ossval 1.2.2

Open Source Software Valuation - Calculate development cost savings from OSS dependencies

OSSVAL - Open Source Software Valuation

Calculate the development cost savings from using open source software by analyzing SBOMs or package lists. Provides comprehensive cost estimation using COCOMO II and SLOCCount models with regional salary data, project type detection, and comprehensive source code analysis.

Features

• Multi-Ecosystem Support: PyPI, npm, Cargo, Maven, Go, RubyGems, and more
• Cost Estimation Models: COCOMO II and SLOCCount with configurable parameters
• Regional Salary Data: 18+ regions with accurate cost calculations
• Project Type Detection: Automatic classification with appropriate multipliers
• Source Code Analysis: SLOC counting, complexity analysis, and health metrics
• Multiple Input Formats: SBOMs (CycloneDX, SPDX) and lockfiles (requirements.txt, package.json, etc.)
• Comprehensive Output: Text, JSON, and CSV formats with detailed breakdowns

Installation

pip install ossval

Quick Start

# Analyze an SBOM file
ossval analyze sbom.json

# Specify region for salary calculations
ossval analyze sbom.json --region us_sf

# Output to JSON
ossval analyze sbom.json --format json --output results.json

# Quick estimate from SLOC
ossval estimate --sloc 50000 --region us_sf --type compiler

Usage

CLI Usage

# Analyze an SBOM or lockfile
ossval analyze pyproject.toml

# With specific region
ossval analyze package.json --region us_sf

# Output formats
ossval analyze sbom.json --format json --output results.json

# Skip repository cloning (faster, but no SLOC analysis)
ossval analyze sbom.json --no-clone

# List supported formats and configurations
ossval formats list              # Show all supported input formats
ossval formats project-types     # Show project types with cost multipliers
ossval formats methodologies     # Show available cost estimation models

# Cache management
ossval cache clear
ossval cache info

Examples

Analyze Python Project

ossval analyze pyproject.toml --region global_average

Analyze npm Project

ossval analyze package-lock.json --format json --output npm-analysis.json

Compare Regions

ossval analyze sbom.json --region us_sf > us_sf_results.txt
ossval analyze sbom.json --region global_average > global_results.txt

Quick Cost Estimate

ossval estimate --sloc 100000 --region us_sf --type framework
# Output:
# Estimated cost: $16,754,251
#   Range: $11,727,975 - $25,131,376
#   Effort: 22.1 person-years
#   Methodology: COCOMO II

Integration with SEMCL.ONE

OSSVAL is a core component of the SEMCL.ONE ecosystem, enabling comprehensive OSS valuation and cost analysis:

• Works with purl2src for repository URL discovery from Package URLs
• Integrates with purl2notices for complete legal compliance workflows
• Supports SBOM analysis from CycloneDX and SPDX formats
• Complements osslili for license analysis and compliance checking

Methodology

Cost Estimation Models

COCOMO II (Primary)

The most sophisticated model, based on Barry Boehm's COCOMO II:

• Effort Formula: Effort = a × (KSLOC)^b × EAF × Complexity × Project_Type
• Cost Formula: Cost = Effort × Monthly_Salary × Region_Multiplier
• Default Parameters: a=2.94, b=1.0997, EAF=1.0
• Confidence Range: 70%-150% of estimate
• Accounts for: Project type, complexity, team experience

SLOCCount (Alternative)

Simpler model based on David Wheeler's SLOCCount:

• Effort Formula: Effort = a × (KSLOC)^b
• Default Parameters: a=2.4, b=1.05
• Faster but less sophisticated
• Lower confidence scores

Project Types and Multipliers

OSSVAL automatically detects project types and applies appropriate cost multipliers:

Project Type	Salary Multiplier	Effort Multiplier	Examples
Cryptography	1.60x	1.26x	openssl, libsodium, bcrypt
Operating System	1.50x	1.22x	kernel, drivers, firmware
Compiler	1.50x	1.22x	gcc, llvm, babel, typescript
Database	1.40x	1.18x	postgres, mysql, redis, mongo
Machine Learning	1.40x	1.18x	tensorflow, pytorch, sklearn
Graphics	1.30x	1.14x	opengl, vulkan, game engines
Embedded	1.25x	1.12x	firmware, rtos, iot, arduino
Networking	1.20x	1.10x	http, grpc, websocket, proxy
Scientific	1.20x	1.10x	scipy, numpy, pandas, matplotlib
Framework	1.15x	1.07x	react, django, rails, spring
DevTools	1.10x	1.05x	linters, formatters, bundlers
Library	1.00x	1.00x	(baseline)
Script	0.70x	0.84x	utilities, helpers, cli tools

Complexity Levels

Code complexity affects cost estimates:

• Trivial: 0.7x multiplier
• Simple: 0.9x multiplier
• Moderate: 1.0x multiplier (baseline)
• Complex: 1.3x multiplier
• Very Complex: 1.7x multiplier

---

Contributing

We welcome contributions! Please see the repository for details on:

• Development setup
• Submitting pull requests
• Reporting issues

Support

For support and questions:

• GitHub Issues - Bug reports and feature requests
• SEMCL.ONE Community - Ecosystem support and discussions

License

Apache License 2.0 - see LICENSE file for details.

---

Part of the SEMCL.ONE ecosystem for comprehensive OSS compliance and code analysis.