pytrilogy 0.0.3.95__tar.gz → 0.0.3.96__tar.gz

{pytrilogy-0.0.3.95/pytrilogy.egg-info → pytrilogy-0.0.3.96}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pytrilogy
-Version: 0.0.3.95
+Version: 0.0.3.96
 Summary: Declarative, typed query language that compiles to SQL.
 Home-page: 
 Author: 
@@ -136,11 +136,11 @@ Versus SQL, Trilogy aims to:
 
 | Backend | Status | Notes |
 |---------|--------|-------|
-| **BigQuery** | ✅ Core | Full support |
-| **DuckDB** | ✅ Core | Full support |
-| **Snowflake** | ✅ Core | Full support |
-| **SQL Server** | ⚠️ Experimental | Limited testing |
-| **Presto** | ⚠️ Experimental | Limited testing |
+| **BigQuery** | Core | Full support |
+| **DuckDB** | Core | Full support |
+| **Snowflake** | Core | Full support |
+| **SQL Server** | Experimental | Limited testing |
+| **Presto** | Experimental | Limited testing |
 
 ## Examples
 
@@ -311,7 +311,30 @@ trilogy fmt <path to trilogy file>
 - [Public model repository](https://github.com/trilogydata/trilogy-public-models) - Great place for modeling examples
 - [Full documentation](https://trilogydata.dev/)
 
-## Syntax Reference
+## Python API Integration
+
+### Root Imports
+
+Are stable and should be sufficient for executing code from Trilogy as text.
+
+```python
+from pytrilogy import Executor, Dialect
+```
+
+### Authoring Imports
+
+Are also stable, and should be used for cases which programatically generate Trilogy statements without a base text format 
+or need to process/transform parsed code in more complicated ways.
+
+```python
+from pytrilogy.authoring import Concept, Function, ...
+```
+
+### Other Imports
+
+Are likely to be unstable. Open an issue if you need to take dependencies on other modules outside those two paths. 
+
+## Trilogy Syntax Reference
 
 ### Import
 ```sql

{pytrilogy-0.0.3.95 → pytrilogy-0.0.3.96}/README.md

@@ -98,11 +98,11 @@ Versus SQL, Trilogy aims to:
 
 | Backend | Status | Notes |
 |---------|--------|-------|
-| **BigQuery** | ✅ Core | Full support |
-| **DuckDB** | ✅ Core | Full support |
-| **Snowflake** | ✅ Core | Full support |
-| **SQL Server** | ⚠️ Experimental | Limited testing |
-| **Presto** | ⚠️ Experimental | Limited testing |
+| **BigQuery** | Core | Full support |
+| **DuckDB** | Core | Full support |
+| **Snowflake** | Core | Full support |
+| **SQL Server** | Experimental | Limited testing |
+| **Presto** | Experimental | Limited testing |
 
 ## Examples
 
@@ -273,7 +273,30 @@ trilogy fmt <path to trilogy file>
 - [Public model repository](https://github.com/trilogydata/trilogy-public-models) - Great place for modeling examples
 - [Full documentation](https://trilogydata.dev/)
 
-## Syntax Reference
+## Python API Integration
+
+### Root Imports
+
+Are stable and should be sufficient for executing code from Trilogy as text.
+
+```python
+from pytrilogy import Executor, Dialect
+```
+
+### Authoring Imports
+
+Are also stable, and should be used for cases which programatically generate Trilogy statements without a base text format 
+or need to process/transform parsed code in more complicated ways.
+
+```python
+from pytrilogy.authoring import Concept, Function, ...
+```
+
+### Other Imports
+
+Are likely to be unstable. Open an issue if you need to take dependencies on other modules outside those two paths. 
+
+## Trilogy Syntax Reference
 
 ### Import
 ```sql

{pytrilogy-0.0.3.95 → pytrilogy-0.0.3.96/pytrilogy.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pytrilogy
-Version: 0.0.3.95
+Version: 0.0.3.96
 Summary: Declarative, typed query language that compiles to SQL.
 Home-page: 
 Author: 
@@ -136,11 +136,11 @@ Versus SQL, Trilogy aims to:
 
 | Backend | Status | Notes |
 |---------|--------|-------|
-| **BigQuery** | ✅ Core | Full support |
-| **DuckDB** | ✅ Core | Full support |
-| **Snowflake** | ✅ Core | Full support |
-| **SQL Server** | ⚠️ Experimental | Limited testing |
-| **Presto** | ⚠️ Experimental | Limited testing |
+| **BigQuery** | Core | Full support |
+| **DuckDB** | Core | Full support |
+| **Snowflake** | Core | Full support |
+| **SQL Server** | Experimental | Limited testing |
+| **Presto** | Experimental | Limited testing |
 
 ## Examples
 
@@ -311,7 +311,30 @@ trilogy fmt <path to trilogy file>
 - [Public model repository](https://github.com/trilogydata/trilogy-public-models) - Great place for modeling examples
 - [Full documentation](https://trilogydata.dev/)
 
-## Syntax Reference
+## Python API Integration
+
+### Root Imports
+
+Are stable and should be sufficient for executing code from Trilogy as text.
+
+```python
+from pytrilogy import Executor, Dialect
+```
+
+### Authoring Imports
+
+Are also stable, and should be used for cases which programatically generate Trilogy statements without a base text format 
+or need to process/transform parsed code in more complicated ways.
+
+```python
+from pytrilogy.authoring import Concept, Function, ...
+```
+
+### Other Imports
+
+Are likely to be unstable. Open an issue if you need to take dependencies on other modules outside those two paths. 
+
+## Trilogy Syntax Reference
 
 ### Import
 ```sql

{pytrilogy-0.0.3.95 → pytrilogy-0.0.3.96}/pytrilogy.egg-info/SOURCES.txt

@@ -36,7 +36,6 @@ tests/test_undefined_concept.py
 tests/test_user_functions.py
 tests/test_where_clause.py
 trilogy/__init__.py
-trilogy/compiler.py
 trilogy/constants.py
 trilogy/engine.py
 trilogy/executor.py
@@ -125,6 +124,7 @@ trilogy/dialect/config.py
 trilogy/dialect/dataframe.py
 trilogy/dialect/duckdb.py
 trilogy/dialect/enums.py
+trilogy/dialect/metadata.py
 trilogy/dialect/postgres.py
 trilogy/dialect/presto.py
 trilogy/dialect/snowflake.py

{pytrilogy-0.0.3.95 → pytrilogy-0.0.3.96}/tests/test_executor.py

@@ -2,7 +2,7 @@ from pathlib import Path
 
 from trilogy import Dialects, parse
 from trilogy.core.models.environment import Environment
-from trilogy.executor import MockResult, MockResultRow
+from trilogy.dialect.metadata import MockResult, MockResultRow
 
 
 def test_file_parsing():

{pytrilogy-0.0.3.95 → pytrilogy-0.0.3.96}/trilogy/__init__.py

@@ -4,6 +4,6 @@ from trilogy.dialect.enums import Dialects
 from trilogy.executor import Executor
 from trilogy.parser import parse
 
-__version__ = "0.0.3.95"
+__version__ = "0.0.3.96"
 
 __all__ = ["parse", "Executor", "Dialects", "Environment", "CONFIG"]

{pytrilogy-0.0.3.95 → pytrilogy-0.0.3.96}/trilogy/authoring/__init__.py

@@ -60,68 +60,82 @@ from trilogy.core.statements.author import (
     RowsetDerivationStatement,
     SelectItem,
     SelectStatement,
+    ShowCategory,
+    ShowStatement,
+    ValidateStatement,
 )
 from trilogy.parsing.common import arbitrary_to_concept, arg_to_datatype
 
 __all__ = [
-    "Concept",
-    "Function",
-    "WhereClause",
-    "Comparison",
-    "FilterItem",
-    "CaseWhen",
-    "CaseElse",
-    "AggregateWrapper",
-    "WindowItem",
-    "WindowOrder",
-    "WindowType",
-    "WindowItemOrder",
-    "WindowItemOver",
-    "DataType",
-    "StructType",
-    "ArrayType",
-    "NumericType",
-    "Grain",
-    "RowsetDerivationStatement",
-    "MapType",
-    "ListWrapper",
+    # trilogy.constants
+    "DEFAULT_NAMESPACE",
+    # trilogy.core.enums
+    "BooleanOperator",
+    "ComparisonOperator",
+    "FunctionClass",
     "FunctionType",
+    "InfiniteFunctionArgs",
+    "Ordering",
+    "Purpose",
+    # trilogy.core.functions
     "FunctionFactory",
-    "ConceptDeclarationStatement",
-    "ConceptTransform",
-    "SelectItem",
-    "SelectStatement",
-    "Environment",
+    # trilogy.core.models.author
+    "AggregateWrapper",
+    "CaseElse",
+    "CaseWhen",
+    "Comparison",
+    "Concept",
     "ConceptRef",
+    "Conditional",
+    "FilterItem",
+    "Function",
+    "FunctionCallWrapper",
     "HavingClause",
     "MagicConstants",
     "Metadata",
+    "MultiSelectLineage",
     "OrderBy",
     "OrderItem",
     "Parenthetical",
+    "RowsetItem",
     "SubselectComparison",
-    "Conditional",
-    "BooleanOperator",
-    "ComparisonOperator",
-    "FunctionClass",
-    "FunctionType",
-    "InfiniteFunctionArgs",
-    "Ordering",
-    "Purpose",
-    "DEFAULT_NAMESPACE",
-    "arbitrary_to_concept",
-    "arg_to_datatype",
-    "MultiSelectStatement",
-    "PersistStatement",
-    "RawSQLStatement",
+    "WhereClause",
+    "WindowItem",
+    "WindowItemOrder",
+    "WindowItemOver",
+    "WindowOrder",
+    "WindowType",
+    # trilogy.core.models.core
+    "ArrayType",
+    "DataType",
+    "ListWrapper",
+    "MapType",
+    "NumericType",
+    "StructType",
+    "TraitDataType",
+    # trilogy.core.models.datasource
+    "Address",
     "Datasource",
     "DatasourceMetadata",
-    "MultiSelectLineage",
-    "RowsetItem",
-    "FunctionCallWrapper",
+    # trilogy.core.models.environment
+    "Environment",
+    # trilogy.core.statements.author
+    "ConceptDeclarationStatement",
+    "ConceptTransform",
     "CopyStatement",
+    "Grain",
     "HasUUID",
     "ImportStatement",
-    "Address",
-    "TraitDataType",
+    "MultiSelectStatement",
+    "PersistStatement",
+    "RawSQLStatement",
+    "RowsetDerivationStatement",
+    "SelectItem",
+    "SelectStatement",
+    "ShowCategory",
+    "ShowStatement",
+    "ValidateStatement",
+    # trilogy.parsing.common
+    "arbitrary_to_concept",
+    "arg_to_datatype",
 ]

{pytrilogy-0.0.3.95 → pytrilogy-0.0.3.96}/trilogy/core/graph_models.py

@@ -64,13 +64,13 @@ def datasource_to_node(input: BuildDatasource) -> str:
 
 
 class ReferenceGraph(nx.DiGraph):
-    def __init__(self, *args, **kwargs):
+    def __init__(self, *args, **kwargs) -> None:
         super().__init__(*args, **kwargs)
         self.concepts: dict[str, BuildConcept] = {}
         self.datasources: dict[str, BuildDatasource] = {}
         self.pseudonyms: set[tuple[str, str]] = set()
 
-    def copy(self):
+    def copy(self) -> "ReferenceGraph":
         g = ReferenceGraph()
         g.concepts = self.concepts.copy()
         g.datasources = self.datasources.copy()
@@ -83,7 +83,7 @@ class ReferenceGraph(nx.DiGraph):
         # g.add_edges_from(self.edges(data=True))
         return g
 
-    def remove_node(self, n):
+    def remove_node(self, n) -> None:
         if n in self.concepts:
             del self.concepts[n]
         if n in self.datasources:
@@ -98,7 +98,7 @@ class ReferenceGraph(nx.DiGraph):
             self.datasources[node_name] = attr["datasource"]
         super().add_node(node_name, **attr)
 
-    def add_datasource_node(self, node_name, datasource):
+    def add_datasource_node(self, node_name, datasource) -> None:
         self.datasources[node_name] = datasource
         super().add_node(node_name, datasource=datasource)
 

{pytrilogy-0.0.3.95 → pytrilogy-0.0.3.96}/trilogy/core/statements/execute.py

@@ -88,6 +88,8 @@ class ProcessedShowStatement:
             BuildConcept,
             BuildDatasource,
             ProcessedQuery,
+            ProcessedQueryPersist,
+            ProcessedCopyStatement,
             ProcessedValidateStatement,
             ProcessedStaticValueOutput,
         ]

{pytrilogy-0.0.3.95 → pytrilogy-0.0.3.96}/trilogy/core/validation/common.py

@@ -27,7 +27,8 @@ class ExpectationType(Enum):
 @dataclass
 class ValidationTest:
     check_type: ExpectationType
-    query: str | None = None
+    raw_query: ProcessedQuery | None = None
+    generated_query: str | None = None
     expected: str | None = None
     result: ModelValidationError | None = None
     ran: bool = True

{pytrilogy-0.0.3.95 → pytrilogy-0.0.3.96}/trilogy/core/validation/concept.py

@@ -1,4 +1,4 @@
-from trilogy import Executor
+from trilogy import Environment, Executor
 from trilogy.core.enums import Derivation, Purpose
 from trilogy.core.exceptions import (
     ConceptModelValidationError,
@@ -12,64 +12,68 @@ from trilogy.core.validation.common import ExpectationType, ValidationTest, easy
 
 
 def validate_property_concept(
-    concept: BuildConcept, generate_only: bool = False
+    concept: BuildConcept, exec: Executor | None = None
 ) -> list[ValidationTest]:
     return []
 
 
 def validate_key_concept(
     concept: BuildConcept,
+    env: Environment,
     build_env: BuildEnvironment,
-    exec: Executor,
-    generate_only: bool = False,
+    exec: Executor | None = None,
 ):
     results: list[ValidationTest] = []
-    seen = {}
+    seen: dict[str, int] = {}
     for datasource in build_env.datasources.values():
         if concept.address in [c.address for c in datasource.concepts]:
             assignment = [
                 x for x in datasource.columns if x.concept.address == concept.address
             ][0]
+            # if it's not a partial, skip it
+            if not assignment.is_complete:
+                continue
             type_query = easy_query(
                 concepts=[
                     # build_env.concepts[concept.address],
                     build_env.concepts[f"grain_check_{concept.safe_address}"],
                 ],
                 datasource=datasource,
-                env=exec.environment,
+                env=env,
                 limit=1,
             )
-            type_sql = exec.generate_sql(type_query)[-1]
+            if exec:
+                type_sql = exec.generate_sql(type_query)[-1]
 
-            rows = exec.execute_raw_sql(type_sql).fetchall()
-            if generate_only and assignment.is_complete:
+                rows = exec.execute_raw_sql(type_sql).fetchall()
+                seen[datasource.name] = rows[0][0] if rows else 0
+            else:
                 results.append(
                     ValidationTest(
-                        query=type_sql,
+                        raw_query=type_query,
                         check_type=ExpectationType.ROWCOUNT,
                         expected=f"equal_max_{concept.safe_address}",
                         result=None,
                         ran=False,
                     )
                 )
-                continue
-            seen[datasource.name] = rows[0][0] if rows else None
-    if generate_only:
+
+    if not exec:
         return results
-    max_seen = max([v for v in seen.values() if v is not None], default=0)
+    max_seen: int = max([v for v in seen.values() if v is not None], default=0)
     for datasource in build_env.datasources.values():
         if concept.address in [c.address for c in datasource.concepts]:
             assignment = [
                 x for x in datasource.columns if x.concept.address == concept.address
             ][0]
             err = None
-            if (seen[datasource.name] or 0) < max_seen and assignment.is_complete:
+            datasource_count: int = seen.get(datasource.name, 0)
+            if datasource_count < max_seen and assignment.is_complete:
                 err = DatasourceModelValidationError(
                     f"Key concept {concept.address} is missing values in datasource {datasource.name} (max cardinality in data {max_seen}, datasource has {seen[datasource.name]} values) but is not marked as partial."
                 )
             results.append(
                 ValidationTest(
-                    query=None,
                     check_type=ExpectationType.ROWCOUNT,
                     expected=str(max_seen),
                     result=err,
@@ -96,7 +100,6 @@ def validate_datasources(
         return []
     return [
         ValidationTest(
-            query=None,
             check_type=ExpectationType.LOGICAL,
             expected=None,
             result=ConceptModelValidationError(
@@ -109,14 +112,14 @@ def validate_datasources(
 
 def validate_concept(
     concept: BuildConcept,
+    env: Environment,
     build_env: BuildEnvironment,
-    exec: Executor,
-    generate_only: bool = False,
+    exec: Executor | None = None,
 ) -> list[ValidationTest]:
     base: list[ValidationTest] = []
     base += validate_datasources(concept, build_env)
     if concept.purpose == Purpose.PROPERTY:
-        base += validate_property_concept(concept, generate_only)
+        base += validate_property_concept(concept)
     elif concept.purpose == Purpose.KEY:
-        base += validate_key_concept(concept, build_env, exec, generate_only)
+        base += validate_key_concept(concept, env, build_env, exec)
     return base

{pytrilogy-0.0.3.95 → pytrilogy-0.0.3.96}/trilogy/core/validation/datasource.py

@@ -2,7 +2,7 @@ from datetime import date, datetime
 from decimal import Decimal
 from typing import Any
 
-from trilogy import Executor
+from trilogy import Environment, Executor
 from trilogy.authoring import (
     ArrayType,
     DataType,
@@ -61,12 +61,12 @@ def type_check(
 
 def validate_datasource(
     datasource: BuildDatasource,
+    env: Environment,
     build_env: BuildEnvironment,
-    exec: Executor,
-    generate_only: bool = False,
+    exec: Executor | None = None,
 ) -> list[ValidationTest]:
     results: list[ValidationTest] = []
-    # we might have merged concepts, where both wil lmap out to the same
+    # we might have merged concepts, where both will map out to the same
     unique_outputs = unique(
         [build_env.concepts[col.concept.address] for col in datasource.columns],
         "address",
@@ -74,18 +74,20 @@ def validate_datasource(
     type_query = easy_query(
         concepts=unique_outputs,
         datasource=datasource,
-        env=exec.environment,
+        env=env,
         limit=100,
     )
-    type_sql = exec.generate_sql(type_query)[-1]
+
     rows = []
-    if not generate_only:
+    if exec:
+        type_sql = exec.generate_sql(type_query)[-1]
         try:
             rows = exec.execute_raw_sql(type_sql).fetchall()
         except Exception as e:
             results.append(
                 ValidationTest(
-                    query=type_sql,
+                    raw_query=type_query,
+                    generated_query=type_sql,
                     check_type=ExpectationType.LOGICAL,
                     expected="valid_sql",
                     result=DatasourceModelValidationError(
@@ -96,9 +98,10 @@ def validate_datasource(
             )
             return results
     else:
+
         results.append(
             ValidationTest(
-                query=type_sql,
+                raw_query=type_query,
                 check_type=ExpectationType.LOGICAL,
                 expected="datatype_match",
                 result=None,
@@ -117,7 +120,6 @@ def validate_datasource(
     cols_with_error = set()
     for row in rows:
         for col in datasource.columns:
-
             actual_address = build_env.concepts[col.concept.address].safe_address
             if actual_address in cols_with_error:
                 continue
@@ -140,7 +142,6 @@ def validate_datasource(
     if failures:
         results.append(
             ValidationTest(
-                query=None,
                 check_type=ExpectationType.LOGICAL,
                 expected="datatype_match",
                 ran=True,
@@ -161,10 +162,10 @@ def validate_datasource(
             operator=ComparisonOperator.GT,
         ),
     )
-    if generate_only:
+    if not exec:
         results.append(
             ValidationTest(
-                query=exec.generate_sql(query)[-1],
+                raw_query=query,
                 check_type=ExpectationType.ROWCOUNT,
                 expected="0",
                 result=None,
@@ -179,7 +180,8 @@ def validate_datasource(
         if rows:
             results.append(
                 ValidationTest(
-                    query=sql,
+                    raw_query=query,
+                    generated_query=sql,
                     check_type=ExpectationType.ROWCOUNT,
                     expected="0",
                     result=DatasourceModelValidationError(

{pytrilogy-0.0.3.95 → pytrilogy-0.0.3.96}/trilogy/core/validation/environment.py

@@ -12,12 +12,12 @@ from trilogy.parsing.common import function_to_concept
 
 def validate_environment(
     env: Environment,
-    exec: Executor,
     scope: ValidationScope = ValidationScope.ALL,
     targets: list[str] | None = None,
-    generate_only: bool = False,
+    exec: Executor | None = None,
 ) -> list[ValidationTest]:
     # avoid mutating the environment for validation
+    generate_only = exec is None
     env = env.duplicate()
     grain_check = function_to_concept(
         parent=Function(
@@ -51,13 +51,13 @@ def validate_environment(
         for datasource in build_env.datasources.values():
             if targets and datasource.name not in targets:
                 continue
-            results += validate_datasource(datasource, build_env, exec, generate_only)
+            results += validate_datasource(datasource, env, build_env, exec)
     if scope == ValidationScope.ALL or scope == ValidationScope.CONCEPTS:
 
         for bconcept in build_env.concepts.values():
             if targets and bconcept.address not in targets:
                 continue
-            results += validate_concept(bconcept, build_env, exec, generate_only)
+            results += validate_concept(bconcept, env, build_env, exec)
 
     # raise a nicely formatted union of all exceptions
     exceptions: list[ModelValidationError] = [e.result for e in results if e.result]

{pytrilogy-0.0.3.95 → pytrilogy-0.0.3.96}/trilogy/dialect/base.py

@@ -76,6 +76,7 @@ from trilogy.core.statements.author import (
 )
 from trilogy.core.statements.execute import (
     PROCESSED_STATEMENT_TYPES,
+    ProcessedCopyStatement,
     ProcessedQuery,
     ProcessedQueryPersist,
     ProcessedRawSQLStatement,
@@ -345,6 +346,7 @@ class BaseDialect:
     COMPLEX_DATATYPE_MAP = COMPLEX_DATATYPE_MAP
     UNNEST_MODE = UnnestMode.CROSS_APPLY
     GROUP_MODE = GroupMode.AUTO
+    EXPLAIN_KEYWORD = "EXPLAIN"
 
     def __init__(self, rendering: Rendering | None = None):
         self.rendering = rendering or CONFIG.rendering
@@ -1135,7 +1137,13 @@ class BaseDialect:
         query: PROCESSED_STATEMENT_TYPES,
     ) -> str:
         if isinstance(query, ProcessedShowStatement):
-            return ";\n".join([str(x) for x in query.output_values])
+            return ";\n".join(
+                [
+                    f'{self.EXPLAIN_KEYWORD} {self.compile_statement(x)}'
+                    for x in query.output_values
+                    if isinstance(x, (ProcessedQuery, ProcessedCopyStatement))
+                ]
+            )
         elif isinstance(query, ProcessedRawSQLStatement):
             return query.text