misata 0.1.0b0__py3-none-any.whl → 0.3.0b0__py3-none-any.whl

misata/streaming.py

@@ -0,0 +1,228 @@
+"""
+Streaming export utilities for Misata.
+
+Provides streaming CSV/Parquet export to handle large datasets
+without loading everything into memory.
+"""
+
+import csv
+import os
+from pathlib import Path
+from typing import Any, Callable, Dict, Generator, Iterator, List, Optional, Union
+
+import numpy as np
+import pandas as pd
+
+from misata.exceptions import ExportError, FileWriteError
+
+
+class StreamingExporter:
+    """Export data in streaming fashion to handle large datasets.
+    
+    Instead of building a full DataFrame and then exporting, this writes
+    batches directly to files as they are generated.
+    
+    Example:
+        exporter = StreamingExporter(output_dir="./data")
+        
+        for table_name, batch_df in simulator.generate_all():
+            exporter.write_batch(table_name, batch_df)
+        
+        exporter.finalize()
+    """
+    
+    def __init__(
+        self,
+        output_dir: str,
+        format: str = "csv",
+        progress_callback: Optional[Callable[[str, int], None]] = None,
+    ):
+        """Initialize the exporter.
+        
+        Args:
+            output_dir: Directory to write files to
+            format: Export format ('csv' or 'parquet')
+            progress_callback: Optional callback(table_name, rows_written)
+        """
+        self.output_dir = Path(output_dir)
+        self.format = format.lower()
+        self.progress_callback = progress_callback
+        
+        self._file_handles: Dict[str, Any] = {}
+        self._csv_writers: Dict[str, csv.writer] = {}
+        self._rows_written: Dict[str, int] = {}
+        self._headers_written: Dict[str, bool] = {}
+        
+        # Create output directory
+        self.output_dir.mkdir(parents=True, exist_ok=True)
+    
+    def write_batch(self, table_name: str, df: pd.DataFrame) -> int:
+        """Write a batch of data to the appropriate file.
+        
+        Args:
+            table_name: Name of the table
+            df: Batch DataFrame to write
+            
+        Returns:
+            Number of rows written
+        """
+        if self.format == "csv":
+            return self._write_csv_batch(table_name, df)
+        elif self.format == "parquet":
+            return self._write_parquet_batch(table_name, df)
+        else:
+            raise ExportError(f"Unsupported format: {self.format}")
+    
+    def _write_csv_batch(self, table_name: str, df: pd.DataFrame) -> int:
+        """Write a batch to CSV file."""
+        file_path = self.output_dir / f"{table_name}.csv"
+        
+        try:
+            # First batch: write header
+            if table_name not in self._headers_written:
+                with open(file_path, 'w', newline='', encoding='utf-8') as f:
+                    writer = csv.writer(f)
+                    writer.writerow(df.columns.tolist())
+                self._headers_written[table_name] = True
+                self._rows_written[table_name] = 0
+            
+            # Append data
+            with open(file_path, 'a', newline='', encoding='utf-8') as f:
+                writer = csv.writer(f)
+                for _, row in df.iterrows():
+                    writer.writerow(row.tolist())
+            
+            rows = len(df)
+            self._rows_written[table_name] = self._rows_written.get(table_name, 0) + rows
+            
+            if self.progress_callback:
+                self.progress_callback(table_name, self._rows_written[table_name])
+            
+            return rows
+            
+        except Exception as e:
+            raise FileWriteError(f"Failed to write CSV: {e}", path=str(file_path))
+    
+    def _write_parquet_batch(self, table_name: str, df: pd.DataFrame) -> int:
+        """Write a batch to Parquet file using append mode."""
+        try:
+            import pyarrow as pa
+            import pyarrow.parquet as pq
+        except ImportError:
+            raise ExportError(
+                "PyArrow required for Parquet export",
+                details={"suggestion": "pip install pyarrow"}
+            )
+        
+        file_path = self.output_dir / f"{table_name}.parquet"
+        
+        try:
+            table = pa.Table.from_pandas(df)
+            
+            if file_path.exists():
+                # Append to existing file
+                existing = pq.read_table(file_path)
+                combined = pa.concat_tables([existing, table])
+                pq.write_table(combined, file_path)
+            else:
+                pq.write_table(table, file_path)
+            
+            rows = len(df)
+            self._rows_written[table_name] = self._rows_written.get(table_name, 0) + rows
+            
+            if self.progress_callback:
+                self.progress_callback(table_name, self._rows_written[table_name])
+            
+            return rows
+            
+        except Exception as e:
+            raise FileWriteError(f"Failed to write Parquet: {e}", path=str(file_path))
+    
+    def finalize(self) -> Dict[str, int]:
+        """Finalize all exports and return summary.
+        
+        Returns:
+            Dict mapping table names to row counts
+        """
+        # Close any open file handles
+        for handle in self._file_handles.values():
+            try:
+                handle.close()
+            except Exception:
+                pass
+        
+        self._file_handles.clear()
+        self._csv_writers.clear()
+        
+        return self._rows_written.copy()
+    
+    def get_file_paths(self) -> Dict[str, Path]:
+        """Get paths to all exported files."""
+        ext = ".csv" if self.format == "csv" else ".parquet"
+        return {
+            table: self.output_dir / f"{table}{ext}"
+            for table in self._rows_written.keys()
+        }
+    
+    def get_stats(self) -> Dict[str, Any]:
+        """Get export statistics."""
+        return {
+            "output_dir": str(self.output_dir),
+            "format": self.format,
+            "tables": self._rows_written.copy(),
+            "total_rows": sum(self._rows_written.values()),
+        }
+
+
+def stream_to_csv(
+    generator: Generator[tuple, None, None],
+    output_dir: str,
+    progress_callback: Optional[Callable[[str, int], None]] = None,
+) -> Dict[str, int]:
+    """Stream data from a generator directly to CSV files.
+    
+    Args:
+        generator: Iterator yielding (table_name, batch_df) tuples
+        output_dir: Directory to write files to
+        progress_callback: Optional callback(table_name, rows_written)
+        
+    Returns:
+        Dict mapping table names to final row counts
+    """
+    exporter = StreamingExporter(
+        output_dir=output_dir,
+        format="csv",
+        progress_callback=progress_callback,
+    )
+    
+    for table_name, batch_df in generator:
+        exporter.write_batch(table_name, batch_df)
+    
+    return exporter.finalize()
+
+
+def stream_to_parquet(
+    generator: Generator[tuple, None, None],
+    output_dir: str,
+    progress_callback: Optional[Callable[[str, int], None]] = None,
+) -> Dict[str, int]:
+    """Stream data from a generator directly to Parquet files.
+    
+    Args:
+        generator: Iterator yielding (table_name, batch_df) tuples
+        output_dir: Directory to write files to
+        progress_callback: Optional callback(table_name, rows_written)
+        
+    Returns:
+        Dict mapping table names to final row counts
+    """
+    exporter = StreamingExporter(
+        output_dir=output_dir,
+        format="parquet",
+        progress_callback=progress_callback,
+    )
+    
+    for table_name, batch_df in generator:
+        exporter.write_batch(table_name, batch_df)
+    
+    return exporter.finalize()

misata/templates/library.py

@@ -0,0 +1,344 @@
+"""
+Pre-built schema templates for common use cases.
+
+Usage:
+    from misata.templates.library import load_template, list_templates
+    
+    # See available templates
+    print(list_templates())
+    
+    # Load a template
+    config = load_template("ecommerce")
+    
+    # Generate data
+    from misata import DataSimulator
+    for table, batch in DataSimulator(config).generate_all():
+        print(f"Generated {len(batch)} rows for {table}")
+"""
+
+from misata.schema import Column, Relationship, SchemaConfig, Table
+
+
+def list_templates() -> list:
+    """List all available built-in templates."""
+    return ["ecommerce", "saas", "healthcare", "fintech"]
+
+
+def load_template(name: str, row_multiplier: float = 1.0) -> SchemaConfig:
+    """
+    Load a pre-built schema template.
+    
+    Args:
+        name: Template name (ecommerce, saas, healthcare, fintech)
+        row_multiplier: Scale row counts (e.g., 0.1 for 10%, 2.0 for 2x)
+        
+    Returns:
+        SchemaConfig ready for DataSimulator
+    """
+    templates = {
+        "ecommerce": _ecommerce_template,
+        "saas": _saas_template,
+        "healthcare": _healthcare_template,
+        "fintech": _fintech_template,
+    }
+    
+    if name not in templates:
+        raise ValueError(f"Unknown template: {name}. Available: {list(templates.keys())}")
+    
+    config = templates[name]()
+    
+    # Apply row multiplier
+    if row_multiplier != 1.0:
+        for table in config.tables:
+            if not table.is_reference:
+                table.row_count = int(table.row_count * row_multiplier)
+    
+    return config
+
+
+def _ecommerce_template() -> SchemaConfig:
+    """E-commerce platform with products, orders, reviews."""
+    return SchemaConfig(
+        name="E-commerce Platform",
+        description="Complete e-commerce dataset with products, orders, and reviews",
+        seed=42,
+        tables=[
+            # Reference tables
+            Table(
+                name="categories",
+                is_reference=True,
+                inline_data=[
+                    {"id": 1, "name": "Electronics", "margin_pct": 15},
+                    {"id": 2, "name": "Clothing", "margin_pct": 40},
+                    {"id": 3, "name": "Home & Garden", "margin_pct": 25},
+                    {"id": 4, "name": "Sports", "margin_pct": 30},
+                    {"id": 5, "name": "Books", "margin_pct": 35},
+                    {"id": 6, "name": "Beauty", "margin_pct": 50},
+                ],
+            ),
+            Table(
+                name="shipping_methods",
+                is_reference=True,
+                inline_data=[
+                    {"id": 1, "name": "Standard", "days": 5, "cost": 4.99},
+                    {"id": 2, "name": "Express", "days": 2, "cost": 9.99},
+                    {"id": 3, "name": "Next Day", "days": 1, "cost": 19.99},
+                    {"id": 4, "name": "Free Shipping", "days": 7, "cost": 0.00},
+                ],
+            ),
+            # Transactional tables
+            Table(name="customers", row_count=10000),
+            Table(name="products", row_count=500),
+            Table(name="orders", row_count=50000),
+            Table(name="order_items", row_count=150000),
+            Table(name="reviews", row_count=20000),
+        ],
+        columns={
+            "customers": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 10000}, unique=True),
+                Column(name="name", type="text", distribution_params={"text_type": "name"}),
+                Column(name="email", type="text", distribution_params={"text_type": "email"}),
+                Column(name="city", type="text", distribution_params={"text_type": "word", "smart_generate": True}),
+                Column(name="created_at", type="date", distribution_params={"start": "2020-01-01", "end": "2024-12-31"}),
+                Column(name="is_premium", type="boolean", distribution_params={"probability": 0.15}),
+            ],
+            "products": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 500}, unique=True),
+                Column(name="name", type="text", distribution_params={"text_type": "sentence"}),
+                Column(name="category_id", type="foreign_key", distribution_params={}),
+                Column(name="price", type="float", distribution_params={"distribution": "uniform", "min": 9.99, "max": 299.99, "decimals": 2}),
+                Column(name="stock", type="int", distribution_params={"distribution": "poisson", "lambda": 50}),
+            ],
+            "orders": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 50000}, unique=True),
+                Column(name="customer_id", type="foreign_key", distribution_params={}),
+                Column(name="shipping_method_id", type="foreign_key", distribution_params={}),
+                Column(name="order_date", type="date", distribution_params={"start": "2023-01-01", "end": "2024-12-31"}),
+                Column(name="status", type="categorical", distribution_params={"choices": ["completed", "pending", "shipped", "cancelled"], "probabilities": [0.6, 0.15, 0.2, 0.05]}),
+                Column(name="total", type="float", distribution_params={"distribution": "exponential", "scale": 75, "min": 10, "decimals": 2}),
+            ],
+            "order_items": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 150000}, unique=True),
+                Column(name="order_id", type="foreign_key", distribution_params={}),
+                Column(name="product_id", type="foreign_key", distribution_params={}),
+                Column(name="quantity", type="int", distribution_params={"distribution": "poisson", "lambda": 2, "min": 1}),
+                Column(name="unit_price", type="float", distribution_params={"distribution": "uniform", "min": 5.0, "max": 200.0, "decimals": 2}),
+            ],
+            "reviews": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 20000}, unique=True),
+                Column(name="product_id", type="foreign_key", distribution_params={}),
+                Column(name="customer_id", type="foreign_key", distribution_params={}),
+                Column(name="rating", type="int", distribution_params={"distribution": "categorical", "choices": [1, 2, 3, 4, 5], "probabilities": [0.05, 0.08, 0.15, 0.32, 0.40]}),
+                Column(name="title", type="text", distribution_params={"text_type": "sentence", "smart_generate": True}),
+                Column(name="created_at", type="date", distribution_params={"start": "2023-01-01", "end": "2024-12-31"}),
+            ],
+        },
+        relationships=[
+            Relationship(parent_table="categories", child_table="products", parent_key="id", child_key="category_id"),
+            Relationship(parent_table="customers", child_table="orders", parent_key="id", child_key="customer_id"),
+            Relationship(parent_table="shipping_methods", child_table="orders", parent_key="id", child_key="shipping_method_id"),
+            Relationship(parent_table="orders", child_table="order_items", parent_key="id", child_key="order_id"),
+            Relationship(parent_table="products", child_table="order_items", parent_key="id", child_key="product_id"),
+            Relationship(parent_table="products", child_table="reviews", parent_key="id", child_key="product_id"),
+            Relationship(parent_table="customers", child_table="reviews", parent_key="id", child_key="customer_id"),
+        ],
+    )
+
+
+def _saas_template() -> SchemaConfig:
+    """SaaS platform with users, subscriptions, and usage events."""
+    return SchemaConfig(
+        name="SaaS Platform",
+        description="B2B SaaS with companies, users, subscriptions, and usage tracking",
+        seed=42,
+        tables=[
+            Table(
+                name="plans",
+                is_reference=True,
+                inline_data=[
+                    {"id": 1, "name": "Free", "price": 0, "seats": 1, "features": "Basic"},
+                    {"id": 2, "name": "Starter", "price": 29, "seats": 5, "features": "Core features"},
+                    {"id": 3, "name": "Professional", "price": 99, "seats": 20, "features": "All features"},
+                    {"id": 4, "name": "Enterprise", "price": 299, "seats": 100, "features": "Custom"},
+                ],
+            ),
+            Table(name="companies", row_count=1000),
+            Table(name="users", row_count=25000),
+            Table(name="subscriptions", row_count=1200),
+            Table(name="usage_events", row_count=500000),
+        ],
+        columns={
+            "companies": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 1000}, unique=True),
+                Column(name="name", type="text", distribution_params={"text_type": "company"}),
+                Column(name="industry", type="text", distribution_params={"text_type": "word", "smart_generate": True}),
+                Column(name="employee_count", type="int", distribution_params={"distribution": "exponential", "scale": 50, "min": 1}),
+                Column(name="created_at", type="date", distribution_params={"start": "2020-01-01", "end": "2024-06-30"}),
+            ],
+            "users": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 25000}, unique=True),
+                Column(name="company_id", type="foreign_key", distribution_params={}),
+                Column(name="name", type="text", distribution_params={"text_type": "name"}),
+                Column(name="email", type="text", distribution_params={"text_type": "email"}),
+                Column(name="role", type="categorical", distribution_params={"choices": ["admin", "member", "viewer"], "probabilities": [0.1, 0.6, 0.3]}),
+                Column(name="is_active", type="boolean", distribution_params={"probability": 0.85}),
+                Column(name="last_login", type="date", distribution_params={"start": "2024-01-01", "end": "2024-12-31"}),
+            ],
+            "subscriptions": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 1200}, unique=True),
+                Column(name="company_id", type="foreign_key", distribution_params={}),
+                Column(name="plan_id", type="foreign_key", distribution_params={}),
+                Column(name="status", type="categorical", distribution_params={"choices": ["active", "cancelled", "trial", "past_due"], "probabilities": [0.7, 0.1, 0.15, 0.05]}),
+                Column(name="start_date", type="date", distribution_params={"start": "2022-01-01", "end": "2024-12-31"}),
+                Column(name="mrr", type="float", distribution_params={"distribution": "exponential", "scale": 100, "min": 0, "decimals": 2}),
+            ],
+            "usage_events": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 500000}, unique=True),
+                Column(name="user_id", type="foreign_key", distribution_params={}),
+                Column(name="event_type", type="categorical", distribution_params={"choices": ["page_view", "api_call", "export", "login", "feature_use"], "probabilities": [0.4, 0.3, 0.1, 0.1, 0.1]}),
+                Column(name="timestamp", type="datetime", distribution_params={"start": "2024-01-01", "end": "2024-12-31"}),
+            ],
+        },
+        relationships=[
+            Relationship(parent_table="companies", child_table="users", parent_key="id", child_key="company_id"),
+            Relationship(parent_table="companies", child_table="subscriptions", parent_key="id", child_key="company_id"),
+            Relationship(parent_table="plans", child_table="subscriptions", parent_key="id", child_key="plan_id"),
+            Relationship(parent_table="users", child_table="usage_events", parent_key="id", child_key="user_id"),
+        ],
+    )
+
+
+def _healthcare_template() -> SchemaConfig:
+    """Healthcare system with patients, doctors, appointments, prescriptions."""
+    return SchemaConfig(
+        name="Healthcare System",
+        description="Hospital management with patients, appointments, and prescriptions",
+        seed=42,
+        tables=[
+            Table(
+                name="specialties",
+                is_reference=True,
+                inline_data=[
+                    {"id": 1, "name": "General Practice", "avg_consult_mins": 15},
+                    {"id": 2, "name": "Cardiology", "avg_consult_mins": 30},
+                    {"id": 3, "name": "Dermatology", "avg_consult_mins": 20},
+                    {"id": 4, "name": "Orthopedics", "avg_consult_mins": 25},
+                    {"id": 5, "name": "Pediatrics", "avg_consult_mins": 20},
+                    {"id": 6, "name": "Psychiatry", "avg_consult_mins": 45},
+                    {"id": 7, "name": "Neurology", "avg_consult_mins": 30},
+                ],
+            ),
+            Table(name="patients", row_count=10000),
+            Table(name="doctors", row_count=100),
+            Table(name="appointments", row_count=50000),
+            Table(name="prescriptions", row_count=75000),
+        ],
+        columns={
+            "patients": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 10000}, unique=True),
+                Column(name="name", type="text", distribution_params={"text_type": "name"}),
+                Column(name="date_of_birth", type="date", distribution_params={"start": "1940-01-01", "end": "2020-12-31"}),
+                Column(name="gender", type="categorical", distribution_params={"choices": ["M", "F", "Other"], "probabilities": [0.48, 0.48, 0.04]}),
+                Column(name="phone", type="text", distribution_params={"text_type": "phone"}),
+                Column(name="insurance_id", type="text", distribution_params={"text_type": "word"}),
+            ],
+            "doctors": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 100}, unique=True),
+                Column(name="name", type="text", distribution_params={"text_type": "name"}),
+                Column(name="specialty_id", type="foreign_key", distribution_params={}),
+                Column(name="years_experience", type="int", distribution_params={"distribution": "normal", "mean": 15, "std": 8, "min": 1, "max": 40}),
+                Column(name="is_accepting_patients", type="boolean", distribution_params={"probability": 0.8}),
+            ],
+            "appointments": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 50000}, unique=True),
+                Column(name="patient_id", type="foreign_key", distribution_params={}),
+                Column(name="doctor_id", type="foreign_key", distribution_params={}),
+                Column(name="appointment_date", type="datetime", distribution_params={"start": "2023-01-01", "end": "2024-12-31"}),
+                Column(name="duration_mins", type="int", distribution_params={"distribution": "normal", "mean": 25, "std": 10, "min": 10, "max": 60}),
+                Column(name="status", type="categorical", distribution_params={"choices": ["completed", "scheduled", "cancelled", "no_show"], "probabilities": [0.65, 0.2, 0.1, 0.05]}),
+                Column(name="notes", type="text", distribution_params={"text_type": "sentence"}),
+            ],
+            "prescriptions": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 75000}, unique=True),
+                Column(name="appointment_id", type="foreign_key", distribution_params={}),
+                Column(name="medication", type="text", distribution_params={"text_type": "word", "smart_generate": True}),
+                Column(name="dosage", type="text", distribution_params={"text_type": "word"}),
+                Column(name="duration_days", type="int", distribution_params={"distribution": "categorical", "choices": [7, 14, 30, 60, 90], "probabilities": [0.3, 0.25, 0.25, 0.1, 0.1]}),
+            ],
+        },
+        relationships=[
+            Relationship(parent_table="specialties", child_table="doctors", parent_key="id", child_key="specialty_id"),
+            Relationship(parent_table="patients", child_table="appointments", parent_key="id", child_key="patient_id"),
+            Relationship(parent_table="doctors", child_table="appointments", parent_key="id", child_key="doctor_id"),
+            Relationship(parent_table="appointments", child_table="prescriptions", parent_key="id", child_key="appointment_id"),
+        ],
+    )
+
+
+def _fintech_template() -> SchemaConfig:
+    """Fintech platform with accounts, transactions, and fraud detection."""
+    return SchemaConfig(
+        name="Fintech Platform",
+        description="Banking/payments platform with accounts, transactions, and fraud labels",
+        seed=42,
+        tables=[
+            Table(
+                name="account_types",
+                is_reference=True,
+                inline_data=[
+                    {"id": 1, "name": "Checking", "min_balance": 0, "monthly_fee": 0},
+                    {"id": 2, "name": "Savings", "min_balance": 100, "monthly_fee": 0},
+                    {"id": 3, "name": "Premium", "min_balance": 5000, "monthly_fee": 15},
+                    {"id": 4, "name": "Business", "min_balance": 1000, "monthly_fee": 25},
+                ],
+            ),
+            Table(
+                name="transaction_types",
+                is_reference=True,
+                inline_data=[
+                    {"id": 1, "name": "deposit", "direction": "in"},
+                    {"id": 2, "name": "withdrawal", "direction": "out"},
+                    {"id": 3, "name": "transfer", "direction": "both"},
+                    {"id": 4, "name": "payment", "direction": "out"},
+                    {"id": 5, "name": "refund", "direction": "in"},
+                ],
+            ),
+            Table(name="customers", row_count=25000),
+            Table(name="accounts", row_count=35000),
+            Table(name="transactions", row_count=500000),
+        ],
+        columns={
+            "customers": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 25000}, unique=True),
+                Column(name="name", type="text", distribution_params={"text_type": "name"}),
+                Column(name="email", type="text", distribution_params={"text_type": "email"}),
+                Column(name="phone", type="text", distribution_params={"text_type": "phone"}),
+                Column(name="created_at", type="date", distribution_params={"start": "2018-01-01", "end": "2024-12-31"}),
+                Column(name="risk_score", type="int", distribution_params={"distribution": "normal", "mean": 30, "std": 20, "min": 0, "max": 100}),
+                Column(name="is_verified", type="boolean", distribution_params={"probability": 0.92}),
+            ],
+            "accounts": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 35000}, unique=True),
+                Column(name="customer_id", type="foreign_key", distribution_params={}),
+                Column(name="account_type_id", type="foreign_key", distribution_params={}),
+                Column(name="balance", type="float", distribution_params={"distribution": "exponential", "scale": 5000, "min": 0, "decimals": 2}),
+                Column(name="opened_date", type="date", distribution_params={"start": "2018-01-01", "end": "2024-12-31"}),
+                Column(name="is_active", type="boolean", distribution_params={"probability": 0.88}),
+            ],
+            "transactions": [
+                Column(name="id", type="int", distribution_params={"min": 1, "max": 500000}, unique=True),
+                Column(name="account_id", type="foreign_key", distribution_params={}),
+                Column(name="transaction_type_id", type="foreign_key", distribution_params={}),
+                Column(name="amount", type="float", distribution_params={"distribution": "exponential", "scale": 150, "min": 0.01, "decimals": 2}),
+                Column(name="timestamp", type="datetime", distribution_params={"start": "2024-01-01", "end": "2024-12-31"}),
+                Column(name="merchant", type="text", distribution_params={"text_type": "company"}),
+                Column(name="is_fraud", type="boolean", distribution_params={"probability": 0.012}),  # 1.2% fraud rate
+            ],
+        },
+        relationships=[
+            Relationship(parent_table="customers", child_table="accounts", parent_key="id", child_key="customer_id"),
+            Relationship(parent_table="account_types", child_table="accounts", parent_key="id", child_key="account_type_id"),
+            Relationship(parent_table="accounts", child_table="transactions", parent_key="id", child_key="account_id"),
+            Relationship(parent_table="transaction_types", child_table="transactions", parent_key="id", child_key="transaction_type_id"),
+        ],
+    )

{misata-0.1.0b0.dist-info → misata-0.3.0b0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: misata
-Version: 0.1.0b0
+Version: 0.3.0b0
 Summary: AI-Powered Synthetic Data Engine - Generate realistic multi-table datasets from natural language
 Author-email: Muhammed Rasin <rasinbinabdulla@gmail.com>
 License: MIT
@@ -23,6 +23,7 @@ Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Classifier: Topic :: Database
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
+License-File: LICENSE
 Requires-Dist: pandas>=2.0.0
 Requires-Dist: numpy>=1.24.0
 Requires-Dist: pydantic>=2.0.0
@@ -41,6 +42,7 @@ Requires-Dist: pytest-benchmark>=4.0.0; extra == "dev"
 Requires-Dist: black>=23.0.0; extra == "dev"
 Requires-Dist: ruff>=0.1.0; extra == "dev"
 Requires-Dist: mypy>=1.5.0; extra == "dev"
+Dynamic: license-file
 
 # 🧠 Misata
 
@@ -48,7 +50,7 @@ Requires-Dist: mypy>=1.5.0; extra == "dev"
 
 No schema writing. No training data. Just describe what you need.
 
-[![Version](https://img.shields.io/badge/version-0.1.0--beta-purple.svg)]()
+[![Version](https://img.shields.io/badge/version-0.2.0--beta-purple.svg)]()
 [![License](https://img.shields.io/badge/license-MIT-blue.svg)]()
 [![Python](https://img.shields.io/badge/python-3.10+-green.svg)]()
 

misata-0.3.0b0.dist-info/RECORD

@@ -0,0 +1,37 @@
+misata/__init__.py,sha256=Vra5zMkd5Y6HTzhGRc76jTv10Z0yuhw33MDUoLpACrE,3144
+misata/api.py,sha256=Wq2H3iJzocNTsCzb9vhYJxDyag3Yiucvb-GVF0tdKhI,14999
+misata/audit.py,sha256=4eUCHT2STptemfakWeNODbVuBRhyD8Q32LlB2eufvuw,12291
+misata/benchmark.py,sha256=Y1-tuKegJyAlTneROQpPo276qnfmMmupGDbVDs9k5J8,12358
+misata/cache.py,sha256=fuLk7cQ7hOEmlqEWmm-O516L26btZ6zFO8FdrqFCRLg,7087
+misata/cli.py,sha256=a7YijZCUYrkCYGVYJ2nZSL9J3JfFqbXQQOad6bhy7zM,22642
+misata/codegen.py,sha256=m7ykTtLgITvaqzVB1cVhs1b9Puo2X4uyzngZ85wi6J0,5791
+misata/constraints.py,sha256=8jUKlA2VyVomDnl2zz0RDkUqxEkxlwUnBbHTXr_SA5g,9937
+misata/context.py,sha256=tjYrU67wjII07Pl3MKV_uCMl_s55DIOQZCouxAryyzE,8509
+misata/curve_fitting.py,sha256=gLj4BkIxNWKkfo3QKZFI_aq60bsXlI53K5yZX4hc9EU,4126
+misata/customization.py,sha256=pw-BEsPKN091hyOrQWWQoRhTrlmQ9_PXXopm2FZSEvs,8551
+misata/exceptions.py,sha256=C3IGMk8xAy9AmRVWeSAnLHHui7drv6rzgzvOmr6gh50,8335
+misata/feedback.py,sha256=HBEsoKi_vdRqwRzMoVFVj_cjfzQ5SUAaGz40s1HMD50,13313
+misata/formulas.py,sha256=KOTq5YN_19vv1ERd92bdzKot9yo9rrrwjOuWO13nFCg,11210
+misata/generators.py,sha256=NrMF12i6CB7K6fUsqcqurmZBBQ382ZhVnYB9oMBIZCE,8844
+misata/hybrid.py,sha256=5oopAdfOLWUYzdRWlc0plVeVEVg7Nu1CVGNNCDSjQt8,13104
+misata/llm_parser.py,sha256=2SVozbKtb0kaPaR4ERz9FtIIxK5jQVaYJ8L_xC6gU10,20662
+misata/noise.py,sha256=UO7MokzQ5Y5Vj7JaayDUG0JwCLnpHtnpQTcJ4UHWibo,10460
+misata/profiles.py,sha256=0djys8wWvH8VP74KmGn6cGLuOb64h9Hk0g0bkXOfxP4,9578
+misata/quality.py,sha256=VSntJfMnF1tVWJ05fvbVJOMcAPEB7QtuEg18k6aEwhA,11685
+misata/schema.py,sha256=zMYDPCgPfcy_STgANiS-Ow3dUETpW3Ayo02G88jmBe0,8954
+misata/semantic.py,sha256=0fauGWJ75wlbHVqT0hohYTN4m_nscdaMaVAIfkhTZXk,7087
+misata/simulator.py,sha256=nq9KxOS-4oUMNu7a2Ten0TQyhT2u_rTo2ImmvdkMRbU,34037
+misata/smart_values.py,sha256=8-TYBK5cVBst9tfGuQXXetOLSqgns_NKnIl14rpVrbk,35870
+misata/story_parser.py,sha256=7N7so3KWisl2UxkOtENQwP-4hN2cs9vTKsPHVRZB2Mc,15964
+misata/streaming.py,sha256=qbEnoFRfn9a7H_gWlq5C3TwbNUnP5U98OPo1EdU_cQ0,7578
+misata/validation.py,sha256=5yJSN7jecVNLJ8ss6y7l2U4hF1Ljn27Q6Xs9N1iDPcw,10791
+misata/generators/__init__.py,sha256=V4I_1IucuywRJZH3cLxKvBd2Ib7kE0WIJ7tq8y4lkx8,568
+misata/generators/base.py,sha256=iON9iAONMEQdbq2Fdric3V3bWn3caD1ITC16DTCK0Og,21329
+misata/templates/__init__.py,sha256=0RcZz9d4bmCqLAr77h0gpMfHncqAPeZCguqsuGCz7rE,25245
+misata/templates/library.py,sha256=eMex18ZKlzQqIkGFgs1uy9QGs7PmUN_VVL4txKvxynM,20930
+misata-0.3.0b0.dist-info/licenses/LICENSE,sha256=oagkechmfr9iT214N871zCm7TnB0KTfPjAUWxHsYJ4I,1071
+misata-0.3.0b0.dist-info/METADATA,sha256=Wxpa2V0Sum-CFOpNnmRd27eEDfyT9CKIy-4nGZnrCys,8114
+misata-0.3.0b0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+misata-0.3.0b0.dist-info/entry_points.txt,sha256=k3SDuju7VnqB4AcY0Vufw-j1tWU3Ay612G3DGqoNs0U,43
+misata-0.3.0b0.dist-info/top_level.txt,sha256=dpwR99XWKUAXqNg7WiNLu_XYd7WYGmZpJzrfQXbAZFs,7
+misata-0.3.0b0.dist-info/RECORD,,