Skip to content

README_modular.md

Latest commit

 

History

History
executable file
·
156 lines (116 loc) · 5.65 KB

README_modular.md

File metadata and controls

executable file
·
156 lines (116 loc) · 5.65 KB

OraSchemaGen - Modular Oracle Schema Generator

OraSchemaGen is a powerful Python utility for generating Oracle database schema exports with realistic tables, data, and advanced PL/SQL objects. It's particularly useful for development environments, testing, and educational purposes where an actual Oracle database may not be available.

Key Features

  • Complete Schema Generation: Creates tables, constraints, indexes, and sequences that mimic real Oracle databases
  • Realistic Test Data: Generates appropriate sample data based on column names and types
  • Advanced PL/SQL Objects: Produces triggers, procedures, functions, packages, and LOB operations
  • Japanese Language Support: Full support for Japanese text with Shift-JIS encoding
  • Modular Architecture: Select which object types to generate based on your needs
  • Progress Tracking: Visual progress indicators for lengthy generation tasks
  • Flexible Output Options: Generate a single file or split by object type

Installation

# Clone the repository
git clone https://github.com/yourusername/OraSchemaGen.git
cd OraSchemaGen

# Install required dependencies
pip install -r requirements.txt

Requirements

  • Python 3.6+
  • Required packages:
    • faker
    • tqdm

Usage

Basic Usage

Generate a complete Oracle schema export with default settings:

python modular_OraSchemaGen.py

This will create oracle_export.sql with 6 tables, 20 rows per table, and all PL/SQL object types in Shift-JIS encoding.

Command-line Options

usage: modular_OraSchemaGen.py [-h] [--tables TABLES] [--rows ROWS] [--output OUTPUT]
                  [--encoding {shift-jis,utf-8}] [--objects OBJECTS] [--no-header]
                  [--no-storage] [--split] [--output-dir OUTPUT_DIR]

Generate Oracle schema and objects with Shift-JIS encoding

optional arguments:
  -h, --help                Show this help message and exit
  --tables TABLES           Number of tables to generate (max 8) [default: 6]
  --rows ROWS               Number of rows per table [default: 20]
  --output OUTPUT           Output SQL file name [default: oracle_export.sql]
  --encoding {shift-jis,utf-8}
                            Output file encoding [default: shift-jis]
  --objects OBJECTS         Comma-separated list of object types to generate
                            (table,data,trigger,procedure,function,package,lob,all)
                            [default: all]
  --no-header               Skip export header and footer
  --no-storage              Skip storage clauses in CREATE statements
  --split                   Split output into multiple files by object type
  --output-dir OUTPUT_DIR   Output directory for split files

Examples

Generate a large schema with 1,000 rows per table:

python modular_OraSchemaGen.py --rows 1000 --output large_export.sql

Generate only tables and data, no PL/SQL objects:

python modular_OraSchemaGen.py --objects table,data --output simple_export.sql

Generate only tables with triggers and procedures:

python modular_OraSchemaGen.py --objects table,trigger,procedure

Generate a schema with separate files for each object type:

python modular_OraSchemaGen.py --split --output-dir ./export_files

Generate a UTF-8 encoded export instead of Shift-JIS:

python modular_OraSchemaGen.py --encoding utf-8

Create a clean schema without storage clauses:

python modular_OraSchemaGen.py --no-storage

Generated Objects

The following object types can be generated:

Tables

  • EMPLOYEES - Employee information with Japanese name fields
  • DEPARTMENTS - Department information
  • JOBS - Job titles and salary ranges
  • LOCATIONS - Location information
  • PRODUCTS - Product details
  • CUSTOMERS - Customer information with Japanese name fields
  • ORDERS - Order header information
  • ORDER_ITEMS - Order line items

Advanced PL/SQL Objects

  • Triggers: Complex compound triggers, before/after triggers, system event triggers
  • Procedures: Employee management, department operations, data validation, test data generation
  • Functions: Date handling, financial calculations, employee metrics, customer analysis
  • Packages: Employee management, Japanese text utilities, data processing utilities
  • LOB Handling: Document storage and retrieval, text extraction, thumbnail generation

Architecture

OraSchemaGen employs a modular design with these key components:

  1. Generators: Specialized classes for each object type (SchemaGenerator, TriggerGenerator, etc.)
  2. OracleObject: Represents an individual Oracle database object
  3. OracleObjectFactory: Creates appropriate generators based on requested object types
  4. OutputHandler: Manages writing objects to files with proper dependency ordering
  5. EncodingHandler: Handles character encoding conversion

Customization

The code is designed to be easily extensible:

  1. Add new object generators by extending the OracleObjectGenerator base class
  2. Modify existing generators to add new object types
  3. Expand the data generation logic to produce more specialized data

Use Cases

  • Development Environments: Create test Oracle schemas without needing a full Oracle installation
  • Migration Testing: Test Oracle-to-PostgreSQL migrations with ora2pg
  • Education: Learn Oracle database programming with realistic examples
  • Demos: Generate sample data for application demos
  • Localization Testing: Test applications with mixed English/Japanese data

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

  • The Faker library for realistic data generation
  • The tqdm library for progress visualization