Merge pull request #4816 from yandex/add-setting-to-disable-hyperscan

Added a setting to disallow Hyperscan.

dbms/src/Common/ErrorCodes.cpp

@@ -420,6 +420,7 @@ namespace ErrorCodes
     extern const int NO_COMMON_COLUMNS_WITH_PROTOBUF_SCHEMA = 443;
     extern const int UNKNOWN_PROTOBUF_FORMAT = 444;
     extern const int CANNOT_MPROTECT = 445;
+    extern const int FUNCTION_NOT_ALLOWED = 446;
 
     extern const int KEEPER_EXCEPTION = 999;
     extern const int POCO_EXCEPTION = 1000;

dbms/src/Core/Settings.h

@@ -305,7 +305,9 @@ struct Settings
     M(SettingBool, allow_experimental_cross_to_join_conversion, true, "Convert CROSS JOIN to INNER JOIN if possible") \
     M(SettingBool, cancel_http_readonly_queries_on_client_close, false, "Cancel HTTP readonly queries when a client closes the connection without waiting for response.") \
     M(SettingBool, external_table_functions_use_nulls, true, "If it is set to true, external table functions will implicitly use Nullable type if needed. Otherwise NULLs will be substituted with default values. Currently supported only for 'mysql' table function.") \
-    M(SettingBool, allow_experimental_data_skipping_indices, false, "If it is set to true, data skipping indices can be used in CREATE TABLE/ALTER TABLE queries.")\
+    M(SettingBool, allow_experimental_data_skipping_indices, false, "If it is set to true, data skipping indices can be used in CREATE TABLE/ALTER TABLE queries.") \
+    \
+    M(SettingBool, allow_hyperscan, true, "Allow functions that use Hyperscan library. Disable to avoid potentially long compilation times and excessive resource usage.") \
 
 #define DECLARE(TYPE, NAME, DEFAULT, DESCRIPTION) \
     TYPE NAME {DEFAULT};

dbms/src/Functions/FunctionsStringSearch.cpp

@@ -340,6 +340,7 @@ template <typename Impl>
 struct MultiSearchImpl
 {
     using ResultType = UInt8;
+    static constexpr bool is_using_hyperscan = false;
 
     static void vector_constant(
         const ColumnString::Chars & haystack_data,
@@ -355,6 +356,7 @@ template <typename Impl>
 struct MultiSearchFirstPositionImpl
 {
     using ResultType = UInt64;
+    static constexpr bool is_using_hyperscan = false;
 
     static void vector_constant(
         const ColumnString::Chars & haystack_data,
@@ -374,6 +376,7 @@ template <typename Impl>
 struct MultiSearchFirstIndexImpl
 {
     using ResultType = UInt64;
+    static constexpr bool is_using_hyperscan = false;
 
     static void vector_constant(
         const ColumnString::Chars & haystack_data,
@@ -610,6 +613,7 @@ struct MultiMatchAnyImpl
 {
     static_assert(static_cast<int>(FindAny) + static_cast<int>(FindAnyIndex) == 1);
     using ResultType = Type;
+    static constexpr bool is_using_hyperscan = true;
 
     static void vector_constant(
         const ColumnString::Chars & haystack_data,

dbms/src/Functions/FunctionsStringSearch.h

@@ -11,8 +11,10 @@
 #include <Functions/FunctionHelpers.h>
 #include <Functions/IFunction.h>
 #include <IO/WriteHelpers.h>
+#include <Interpreters/Context.h>
 #include <common/StringRef.h>
 
+
 namespace DB
 {
 /** Search and replace functions in strings:
@@ -67,6 +69,7 @@ namespace ErrorCodes
     extern const int ILLEGAL_TYPE_OF_ARGUMENT;
     extern const int ILLEGAL_COLUMN;
     extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH;
+    extern const int FUNCTION_NOT_ALLOWED;
 }
 
 template <typename Impl, typename Name>
@@ -285,7 +288,13 @@ class FunctionsMultiStringSearch : public IFunction
 
 public:
     static constexpr auto name = Name::name;
-    static FunctionPtr create(const Context &) { return std::make_shared<FunctionsMultiStringSearch>(); }
+    static FunctionPtr create(const Context & context)
+    {
+        if (Impl::is_using_hyperscan && !context.getSettingsRef().allow_hyperscan)
+            throw Exception("Hyperscan functions are disabled, because setting 'allow_hyperscan' is set to 0", ErrorCodes::FUNCTION_NOT_ALLOWED);
+
+        return std::make_shared<FunctionsMultiStringSearch>();
+    }
 
     String getName() const override { return name; }
 

dbms/tests/queries/0_stateless/00927_disable_hyperscan.reference

+1
+1
+1
+1
+0
+1
+1
+1
+0

dbms/tests/queries/0_stateless/00927_disable_hyperscan.sql

+SET allow_hyperscan = 1;
+SELECT multiMatchAny(arrayJoin(['hello', 'world', 'hellllllllo', 'wororld', 'abc']), ['hel+o', 'w(or)*ld']);
+SET allow_hyperscan = 0;
+SELECT multiMatchAny(arrayJoin(['hello', 'world', 'hellllllllo', 'wororld', 'abc']), ['hel+o', 'w(or)*ld']); -- { serverError 446 }
+
+SELECT multiSearchAny(arrayJoin(['hello', 'world', 'hello, world', 'abc']), ['hello', 'world']);