Merge branch 'master' into fix-inconsistent-formatting-except

Author: alexey-milovidov

README.md

@@ -52,6 +52,7 @@ Upcoming meetups
 * [Istanbul Meetup](https://www.meetup.com/clickhouse-turkiye-meetup-group/events/306978337/) - May 15, 2025
 * [Shenzhen Meetup](https://www.huodongxing.com/event/7803892350511) - May 17, 2025
 * [Seattle Meetup](https://www.meetup.com/clickhouse-seattle-user-group/events/307622716/) - May 20, 2025
+* [Tokyo Meetup](https://www.meetup.com/clickhouse-tokyo-user-group/events/307689645/) - June 5, 2025
 * [Washington DC Meetup](https://www.meetup.com/clickhouse-dc-user-group/events/307622954/) - June 12, 2025
 * [Atlanta Meetup](https://www.meetup.com/clickhouse-atlanta-meetup-group/events/307627590/) - July 8, 2025
 * [New York Meetup](https://www.meetup.com/clickhouse-new-york-user-group/events/307627675/) - July 15, 2025

docs/en/sql-reference/functions/splitting-merging-functions.md

@@ -21,7 +21,7 @@ splitByChar(separator, s[, max_substrings]))
 
 **Arguments**
 
-- `separator` — The separator which should contain exactly one character. [String](../data-types/string.md).
+- `separator` — The separator must be a single-byte character. [String](../data-types/string.md).
 - `s` — The string to split. [String](../data-types/string.md).
 - `max_substrings` — An optional `Int64` defaulting to 0. If `max_substrings` > 0, the returned array will contain at most `max_substrings` substrings, otherwise the function will return as many substrings as possible.
 

programs/client/Client.cpp

@@ -353,7 +353,6 @@ try
     initTTYBuffer(
         toProgressOption(config().getString("progress", "default")), toProgressOption(config().getString("progress-table", "default")));
     initKeystrokeInterceptor();
-    ASTAlterCommand::setFormatAlterCommandsWithParentheses(true);
 
     {
         // All that just to set DB::CurrentThread::get().getGlobalContext()

programs/format/Format.cpp

@@ -107,7 +107,6 @@ int mainEntryClickHouseFormat(int argc, char ** argv)
             ("backslash", "add a backslash at the end of each line of the formatted query")
             ("allow_settings_after_format_in_insert", "Allow SETTINGS after FORMAT, but note, that this is not always safe")
             ("seed", po::value<std::string>(), "seed (arbitrary string) that determines the result of obfuscation")
-            ("format_alter_operations_with_parentheses", po::value<bool>()->default_value(true), "If set to `true`, then alter operations will be surrounded by parentheses in formatted queries. This makes the parsing of formatted alter queries less ambiguous.")
         ;
 
         Settings cmd_settings;
@@ -134,7 +133,6 @@ int mainEntryClickHouseFormat(int argc, char ** argv)
         bool obfuscate = options.count("obfuscate");
         bool backslash = options.count("backslash");
         bool allow_settings_after_format_in_insert = options.count("allow_settings_after_format_in_insert");
-        bool format_alter_operations_with_parentheses = options["format_alter_operations_with_parentheses"].as<bool>();
 
         if (quiet && (hilite || oneline || obfuscate))
         {
@@ -160,8 +158,6 @@ int mainEntryClickHouseFormat(int argc, char ** argv)
             return 2;
         }
 
-        ASTAlterCommand::setFormatAlterCommandsWithParentheses(format_alter_operations_with_parentheses);
-
         String query;
 
         if (options.count("query"))

programs/local/LocalServer.cpp

@@ -617,7 +617,6 @@ try
     initTTYBuffer(toProgressOption(getClientConfiguration().getString("progress", "default")),
         toProgressOption(config().getString("progress-table", "default")));
     initKeystrokeInterceptor();
-    ASTAlterCommand::setFormatAlterCommandsWithParentheses(true);
 
     /// try to load user defined executable functions, throw on error and die
     try

programs/server/Server.cpp

@@ -217,7 +217,6 @@ namespace ServerSetting
     extern const ServerSettingsInt32 dns_cache_update_period;
     extern const ServerSettingsUInt32 dns_max_consecutive_failures;
     extern const ServerSettingsBool enable_azure_sdk_logging;
-    extern const ServerSettingsBool format_alter_operations_with_parentheses;
     extern const ServerSettingsUInt64 global_profiler_cpu_time_period_ns;
     extern const ServerSettingsUInt64 global_profiler_real_time_period_ns;
     extern const ServerSettingsUInt64 http_connections_soft_limit;
@@ -984,8 +983,6 @@ try
     ServerSettings server_settings;
     server_settings.loadSettingsFromConfig(config());
 
-    ASTAlterCommand::setFormatAlterCommandsWithParentheses(server_settings[ServerSetting::format_alter_operations_with_parentheses]);
-
     StackTrace::setShowAddresses(server_settings[ServerSetting::show_addresses_in_stack_traces]);
 
 #if USE_HDFS

src/Common/ProfileEvents.cpp

@@ -136,8 +136,16 @@
     M(LocalWriteThrottlerBytes, "Bytes passed through 'max_local_write_bandwidth_for_server'/'max_local_write_bandwidth' throttler.", ValueType::Bytes) \
     M(LocalWriteThrottlerSleepMicroseconds, "Total time a query was sleeping to conform 'max_local_write_bandwidth_for_server'/'max_local_write_bandwidth' throttling.", ValueType::Microseconds) \
     M(ThrottlerSleepMicroseconds, "Total time a query was sleeping to conform all throttling settings.", ValueType::Microseconds) \
-    M(ReadTasksWithAppliedMutationsOnFly, "Total number of parts for which there was any mutation applied on fly", ValueType::Number) \
-    M(MutationsAppliedOnFlyInAllReadTasks, "The sum of number of applied mutations on-fly for part among all read parts", ValueType::Number) \
+    M(ReadTasksWithAppliedPatches, "Total number of read tasks for which there was any patch part applied", ValueType::Number) \
+    M(PatchesAppliedInAllReadTasks, "Total number of applied patch parts among all read tasks", ValueType::Number) \
+    M(PatchesMergeAppliedInAllReadTasks, "Total number of applied patch parts with Merge mode among all read tasks", ValueType::Number) \
+    M(PatchesJoinAppliedInAllReadTasks, "Total number of applied patch parts with Join mode among all read tasks", ValueType::Number) \
+    M(ApplyPatchesMicroseconds, "Total time spent applying patch parts", ValueType::Number) \
+    M(ApplyPatchesMergeMicroseconds, "Total time spent applying patch parts with Merge mode", ValueType::Number) \
+    M(ApplyPatchesJoinMicroseconds, "Total time spent applying patch parts with Join mode", ValueType::Number) \
+    M(AnalyzePatchRangesMicroseconds, "Total time spent analyzing index of patch parts", ValueType::Number) \
+    M(ReadTasksWithAppliedMutationsOnFly, "Total number of read tasks for which there was any mutation applied on fly", ValueType::Number) \
+    M(MutationsAppliedOnFlyInAllReadTasks, "Total number of applied mutations on-fly among all read tasks", ValueType::Number) \
     \
     M(SchedulerIOReadRequests, "Resource requests passed through scheduler for IO reads.", ValueType::Number) \
     M(SchedulerIOReadBytes, "Bytes passed through scheduler for IO reads.", ValueType::Bytes) \
@@ -1135,6 +1143,8 @@ Event end() { return END; }
 
 bool checkCPUOverload(Int64 os_cpu_busy_time_threshold, double min_ratio, double max_ratio, bool should_throw)
 {
+    if ((max_ratio <= 0.0) || (max_ratio <= min_ratio))
+        return false;
     double cpu_load = global_counters.getCPUOverload(os_cpu_busy_time_threshold);
 
     if (cpu_load > DBL_EPSILON)

src/Compression/CompressedReadBufferBase.cpp

@@ -1,4 +1,4 @@
-#include "CompressedReadBufferBase.h"
+#include <Compression/CompressedReadBufferBase.h>
 
 #include <bit>
 #include <cstring>
@@ -179,8 +179,8 @@ size_t CompressedReadBufferBase::readCompressedData(size_t & size_decompressed,
 
     UInt8 header_size = ICompressionCodec::getHeaderSize();
     own_compressed_buffer.resize(header_size + sizeof(Checksum));
-
     compressed_in->readStrict(own_compressed_buffer.data(), sizeof(Checksum) + header_size);
+    own_compressed_buffer_header_init = true;
 
     readHeaderAndGetCodecAndSize(
         own_compressed_buffer.data() + sizeof(Checksum),
@@ -233,6 +233,7 @@ size_t CompressedReadBufferBase::readCompressedDataBlockForAsynchronous(size_t &
 
     own_compressed_buffer.resize(header_size + sizeof(Checksum));
     compressed_in->readStrict(own_compressed_buffer.data(), sizeof(Checksum) + header_size);
+    own_compressed_buffer_header_init = true;
 
     readHeaderAndGetCodecAndSize(
         own_compressed_buffer.data() + sizeof(Checksum),
@@ -337,7 +338,9 @@ void CompressedReadBufferBase::addDiagnostics(Exception & e) const
     if (auto * seekable_in = dynamic_cast<SeekableReadBuffer *>(compressed_in))
         current_pos = seekable_in->tryGetPosition();
     UInt8 header_size = ICompressionCodec::getHeaderSize();
-    String header_hex = hexString(own_compressed_buffer.data(), std::min(own_compressed_buffer.size(), sizeof(Checksum) + header_size));
+    String header_hex = own_compressed_buffer_header_init ?
+        hexString(own_compressed_buffer.data(), std::min(own_compressed_buffer.size(), sizeof(Checksum) + header_size)) :
+        String("<uninitialized>"); // We do not print uninitialized memory because it's a security vulnerability and triggers msan
 
     e.addMessage("While reading or decompressing {} (position: {}, typename: {}, compressed data header: {})",
                  getFileNameFromReadBuffer(*compressed_in),

src/Compression/CompressedReadBufferBase.h

@@ -22,6 +22,7 @@ class CompressedReadBufferBase
 
     /// If 'compressed_in' buffer has whole compressed block - then use it. Otherwise copy parts of data to 'own_compressed_buffer'.
     PODArray<char> own_compressed_buffer;
+    bool own_compressed_buffer_header_init = false; // true if own_compressed_buffer header was initialized
     /// Points to memory, holding compressed block.
     char * compressed_buffer = nullptr;
 

src/Core/ServerSettings.cpp

@@ -898,7 +898,6 @@ The policy on how to perform a scheduling of CPU slots specified by `concurrent_
     )", 0) \
     DECLARE(UInt32, max_database_replicated_create_table_thread_pool_size, 1, R"(The number of threads to create tables during replica recovery in DatabaseReplicated. Zero means number of threads equal number of cores.)", 0) \
     DECLARE(Bool, database_replicated_allow_detach_permanently, true, R"(Allow detaching tables permanently in Replicated databases)", 0) \
-    DECLARE(Bool, format_alter_operations_with_parentheses, true, R"(If set to `true`, then alter operations will be surrounded by parentheses in formatted queries. This makes the parsing of formatted alter queries less ambiguous.)", 0) \
     DECLARE(String, default_replica_path, "/clickhouse/tables/{uuid}/{shard}", R"(
     The path to the table in ZooKeeper.