[refactor](functions) clean some ip functions code and make IS_IP_ADD…

…RESS_IN_RANGE DEPENDS_ON_ARGUMENT (apache#35239)
sjyango · Aug 7, 2024 · f871e59 · f871e59
1 parent 0a3874f
commit f871e59
Show file tree

Hide file tree

Showing 10 changed files with 117 additions and 181 deletions.
diff --git a/be/src/agent/be_exec_version_manager.h b/be/src/agent/be_exec_version_manager.h
@@ -70,7 +70,12 @@ class BeExecVersionManager {
  *    f. shrink some function's nullable mode.
  *    g. do local merge of remote runtime filter
  *    h. "now": ALWAYS_NOT_NULLABLE -> DEPEND_ON_ARGUMENTS
-*/
+ *
+ * 5: start from doris 3.0.0
+ *    a. change the impl of percentile (need fix)
+ *    b. clear old version of version 3->4
+ *    c. change FunctionIsIPAddressInRange from AlwaysNotNullable to DependOnArguments
+ */
 constexpr inline int BeExecVersionManager::max_be_exec_version = 4;
 constexpr inline int BeExecVersionManager::min_be_exec_version = 0;
 

diff --git a/be/src/vec/columns/column.h b/be/src/vec/columns/column.h
@@ -22,10 +22,9 @@
 
 #include <fmt/format.h>
 #include <glog/logging.h>
-#include <stdint.h>
 #include <sys/types.h>
 
-#include <algorithm>
+#include <cstdint>
 #include <functional>
 #include <ostream>
 #include <string>
@@ -743,6 +742,7 @@ struct IsMutableColumns<> {
     static const bool value = true;
 };
 
+// prefer assert_cast than check_and_get
 template <typename Type>
 const Type* check_and_get_column(const IColumn& column) {
     return typeid_cast<const Type*>(&column);

diff --git a/be/src/vec/functions/function_ip.cpp b/be/src/vec/functions/function_ip.cpp
@@ -17,6 +17,8 @@
 
 #include "vec/functions/function_ip.h"
 
+#include "vec/functions/simple_function_factory.h"
+
 namespace doris::vectorized {
 
 void register_function_ip(SimpleFunctionFactory& factory) {

diff --git a/be/src/vec/functions/function_ip.h b/be/src/vec/functions/function_ip.h
diff --git a/be/test/vec/function/function_ip_test.cpp b/be/test/vec/function/function_ip_test.cpp
@@ -56,16 +56,16 @@ TEST(FunctionIpTest, FunctionIsIPAddressInRangeTest) {
     {
         // vector vs vector
         InputTypeSet input_types = {TypeIndex::String, TypeIndex::String};
-        static_cast<void>(check_function<DataTypeUInt8, false>(func_name, input_types, data_set));
+        static_cast<void>(check_function<DataTypeUInt8, true>(func_name, input_types, data_set));
     }
 
     {
         // vector vs scalar
         InputTypeSet input_types = {TypeIndex::String, Consted {TypeIndex::String}};
         for (const auto& line : data_set) {
             DataSet const_cidr_dataset = {line};
-            static_cast<void>(check_function<DataTypeUInt8, false>(func_name, input_types,
-                                                                   const_cidr_dataset));
+            static_cast<void>(check_function<DataTypeUInt8, true>(func_name, input_types,
+                                                                  const_cidr_dataset));
         }
     }
 
@@ -74,8 +74,8 @@ TEST(FunctionIpTest, FunctionIsIPAddressInRangeTest) {
         InputTypeSet input_types = {Consted {TypeIndex::String}, TypeIndex::String};
         for (const auto& line : data_set) {
             DataSet const_addr_dataset = {line};
-            static_cast<void>(check_function<DataTypeUInt8, false>(func_name, input_types,
-                                                                   const_addr_dataset));
+            static_cast<void>(check_function<DataTypeUInt8, true>(func_name, input_types,
+                                                                  const_addr_dataset));
         }
     }
 }

diff --git a/...ain/java/org/apache/doris/nereids/trees/expressions/functions/scalar/Ipv6NumToString.java b/...ain/java/org/apache/doris/nereids/trees/expressions/functions/scalar/Ipv6NumToString.java
@@ -41,8 +41,7 @@ public class Ipv6NumToString extends ScalarFunction
     public static final List<FunctionSignature> SIGNATURES = ImmutableList.of(
             FunctionSignature.ret(VarcharType.SYSTEM_DEFAULT).args(VarcharType.SYSTEM_DEFAULT),
             FunctionSignature.ret(StringType.INSTANCE).args(StringType.INSTANCE),
-            FunctionSignature.ret(VarcharType.SYSTEM_DEFAULT).args(IPv6Type.INSTANCE),
-            FunctionSignature.ret(StringType.INSTANCE).args(IPv6Type.INSTANCE));
+            FunctionSignature.ret(VarcharType.SYSTEM_DEFAULT).args(IPv6Type.INSTANCE));
 
     public Ipv6NumToString(Expression arg0) {
         super("ipv6_num_to_string", arg0);

diff --git a/.../java/org/apache/doris/nereids/trees/expressions/functions/scalar/IsIpAddressInRange.java b/.../java/org/apache/doris/nereids/trees/expressions/functions/scalar/IsIpAddressInRange.java
@@ -19,8 +19,8 @@
 
 import org.apache.doris.catalog.FunctionSignature;
 import org.apache.doris.nereids.trees.expressions.Expression;
-import org.apache.doris.nereids.trees.expressions.functions.AlwaysNotNullable;
 import org.apache.doris.nereids.trees.expressions.functions.ExplicitlyCastableSignature;
+import org.apache.doris.nereids.trees.expressions.functions.PropagateNullable;
 import org.apache.doris.nereids.trees.expressions.shape.BinaryExpression;
 import org.apache.doris.nereids.trees.expressions.visitor.ExpressionVisitor;
 import org.apache.doris.nereids.types.BooleanType;
@@ -36,7 +36,7 @@
  * scalar function `is_ip_address_in_range`
  */
 public class IsIpAddressInRange extends ScalarFunction
-        implements BinaryExpression, ExplicitlyCastableSignature, AlwaysNotNullable {
+        implements BinaryExpression, ExplicitlyCastableSignature, PropagateNullable {
 
     public static final List<FunctionSignature> SIGNATURES = ImmutableList.of(
             FunctionSignature.ret(BooleanType.INSTANCE).args(VarcharType.SYSTEM_DEFAULT, VarcharType.SYSTEM_DEFAULT),

diff --git a/gensrc/script/doris_builtins_functions.py b/gensrc/script/doris_builtins_functions.py
@@ -2050,7 +2050,6 @@
         [['ipv6_num_to_string','inet6_ntoa'], 'VARCHAR', ['VARCHAR'], 'ALWAYS_NULLABLE'],     
         [['ipv6_num_to_string','inet6_ntoa'], 'STRING', ['STRING'], 'ALWAYS_NULLABLE'],
         [['ipv6_num_to_string','inet6_ntoa'], 'VARCHAR', ['IPV6'], 'ALWAYS_NULLABLE'],
-        [['ipv6_num_to_string','inet6_ntoa'], 'STRING', ['IPV6'], 'ALWAYS_NULLABLE'],
         [['ipv6_string_to_num'], 'VARCHAR', ['VARCHAR'], 'ALWAYS_NOT_NULLABLE'],
         [['ipv6_string_to_num'], 'STRING', ['STRING'], 'ALWAYS_NOT_NULLABLE'],
         [['ipv6_string_to_num_or_default'], 'VARCHAR', ['VARCHAR'], 'ALWAYS_NOT_NULLABLE'],
@@ -2065,12 +2064,12 @@
         [['is_ipv4_string'], 'BOOLEAN', ['STRING'], ''],
         [['is_ipv6_string'], 'BOOLEAN', ['VARCHAR'], ''],
         [['is_ipv6_string'], 'BOOLEAN', ['STRING'], ''],
-        [['is_ip_address_in_range'], 'BOOLEAN', ['VARCHAR', 'VARCHAR'], 'ALWAYS_NOT_NULLABLE'],
-        [['is_ip_address_in_range'], 'BOOLEAN', ['STRING', 'STRING'], 'ALWAYS_NOT_NULLABLE'],
-        [['ipv4_cidr_to_range'], 'STRUCT<IPV4, IPV4>', ['IPV4', 'SMALLINT'], ''],
-        [['ipv6_cidr_to_range'], 'STRUCT<IPV6, IPV6>', ['IPV6', 'SMALLINT'], ''],
-        [['ipv6_cidr_to_range'], 'STRUCT<IPV6, IPV6>', ['VARCHAR', 'SMALLINT'], ''],
-        [['ipv6_cidr_to_range'], 'STRUCT<IPV6, IPV6>', ['STRING', 'SMALLINT'], ''],
+        [['is_ip_address_in_range'], 'BOOLEAN', ['VARCHAR', 'VARCHAR'], 'DEPEND_ON_ARGUMENT'],
+        [['is_ip_address_in_range'], 'BOOLEAN', ['STRING', 'STRING'], 'DEPEND_ON_ARGUMENT'],
+        [['ipv4_cidr_to_range'], 'STRUCT<IPV4, IPV4>', ['IPV4', 'SMALLINT'], 'DEPEND_ON_ARGUMENT'],
+        [['ipv6_cidr_to_range'], 'STRUCT<IPV6, IPV6>', ['IPV6', 'SMALLINT'], 'DEPEND_ON_ARGUMENT'],
+        [['ipv6_cidr_to_range'], 'STRUCT<IPV6, IPV6>', ['VARCHAR', 'SMALLINT'], 'DEPEND_ON_ARGUMENT'],
+        [['ipv6_cidr_to_range'], 'STRUCT<IPV6, IPV6>', ['STRING', 'SMALLINT'], 'DEPEND_ON_ARGUMENT'],
         [['to_ipv4'], 'IPV4', ['VARCHAR'], 'ALWAYS_NOT_NULLABLE'],
         [['to_ipv4'], 'IPV4', ['STRING'], 'ALWAYS_NOT_NULLABLE'],
         [['to_ipv4_or_default'], 'IPV4', ['VARCHAR'], 'ALWAYS_NOT_NULLABLE'],

diff --git a/...on-test/data/query_p0/sql_functions/ip_functions/test_is_ip_address_in_range_function.out b/...on-test/data/query_p0/sql_functions/ip_functions/test_is_ip_address_in_range_function.out
@@ -1,3 +1,4 @@
+-- This file is automatically generated. You should know what you did if you want to edit this
 -- !sql --
 1	true
 2	false
@@ -80,4 +81,14 @@
 23	false
 24	false
 25	false
-26	false
+26	false
+
+-- !sql --
+\N
+
+-- !sql --
+\N
+
+-- !sql --
+\N
+
diff --git a/...st/suites/query_p0/sql_functions/ip_functions/test_is_ip_address_in_range_function.groovy b/...st/suites/query_p0/sql_functions/ip_functions/test_is_ip_address_in_range_function.groovy
@@ -73,22 +73,9 @@ suite("test_is_ip_address_in_range_function") {
     // scalar vs vector
     qt_sql "select id, is_ip_address_in_range('192.168.100.0', cidr) from test_is_ip_address_in_range_function order by id"
 
-    test {
-        sql "SELECT is_ip_address_in_range('::ffff:192.168.0.1', NULL)"
-        // check exception message contains
-        exception "The arguments of function is_ip_address_in_range must be String, not NULL"
-    }
+    qt_sql "SELECT is_ip_address_in_range('::ffff:192.168.0.1', NULL)"
 
-    test {
-        sql "SELECT is_ip_address_in_range(NULL, '::ffff:192.168.0.4/128')"
-        // check exception message contains
-        exception "The arguments of function is_ip_address_in_range must be String, not NULL"
-    }
+    qt_sql "SELECT is_ip_address_in_range(NULL, '::ffff:192.168.0.4/128')"
 
-    test {
-        sql "SELECT is_ip_address_in_range(NULL, NULL)"
-        // check exception message contains
-        exception "The arguments of function is_ip_address_in_range must be String, not NULL"
-    }
-    sql """ DROP TABLE IF EXISTS test_is_ip_address_in_range_function """
+    qt_sql "SELECT is_ip_address_in_range(NULL, NULL)"
 }