Add test for all required list function except for list_zip, modify list_insersect and list_where to make them act similar to those in duckdb

Author: Jasonlmx

src/function/list/list_select_function.cpp

@@ -24,18 +24,23 @@ struct ListSelect {
         auto rightDataVector = common::ListVector::getDataVector(&rightVector);
         auto rightPos = right.offset;
         for (auto i=0u; i < right.size; i++) {
-            auto leftIndexPos=rightDataVector->getValue<long long>(rightPos+i);
+            auto leftIndexPos=rightDataVector->getValue<long long>(rightPos+i)-1;
             if ((leftIndexPos<0)||(leftIndexPos>=left.size)) {
-                throw BinderException(stringFormat("LIST_SELECTION encounters index out of range : {} , min: {}, max: {}", leftIndexPos, 0, left.size-1));
+                // append null to result if out of index
+                resultDataVector->setNull(resultPos++, true);
+            } else {
+                resultDataVector->copyFromVectorData(resultPos++, leftDataVector, leftIndexPos);
             }
-            resultDataVector->copyFromVectorData(resultPos++, leftDataVector, leftIndexPos);
         }
     }
 };
 static std::unique_ptr<FunctionBindData> bindFunc(const ScalarBindFuncInput& input) {
     std::vector<LogicalType> types;
     types.push_back(input.arguments[0]->getDataType().copy());
     types.push_back(input.arguments[1]->getDataType().copy());
+    if (types[0].getPhysicalType()!=PhysicalTypeID::LIST) {
+        throw BinderException("LIST_SELECT expecting argument type: LIST of ANY, LIST of INT");
+    }
     if (types[1].getPhysicalType()!=PhysicalTypeID::LIST) {
         throw BinderException(ExceptionMessage::listFunctionIncompatibleChildrenType(
             ListIntersectFunction::name, types[0].toString(), types[1].toString()));

src/function/list/list_where_function.cpp

@@ -17,29 +17,34 @@ struct ListWhere {
     static void operation(common::list_entry_t& left, common::list_entry_t& right,
         common::list_entry_t& result, common::ValueVector& leftVector,
         common::ValueVector& rightVector, common::ValueVector& resultVector) {
-        if (right.size!=left.size) {
-            throw BinderException(stringFormat("LIST_WHERE expecting lists of same size, receiving size {} and size {}", left.size, left.size));
-        }
         auto leftDataVector = common::ListVector::getDataVector(&leftVector);
         auto leftPos = left.offset;
         auto rightDataVector = common::ListVector::getDataVector(&rightVector);
         auto rightPos = right.offset;
         list_size_t resultSize=0;
         std::vector<bool> maskListBools;
         for (auto i=0u; i < right.size; i++) {
+            if (rightDataVector->isNull(rightPos+i)) {
+                throw BinderException("NULLs are not allowed as list elements in the second input parameter.");
+            }
             auto maskBool=rightDataVector->getValue<bool>(rightPos+i);
-            maskListBools.push_back(maskBool);
-            if (maskBool) {
+            if (maskBool){
                 resultSize++;
             }
+            maskListBools.push_back(maskBool);
         }
         result = common::ListVector::addList(&resultVector, resultSize);
         auto resultDataVector = common::ListVector::getDataVector(&resultVector);
         auto resultPos = result.offset;
         for (auto i=0u; i < right.size; i++) {
             auto maskBool=maskListBools.at(i);
             if (maskBool) {
-                resultDataVector->copyFromVectorData(resultPos++, leftDataVector, leftPos+i);
+                if (leftPos+i<left.size) {
+                    resultDataVector->copyFromVectorData(resultPos++, leftDataVector, leftPos+i);
+                } else {
+                    resultDataVector->setNull(resultPos++,true);
+                }
+                
             }
         }
     }
@@ -48,14 +53,17 @@ static std::unique_ptr<FunctionBindData> bindFunc(const ScalarBindFuncInput& inp
     std::vector<LogicalType> types;
     types.push_back(input.arguments[0]->getDataType().copy());
     types.push_back(input.arguments[1]->getDataType().copy());
+    if (types[0].getPhysicalType()!=PhysicalTypeID::LIST) {
+        throw BinderException("LIST_WHERE expecting argument type: LIST of ANY, LIST of BOOL");
+    }
     if (types[1].getPhysicalType()!=PhysicalTypeID::LIST) {
         throw BinderException(ExceptionMessage::listFunctionIncompatibleChildrenType(
             ListIntersectFunction::name, types[0].toString(), types[1].toString()));
     } else {
         auto thisExtraTypeInfo=types[1].getExtraTypeInfo();
         auto thisListTypeInfo=ku_dynamic_cast<const ListTypeInfo*>(thisExtraTypeInfo);
         if (thisListTypeInfo->getChildType().getPhysicalType()!=PhysicalTypeID::BOOL) {
-            throw BinderException("LIST_SELECT expecting argument type: LIST of ANY, LIST of BOOL");
+            throw BinderException("LIST_WHERE expecting argument type: LIST of ANY, LIST of BOOL");
         }
     }
     return std::make_unique<FunctionBindData>(std::move(types), types[0].copy());

test/test_files/function/list.test

@@ -2306,3 +2306,80 @@ False
 ---- 1
 [False]
 
+-CASE ListSelect
+-STATEMENT RETURN list_select([1,2,3,4,5],[1,2,3])
+---- 1
+[1,2,3]
+
+-STATEMENT RETURN list_select(['s','t','d','u','e','f','~'],[2,2,3,4,5])
+---- 1
+[t,t,d,u,e]
+
+-STATEMENT RETURN list_select([1,2,3,4,5],[1.0,2.0,3.0,5.0])
+---- error
+Binder exception: LIST_SELECT expecting argument type: LIST of ANY, LIST of INT
+
+-STATEMENT RETURN list_select([[1],[2,3],[2,3,4],[1,2,3],[1]],[1,2,3,4,2,3,2])
+---- 1 
+[[1],[2,3],[2,3,4],[1,2,3],[2,3],[2,3,4],[2,3]]
+
+-STATEMENT Return list_select([1,2,3,4,5],[0,1,2,3,4,5,6])
+---- 1
+[,1,2,3,4,5,]
+
+-STATEMENT Return list_select([1,2,3,4,5],[null])
+---- 1
+[]
+
+-STATEMENT Return list_select([1,2,3,4,5],null)
+---- error
+Binder exception: Cannot bind LIST_INTERSECT with parameter type INT64[] and ANY.
+
+-STATEMENT Return list_select(null,[0,1,2,3,4,5])
+---- error
+Binder exception: LIST_SELECT expecting argument type: LIST of ANY, LIST of INT
+
+-STATEMENT RETURN list_select([{q:1, p:2},{q:2, p:3},{a: 3, q: 4}],[1,2,2,1])
+---- 1
+[{a: 1, q: 2},{a: 2, q: 3},{a: 2, q: 3},{a: 1, q: 2}]
+
+-STATEMENT RETURN list_select([1,2,3,4,5],[null, 2, null, 4, 5])
+---- 1
+[,2,,4,5]
+
+-CASE ListWhere
+-STATEMENT RETURN list_where([1,2,3,4,5],[true, false, true, true, false])
+---- 1
+[1,3,4]
+
+-STATEMENT RETURN list_where([{a:2, b:3},{a:3, b:5},{a:4, c:1},{a:21, u:33}],[true, false, true, true, false])
+---- 1
+[{a: 2, u: 3},{a: 4, u: 1},{a: 21, u: 33}]
+
+-STATEMENT RETURN list_where([1,2,3,4,null],[true, false, true, true, true])
+---- 1
+[1,3,4,]
+
+-STATEMENT RETURN list_where([1,2,3,4],[true, false, true])
+---- 1
+[1,3]
+
+-STATEMENT RETURN list_where([1,2,3,4],[true, false, true,false, true, true])
+---- 1
+[1,3,,]
+
+-STATEMENT RETURN list_where(['a','b','c','6','r'],[false, false, true,false, true])
+---- 1
+[c,r]
+
+-STATEMENT RETURN list_where(null,[false, false, true,false, true])
+---- error
+Binder exception: LIST_WHERE expecting argument type: LIST of ANY, LIST of BOOL
+
+-STATEMENT RETURN list_where([1,2,3,4],null)
+---- error
+Binder exception: Cannot bind LIST_INTERSECT with parameter type INT64[] and ANY.
+
+-STATEMENT RETURN list_where([1,2,3,4],[true, false, null, true])
+---- error
+Binder exception: NULLs are not allowed as list elements in the second input parameter.