Accept casting Scalars inside Sub operator (DO NOT MERGE)

src/vtlengine/AST/ASTConstructorModules/Expr.py

@@ -1886,6 +1886,8 @@ def visitSubspaceClauseItem(self, ctx: Parser.SubspaceClauseItemContext):
         op_node = ctx_list[1].getSymbol().text
         if isinstance(ctx_list[2], Parser.ScalarWithCastContext):
             right_node = Terminals().visitScalarWithCast(ctx_list[2])
+        elif isinstance(ctx_list[2], Parser.ScalarVarWithCastContext):
+            right_node = Terminals().visitScalarVarWithCast(ctx_list[2])
         elif isinstance(ctx_list[2], Parser.ScalarItemContext):
             right_node = Terminals().visitScalarItem(ctx_list[2])
         else:

src/vtlengine/AST/ASTConstructorModules/Terminals.py

@@ -512,6 +512,8 @@ def visitScalarItem(self, ctx: Parser.ScalarItemContext):
             return self.visitConstant(c)
         elif isinstance(c, Parser.ScalarWithCastContext):
             return self.visitScalarWithCast(c)
+        elif isinstance(c, Parser.ScalarVarWithCastContext):
+            return self.visitScalarVarWithCast(c)
         else:
             raise NotImplementedError
 
@@ -549,6 +551,40 @@ def visitScalarWithCast(self, ctx: Parser.ScalarWithCastContext):
             # AST_ASTCONSTRUCTOR.14
             raise NotImplementedError
 
+    def visitScalarVarWithCast(self, ctx: Parser.ScalarVarWithCastContext):
+        """
+        |  CAST LPAREN varID COMMA (basicScalarType) (COMMA STRING_CONSTANT)? RPAREN    #scalarVarWithCast
+        """  # noqa E501
+        ctx_list = list(ctx.getChildren())
+        c = ctx_list[0]
+
+        token = c.getSymbol()
+
+        op = token.text
+        var_node = self.visitVarID(ctx_list[2])
+        basic_scalar_type = [self.visitBasicScalarType(ctx_list[4])]
+
+        param_node = (
+            [
+                ParamConstant(
+                    type_="PARAM_CAST", value=ctx_list[6], **extract_token_info(ctx_list[6])
+                )
+            ]
+            if len(ctx_list) > 6
+            else []
+        )
+
+        if len(basic_scalar_type) == 1:
+            children_nodes = [var_node, basic_scalar_type[0]]
+
+            return ParamOp(
+                op=op, children=children_nodes, params=param_node, **extract_token_info(ctx)
+            )
+
+        else:
+            # AST_ASTCONSTRUCTOR.14
+            raise NotImplementedError
+
     def visitScalarSetType(self, ctx: Parser.ScalarSetTypeContext):
         """
         scalarSetType: SET ('<' scalarType '>')? ;

src/vtlengine/AST/Grammar/Vtl.g4

@@ -369,6 +369,7 @@ subspaceClauseItem:
 scalarItem:
  constant                                                                           #simpleScalar
  |  CAST LPAREN constant COMMA (basicScalarType) (COMMA STRING_CONSTANT)? RPAREN    #scalarWithCast
+ |  CAST LPAREN varID COMMA (basicScalarType) (COMMA STRING_CONSTANT)? RPAREN       #scalarVarWithCast
 ;
 /*END SUBSPACE CLAUSE*/
 /*----------------------------------------------END CLAUSE EXPRESSION--------------------------------------*/

src/vtlengine/AST/Grammar/lexer.py

@@ -1,4 +1,4 @@
-# Generated from Vtl.g4 by ANTLR 4.9.2
+# Generated from Vtl.g4 by ANTLR 4.9.3
 from antlr4 import (
     Lexer as ANTLRLexer,
     ATNDeserializer,
@@ -2132,6 +2132,7 @@ class Lexer(ANTLRLexer):
 
     def __init__(self, input=None, output: TextIO = sys.stdout):
         super().__init__(input, output)
+        self.checkVersion("4.9.3")
         self._interp = LexerATNSimulator(
             self, self.atn, self.decisionsToDFA, PredictionContextCache()
         )

src/vtlengine/AST/VtlVisitor.py

@@ -532,6 +532,10 @@ def visitSimpleScalar(self, ctx: Parser.SimpleScalarContext):
     def visitScalarWithCast(self, ctx: Parser.ScalarWithCastContext):
         return self.visitChildren(ctx)
 
+    # Visit a parse tree produced by Parser#scalarVarWithCast.
+    def visitScalarVarWithCast(self, ctx: Parser.ScalarVarWithCastContext):
+        return self.visitChildren(ctx)
+
     # Visit a parse tree produced by Parser#joinClauseWithoutUsing.
     def visitJoinClauseWithoutUsing(self, ctx: Parser.JoinClauseWithoutUsingContext):
         return self.visitChildren(ctx)

src/vtlengine/Exceptions/messages.py

@@ -264,6 +264,12 @@
         "message": "Component {comp_name} not found in Dataset {dataset_name}.",
         "description": "Occurs when a referenced component is missing from the Dataset.",
     },
+    "1-1-1-11": {
+        "message": "At op {op}: Sub comparison is not allowed between components. "
+        "Please check sub comparison in Dataset {dataset_name} is between a Component and a "
+        "Scalar/Constant.",
+        "description": "Occurs when a sub operation attempts to compare components.",
+    },
     "1-1-1-13": {
         "message": "At op {op}: Component {comp_name} role must be '{role_1}', found '{role_2}'.",
         "description": "Raised when a Dataset component has an unexpected role for the operation.",

src/vtlengine/Operators/Clause.py

@@ -1,3 +1,4 @@
+import re
 from copy import copy
 from typing import List, Type, Union
 
@@ -318,6 +319,12 @@ def validate(cls, operands: List[DataComponent], dataset: Dataset) -> Dataset:
             raise SemanticError("1-2-10", op=cls.op)
         for operand in operands:
             if operand.name not in dataset.components:
+                if re.match(r"__VDC_\d+__", operand.name):
+                    raise SemanticError(
+                        "1-1-1-11",
+                        op=cls.op,
+                        dataset_name=dataset_name,
+                    )
                 raise SemanticError(
                     "1-1-1-10",
                     op=cls.op,

src/vtlengine/Operators/Numeric.py

@@ -47,7 +47,7 @@ def apply_operation_component(cls, series: Any) -> Any:
         if cls.pc_func is not None:
             arr = series.values._pa_array
             return pd.Series(
-                pd.arrays.ArrowExtensionArray(cls.pc_func(arr)),  # type: ignore[attr-defined]
+                pd.arrays.ArrowExtensionArray(cls.pc_func(arr)),  # type: ignore[attr-defined,unused-ignore]
                 index=series.index,
             )
         return super().apply_operation_component(series)

tests/Cast/data/DataStructure/input/GH_539_1-1.json

@@ -0,0 +1,27 @@
+{
+  "datasets": [
+    {
+      "name": "DS_1",
+      "DataStructure": [
+        {
+          "name": "Id_1",
+          "type": "Integer",
+          "nullable": false,
+          "role": "Identifier"
+        },
+        {
+          "name": "Id_2",
+          "type": "String",
+          "nullable": false,
+          "role": "Identifier"
+        },
+        {
+          "name": "Me_1",
+          "type": "Integer",
+          "nullable": true,
+          "role": "Measure"
+        }
+      ]
+    }
+  ]
+}

tests/Cast/data/DataStructure/input/GH_539_1-2.json

@@ -0,0 +1,8 @@
+{
+  "scalars": [
+    {
+      "name": "sc_1",
+      "type": "Integer"
+    }
+  ]
+}

tests/Cast/data/DataStructure/input/GH_539_2-1.json

@@ -0,0 +1,27 @@
+{
+  "datasets": [
+    {
+      "name": "DS_1",
+      "DataStructure": [
+        {
+          "name": "Id_1",
+          "type": "Integer",
+          "nullable": false,
+          "role": "Identifier"
+        },
+        {
+          "name": "Id_2",
+          "type": "String",
+          "nullable": false,
+          "role": "Identifier"
+        },
+        {
+          "name": "Me_1",
+          "type": "Integer",
+          "nullable": true,
+          "role": "Measure"
+        }
+      ]
+    }
+  ]
+}

tests/Cast/data/DataStructure/output/GH_539_1-1.json

@@ -0,0 +1,21 @@
+{
+  "datasets": [
+    {
+      "name": "DS_r",
+      "DataStructure": [
+        {
+          "name": "Id_1",
+          "type": "Integer",
+          "nullable": false,
+          "role": "Identifier"
+        },
+        {
+          "name": "Me_1",
+          "type": "Integer",
+          "nullable": true,
+          "role": "Measure"
+        }
+      ]
+    }
+  ]
+}

tests/Cast/data/Dataset/input/GH_539_1-1.csv

@@ -0,0 +1,5 @@
+Id_1,Id_2,Me_1
+1,1,10
+2,1,20
+1,2,30
+2,2,40

tests/Cast/data/Dataset/output/GH_539_1-1.csv

@@ -0,0 +1,3 @@
+Id_1,Me_1
+1,10
+2,20

tests/Cast/data/vtl/GH_539_1.vtl

@@ -0,0 +1 @@
+DS_r <- DS_1[sub Id_2 = cast(sc_1, string)];

tests/Cast/data/vtl/GH_539_2.vtl

@@ -0,0 +1 @@
+DS_r <- DS_1[sub Id_2 = cast(Id_1, string)];

tests/Cast/test_cast.py

@@ -69,6 +69,27 @@ def test_GH_537_1(self):
 
         self.BaseTest(code, number_inputs, references_names=reference_names)
 
+    def test_GH_539_1(self):
+        """
+        Solves bug report in github issue #539: sub fails whith scalar as VarID casting
+        """
+        code = "GH_539_1"
+        number_inputs = 2
+        reference_names = ["1"]
+        scalars = {"sc_1": 1}
+
+        self.BaseTest(code, number_inputs, references_names=reference_names, scalars=scalars)
+
+    def test_GH_539_2(self):
+        """
+        Solves bug report in github issue #539: sub fails whith scalar as VarID casting
+        """
+        code = "GH_539_2"
+        number_inputs = 1
+        exception_code = "1-1-1-11"  # sub comparison is not allowed between components
+
+        self.NewSemanticExceptionTest(code, number_inputs, exception_code)
+
 
 # ===========================================================================
 # Comprehensive explicit cast tests (VTL 2.2) - Without mask