functions/type-conversion-functions.md
View File

@ -36,10 +36,14 @@ toc_title: "\u0424\u0443\u043d\u043a\u0446\u0438\u0438\u0020\u043f\u0440\u0435\u
**Пример**
Запрос:
``` sql
SELECT toInt64(nan), toInt32(32), toInt16('16'), toInt8(8.8)
SELECT toInt64(nan), toInt32(32), toInt16('16'), toInt8(8.8);
```
Результат:
``` text
┌─────────toInt64(nan)─┬─toInt32(32)─┬─toInt16('16')─┬─toInt8(8.8)─┐
│ -9223372036854775808 │ 32 │ 16 │ 8 │
@ -52,10 +56,14 @@ SELECT toInt64(nan), toInt32(32), toInt16('16'), toInt8(8.8)
**Пример**
Запрос:
``` sql
select toInt64OrZero('123123'), toInt8OrZero('123qwe123')
SELECT toInt64OrZero('123123'), toInt8OrZero('123qwe123');
```
Результат:
``` text
┌─toInt64OrZero('123123')─┬─toInt8OrZero('123qwe123')─┐
│ 123123 │ 0 │
@ -68,10 +76,14 @@ select toInt64OrZero('123123'), toInt8OrZero('123qwe123')
**Пример**
Запрос:
``` sql
select toInt64OrNull('123123'), toInt8OrNull('123qwe123')
SELECT toInt64OrNull('123123'), toInt8OrNull('123qwe123');
```
Результат:
``` text
┌─toInt64OrNull('123123')─┬─toInt8OrNull('123qwe123')─┐
│ 123123 │ ᴺᵁᴸᴸ │
@ -102,10 +114,14 @@ select toInt64OrNull('123123'), toInt8OrNull('123qwe123')
**Пример**
Запрос:
``` sql
SELECT toUInt64(nan), toUInt32(-32), toUInt16('16'), toUInt8(8.8)
SELECT toUInt64(nan), toUInt32(-32), toUInt16('16'), toUInt8(8.8);
```
Результат:
``` text
┌───────toUInt64(nan)─┬─toUInt32(-32)─┬─toUInt16('16')─┬─toUInt8(8.8)─┐
│ 9223372036854775808 │ 4294967264 │ 16 │ 8 │
@ -124,6 +140,8 @@ SELECT toUInt64(nan), toUInt32(-32), toUInt16('16'), toUInt8(8.8)
## toDate {#todate}
Cиноним: `DATE`.
## toDateOrZero {#todateorzero}
## toDateOrNull {#todateornull}
@ -168,20 +186,28 @@ SELECT toUInt64(nan), toUInt32(-32), toUInt16('16'), toUInt8(8.8)
**Примеры**
Запрос:
``` sql
SELECT toDecimal32OrNull(toString(-1.111), 5) AS val, toTypeName(val)
SELECT toDecimal32OrNull(toString(-1.111), 5) AS val, toTypeName(val);
```
Результат:
``` text
┌──────val─┬─toTypeName(toDecimal32OrNull(toString(-1.111), 5))─┐
│ -1.11100 │ Nullable(Decimal(9, 5)) │
└──────────┴────────────────────────────────────────────────────┘
```
Запрос:
``` sql
SELECT toDecimal32OrNull(toString(-1.111), 2) AS val, toTypeName(val)
SELECT toDecimal32OrNull(toString(-1.111), 2) AS val, toTypeName(val);
```
Результат:
``` text
┌──val─┬─toTypeName(toDecimal32OrNull(toString(-1.111), 2))─┐
│ ᴺᵁᴸᴸ │ Nullable(Decimal(9, 2)) │
@ -213,20 +239,28 @@ SELECT toDecimal32OrNull(toString(-1.111), 2) AS val, toTypeName(val)
**Пример**
Запрос:
``` sql
SELECT toDecimal32OrZero(toString(-1.111), 5) AS val, toTypeName(val)
SELECT toDecimal32OrZero(toString(-1.111), 5) AS val, toTypeName(val);
```
Результат:
``` text
┌──────val─┬─toTypeName(toDecimal32OrZero(toString(-1.111), 5))─┐
│ -1.11100 │ Decimal(9, 5) │
└──────────┴────────────────────────────────────────────────────┘
```
Запрос:
``` sql
SELECT toDecimal32OrZero(toString(-1.111), 2) AS val, toTypeName(val)
SELECT toDecimal32OrZero(toString(-1.111), 2) AS val, toTypeName(val);
```
Результат:
``` text
┌──val─┬─toTypeName(toDecimal32OrZero(toString(-1.111), 2))─┐
│ 0.00 │ Decimal(9, 2) │
@ -258,12 +292,18 @@ YYYY-MM-DD hh:mm:ss
Дополнительно, функция toString от аргумента типа DateTime может принимать второй аргумент String - имя тайм-зоны. Пример: `Asia/Yekaterinburg` В этом случае, форматирование времени производится согласно указанной тайм-зоне.
**Пример**
Запрос:
``` sql
SELECT
now() AS now_local,
toString(now(), 'Asia/Yekaterinburg') AS now_yekat
toString(now(), 'Asia/Yekaterinburg') AS now_yekat;
```
Результат:
``` text
┌───────────now_local─┬─now_yekat───────────┐
│ 2016-06-15 00:11:21 │ 2016-06-15 02:11:21 │
@ -281,22 +321,30 @@ SELECT
Принимает аргумент типа String или FixedString. Возвращает String, вырезая содержимое строки до первого найденного нулевого байта.
Пример:
**Примеры**
Запрос:
``` sql
SELECT toFixedString('foo', 8) AS s, toStringCutToZero(s) AS s_cut
SELECT toFixedString('foo', 8) AS s, toStringCutToZero(s) AS s_cut;
```
Результат:
``` text
┌─s─────────────┬─s_cut─┐
│ foo\0\0\0\0\0 │ foo │
└───────────────┴───────┘
```
Запрос:
``` sql
SELECT toFixedString('foo\0bar', 8) AS s, toStringCutToZero(s) AS s_cut
SELECT toFixedString('foo\0bar', 8) AS s, toStringCutToZero(s) AS s_cut;
```
Результат:
``` text
┌─s──────────┬─s_cut─┐
│ foo\0bar\0 │ foo │
@ -344,7 +392,7 @@ reinterpretAsUUID(fixed_string)
Запрос:
``` sql
SELECT reinterpretAsUUID(reverse(unhex('000102030405060708090a0b0c0d0e0f')))
SELECT reinterpretAsUUID(reverse(unhex('000102030405060708090a0b0c0d0e0f')));
```
Результат:
@ -377,10 +425,15 @@ SELECT uuid = uuid2;
## CAST(x, T) {#type_conversion_function-cast}
Преобразует x в тип данных t.
Поддерживается также синтаксис CAST(x AS t).
Преобразует входное значение `x` в указанный тип данных `T`.
Пример:
Поддерживается также синтаксис `CAST(x AS t)`.
Обратите внимание, что если значение `x` не может быть преобразовано к типу `T`, возникает переполнение. Например, `CAST(-1, 'UInt8')` возвращает 255.
**Пример**
Запрос:
``` sql
SELECT
@ -388,9 +441,11 @@ SELECT
CAST(timestamp AS DateTime) AS datetime,
CAST(timestamp AS Date) AS date,
CAST(timestamp, 'String') AS string,
CAST(timestamp, 'FixedString(22)') AS fixed_string
CAST(timestamp, 'FixedString(22)') AS fixed_string;
```
Результат:
``` text
┌─timestamp───────────┬────────────datetime─┬───────date─┬─string──────────────┬─fixed_string──────────────┐
│ 2016-06-15 23:00:00 │ 2016-06-15 23:00:00 │ 2016-06-15 │ 2016-06-15 23:00:00 │ 2016-06-15 23:00:00\0\0\0 │
@ -399,12 +454,18 @@ SELECT
Преобразование в FixedString(N) работает только для аргументов типа String или FixedString(N).
Поддержано преобразование к типу [Nullable](../../sql-reference/functions/type-conversion-functions.md) и обратно. Пример:
Поддержано преобразование к типу [Nullable](../../sql-reference/functions/type-conversion-functions.md) и обратно.
**Примеры**
Запрос:
``` sql
SELECT toTypeName(x) FROM t_null
SELECT toTypeName(x) FROM t_null;
```
Результат:
``` text
┌─toTypeName(x)─┐
│ Int8 │
@ -412,10 +473,14 @@ SELECT toTypeName(x) FROM t_null
└───────────────┘
```
Запрос:
``` sql
SELECT toTypeName(CAST(x, 'Nullable(UInt16)')) FROM t_null
SELECT toTypeName(CAST(x, 'Nullable(UInt16)')) FROM t_null;
```
Результат:
``` text
┌─toTypeName(CAST(x, 'Nullable(UInt16)'))─┐
│ Nullable(UInt16) │
@ -427,6 +492,93 @@ SELECT toTypeName(CAST(x, 'Nullable(UInt16)')) FROM t_null
- Настройка [cast_keep_nullable](../../operations/settings/settings.md#cast_keep_nullable)
## accurateCast(x, T) {#type_conversion_function-accurate-cast}
Преобразует входное значение `x` в указанный тип данных `T`.
В отличие от функции [cast(x, T)](#type_conversion_function-cast), `accurateCast` не допускает переполнения при преобразовании числовых типов. Например, `accurateCast(-1, 'UInt8')` вызовет исключение.
**Примеры**
Запрос:
``` sql
SELECT cast(-1, 'UInt8') as uint8;
```
Результат:
``` text
┌─uint8─┐
│ 255 │
└─────
Запрос:
```sql
SELECT accurateCast(-1, 'UInt8') as uint8;
```
Результат:
``` text
Code: 70. DB::Exception: Received from localhost:9000. DB::Exception: Value in column Int8 cannot be safely converted into type UInt8: While processing accurateCast(-1, 'UInt8') AS uint8.
```
## accurateCastOrNull(x, T) {#type_conversion_function-accurate-cast_or_null}
Преобразует входное значение `x` в указанный тип данных `T`.
Всегда возвращает тип [Nullable](../../sql-reference/data-types/nullable.md). Если исходное значение не может быть преобразовано к целевому типу, возвращает [NULL](../../sql-reference/syntax.md#null-literal).
**Синтаксис**
```sql
accurateCastOrNull(x, T)
```
**Параметры**
- `x` — входное значение.
- `T` — имя возвращаемого типа данных.
**Возвращаемое значение**
- Значение, преобразованное в указанный тип `T`.
**Примеры**
Запрос:
``` sql
SELECT toTypeName(accurateCastOrNull(5, 'UInt8'));
```
Результат:
``` text
┌─toTypeName(accurateCastOrNull(5, 'UInt8'))─┐
│ Nullable(UInt8) │
└────────────────────────────────────────────┘
```
Запрос:
``` sql
SELECT
accurateCastOrNull(-1, 'UInt8') as uint8,
accurateCastOrNull(128, 'Int8') as int8,
accurateCastOrNull('Test', 'FixedString(2)') as fixed_string;
```
Результат:
``` text
┌─uint8─┬─int8─┬─fixed_string─┐
│ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │
└───────┴──────┴──────────────┘
```
## toInterval(Year\|Quarter\|Month\|Week\|Day\|Hour\|Minute\|Second) {#function-tointerval}
Приводит аргумент из числового типа данных к типу данных [IntervalType](../../sql-reference/data-types/special-data-types/interval.md).
@ -454,6 +606,8 @@ toIntervalYear(number)
**Пример**
Запрос:
``` sql
WITH
toDate('2019-01-01') AS date,
@ -461,9 +615,11 @@ WITH
toIntervalWeek(1) AS interval_to_week
SELECT
date + interval_week,
date + interval_to_week
date + interval_to_week;
```
Результат:
``` text
┌─plus(date, interval_week)─┬─plus(date, interval_to_week)─┐
│ 2019-01-08 │ 2019-01-08 │
@ -479,7 +635,7 @@ SELECT
**Синтаксис**
``` sql
parseDateTimeBestEffort(time_string[, time_zone]);
parseDateTimeBestEffort(time_string[, time_zone])
```
**Параметры**
@ -522,7 +678,7 @@ AS parseDateTimeBestEffort;
``` sql
SELECT parseDateTimeBestEffort('Sat, 18 Aug 2018 07:22:16 GMT', 'Europe/Moscow')
AS parseDateTimeBestEffort
AS parseDateTimeBestEffort;
```
Результат:
@ -537,7 +693,7 @@ AS parseDateTimeBestEffort
``` sql
SELECT parseDateTimeBestEffort('1284101485')
AS parseDateTimeBestEffort
AS parseDateTimeBestEffort;
```
Результат:
@ -552,7 +708,7 @@ AS parseDateTimeBestEffort
``` sql
SELECT parseDateTimeBestEffort('2018-12-12 10:12:12')
AS parseDateTimeBestEffort
AS parseDateTimeBestEffort;
```
Результат:
@ -566,7 +722,7 @@ AS parseDateTimeBestEffort
Запрос:
``` sql
SELECT parseDateTimeBestEffort('10 20:19')
SELECT parseDateTimeBestEffort('10 20:19');
```
Результат:
@ -591,7 +747,7 @@ SELECT parseDateTimeBestEffort('10 20:19')
**Синтаксис**
``` sql
parseDateTimeBestEffortUS(time_string [, time_zone]);
parseDateTimeBestEffortUS(time_string [, time_zone])
```
**Параметры**
@ -620,7 +776,7 @@ SELECT parseDateTimeBestEffortUS('09/12/2020 12:12:57')
AS parseDateTimeBestEffortUS;
```
Ответ:
Результат:
``` text
┌─parseDateTimeBestEffortUS─┐
@ -635,7 +791,7 @@ SELECT parseDateTimeBestEffortUS('09-12-2020 12:12:57')
AS parseDateTimeBestEffortUS;
```
Ответ:
Результат:
``` text
┌─parseDateTimeBestEffortUS─┐
@ -650,7 +806,7 @@ SELECT parseDateTimeBestEffortUS('09.12.2020 12:12:57')
AS parseDateTimeBestEffortUS;
```
Ответ:
Результат:
``` text
┌─parseDateTimeBestEffortUS─┐
@ -658,6 +814,178 @@ AS parseDateTimeBestEffortUS;
└─────────────────────────——┘
```
## parseDateTimeBestEffortUSOrNull {#parsedatetimebesteffortusornull}
Работает аналогично функции [parseDateTimeBestEffortUS](#parsedatetimebesteffortUS), но в отличие от нее возвращает `NULL`, если входная строка не может быть преобразована в тип данных [DateTime](../../sql-reference/data-types/datetime.md).
**Синтаксис**
``` sql
parseDateTimeBestEffortUSOrNull(time_string[, time_zone])
```
**Параметры**
- `time_string` — строка, содержащая дату или дату со временем для преобразования. Дата должна быть в американском формате (`MM/DD/YYYY` и т.д.). [String](../../sql-reference/data-types/string.md).
- `time_zone` — [часовой пояс](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone). Функция анализирует `time_string` в соответствии с заданным часовым поясом. Опциональный параметр. [String](../../sql-reference/data-types/string.md).
**Поддерживаемые нестандартные форматы**
- Строка в формате [unix timestamp](https://en.wikipedia.org/wiki/Unix_time), содержащая 9-10 цифр.
- Строка, содержащая дату и время: `YYYYMMDDhhmmss`, `MM/DD/YYYY hh:mm:ss`, `MM-DD-YY hh:mm`, `YYYY-MM-DD hh:mm:ss` и т.д.
- Строка, содержащая дату без времени: `YYYY`, `YYYYMM`, `YYYY*MM`, `MM/DD/YYYY`, `MM-DD-YY` и т.д.
- Строка, содержащая день и время: `DD`, `DD hh`, `DD hh:mm`. В этом случае `YYYY-MM` заменяется на `2000-01`.
- Строка, содержащая дату и время, а также информацию о часовом поясе: `YYYY-MM-DD hh:mm:ss ±h:mm` и т.д. Например, `2020-12-12 17:36:00 -5:00`.
**Возвращаемые значения**
- `time_string`, преобразованная в тип данных `DateTime`.
- `NULL`, если входная строка не может быть преобразована в тип данных `DateTime`.
**Примеры**
Запрос:
``` sql
SELECT parseDateTimeBestEffortUSOrNull('02/10/2021 21:12:57') AS parseDateTimeBestEffortUSOrNull;
```
Результат:
``` text
┌─parseDateTimeBestEffortUSOrNull─┐
│ 2021-02-10 21:12:57 │
└─────────────────────────────────┘
```
Запрос:
``` sql
SELECT parseDateTimeBestEffortUSOrNull('02-10-2021 21:12:57 GMT', 'Europe/Moscow') AS parseDateTimeBestEffortUSOrNull;
```
Результат:
``` text
┌─parseDateTimeBestEffortUSOrNull─┐
│ 2021-02-11 00:12:57 │
└─────────────────────────────────┘
```
Запрос:
``` sql
SELECT parseDateTimeBestEffortUSOrNull('02.10.2021') AS parseDateTimeBestEffortUSOrNull;
```
Результат:
``` text
┌─parseDateTimeBestEffortUSOrNull─┐
│ 2021-02-10 00:00:00 │
└─────────────────────────────────┘
```
Запрос:
``` sql
SELECT parseDateTimeBestEffortUSOrNull('10.2021') AS parseDateTimeBestEffortUSOrNull;
```
Результат:
``` text
┌─parseDateTimeBestEffortUSOrNull─┐
│ ᴺᵁᴸᴸ │
└─────────────────────────────────┘
```
## parseDateTimeBestEffortUSOrZero {#parsedatetimebesteffortusorzero}
Работает аналогично функции [parseDateTimeBestEffortUS](#parsedatetimebesteffortUS), но в отличие от нее возвращает нулевую дату (`1970-01-01`) или нулевую дату со временем (`1970-01-01 00:00:00`), если входная строка не может быть преобразована в тип данных [DateTime](../../sql-reference/data-types/datetime.md).
**Синтаксис**
``` sql
parseDateTimeBestEffortUSOrZero(time_string[, time_zone])
```
**Параметры**
- `time_string` — строка, содержащая дату или дату со временем для преобразования. Дата должна быть в американском формате (`MM/DD/YYYY` и т.д.). [String](../../sql-reference/data-types/string.md).
- `time_zone` — [часовой пояс](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone). Функция анализирует `time_string` в соответствии с заданным часовым поясом. Опциональный параметр. [String](../../sql-reference/data-types/string.md).
**Поддерживаемые нестандартные форматы**
- Строка в формате [unix timestamp](https://en.wikipedia.org/wiki/Unix_time), содержащая 9-10 цифр.
- Строка, содержащая дату и время: `YYYYMMDDhhmmss`, `MM/DD/YYYY hh:mm:ss`, `MM-DD-YY hh:mm`, `YYYY-MM-DD hh:mm:ss` и т.д.
- Строка, содержащая дату без времени: `YYYY`, `YYYYMM`, `YYYY*MM`, `MM/DD/YYYY`, `MM-DD-YY` и т.д.
- Строка, содержащая день и время: `DD`, `DD hh`, `DD hh:mm`. В этом случае `YYYY-MM` заменяется на `2000-01`.
- Строка, содержащая дату и время, а также информацию о часовом поясе: `YYYY-MM-DD hh:mm:ss ±h:mm` и т.д. Например, `2020-12-12 17:36:00 -5:00`.
**Возвращаемые значения**
- `time_string`, преобразованная в тип данных `DateTime`.
- Нулевая дата или нулевая дата со временем, если входная строка не может быть преобразована в тип данных `DateTime`.
**Примеры**
Запрос:
``` sql
SELECT parseDateTimeBestEffortUSOrZero('02/10/2021 21:12:57') AS parseDateTimeBestEffortUSOrZero;
```
Результат:
``` text
┌─parseDateTimeBestEffortUSOrZero─┐
│ 2021-02-10 21:12:57 │
└─────────────────────────────────┘
```
Запрос:
``` sql
SELECT parseDateTimeBestEffortUSOrZero('02-10-2021 21:12:57 GMT', 'Europe/Moscow') AS parseDateTimeBestEffortUSOrZero;
```
Результат:
``` text
┌─parseDateTimeBestEffortUSOrZero─┐
│ 2021-02-11 00:12:57 │
└─────────────────────────────────┘
```
Запрос:
``` sql
SELECT parseDateTimeBestEffortUSOrZero('02.10.2021') AS parseDateTimeBestEffortUSOrZero;
```
Результат:
``` text
┌─parseDateTimeBestEffortUSOrZero─┐
│ 2021-02-10 00:00:00 │
└─────────────────────────────────┘
```
Запрос:
``` sql
SELECT parseDateTimeBestEffortUSOrZero('02.2021') AS parseDateTimeBestEffortUSOrZero;
```
Результат:
``` text
┌─parseDateTimeBestEffortUSOrZero─┐
│ 1970-01-01 00:00:00 │
└─────────────────────────────────┘
```
## toUnixTimestamp64Milli
## toUnixTimestamp64Micro
## toUnixTimestamp64Nano
@ -685,10 +1013,10 @@ toUnixTimestamp64Milli(value)
``` sql
WITH toDateTime64('2019-09-16 19:20:12.345678910', 6) AS dt64
SELECT toUnixTimestamp64Milli(dt64)
SELECT toUnixTimestamp64Milli(dt64);
```
Ответ:
Результат:
``` text
┌─toUnixTimestamp64Milli(dt64)─┐
@ -700,10 +1028,10 @@ SELECT toUnixTimestamp64Milli(dt64)
``` sql
WITH toDateTime64('2019-09-16 19:20:12.345678910', 6) AS dt64
SELECT toUnixTimestamp64Nano(dt64)
SELECT toUnixTimestamp64Nano(dt64);
```
Ответ:
Результат:
``` text
┌─toUnixTimestamp64Nano(dt64)─┐
@ -738,10 +1066,10 @@ fromUnixTimestamp64Milli(value [, ti])
``` sql
WITH CAST(1234567891011, 'Int64') AS i64
SELECT fromUnixTimestamp64Milli(i64, 'UTC')
SELECT fromUnixTimestamp64Milli(i64, 'UTC');
```
Ответ:
Результат:
``` text
┌─fromUnixTimestamp64Milli(i64, 'UTC')─┐
@ -772,12 +1100,12 @@ toLowCardinality(expr)
Тип: `LowCardinality(expr_result_type)`
**Example**
**Пример**
Запрос:
```sql
SELECT toLowCardinality('1')
SELECT toLowCardinality('1');
```
Результат:
@ -813,10 +1141,10 @@ formatRow(format, x, y, ...)
``` sql
SELECT formatRow('CSV', number, 'good')
FROM numbers(3)
FROM numbers(3);
```
Ответ:
Результат:
``` text
┌─formatRow('CSV', number, 'good')─┐
@ -854,10 +1182,10 @@ formatRowNoNewline(format, x, y, ...)
``` sql
SELECT formatRowNoNewline('CSV', number, 'good')
FROM numbers(3)
FROM numbers(3);
```
Ответ:
Результат:
``` text
┌─formatRowNoNewline('CSV', number, 'good')─┐

20
docs/ru/sql-reference/operators/in.md
View File

@ -13,10 +13,28 @@ SELECT (CounterID, UserID) IN ((34, 123), (101500, 456)) FROM ...
Если слева стоит один столбец, входящий в индекс, а справа - множество констант, то при выполнении запроса, система воспользуется индексом.
Не перечисляйте слишком большое количество значений (миллионы) явно. Если множество большое - лучше загрузить его во временную таблицу (например, смотрите раздел «Внешние данные для обработки запроса»), и затем воспользоваться подзапросом.
Не перечисляйте слишком большое количество значений (миллионы) явно. Если множество большое - лучше загрузить его во временную таблицу (например, смотрите раздел [Внешние данные для обработки запроса](../../engines/table-engines/special/external-data.md)), и затем воспользоваться подзапросом.
В качестве правой части оператора может быть множество константных выражений, множество кортежей с константными выражениями (показано в примерах выше), а также имя таблицы или подзапрос SELECT в скобках.
Если типы данных в левой и правой частях подзапроса `IN` различаются, ClickHouse преобразует значение в левой части к типу данных из правой части. Преобразование выполняется по аналогии с функцией [accurateCastOrNull](../functions/type-conversion-functions.md#type_conversion_function-accurate-cast_or_null), т.е. тип данных становится [Nullable](../../sql-reference/data-types/nullable.md), а если преобразование не может быть выполнено, возвращается значение [NULL](../../sql-reference/syntax.md#null-literal).
**Пример**
Запрос:
``` sql
SELECT '1' IN (SELECT 1);
```
Результат:
``` text
┌─in('1', _subquery49)─┐
│ 1 │
└──────────────────────┘
```
Если в качестве правой части оператора указано имя таблицы (например, `UserID IN users`), то это эквивалентно подзапросу `UserID IN (SELECT * FROM users)`. Это используется при работе с внешними данными, отправляемым вместе с запросом. Например, вместе с запросом может быть отправлено множество идентификаторов посетителей, загруженное во временную таблицу users, по которому следует выполнить фильтрацию.
Если в качестве правой части оператора, указано имя таблицы, имеющий движок Set (подготовленное множество, постоянно находящееся в оперативке), то множество не будет создаваться заново при каждом запросе.

91
docs/ru/sql-reference/table-functions/file.md
View File

@ -5,23 +5,27 @@ toc_title: file
# file {#file}
Создаёт таблицу из файла. Данная табличная функция похожа на табличные функции [file](file.md) и [hdfs](hdfs.md).
Создаёт таблицу из файла. Данная табличная функция похожа на табличные функции [url](../../sql-reference/table-functions/url.md) и [hdfs](../../sql-reference/table-functions/hdfs.md).
Функция `file` может использоваться в запросах `SELECT` и `INSERT` при работе с движком таблиц [File](../../engines/table-engines/special/file.md).
**Синтаксис**
``` sql
file(path, format, structure)
```
**Входные параметры**
**Параметры**
- `path` — относительный путь до файла от [user_files_path](../../sql-reference/table-functions/file.md#server_configuration_parameters-user_files_path). Путь к файлу поддерживает следующие шаблоны в режиме доступа только для чтения `*`, `?`, `{abc,def}` и `{N..M}`, где `N`, `M` — числа, \``'abc', 'def'` — строки.
- `path` — относительный путь до файла от [user_files_path](../../sql-reference/table-functions/file.md#server_configuration_parameters-user_files_path). Путь к файлу поддерживает следующие шаблоны в режиме доступа только для чтения `*`, `?`, `{abc,def}` и `{N..M}`, где `N`, `M` — числа, `'abc', 'def'` — строки.
- `format` — [формат](../../interfaces/formats.md#formats) файла.
- `structure` — структура таблицы. Формат `'colunmn1_name column1_ype, column2_name column2_type, ...'`.
- `structure` — структура таблицы. Формат: `'colunmn1_name column1_ype, column2_name column2_type, ...'`.
**Возвращаемое значение**
Таблица с указанной структурой, предназначенная для чтения или записи данных в указанном файле.
**Пример**
**Примеры**
Настройка `user_files_path` и содержимое файла `test.csv`:
@ -35,12 +39,10 @@ $ cat /var/lib/clickhouse/user_files/test.csv
78,43,45
```
Таблица из `test.csv` и выборка первых двух строк из неё:
Получение данных из таблицы в файле `test.csv` и выборка первых двух строк из неё:
``` sql
SELECT *
FROM file('test.csv', 'CSV', 'column1 UInt32, column2 UInt32, column3 UInt32')
LIMIT 2
SELECT * FROM file('test.csv', 'CSV', 'column1 UInt32, column2 UInt32, column3 UInt32') LIMIT 2;
```
``` text
@ -50,45 +52,61 @@ LIMIT 2
└─────────┴─────────┴─────────┘
```
Шаблоны могут содержаться в нескольких компонентах пути. Обрабатываются только существующие файлы, название которых целиком удовлетворяет шаблону (не только суффиксом или префиксом).
Получение первых 10 строк таблицы, содержащей 3 столбца типа [UInt32](../../sql-reference/data-types/int-uint.md), из CSV-файла:
- `*` — Заменяет любое количество любых символов кроме `/`, включая отсутствие символов.
- `?` — Заменяет ровно один любой символ.
- `{some_string,another_string,yet_another_one}` — Заменяет любую из строк `'some_string', 'another_string', 'yet_another_one'`.
- `{N..M}` — Заменяет любое число в интервале от `N` до `M` включительно (может содержать ведущие нули).
``` sql
SELECT * FROM file('test.csv', 'CSV', 'column1 UInt32, column2 UInt32, column3 UInt32') LIMIT 10;
```
Вставка данных из файла в таблицу:
``` sql
INSERT INTO FUNCTION file('test.csv', 'CSV', 'column1 UInt32, column2 UInt32, column3 UInt32') VALUES (1, 2, 3), (3, 2, 1);
SELECT * FROM file('test.csv', 'CSV', 'column1 UInt32, column2 UInt32, column3 UInt32');
```
``` text
┌─column1─┬─column2─┬─column3─┐
│ 1 │ 2 │ 3 │
│ 3 │ 2 │ 1 │
└─────────┴─────────┴─────────┘
```
## Шаблоны поиска в компонентах пути {#globs-in-path}
При описании пути к файлу могут использоваться шаблоны поиска. Обрабатываются только те файлы, у которых путь и название соответствуют шаблону полностью (а не только префикс или суффикс).
- `*` — заменяет любое количество любых символов кроме `/`, включая отсутствие символов.
- `?` — заменяет ровно один любой символ.
- `{some_string,another_string,yet_another_one}` — заменяет любую из строк `'some_string', 'another_string', 'yet_another_one'`.
- `{N..M}` — заменяет любое число в интервале от `N` до `M` включительно (может содержать ведущие нули).
Конструкция с `{}` аналогична табличной функции [remote](remote.md).
**Пример**
1. Предположим у нас есть несколько файлов со следующими относительными путями:
Предположим, у нас есть несколько файлов со следующими относительными путями:
- ‘some_dir/some_file_1’
- ‘some_dir/some_file_2’
- ‘some_dir/some_file_3’
- ‘another_dir/some_file_1’
- ‘another_dir/some_file_2’
- ‘another_dir/some_file_3’
- 'some_dir/some_file_1'
- 'some_dir/some_file_2'
- 'some_dir/some_file_3'
- 'another_dir/some_file_1'
- 'another_dir/some_file_2'
- 'another_dir/some_file_3'
1. Запросим количество строк в этих файлах:
<!-- -->
Запросим количество строк в этих файлах:
``` sql
SELECT count(*)
FROM file('{some,another}_dir/some_file_{1..3}', 'TSV', 'name String, value UInt32')
SELECT count(*) FROM file('{some,another}_dir/some_file_{1..3}', 'TSV', 'name String, value UInt32');
```
1. Запросим количество строк во всех файлах этих двух директорий:
<!-- -->
Запросим количество строк во всех файлах этих двух директорий:
``` sql
SELECT count(*)
FROM file('{some,another}_dir/*', 'TSV', 'name String, value UInt32')
SELECT count(*) FROM file('{some,another}_dir/*', 'TSV', 'name String, value UInt32');
```
!!! warning "Warning"
!!! warning "Предупреждение"
Если ваш список файлов содержит интервал с ведущими нулями, используйте конструкцию с фигурными скобками для каждой цифры по отдельности или используйте `?`.
**Пример**
@ -96,17 +114,16 @@ FROM file('{some,another}_dir/*', 'TSV', 'name String, value UInt32')
Запрос данных из файлов с именами `file000`, `file001`, … , `file999`:
``` sql
SELECT count(*)
FROM file('big_dir/file{0..9}{0..9}{0..9}', 'CSV', 'name String, value UInt32')
SELECT count(*) FROM file('big_dir/file{0..9}{0..9}{0..9}', 'CSV', 'name String, value UInt32');
```
## Виртуальные столбцы {#virtualnye-stolbtsy}
- `_path` — Путь к файлу.
- `_file` — Имя файла.
- `_path` — путь к файлу.
- `_file` — имя файла.
**Смотрите также**
- [Виртуальные столбцы](index.md#table_engines-virtual_columns)
[Оригинальная статья](https://clickhouse.tech/docs/ru/query_language/table_functions/file/) <!--hide-->
[Оригинальная статья](https://clickhouse.tech/docs/ru/sql-reference/table-functions/file/) <!--hide-->

63
docs/ru/sql-reference/table-functions/mysql.md
View File

@ -7,6 +7,8 @@ toc_title: mysql
Позволяет выполнять запросы `SELECT` над данными, хранящимися на удалённом MySQL сервере.
**Синтаксис**
``` sql
mysql('host:port', 'database', 'table', 'user', 'password'[, replace_query, 'on_duplicate_clause']);
```
@ -23,13 +25,13 @@ mysql('host:port', 'database', 'table', 'user', 'password'[, replace_query, 'on_
- `password` — пароль пользователя.
- `replace_query` — флаг, отвечающий за преобразование запросов `INSERT INTO` в `REPLACE INTO`. Если `replace_query=1`, то запрос заменяется.
- `replace_query` — флаг, отвечающий за преобразование запросов `INSERT INTO` в `REPLACE INTO`. Возможные значения:
- `0` - выполняется запрос `INSERT INTO`.
- `1` - выполняется запрос `REPLACE INTO`.
- `on_duplicate_clause` — выражение `ON DUPLICATE KEY on_duplicate_clause`, добавляемое в запрос `INSERT`.
- `on_duplicate_clause` — выражение `ON DUPLICATE KEY on_duplicate_clause`, добавляемое в запрос `INSERT`. Может быть передано только с помощью `replace_query = 0` (если вы одновременно передадите `replace_query = 1` и `on_duplicate_clause`, будет сгенерировано исключение).
Пример: `INSERT INTO t (c1,c2) VALUES ('a', 2) ON DUPLICATE KEY UPDATE c2 = c2 + 1`, где `on_duplicate_clause` это `UPDATE c2 = c2 + 1`. Чтобы узнать какие `on_duplicate_clause` можно использовать с секцией `ON DUPLICATE KEY` обратитесь к документации MySQL.
Чтобы указать `'on_duplicate_clause'` необходимо передать `0` в параметр `replace_query`. Если одновременно передать `replace_query = 1` и `'on_duplicate_clause'`, то ClickHouse сгенерирует исключение.
Пример: `INSERT INTO t (c1,c2) VALUES ('a', 2) ON DUPLICATE KEY UPDATE c2 = c2 + 1`, где `on_duplicate_clause` это `UPDATE c2 = c2 + 1;`
Простые условия `WHERE` такие как `=, !=, >, >=, <, =` выполняются на стороне сервера MySQL.
@ -39,46 +41,59 @@ mysql('host:port', 'database', 'table', 'user', 'password'[, replace_query, 'on_
Объект таблицы с теми же столбцами, что и в исходной таблице MySQL.
## Пример использования {#primer-ispolzovaniia}
!!! note "Примечание"
Чтобы отличить табличную функцию `mysql (...)` в запросе `INSERT` от имени таблицы со списком имен столбцов, используйте ключевые слова `FUNCTION` или `TABLE FUNCTION`. См. примеры ниже.
**Примеры**
Таблица в MySQL:
``` text
mysql> CREATE TABLE `test`.`test` (
-> `int_id` INT NOT NULL AUTO_INCREMENT,
-> `int_nullable` INT NULL DEFAULT NULL,
-> `float` FLOAT NOT NULL,
-> `float_nullable` FLOAT NULL DEFAULT NULL,
-> PRIMARY KEY (`int_id`));
Query OK, 0 rows affected (0,09 sec)
mysql> insert into test (`int_id`, `float`) VALUES (1,2);
Query OK, 1 row affected (0,00 sec)
mysql> INSERT INTO test (`int_id`, `float`) VALUES (1,2);
mysql> select * from test;
+--------+--------------+-------+----------------+
| int_id | int_nullable | float | float_nullable |
+--------+--------------+-------+----------------+
| 1 | NULL | 2 | NULL |
+--------+--------------+-------+----------------+
1 row in set (0,00 sec)
mysql> SELECT * FROM test;
+--------+-------+
| int_id | float |
+--------+-------+
| 1 | 2 |
+--------+-------+
```
Получение данных в ClickHouse:
``` sql
SELECT * FROM mysql('localhost:3306', 'test', 'test', 'bayonet', '123')
SELECT * FROM mysql('localhost:3306', 'test', 'test', 'bayonet', '123');
```
``` text
┌─int_id─┬─int_nullable─┬─float─┬─float_nullable─┐
│ 1 │ ᴺᵁᴸᴸ │ 2 │ ᴺᵁᴸᴸ │
└────────┴──────────────┴───────┴────────────────┘
┌─int_id─┬─float─┐
│ 1 │ 2 │
└────────┴───────┘
```
## Смотрите также {#smotrite-takzhe}
Замена и вставка:
```sql
INSERT INTO FUNCTION mysql('localhost:3306', 'test', 'test', 'bayonet', '123', 1) (int_id, float) VALUES (1, 3);
INSERT INTO TABLE FUNCTION mysql('localhost:3306', 'test', 'test', 'bayonet', '123', 0, 'UPDATE int_id = int_id + 1') (int_id, float) VALUES (1, 4);
SELECT * FROM mysql('localhost:3306', 'test', 'test', 'bayonet', '123');
```
``` text
┌─int_id─┬─float─┐
│ 1 │ 3 │
│ 2 │ 4 │
└────────┴───────┘
```
**Смотрите также**
- [Движок таблиц ‘MySQL’](../../sql-reference/table-functions/mysql.md)
- [Использование MySQL как источника данных для внешнего словаря](../../sql-reference/table-functions/mysql.md#dicts-external_dicts_dict_sources-mysql)
[Оригинальная статья](https://clickhouse.tech/docs/ru/query_language/table_functions/mysql/) <!--hide-->
[Оригинальная статья](https://clickhouse.tech/docs/ru/sql-reference/table_functions/mysql/) <!--hide-->

78
docs/ru/sql-reference/table-functions/remote.md
View File

@ -5,9 +5,11 @@ toc_title: remote
# remote, remoteSecure {#remote-remotesecure}
Позволяет обратиться к удалённым серверам без создания таблицы типа `Distributed`.
Позволяет обратиться к удалённым серверам без создания таблицы типа [Distributed](../../engines/table-engines/special/distributed.md). Функция `remoteSecure` работает аналогично `remote`, но использует защищенное соединение.
Сигнатуры:
Обе функции могут использоваться в запросах `SELECT` и `INSERT`.
**Синтаксис**
``` sql
remote('addresses_expr', db, table[, 'user'[, 'password']])
@ -16,12 +18,40 @@ remoteSecure('addresses_expr', db, table[, 'user'[, 'password']])
remoteSecure('addresses_expr', db.table[, 'user'[, 'password']])
```
`addresses_expr` - выражение, генерирующее адреса удалённых серверов. Это может быть просто один адрес сервера. Адрес сервера - это `хост:порт`, или только `хост`. Хост может быть указан в виде имени сервера, или в виде IPv4 или IPv6 адреса. IPv6 адрес указывается в квадратных скобках. Порт - TCP-порт удалённого сервера. Если порт не указан, используется `tcp_port` из конфигурационного файла сервера (по умолчанию - 9000).
**Параметры**
- `addresses_expr` — выражение, генерирующее адреса удалённых серверов. Это может быть просто один адрес сервера. Адрес сервера — это `host:port` или только `host`.
Вместо параметра `host` может быть указано имя сервера или его адрес в формате IPv4 или IPv6. IPv6 адрес указывается в квадратных скобках.
`port` — TCP-порт удалённого сервера. Если порт не указан, используется [tcp_port](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-tcp_port) из конфигурационного файла сервера, к которому обратились через функцию `remote` (по умолчанию - 9000), и [tcp_port_secure](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-tcp_port_secure), к которому обратились через функцию `remoteSecure` (по умолчанию — 9440).
!!! important "Важно"
С IPv6-адресом обязательно нужно указывать порт.
Примеры:
Тип: [String](../../sql-reference/data-types/string.md).
- `db` — имя базы данных. Тип: [String](../../sql-reference/data-types/string.md).
- `table` — имя таблицы. Тип: [String](../../sql-reference/data-types/string.md).
- `user` — имя пользователя. Если пользователь не указан, то по умолчанию `default`. Тип: [String](../../sql-reference/data-types/string.md).
- `password` — пароль. Если пароль не указан, то используется пустой пароль. Тип: [String](../../sql-reference/data-types/string.md).
- `sharding_key` — ключ шардирования для поддержки распределения данных между узлами. Например: `insert into remote('127.0.0.1:9000,127.0.0.2', db, table, 'default', rand())`. Тип: [UInt32](../../sql-reference/data-types/int-uint.md).
**Возвращаемое значение**
Набор данных с удаленных серверов.
**Использование**
Использование табличной функции `remote` менее оптимально, чем создание таблицы типа `Distributed`, так как в этом случае соединения с серверами устанавливаются заново при каждом запросе. Если указываются имена серверов, то приходится также выполнять поиск сервера по имени. Кроме того, не ведётся сквозной подсчёт ошибок при работе с разными репликами. При обработке большого количества запросов всегда создавайте таблицу типа `Distributed`, использовать табличную функцию `remote` в таких случаях не рекомендуется.
Табличная функция `remote` может быть полезна в следующих случаях:
- Обращение на конкретный сервер для сравнения данных, отладки и тестирования.
- Запросы между разными кластерами ClickHouse для исследований.
- Нечастые распределённые запросы, задаваемые вручную.
- Распределённые запросы, где набор серверов определяется каждый раз заново.
**Адреса**
``` text
example01-01-1
@ -32,9 +62,7 @@ localhost
[2a02:6b8:0:1111::11]:9000
```
Адреса можно указать через запятую, в этом случае ClickHouse обработает запрос как распределённый, т.е. отправит его по всем указанным адресам как на шарды с разными данными.
Пример:
Адреса можно указать через запятую. В этом случае ClickHouse обработает запрос как распределённый, т.е. отправит его по всем указанным адресам как на шарды с разными данными. Пример:
``` text
example01-01-1,example01-02-1
@ -46,38 +74,36 @@ example01-01-1,example01-02-1
example01-0{1,2}-1
```
В фигурных скобках может быть указан диапазон (неотрицательных целых) чисел через две точки. В этом случае, диапазон раскрывается в множество значений, генерирующих адреса шардов. Если запись первого числа начинается с нуля, то значения формируются с таким же выравниванием нулями. Предыдущий пример может быть записан следующим образом:
В фигурных скобках может быть указан диапазон (неотрицательных целых) чисел через две точки. В этом случае диапазон раскрывается в множество значений, генерирующих адреса шардов. Если запись первого числа начинается с нуля, то значения формируются с таким же выравниванием нулями. Предыдущий пример может быть записан следующим образом:
``` text
example01-{01..02}-1
```
При наличии нескольких пар фигурных скобок, генерируется прямое произведение соответствующих множеств.
При наличии нескольких пар фигурных скобок генерируется прямое произведение соответствующих множеств.
Адреса или их фрагменты в фигурных скобках можно указать через символ \|. В этом случае, соответствующие множества адресов понимаются как реплики - запрос будет отправлен на первую живую реплику. При этом, реплики перебираются в порядке, согласно текущей настройке [load_balancing](../../operations/settings/settings.md).
Пример:
Адреса или их фрагменты в фигурных скобках можно указать через символ \|. В этом случае соответствующие множества адресов понимаются как реплики — запрос будет отправлен на первую живую реплику. При этом реплики перебираются в порядке, согласно текущей настройке [load_balancing](../../operations/settings/settings.md#settings-load_balancing). В этом примере указаны два шарда, в каждом из которых имеются две реплики:
``` text
example01-{01..02}-{1|2}
```
В этом примере указано два шарда, в каждом из которых имеется две реплики.
Количество генерируемых адресов ограничено константой. Сейчас это 1000 адресов.
Количество генерируемых адресов ограничено константой - сейчас это 1000 штук.
**Примеры**
Использование табличной функции `remote` менее оптимально, чем создание таблицы типа `Distributed`, так как в этом случае, соединения с серверами устанавливаются заново при каждом запросе, в случае задания имён хостов, делается резолвинг имён, а также не ведётся подсчёт ошибок при работе с разными репликами. При обработке большого количества запросов, всегда создавайте `Distributed` таблицу заранее, не используйте табличную функцию `remote`.
Выборка данных с удаленного сервера:
Табличная функция `remote` может быть полезна для следующих случаях:
``` sql
SELECT * FROM remote('127.0.0.1', db.remote_engine_table) LIMIT 3;
```
- обращение на конкретный сервер в целях сравнения данных, отладки и тестирования;
- запросы между разными кластерами ClickHouse в целях исследований;
- нечастых распределённых запросов, задаваемых вручную;
- распределённых запросов, где набор серверов определяется каждый раз заново.
Вставка данных с удаленного сервера в таблицу:
Если пользователь не задан,то используется `default`.
Если пароль не задан, то используется пустой пароль.
``` sql
CREATE TABLE remote_table (name String, value UInt32) ENGINE=Memory;
INSERT INTO FUNCTION remote('127.0.0.1', currentDatabase(), 'remote_table') VALUES ('test', 42);
SELECT * FROM remote_table;
```
`remoteSecure` - аналогично функции `remote`, но с соединением по шифрованному каналу. Порт по умолчанию - `tcp_port_secure` из конфига или 9440.
[Оригинальная статья](https://clickhouse.tech/docs/ru/query_language/table_functions/remote/) <!--hide-->
[Оригинальная статья](https://clickhouse.tech/docs/ru/sql-reference/table-functions/remote/) <!--hide-->

43
docs/ru/sql-reference/table-functions/url.md
View File

@ -5,21 +5,40 @@ toc_title: url
# url {#url}
`url(URL, format, structure)` - возвращает таблицу со столбцами, указанными в
`structure`, созданную из данных находящихся по `URL` в формате `format`.
Функция `url` берет данные по указанному адресу `URL` и создает из них таблицу указанной структуры со столбцами указанного формата.
URL - адрес, по которому сервер принимает `GET` и/или `POST` запросы по
протоколу HTTP или HTTPS.
Функция `url` может быть использована в запросах `SELECT` и `INSERT` с таблицами на движке [URL](../../engines/table-engines/special/url.md).
format - [формат](../../interfaces/formats.md#formats) данных.
structure - структура таблицы в форме `'UserID UInt64, Name String'`. Определяет имена и типы столбцов.
**Пример**
**Синтаксис**
``` sql
-- получение 3-х строк таблицы, состоящей из двух колонк типа String и UInt32 от сервера, отдающего данные в формате CSV
SELECT * FROM url('http://127.0.0.1:12345/', CSV, 'column1 String, column2 UInt32') LIMIT 3
url(URL, format, structure)
```
[Оригинальная статья](https://clickhouse.tech/docs/ru/query_language/table_functions/url/) <!--hide-->
**Параметры**
- `URL` — HTTP или HTTPS-адрес сервера, который может принимать запросы `GET` или `POST` (для запросов `SELECT` или `INSERT` соответственно). Тип: [String](../../sql-reference/data-types/string.md).
- `format` — [формат](../../interfaces/formats.md#formats) данных. Тип: [String](../../sql-reference/data-types/string.md).
- `structure` — структура таблицы в формате `'UserID UInt64, Name String'`. Определяет имена и типы столбцов. Тип: [String](../../sql-reference/data-types/string.md).
**Возвращаемое значение**
Таблица с указанными форматом и структурой, а также с данными, полученными из указанного адреса `URL`.
**Примеры**
Получение с HTTP-сервера первых 3 строк таблицы с данными в формате [CSV](../../interfaces/formats.md/#csv), содержащей столбцы типа [String](../../sql-reference/data-types/string.md) и [UInt32](../../sql-reference/data-types/int-uint.md).
``` sql
SELECT * FROM url('http://127.0.0.1:12345/', CSV, 'column1 String, column2 UInt32') LIMIT 3;
```
Вставка данных в таблицу:
``` sql
CREATE TABLE test_table (column1 String, column2 UInt32) ENGINE=Memory;
INSERT INTO FUNCTION url('http://127.0.0.1:8123/?query=INSERT+INTO+test_table+FORMAT+CSV', 'CSV', 'column1 String, column2 UInt32') VALUES ('http interface', 42);
SELECT * FROM test_table;
```
[Оригинальная статья](https://clickhouse.tech/docs/ru/sql-reference/table-functions/url/) <!--hide-->

22
programs/odbc-bridge/ColumnInfoHandler.cpp
View File

@ -4,14 +4,14 @@
# include <DataTypes/DataTypeFactory.h>
# include <DataTypes/DataTypeNullable.h>
# include <IO/WriteBufferFromHTTPServerResponse.h>
# include <Server/HTTP/WriteBufferFromHTTPServerResponse.h>
# include <IO/WriteHelpers.h>
# include <Parsers/ParserQueryWithOutput.h>
# include <Parsers/parseQuery.h>
# include <Poco/Data/ODBC/ODBCException.h>
# include <Poco/Data/ODBC/SessionImpl.h>
# include <Poco/Data/ODBC/Utility.h>
# include <Poco/Net/HTMLForm.h>
# include <Server/HTTP/HTMLForm.h>
# include <Poco/Net/HTTPServerRequest.h>
# include <Poco/Net/HTTPServerResponse.h>
# include <Poco/NumberParser.h>
@ -59,16 +59,16 @@ namespace
}
}
void ODBCColumnsInfoHandler::handleRequest(Poco::Net::HTTPServerRequest & request, Poco::Net::HTTPServerResponse & response)
void ODBCColumnsInfoHandler::handleRequest(HTTPServerRequest & request, HTTPServerResponse & response)
{
Poco::Net::HTMLForm params(request, request.stream());
HTMLForm params(request, request.getStream());
LOG_TRACE(log, "Request URI: {}", request.getURI());
auto process_error = [&response, this](const std::string & message)
{
response.setStatusAndReason(Poco::Net::HTTPResponse::HTTP_INTERNAL_SERVER_ERROR);
if (!response.sent())
response.send() << message << std::endl;
*response.send() << message << std::endl;
LOG_WARNING(log, message);
};
@ -159,8 +159,16 @@ void ODBCColumnsInfoHandler::handleRequest(Poco::Net::HTTPServerRequest & reques
columns.emplace_back(reinterpret_cast<char *>(column_name), std::move(column_type));
}
WriteBufferFromHTTPServerResponse out(request, response, keep_alive_timeout);
writeStringBinary(columns.toString(), out);
WriteBufferFromHTTPServerResponse out(response, request.getMethod() == Poco::Net::HTTPRequest::HTTP_HEAD, keep_alive_timeout);
try
{
writeStringBinary(columns.toString(), out);
out.finalize();
}
catch (...)
{
out.finalize();
}
}
catch (...)
{

9
programs/odbc-bridge/ColumnInfoHandler.h
View File

@ -3,10 +3,11 @@
#if USE_ODBC
# include <Interpreters/Context.h>
# include <Poco/Logger.h>
# include <Poco/Net/HTTPRequestHandler.h>
# include <Server/HTTP/HTTPRequestHandler.h>
# include <Common/config.h>
# include <Poco/Logger.h>
/** The structure of the table is taken from the query "SELECT * FROM table WHERE 1=0".
* TODO: It would be much better to utilize ODBC methods dedicated for columns description.
* If there is no such table, an exception is thrown.
@ -14,7 +15,7 @@
namespace DB
{
class ODBCColumnsInfoHandler : public Poco::Net::HTTPRequestHandler
class ODBCColumnsInfoHandler : public HTTPRequestHandler
{
public:
ODBCColumnsInfoHandler(size_t keep_alive_timeout_, Context & context_)
@ -22,7 +23,7 @@ public:
{
}
void handleRequest(Poco::Net::HTTPServerRequest & request, Poco::Net::HTTPServerResponse & response) override;
void handleRequest(HTTPServerRequest & request, HTTPServerResponse & response) override;
private:
Poco::Logger * log;

15
programs/odbc-bridge/HandlerFactory.cpp
View File

@ -7,39 +7,40 @@
namespace DB
{
Poco::Net::HTTPRequestHandler * HandlerFactory::createRequestHandler(const Poco::Net::HTTPServerRequest & request)
std::unique_ptr<HTTPRequestHandler> HandlerFactory::createRequestHandler(const HTTPServerRequest & request)
{
Poco::URI uri{request.getURI()};
LOG_TRACE(log, "Request URI: {}", uri.toString());
if (uri.getPath() == "/ping" && request.getMethod() == Poco::Net::HTTPRequest::HTTP_GET)
return new PingHandler(keep_alive_timeout);
return std::make_unique<PingHandler>(keep_alive_timeout);
if (request.getMethod() == Poco::Net::HTTPRequest::HTTP_POST)
{
if (uri.getPath() == "/columns_info")
#if USE_ODBC
return new ODBCColumnsInfoHandler(keep_alive_timeout, context);
return std::make_unique<ODBCColumnsInfoHandler>(keep_alive_timeout, context);
#else
return nullptr;
#endif
else if (uri.getPath() == "/identifier_quote")
#if USE_ODBC
return new IdentifierQuoteHandler(keep_alive_timeout, context);
return std::make_unique<IdentifierQuoteHandler>(keep_alive_timeout, context);
#else
return nullptr;
#endif
else if (uri.getPath() == "/schema_allowed")
#if USE_ODBC
return new SchemaAllowedHandler(keep_alive_timeout, context);
return std::make_unique<SchemaAllowedHandler>(keep_alive_timeout, context);
#else
return nullptr;
#endif
else if (uri.getPath() == "/write")
return new ODBCHandler(pool_map, keep_alive_timeout, context, "write");
return std::make_unique<ODBCHandler>(pool_map, keep_alive_timeout, context, "write");
else
return new ODBCHandler(pool_map, keep_alive_timeout, context, "read");
return std::make_unique<ODBCHandler>(pool_map, keep_alive_timeout, context, "read");
}
return nullptr;
}

15
programs/odbc-bridge/HandlerFactory.h
View File

@ -1,16 +1,17 @@
#pragma once
#include <Interpreters/Context.h>
#include <Poco/Logger.h>
#include <Poco/Net/HTTPRequestHandler.h>
#include <Poco/Net/HTTPRequestHandlerFactory.h>
#include "MainHandler.h"
#include <Server/HTTP/HTTPRequestHandlerFactory.h>
#include "ColumnInfoHandler.h"
#include "IdentifierQuoteHandler.h"
#include "MainHandler.h"
#include "SchemaAllowedHandler.h"
#include <Poco/Logger.h>
#pragma GCC diagnostic push
#pragma GCC diagnostic ignored "-Wunused-parameter"
#include <Poco/Data/SessionPool.h>
#include <Poco/Data/SessionPool.h>
#pragma GCC diagnostic pop
@ -19,7 +20,7 @@ namespace DB
/** Factory for '/ping', '/', '/columns_info', '/identifier_quote', '/schema_allowed' handlers.
* Also stores Session pools for ODBC connections
*/
class HandlerFactory : public Poco::Net::HTTPRequestHandlerFactory
class HandlerFactory : public HTTPRequestHandlerFactory
{
public:
HandlerFactory(const std::string & name_, size_t keep_alive_timeout_, Context & context_)
@ -28,7 +29,7 @@ public:
pool_map = std::make_shared<ODBCHandler::PoolMap>();
}
Poco::Net::HTTPRequestHandler * createRequestHandler(const Poco::Net::HTTPServerRequest & request) override;
std::unique_ptr<HTTPRequestHandler> createRequestHandler(const HTTPServerRequest & request) override;
private:
Poco::Logger * log;

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