0.12.0-a6

.github/workflows/regression_suite_arm.yaml

@@ -38,7 +38,7 @@ jobs:
           python setup.py build_ext --inplace
 
       - name: "Authenticate to Google Cloud"
-        uses: google-github-actions/auth@v0.4.0
+        uses: google-github-actions/auth@v1
         with:
           credentials_json: '${{ secrets.GCP_KEY }}'
 

.github/workflows/regression_suite_mac_x86.yaml

@@ -37,7 +37,7 @@ jobs:
           python setup.py build_ext --inplace
 
       - name: "Authenticate to Google Cloud"
-        uses: google-github-actions/auth@v0.4.0
+        uses: google-github-actions/auth@v1
         with:
           credentials_json: '${{ secrets.GCP_KEY }}'
 

.github/workflows/regression_suite_windows.yaml

@@ -38,7 +38,7 @@ jobs:
           python setup.py build_ext --inplace
 
       - name: "Authenticate to Google Cloud"
-        uses: google-github-actions/auth@v0.4.0
+        uses: google-github-actions/auth@v1
         with:
           credentials_json: '${{ secrets.GCP_KEY }}'
 

opteryx/__version__.py

@@ -17,4 +17,4 @@
 """
 
 # __version__ = "0.4.0-alpha.6"
-__version__ = "0.12.0-alpha.3"
+__version__ = "0.12.0-alpha.6"

opteryx/components/binder/binder_visitor.py

@@ -395,11 +395,6 @@ def visit_join(self, node: Node, context: BindingContext) -> Tuple[Node, Binding
         if node.column:
             if not node.alias:
                 node.alias = f"UNNEST({node.column.query_column})"
-            # Find which relation on the left side of the plan the field is in
-            for left_relation_name in node.left_relation_names:
-                if context.schemas[left_relation_name].find_column(node.column.value):
-                    node.source = left_relation_name
-                    break
             # this is the column which is being unnested
             node.column, context = inner_binder(node.column, context, node.identity)
             # this is the column that is being created - find it from it's name
@@ -418,6 +413,25 @@ def visit_join(self, node: Node, context: BindingContext) -> Tuple[Node, Binding
         return node, context
 
     def visit_project(self, node: Node, context: BindingContext) -> Tuple[Node, BindingContext]:
+        columns = []
+        for column in node.columns:
+            if not column.node_type == NodeType.WILDCARD:
+                columns.append(column)
+            else:
+                schema = context.schemas[column.value[0]]
+
+                for schema_column in schema.columns:
+                    column_reference = Node(
+                        node_type=NodeType.IDENTIFIER,
+                        name=schema_column.name,
+                        schema_column=schema_column,
+                        type=schema_column.type,
+                        query_column=f"{column.value[0]}.{schema_column.name}",
+                    )
+                    columns.append(column_reference)
+
+        node.columns = columns
+
         node.columns, group_contexts = zip(
             *(inner_binder(col, context, node.identity) for col in node.columns)
         )

opteryx/components/logical_planner.py

@@ -306,7 +306,11 @@ def inner_query_planner(ast_branch):
             inner_plan.add_edge(previous_step_id, step_id)
 
     # projection
-    if not (len(_projection) == 1 and _projection[0].node_type == NodeType.WILDCARD):
+    if not (
+        len(_projection) == 1
+        and _projection[0].node_type == NodeType.WILDCARD
+        and _projection[0].value is None
+    ):
         project_step = LogicalPlanNode(node_type=LogicalPlanStepType.Project)
         project_step.columns = _projection
         previous_step_id, step_id = step_id, random_string()
@@ -443,7 +447,9 @@ def create_node_relation(relation):
         if function["alias"] is None:
             from opteryx.exceptions import UnnamedColumnError
 
-            raise UnnamedColumnError(f"Column created by {function_name} has no name.")
+            raise UnnamedColumnError(
+                f"Column created by {function_name} has no name, use AS to name the column."
+            )
 
         function = relation["relation"]["Table"]
         function_name = function["name"][0]["value"].upper()

opteryx/managers/expression/formatter.py

@@ -50,6 +50,8 @@ def format_expression(root):
                 return f"{root.value.upper()}({distinct}{format_expression(root.expression)}{order}{limit})"
             return f"{root.value.upper()}({distinct}{','.join([format_expression(e) for e in root.parameters])}{order})"
         if node_type == NodeType.WILDCARD:
+            if root.value:
+                return f"{root.value[0]}.*"
             return "*"
         if node_type == NodeType.BINARY_OPERATOR:
             _map = {

opteryx/models/node.py

@@ -29,8 +29,6 @@
 from typing import Any
 from typing import Dict
 
-from orso.tools import random_string
-
 
 class Node:
     def __init__(self, node_type: str = None, **kwargs: Any):
@@ -44,9 +42,6 @@ def __init__(self, node_type: str = None, **kwargs: Any):
                 Dynamic attributes for the node.
         """
         object.__setattr__(self, "_internal", {})  # Directly set _internal using the base method
-        object.__setattr__(
-            self, "_identity", random_string()
-        )  # Generate and set immutable identifier
         if node_type:
             self._internal["node_type"] = node_type
         for k, v in kwargs.items():
@@ -74,16 +69,6 @@ def properties(self) -> Dict[str, Any]:
         """
         return self._internal
 
-    @property
-    def identity(self) -> str:
-        """
-        Get the immutable unique identity for the Node.
-
-        Returns:
-            str: The immutable identity.
-        """
-        return self._identity
-
     def __getattr__(self, name: str) -> Any:
         """
         Retrieve attribute from the internal dictionary or the _identity.
@@ -107,9 +92,7 @@ def __setattr__(self, name: str, value: Any) -> None:
             value: Any
                 The value to set.
         """
-        if name == "identity":
-            raise AttributeError("The 'identity' attribute is immutable.")
-        elif name == "_internal":
+        if name == "_internal":
             self._internal = value
         else:
             self._is_valid_key(name)

tests/misc/test_node.py

@@ -119,58 +119,6 @@ def __str__(self):
     assert '"custom":"CustomObject"' in str(n)
 
 
-import pytest
-
-
-def test_node_identifier_uniqueness():
-    """
-    Test that each Node instance gets a unique identifier.
-    """
-    node1 = Node()
-    node2 = Node()
-    node3 = Node(node_type="test")
-
-    assert node1.identifier is not None
-    assert node2.identifier is not None
-    assert node3.identifier is not None
-
-    # Test that identifiers are unique
-    assert node1.identifier != node2.identifier
-    assert node1.identifier != node3.identifier
-    assert node2.identifier != node3.identifier
-
-
-def test_node_identifier_immutable():
-    """
-    Test that the identifier attribute cannot be changed.
-    """
-    node = Node()
-
-    # Attempt to change the identifier, should raise an error
-    with pytest.raises(AttributeError):
-        node.identifier = "new_id"
-
-    # Attempt to delete the identifier, should raise an error
-    with pytest.raises(AttributeError):
-        del node.identifier
-
-
-def test_node_identifier_consistency():
-    """
-    Test that the identifier remains the same across the lifetime of the Node object.
-    """
-    node = Node()
-    initial_identifier = node.identifier
-
-    # Perform some operations on the Node object
-    node._internal["new_key"] = "new_value"
-    node_copy = node.copy()
-
-    # Check that the identifier remains unchanged
-    assert node.identifier == initial_identifier
-    assert node_copy.identifier != initial_identifier  # Copy should have its own unique identifier
-
-
 if __name__ == "__main__":  # pragma: no cover
     from tests.tools import run_tests
 

tests/sql_battery/test_shapes_and_errors_battery.py

@@ -644,7 +644,7 @@
         ("SELECT * FROM $satellites INNER JOIN $planets WITH (NO_CACHE) USING (id)", 9, 27, None),
         ("SELECT * FROM $satellites JOIN $planets USING (id)", 9, 27, None),
         ("SELECT * FROM $astronauts CROSS JOIN UNNEST(missions) AS mission WHERE mission = 'Apollo 11'", 3, 20, None),
-        ("SELECT * FROM $astronauts CROSS JOIN UNNEST(missions)", 869, 20, None),
+        ("SELECT * FROM $astronauts CROSS JOIN UNNEST(missions) AS m", 869, 20, None),
         ("SELECT * FROM $planets INNER JOIN $satellites ON $planets.id = $satellites.planetId", 177, 28, None),
         ("SELECT DISTINCT planetId FROM $satellites LEFT OUTER JOIN $planets ON $satellites.planetId = $planets.id", 7, 1, None),
         ("SELECT DISTINCT planetId FROM $satellites LEFT JOIN $planets ON $satellites.planetId = $planets.id", 7, 1, None),

tests/storage/test_sql_biq_query.py

@@ -14,7 +14,7 @@
 
 
 # skip to reduce billing
-# @skip_if(is_arm() or is_windows() or is_mac())
+@skip_if(is_arm() or is_windows() or is_mac())
 def test_bigquery_storage():
     from sqlalchemy.engine import create_engine