additory 0.1.0a2__tar.gz → 0.1.0a3__tar.gz

{additory-0.1.0a2 → additory-0.1.0a3}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: additory
-Version: 0.1.0a2
-Summary: A semantic, extensible dataframe transformation engine with expressions, lookup, synthetic data, and sample-data support.
+Version: 0.1.0a3
+Summary: A semantic, extensible dataframe transformation engine with expressions, lookup, and synthetic data generation support.
 Author: Krishnamoorthy Sankaran
 License: MIT
 Project-URL: homepage, https://github.com/sekarkrishna/additory
@@ -13,6 +13,7 @@ Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: pandas>=1.5
 Requires-Dist: polars>=0.20
+Requires-Dist: pyarrow>=10.0
 Requires-Dist: pyyaml>=6.0
 Requires-Dist: requests>=2.31
 Requires-Dist: toml>=0.10
@@ -34,11 +35,11 @@ Dynamic: license-file
 
 # Additory
 
-**A semantic, extensible dataframe transformation engine with expressions, lookup, synthetic data, and sample-data support.**
+**A semantic, extensible dataframe transformation engine with expressions, lookup, and augmentation support.**
 
 [![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg)](https://www.python.org/downloads/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
-[![Version](https://img.shields.io/badge/version-0.1.0a1-orange.svg)](https://github.com/sekarkrishna/additory/tree/main/V0.1.0a1/)
+[![Version](https://img.shields.io/badge/version-0.1.0a2-orange.svg)](https://github.com/sekarkrishna/additory/tree/main/V0.1.0a1/)
 
 **Author:** Krishnamoorthy Sankaran
 
@@ -51,17 +52,17 @@ Dynamic: license-file
 ## 📦 Installation
 
 ```bash
-pip install additory==0.1.0a1
+pip install additory==0.1.0a2
 ```
 
 **Optional GPU support:**
 ```bash
-pip install additory[gpu]==0.1.0a1  # Includes cuDF for GPU acceleration
+pip install additory[gpu]==0.1.0a2  # Includes cuDF for GPU acceleration
 ```
 
 **Development installation:**
 ```bash
-pip install additory[dev]==0.1.0a1  # Includes testing and development tools
+pip install additory[dev]==0.1.0a2  # Includes testing and development tools
 ```
 
 ## 🎯 Core Functions
@@ -70,7 +71,6 @@ pip install additory[dev]==0.1.0a1  # Includes testing and development tools
 |----------|---------|---------|
 | `add.to()` | Lookup/join operations | `add.to(df1, from_df=df2, bring='col', against='key')` |
 | `add.augment()` | Generate additional data | `add.augment(df, n_rows=1000)` |
-| `add.synth()` | Synthetic data from schemas | `add.synth("schema.toml", rows=5000)` |
 | `add.scan()` | Data profiling & analysis | `add.scan(df, preset="full")` |
 
 ## 🧬 Available Expressions
@@ -193,13 +193,9 @@ patients_with_bsa = add.bsa(patients)
 result = add.fitness_score(add.bmr(add.bmi(patients)))
 ```
 
-### 🔄 Augment and Synthetic Data
+### 🔄 Augment Data Generation
 
-**Augment** generates more data similar to your existing dataset, while **Synthetic** creates entirely new datasets from schema definitions.
-
-**Key Differences:**
-- **Augment**: Learns patterns from existing data to create similar rows
-- **Synthetic**: Uses predefined schemas to generate structured data
+**Augment** generates additional data similar to your existing dataset using inline strategies.
 
 ```python
 # Augment existing data (learns from patterns)
@@ -211,9 +207,6 @@ new_data = add.augment("@new", n_rows=500, strategy={
     'name': 'choice:[John,Jane,Bob]',
     'age': 'range:18-65'
 })
-
-# Generate from schema file (structured approach)
-customers = add.synth("customer_schema.toml", rows=10000)
 ```
 
 ## 🧪 Examples

{additory-0.1.0a2 → additory-0.1.0a3}/README.md

@@ -1,10 +1,10 @@
 # Additory
 
-**A semantic, extensible dataframe transformation engine with expressions, lookup, synthetic data, and sample-data support.**
+**A semantic, extensible dataframe transformation engine with expressions, lookup, and augmentation support.**
 
 [![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg)](https://www.python.org/downloads/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
-[![Version](https://img.shields.io/badge/version-0.1.0a1-orange.svg)](https://github.com/sekarkrishna/additory/tree/main/V0.1.0a1/)
+[![Version](https://img.shields.io/badge/version-0.1.0a2-orange.svg)](https://github.com/sekarkrishna/additory/tree/main/V0.1.0a1/)
 
 **Author:** Krishnamoorthy Sankaran
 
@@ -17,17 +17,17 @@
 ## 📦 Installation
 
 ```bash
-pip install additory==0.1.0a1
+pip install additory==0.1.0a2
 ```
 
 **Optional GPU support:**
 ```bash
-pip install additory[gpu]==0.1.0a1  # Includes cuDF for GPU acceleration
+pip install additory[gpu]==0.1.0a2  # Includes cuDF for GPU acceleration
 ```
 
 **Development installation:**
 ```bash
-pip install additory[dev]==0.1.0a1  # Includes testing and development tools
+pip install additory[dev]==0.1.0a2  # Includes testing and development tools
 ```
 
 ## 🎯 Core Functions
@@ -36,7 +36,6 @@ pip install additory[dev]==0.1.0a1  # Includes testing and development tools
 |----------|---------|---------|
 | `add.to()` | Lookup/join operations | `add.to(df1, from_df=df2, bring='col', against='key')` |
 | `add.augment()` | Generate additional data | `add.augment(df, n_rows=1000)` |
-| `add.synth()` | Synthetic data from schemas | `add.synth("schema.toml", rows=5000)` |
 | `add.scan()` | Data profiling & analysis | `add.scan(df, preset="full")` |
 
 ## 🧬 Available Expressions
@@ -159,13 +158,9 @@ patients_with_bsa = add.bsa(patients)
 result = add.fitness_score(add.bmr(add.bmi(patients)))
 ```
 
-### 🔄 Augment and Synthetic Data
+### 🔄 Augment Data Generation
 
-**Augment** generates more data similar to your existing dataset, while **Synthetic** creates entirely new datasets from schema definitions.
-
-**Key Differences:**
-- **Augment**: Learns patterns from existing data to create similar rows
-- **Synthetic**: Uses predefined schemas to generate structured data
+**Augment** generates additional data similar to your existing dataset using inline strategies.
 
 ```python
 # Augment existing data (learns from patterns)
@@ -177,9 +172,6 @@ new_data = add.augment("@new", n_rows=500, strategy={
     'name': 'choice:[John,Jane,Bob]',
     'age': 'range:18-65'
 })
-
-# Generate from schema file (structured approach)
-customers = add.synth("customer_schema.toml", rows=10000)
 ```
 
 ## 🧪 Examples

{additory-0.1.0a2 → additory-0.1.0a3}/additory/__init__.py

@@ -2,6 +2,9 @@
 
 from .dynamic_api import add as _api_instance
 
+# Version information
+__version__ = "0.1.0a3"
+
 # Expose the API instance normally
 add = _api_instance
 
@@ -12,4 +15,5 @@ def __getattr__(name):
 
 __all__ = [
     "add",
+    "__version__",
 ]

{additory-0.1.0a2 → additory-0.1.0a3}/additory/common/__init__.py

@@ -1,14 +1,14 @@
 """
 Common Utilities Module
 
-Shared functionality used by both augment and synthetic modules:
+Shared functionality used by both synthetic and expressions modules:
 - Distribution functions (normal, uniform, skewed, etc.)
 - List file management (.list format)
 - Pattern file management (.properties format)
 - Fallback resolution logic
 
 This module eliminates code duplication and provides consistent behavior
-across augment and synthetic data generation.
+across synthetic and expression data generation.
 """
 
 from .distributions import (

{additory-0.1.0a2 → additory-0.1.0a3}/additory/common/backend.py

@@ -180,11 +180,14 @@ def get_arrow_bridge():
         - Use for all cross-backend conversions
         - Handles pandas/polars/cuDF via Arrow
     """
-    from additory.core.backends.arrow_bridge import EnhancedArrowBridge
+    from additory.core.backends.arrow_bridge import EnhancedArrowBridge, ArrowBridgeError
     
     # Singleton pattern
     if not hasattr(get_arrow_bridge, '_instance'):
-        get_arrow_bridge._instance = EnhancedArrowBridge()
+        try:
+            get_arrow_bridge._instance = EnhancedArrowBridge()
+        except ArrowBridgeError:
+            get_arrow_bridge._instance = None
     
     return get_arrow_bridge._instance
 
@@ -194,7 +197,7 @@ def to_polars(df: Any, backend_type: BackendType = None) -> 'pl.DataFrame':
     Convert any dataframe to Polars via Arrow bridge.
     
     This is the primary conversion function for the Polars-only architecture.
-    All operations (expressions, augment, etc.) use this to convert input
+    All operations (expressions, synthetic, etc.) use this to convert input
     dataframes to Polars for processing.
     
     Args:
@@ -224,7 +227,7 @@ def to_polars(df: Any, backend_type: BackendType = None) -> 'pl.DataFrame':
         )
     
     # Fast path: already Polars
-    if isinstance(df, pl.DataFrame):
+    if HAS_POLARS and isinstance(df, pl.DataFrame):
         return df
     
     # Validate input
@@ -240,6 +243,13 @@ def to_polars(df: Any, backend_type: BackendType = None) -> 'pl.DataFrame':
     # Convert via Arrow bridge
     try:
         bridge = get_arrow_bridge()
+        if bridge is None:
+            # Fallback: direct conversion for pandas
+            if backend_type == "pandas":
+                if isinstance(df, pd.DataFrame):
+                    return pl.from_pandas(df)
+            raise RuntimeError("Arrow bridge not available and cannot convert non-pandas DataFrame")
+        
         arrow_table = bridge.to_arrow(df, backend_type)
         pl_df = bridge.from_arrow(arrow_table, "polars")
         return pl_df
@@ -309,6 +319,12 @@ def from_polars(pl_df: 'pl.DataFrame', target_backend: BackendType) -> Any:
     # Convert via Arrow bridge
     try:
         bridge = get_arrow_bridge()
+        if bridge is None:
+            # Fallback: direct conversion for pandas
+            if target_backend == "pandas":
+                return pl_df.to_pandas()
+            raise RuntimeError("Arrow bridge not available and cannot convert to non-pandas DataFrame")
+        
         arrow_table = bridge.to_arrow(pl_df, "polars")
         result_df = bridge.from_arrow(arrow_table, target_backend)
         return result_df

{additory-0.1.0a2 → additory-0.1.0a3}/additory/common/distributions.py

@@ -1,5 +1,5 @@
 """
-Distribution Strategies for Data Augmentation
+Distribution Strategies for Synthetic Data Generation
 
 Provides statistical distribution-based data generation:
 - Normal (Gaussian) distribution

{additory-0.1.0a2 → additory-0.1.0a3}/additory/common/sample_data.py

@@ -8,8 +8,8 @@ loaded on-demand using the existing .add file parser.
 Usage:
     from additory.common.sample_data import get_sample_dataset
     
-    # For augment
-    df = get_sample_dataset("augment", "sample")
+    # For synthetic
+    df = get_sample_dataset("synthetic", "sample")
     
     # For expressions (future)
     df = get_sample_dataset("expressions", "sample")
@@ -25,7 +25,7 @@ from additory.common.exceptions import ValidationError
 
 
 def get_sample_dataset(
-    module: str = "augment",
+    module: str = "synthetic",
     block: str = "sample",
     dataset_type: str = "clean"
 ) -> pl.DataFrame:
@@ -33,12 +33,12 @@ def get_sample_dataset(
     Load a sample dataset from .add files.
     
     This function provides centralized access to sample datasets across
-    all additory modules (augment, expressions, utilities). Sample datasets
+    all additory modules (synthetic, expressions, utilities). Sample datasets
     are stored as .add files in the reference/ directory structure.
     
     Args:
-        module: Module name ("augment", "expressions", "utilities")
-        block: Block name within the .add file ("sample" for augment)
+        module: Module name ("synthetic", "expressions", "utilities")
+        block: Block name within the .add file ("sample" for synthetic)
         dataset_type: Type of sample data ("clean" or "unclean")
     
     Returns:
@@ -48,8 +48,8 @@ def get_sample_dataset(
         ValidationError: If module, block, or dataset_type not found
     
     Examples:
-        >>> # Load augment sample dataset
-        >>> df = get_sample_dataset("augment", "sample")
+        >>> # Load synthetic sample dataset
+        >>> df = get_sample_dataset("synthetic", "sample")
         >>> print(df.shape)
         (50, 10)
         
@@ -57,7 +57,7 @@ def get_sample_dataset(
         >>> df = get_sample_dataset("expressions", "sample", "clean")
         >>> df_unclean = get_sample_dataset("expressions", "sample", "unclean")
     
-    Sample Dataset Structure (augment):
+    Sample Dataset Structure (synthetic):
         - id: Sequential numeric IDs (1-50)
         - emp_id: Employee IDs with pattern (EMP_001 - EMP_050)
         - order_id: Order IDs with different padding (ORD_0001 - ORD_0050)
@@ -72,8 +72,8 @@ def get_sample_dataset(
     # Construct path to .add file
     base_path = Path(__file__).parent.parent.parent / "reference"
     
-    if module == "augment":
-        add_file_path = base_path / "augment_definitions" / f"{block}_0.1.add"
+    if module == "synthetic":
+        add_file_path = base_path / "synthetic_definitions" / f"{block}_0.1.add"
     elif module == "expressions":
         add_file_path = base_path / "expressions_definitions" / f"{block}_0.1.add"
     elif module == "utilities":
@@ -81,7 +81,7 @@ def get_sample_dataset(
     else:
         raise ValidationError(
             f"Unknown module '{module}'. "
-            f"Valid modules: augment, expressions, utilities"
+            f"Valid modules: synthetic, expressions, utilities"
         )
     
     # Check if file exists
@@ -141,7 +141,7 @@ def list_available_samples() -> dict:
         >>> samples = list_available_samples()
         >>> print(samples)
         {
-            'augment': ['sample'],
+            'synthetic': ['sample'],
             'expressions': ['sample'],
             'utilities': []
         }
@@ -149,15 +149,15 @@ def list_available_samples() -> dict:
     base_path = Path(__file__).parent.parent.parent / "reference"
     available = {}
     
-    # Check augment
-    augment_path = base_path / "augment_definitions"
-    if augment_path.exists():
-        available['augment'] = [
+    # Check synthetic
+    synthetic_path = base_path / "synthetic_definitions"
+    if synthetic_path.exists():
+        available['synthetic'] = [
             f.stem.rsplit('_', 1)[0]  # Remove version suffix
-            for f in augment_path.glob("*.add")
+            for f in synthetic_path.glob("*.add")
         ]
     else:
-        available['augment'] = []
+        available['synthetic'] = []
     
     # Check expressions
     expressions_path = base_path / "expressions_definitions"

{additory-0.1.0a2 → additory-0.1.0a3}/additory/core/backends/arrow_bridge.py

@@ -16,6 +16,13 @@ try:
 except ImportError as e:
     ARROW_AVAILABLE = False
     IMPORT_ERROR = str(e)
+    # Create dummy classes for type annotations
+    class pa:
+        Table = Any
+    class pl:
+        DataFrame = Any
+    class pd:
+        DataFrame = Any
 
 from ..logging import log_info, log_warning
 from .cudf_bridge import get_cudf_bridge

{additory-0.1.0a2 → additory-0.1.0a3}/additory/core/polars_expression_engine.py

@@ -32,7 +32,10 @@ class PolarsExpressionEngine:
     """Exclusive Polars-based expression processing engine"""
     
     def __init__(self):
-        self.arrow_bridge = EnhancedArrowBridge()
+        try:
+            self.arrow_bridge = EnhancedArrowBridge()
+        except ArrowBridgeError:
+            self.arrow_bridge = None
         self.execution_stats = {
             "total_executions": 0,
             "total_time_ms": 0.0,
@@ -68,14 +71,28 @@ class PolarsExpressionEngine:
             try:
                 # Auto-detect backend if not specified
                 if backend_type is None:
-                    backend_type = self.arrow_bridge.detect_backend(df)
+                    if self.arrow_bridge:
+                        backend_type = self.arrow_bridge.detect_backend(df)
+                    else:
+                        backend_type = "pandas"  # fallback
                 
                 # Get memory usage before processing
-                memory_before = self.arrow_bridge._get_memory_usage_mb()
+                if self.arrow_bridge:
+                    memory_before = self.arrow_bridge._get_memory_usage_mb()
+                else:
+                    memory_before = 0
                 
                 # 1. Convert input to Arrow
                 log_info(f"[polars_engine] Converting {backend_type} to Arrow")
-                arrow_table = self.arrow_bridge.to_arrow(df, backend_type)
+                if self.arrow_bridge:
+                    arrow_table = self.arrow_bridge.to_arrow(df, backend_type)
+                else:
+                    # Fallback: assume pandas and convert directly
+                    import pandas as pd
+                    if isinstance(df, pd.DataFrame):
+                        arrow_table = pl.from_pandas(df).to_arrow()
+                    else:
+                        raise RuntimeError("Arrow bridge not available and input is not pandas DataFrame")
                 
                 # 2. Convert Arrow to Polars
                 log_info("[polars_engine] Converting Arrow to Polars")
@@ -93,11 +110,18 @@ class PolarsExpressionEngine:
                 
                 # 5. Convert to original backend format
                 log_info(f"[polars_engine] Converting Arrow to {backend_type}")
-                final_result = self.arrow_bridge.from_arrow(result_arrow, backend_type)
+                if self.arrow_bridge:
+                    final_result = self.arrow_bridge.from_arrow(result_arrow, backend_type)
+                else:
+                    # Fallback: convert back to pandas
+                    final_result = pl.from_arrow(result_arrow).to_pandas()
                 
                 # Calculate execution statistics
                 execution_time = (datetime.now() - start_time).total_seconds() * 1000
-                memory_after = self.arrow_bridge._get_memory_usage_mb()
+                if self.arrow_bridge:
+                    memory_after = self.arrow_bridge._get_memory_usage_mb()
+                else:
+                    memory_after = 0
                 memory_used = max(0, memory_after - memory_before)
                 
                 # Update global statistics
@@ -122,7 +146,8 @@ class PolarsExpressionEngine:
                 
             finally:
                 # 6. Always cleanup Arrow memory
-                self.arrow_bridge.cleanup_arrow_memory()
+                if self.arrow_bridge:
+                    self.arrow_bridge.cleanup_arrow_memory()
     
     def _execute_polars_expression(self, polars_df: pl.DataFrame, 
                                  expression: str, output_column: str) -> pl.DataFrame:
@@ -381,14 +406,28 @@ class PolarsExpressionEngine:
         try:
             # Auto-detect backend if not specified
             if backend_type is None:
-                backend_type = self.arrow_bridge.detect_backend(df)
+                if self.arrow_bridge:
+                    backend_type = self.arrow_bridge.detect_backend(df)
+                else:
+                    backend_type = "pandas"
             
             # Get memory usage before processing
-            memory_before = self.arrow_bridge._get_memory_usage_mb()
+            if self.arrow_bridge:
+                memory_before = self.arrow_bridge._get_memory_usage_mb()
+            else:
+                memory_before = 0
             
             # Convert to Polars via Arrow
-            arrow_table = self.arrow_bridge.to_arrow(df, backend_type)
-            polars_df = pl.from_arrow(arrow_table)
+            if self.arrow_bridge:
+                arrow_table = self.arrow_bridge.to_arrow(df, backend_type)
+                polars_df = pl.from_arrow(arrow_table)
+            else:
+                # Fallback: assume pandas
+                import pandas as pd
+                if isinstance(df, pd.DataFrame):
+                    polars_df = pl.from_pandas(df)
+                else:
+                    raise RuntimeError("Arrow bridge not available and input is not pandas DataFrame")
             
             # Execute using AST
             polars_expr = self._ast_to_polars_expr(ast_tree)
@@ -396,11 +435,17 @@ class PolarsExpressionEngine:
             
             # Convert back to original format
             result_arrow = result_df.to_arrow()
-            final_result = self.arrow_bridge.from_arrow(result_arrow, backend_type)
+            if self.arrow_bridge:
+                final_result = self.arrow_bridge.from_arrow(result_arrow, backend_type)
+            else:
+                final_result = pl.from_arrow(result_arrow).to_pandas()
             
             # Calculate statistics
             execution_time = (datetime.now() - start_time).total_seconds() * 1000
-            memory_after = self.arrow_bridge._get_memory_usage_mb()
+            if self.arrow_bridge:
+                memory_after = self.arrow_bridge._get_memory_usage_mb()
+            else:
+                memory_after = 0
             memory_used = max(0, memory_after - memory_before)
             
             # Update statistics
@@ -422,7 +467,8 @@ class PolarsExpressionEngine:
             raise PolarsExpressionError(f"AST execution failed: {e}")
             
         finally:
-            self.arrow_bridge.cleanup_arrow_memory()
+            if self.arrow_bridge:
+                self.arrow_bridge.cleanup_arrow_memory()
     
     def validate_expression(self, expression: str) -> bool:
         """
@@ -489,7 +535,10 @@ class PolarsExpressionEngine:
             Benchmark results
         """
         times = []
-        backend_type = self.arrow_bridge.detect_backend(df)
+        if self.arrow_bridge:
+            backend_type = self.arrow_bridge.detect_backend(df)
+        else:
+            backend_type = "pandas"
         
         for i in range(iterations):
             try:
@@ -532,7 +581,8 @@ class PolarsExpressionEngine:
         """Cleanup callback for memory manager"""
         try:
             # Cleanup Arrow bridge memory
-            self.arrow_bridge.cleanup_arrow_memory()
+            if self.arrow_bridge:
+                self.arrow_bridge.cleanup_arrow_memory()
             
             # Reset statistics if they get too large
             if self.execution_stats["total_executions"] > 10000: