#2193 Optimizer isn't cost-based

opteryx/__version__.py

@@ -1,4 +1,4 @@
-__build__ = 946
+__build__ = 947
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

opteryx/connectors/__init__.py

@@ -13,6 +13,7 @@
 from opteryx.connectors.disk_connector import DiskConnector
 from opteryx.connectors.gcp_cloudstorage_connector import GcpCloudStorageConnector
 from opteryx.connectors.gcp_firestore_connector import GcpFireStoreConnector
+from opteryx.connectors.iceberg_connector import IcebergConnector
 from opteryx.connectors.mongodb_connector import MongoDbConnector
 from opteryx.connectors.sql_connector import SqlConnector
 from opteryx.shared import MaterializedDatasets
@@ -35,6 +36,7 @@
     "DiskConnector",
     "GcpCloudStorageConnector",
     "GcpFireStoreConnector",
+    "IcebergConnector",
     "MongoDbConnector",
     "SqlConnector",
 )

opteryx/connectors/disk_connector.py

@@ -37,6 +37,10 @@
 
 
 class DiskConnector(BaseConnector, Cacheable, Partitionable, PredicatePushable, Asynchronous):
+    """
+    Connector for reading datasets from files on local storage.
+    """
+
     __mode__ = "Blob"
     __type__ = "LOCAL"
 

opteryx/connectors/file_connector.py

@@ -28,6 +28,10 @@
 
 
 class FileConnector(BaseConnector, PredicatePushable):
+    """
+    Connector for reading datasets from a file.
+    """
+
     __mode__ = "Blob"
     __type__ = "FILE"
     _byte_array: Optional[bytes] = None  # Instance attribute to store file bytes

opteryx/connectors/iceberg_connector.py

@@ -0,0 +1,52 @@
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# See the License at http://www.apache.org/licenses/LICENSE-2.0
+# Distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND.
+
+"""
+Arrow Reader
+
+Used to read datasets registered using the register_arrow or register_df functions.
+"""
+
+import pyarrow
+from orso.schema import FlatColumn
+from orso.schema import RelationSchema
+
+from opteryx.connectors.base.base_connector import DEFAULT_MORSEL_SIZE
+from opteryx.connectors.base.base_connector import BaseConnector
+from opteryx.shared import MaterializedDatasets
+from opteryx.utils import arrow
+
+
+class IcebergConnector(BaseConnector):
+    __mode__ = "Internal"
+    __type__ = "ARROW"
+
+    def __init__(self, *args, **kwargs):
+        BaseConnector.__init__(self, **kwargs)
+
+        self.dataset = self.dataset.lower()
+        self._datasets = MaterializedDatasets()
+
+    def get_dataset_schema(self) -> RelationSchema:
+        dataset = self._datasets[self.dataset]
+        arrow_schema = dataset.schema
+
+        self.schema = RelationSchema(
+            name=self.dataset,
+            columns=[FlatColumn.from_arrow(field) for field in arrow_schema],
+        )
+
+        return self.schema
+
+    def read_dataset(self, columns: list = None, **kwargs) -> pyarrow.Table:
+        dataset = self._datasets[self.dataset]
+
+        batch_size = DEFAULT_MORSEL_SIZE // (dataset.nbytes / dataset.num_rows)
+
+        for batch in dataset.to_batches(max_chunksize=batch_size):
+            morsel = pyarrow.Table.from_batches([batch], schema=dataset.schema)
+            if columns:
+                morsel = arrow.post_read_projector(morsel, columns)
+            yield morsel

opteryx/planner/__init__.py

@@ -123,9 +123,9 @@ def query_planner(
     from opteryx.models import QueryProperties
     from opteryx.planner.ast_rewriter import do_ast_rewriter
     from opteryx.planner.binder import do_bind_phase
-    from opteryx.planner.cost_based_optimizer import do_cost_based_optimizer
     from opteryx.planner.logical_planner import apply_visibility_filters
     from opteryx.planner.logical_planner import do_logical_planning_phase
+    from opteryx.planner.optimizer import do_optimizer
     from opteryx.planner.physical_planner import create_physical_plan
     from opteryx.planner.sql_rewriter import do_sql_rewrite
     from opteryx.third_party import sqloxide
@@ -182,7 +182,7 @@ def query_planner(
     statistics.time_planning_binder += time.monotonic_ns() - start
 
     start = time.monotonic_ns()
-    optimized_plan = do_cost_based_optimizer(bound_plan, statistics)
+    optimized_plan = do_optimizer(bound_plan, statistics)
     statistics.time_planning_optimizer += time.monotonic_ns() - start
 
     # before we write the new optimizer and execution engine, convert to a V1 plan

opteryx/planner/executor/v2_coordinator.py

@@ -75,9 +75,9 @@ def query_planner(
     from opteryx.models import QueryProperties
     from opteryx.planner.ast_rewriter import do_ast_rewriter
     from opteryx.planner.binder import do_bind_phase
-    from opteryx.planner.cost_based_optimizer import do_cost_based_optimizer
     from opteryx.planner.logical_planner import LogicalPlan
     from opteryx.planner.logical_planner import do_logical_planning_phase
+    from opteryx.planner.optimizer import do_optimizer
     from opteryx.planner.sql_rewriter import do_sql_rewrite
     from opteryx.planner.temporary_physical_planner import create_physical_plan
     from opteryx.third_party import sqloxide
@@ -130,7 +130,7 @@ def query_planner(
             # common_table_expressions=ctes,
         )
 
-        optimized_plan = do_cost_based_optimizer(bound_plan, statistics)
+        optimized_plan = do_optimizer(bound_plan, statistics)
 
         # before we write the new optimizer and execution engine, convert to a V1 plan
         query_properties = QueryProperties(qid=qid, variables=conn.context.variables)

opteryx/planner/cost_based_optimizer/__init__.py → opteryx/planner/optimizer/__init__.py

@@ -48,22 +48,22 @@
 - Context: Maintains the state during optimization, including the pre-optimized and optimized plans.
 
 The `CostBasedOptimizerVisitor` class orchestrates the optimization process by applying each strategy
-in sequence. The `do_cost_based_optimizer` function serves as the entry point for optimizing a logical plan.
+in sequence. The `do_optimizer` function serves as the entry point for optimizing a logical plan.
 
 Example Usage:
-    optimized_plan = do_cost_based_optimizer(logical_plan)
+    optimized_plan = do_optimizer(logical_plan)
 
 This module aims to enhance query performance through systematic and incremental optimization steps.
 """
 
 from opteryx.config import DISABLE_OPTIMIZER
 from opteryx.models import QueryStatistics
-from opteryx.planner.cost_based_optimizer.strategies import *
 from opteryx.planner.logical_planner import LogicalPlan
+from opteryx.planner.optimizer.strategies import *
 
 from .strategies.optimization_strategy import OptimizerContext
 
-__all__ = "do_cost_based_optimizer"
+__all__ = "do_optimizer"
 
 
 class CostBasedOptimizerVisitor:
@@ -134,9 +134,9 @@ def optimize(self, plan: LogicalPlan) -> LogicalPlan:
         return current_plan
 
 
-def do_cost_based_optimizer(plan: LogicalPlan, statistics: QueryStatistics) -> LogicalPlan:
+def do_optimizer(plan: LogicalPlan, statistics: QueryStatistics) -> LogicalPlan:
     """
-    Perform cost-based optimization on the given logical plan.
+    Perform optimization on the given logical plan.
 
     Parameters:
         plan (LogicalPlan): The logical plan to optimize.

pyproject.toml

@@ -45,5 +45,5 @@ select = ["SIM"]
 ignore = []
 
 [tool.ruff.lint.per-file-ignores]
-"**/cost_based_optimizer/**" = ["SIM102"]
+"**/optimizer/**" = ["SIM102"]
 "opteryx/managers/expression/ops.py" = ["SIM118"]

tests/catalog/test_iceberg.py

@@ -1,11 +1,17 @@
 
 import os
+import sys
+
+sys.path.insert(1, os.path.join(sys.path[0], "../.."))
+
+from tests.tools import is_arm, is_mac, is_windows, skip_if
 import opteryx
-from pyiceberg.catalog.sql import SqlCatalog
-from pyiceberg.catalog import load_catalog
+from opteryx.connectors import DiskConnector
+from opteryx.connectors import IcebergConnector
 
 BASE_PATH: str = "tmp/iceberg"
 
+@skip_if(is_arm() or is_windows() or is_mac())
 def set_up_iceberg():
     """
     Set up a local Iceberg catalog for testing with NVD data.
@@ -19,6 +25,9 @@ def set_up_iceberg():
     Returns:
         str: Path to the created Iceberg table.
     """
+
+    from pyiceberg.catalog.sql import SqlCatalog
+
     # Clean up previous test runs if they exist
     if os.path.exists(BASE_PATH):
         import shutil
@@ -47,15 +56,28 @@ def set_up_iceberg():
     print(f"Iceberg table set up at {BASE_PATH}")
     return BASE_PATH
 
-set_up_iceberg()
 
-catalog = load_catalog(
-        "default",
-        **{
-            "uri": f"sqlite:///{BASE_PATH}/pyiceberg_catalog.db",
-            "warehouse": f"file://{BASE_PATH}",
-        },
-)
+def test_iceberg_basic():
+
+    from pyiceberg.catalog import load_catalog
+
+    set_up_iceberg()
+
+    catalog = load_catalog(
+            "default",
+            **{
+                "uri": f"sqlite:///{BASE_PATH}/pyiceberg_catalog.db",
+                "warehouse": f"file://{BASE_PATH}",
+            },
+    )
+
+    opteryx.register_store("iceberg", IcebergConnector, io=DiskConnector)
+
+    table = catalog.load_table("iceberg.tweets")
+    print(table.scan().to_arrow())
+
+
+if __name__ == "__main__":  # pragma: no cover
+    from tests.tools import run_tests
 
-table = catalog.load_table("iceberg.tweets")
-print(table.scan().to_arrow())
+    run_tests()