Merge pull request #26180 from ClickHouse/azat-subdomains

Merging #26041.

src/Common/HashTable/StringHashTable.h

@@ -237,7 +237,12 @@ public:
     // 1. Always memcpy 8 times bytes
     // 2. Use switch case extension to generate fast dispatching table
     // 3. Funcs are named callables that can be force_inlined
+    //
     // NOTE: It relies on Little Endianness
+    //
+    // NOTE: It requires padded to 8 bytes keys (IOW you cannot pass
+    // std::string here, but you can pass i.e. ColumnString::getDataAt()),
+    // since it copies 8 bytes at a time.
     template <typename Self, typename KeyHolder, typename Func>
     static auto ALWAYS_INLINE dispatch(Self & self, KeyHolder && key_holder, Func && func)
     {

src/Functions/IFunction.h

@@ -358,6 +358,10 @@ public:
       */
     virtual bool useDefaultImplementationForConstants() const { return false; }
 
+    /** Some arguments could remain constant during this implementation.
+      */
+    virtual ColumnNumbers getArgumentsThatAreAlwaysConstant() const { return {}; }
+
     /** If function arguments has single low cardinality column and all other arguments are constants, call function on nested column.
       * Otherwise, convert all low cardinality columns to ordinary columns.
       * Returns ColumnLowCardinality if at least one argument is ColumnLowCardinality.
@@ -367,10 +371,6 @@ public:
     /// If it isn't, will convert all ColumnLowCardinality arguments to full columns.
     virtual bool canBeExecutedOnLowCardinalityDictionary() const { return true; }
 
-    /** Some arguments could remain constant during this implementation.
-      */
-    virtual ColumnNumbers getArgumentsThatAreAlwaysConstant() const { return {}; }
-
     /** True if function can be called on default arguments (include Nullable's) and won't throw.
       * Counterexample: modulo(0, 0)
       */

src/Functions/URL/FirstSignificantSubdomainCustomImpl.h

@@ -41,6 +41,9 @@ public:
     String getName() const override { return name; }
     size_t getNumberOfArguments() const override { return 2; }
 
+    bool useDefaultImplementationForConstants() const override { return true; }
+    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
+
     DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
     {
         if (!isString(arguments[0].type))
@@ -65,9 +68,7 @@ public:
         const ColumnConst * column_tld_list_name = checkAndGetColumnConstStringOrFixedString(arguments[1].column.get());
         FirstSignificantSubdomainCustomLookup tld_lookup(column_tld_list_name->getValue<String>());
 
-        /// FIXME: convertToFullColumnIfConst() is suboptimal
-        auto column = arguments[0].column->convertToFullColumnIfConst();
-        if (const ColumnString * col = checkAndGetColumn<ColumnString>(*column))
+        if (const ColumnString * col = checkAndGetColumn<ColumnString>(*arguments[0].column))
         {
             auto col_res = ColumnString::create();
             vector(tld_lookup, col->getChars(), col->getOffsets(), col_res->getChars(), col_res->getOffsets());

tests/queries/0_stateless/01601_custom_tld.reference

@@ -22,3 +22,9 @@ foobar.com
 foobar.com
 foobar.com
 xx.blogspot.co.at
+-- www
+www.foo
+foo
+-- vector
+xx.blogspot.co.at
+

tests/queries/0_stateless/01601_custom_tld.sql

@@ -29,3 +29,11 @@ select cutToFirstSignificantSubdomainCustom('http://foobar.com', 'public_suffix_
 select cutToFirstSignificantSubdomainCustom('http://foobar.com/foo', 'public_suffix_list');
 select cutToFirstSignificantSubdomainCustom('http://bar.foobar.com/foo', 'public_suffix_list');
 select cutToFirstSignificantSubdomainCustom('http://xx.blogspot.co.at', 'public_suffix_list');
+
+select '-- www';
+select cutToFirstSignificantSubdomainCustomWithWWW('http://www.foo', 'public_suffix_list');
+select cutToFirstSignificantSubdomainCustom('http://www.foo', 'public_suffix_list');
+
+select '-- vector';
+select cutToFirstSignificantSubdomainCustom('http://xx.blogspot.co.at/' || toString(number), 'public_suffix_list') from numbers(1);
+select cutToFirstSignificantSubdomainCustom('there-is-no-such-domain' || toString(number), 'public_suffix_list') from numbers(1);

tests/queries/0_stateless/01940_custom_tld_sharding_key.reference

@@ -0,0 +1 @@
+foo.com

tests/queries/0_stateless/01940_custom_tld_sharding_key.sql

@@ -0,0 +1,2 @@
+select * from remote('127.{1,2}', view(select 'foo.com' key), cityHash64(key)) where key = cutToFirstSignificantSubdomainCustom('foo.com', 'public_suffix_list') settings optimize_skip_unused_shards=1, force_optimize_skip_unused_shards=1;
+select * from remote('127.{1,2}', view(select 'foo.com' key), cityHash64(key)) where key = cutToFirstSignificantSubdomainCustom('bar.com', 'public_suffix_list') settings optimize_skip_unused_shards=1, force_optimize_skip_unused_shards=1;