utils/generator.py

"""
Generator utilities for creating JSON and Markdown outputs
"""

import json
from datetime import datetime

def generate_json_output(metadata, top_level, processes, generator_version):
    """
    Generate JSON output for the model card
    
    Args:
        metadata: Dictionary of metadata fields
        top_level: Dictionary of top-level elements (Figure 6)
        processes: Dictionary of data & process elements (Figure 7)
        generator_version: Version of the generator
    
    Returns:
        String containing JSON data ready for file download
    """
    output = {
        "df_model_card_metadata": metadata,
        "df_mc_0_top_level": {
            k: v for k, v in top_level.items() 
            if v.get("applicable") and v.get("description")
        },
        "df_mc_1_data_processes": {
            k: v for k, v in processes.items()
            if v.get("applicable") and v.get("description")
        },
        "generated_at": datetime.utcnow().isoformat() + "Z",
        "generator_version": generator_version,
        "schema_version": "1.0"
    }
    
    return json.dumps(output, indent=2)


def generate_markdown_output(metadata, top_level, processes, generator_version):
    """
    Generate Markdown README output for the model card
    
    Args:
        metadata: Dictionary of metadata fields
        top_level: Dictionary of top-level elements (Figure 6)
        processes: Dictionary of data & process elements (Figure 7)
        generator_version: Version of the generator
    
    Returns:
        String containing Markdown content ready for file download
    """
    
    md = []
    
    # Header
    md.append("# Digital Forensics Model Card\n")
    md.append(f"**Generated:** {datetime.utcnow().strftime('%Y-%m-%d %H:%M:%S')} UTC\n")
    md.append("---\n")
    
    # Metadata Section
    md.append("## Metadata\n")
    md.append(f"- **Identifier (MMCID):** {metadata.get('mmcid', 'Not specified')}")
    md.append(f"- **Version (MCV):** {metadata.get('version', 'N/A')}")
    md.append(f"- **Owner:** {metadata.get('owner', 'Not specified')}")
    md.append(f"- **Usage Context:** {metadata.get('use_context', 'Not specified')}")
    md.append(f"- **Layer/Stage (Ln):** {metadata.get('layer_n', 'N/A')}\n")
    
    # Case Statement
    if metadata.get('case_statement'):
        md.append("### Case Statement")
        md.append(f"{metadata['case_statement']}\n")
    
    # Hypothesis
    if metadata.get('hypothesis'):
        md.append("### Hypothesis")
        md.append(f"{metadata['hypothesis']}\n")
    
    # Classification
    if metadata.get('classification'):
        md.append("### Classification")
        for item in metadata['classification']:
            md.append(f"- {item}")
        md.append("")
    
    # Reasoning Type
    if metadata.get('reasoning_type'):
        md.append("### Type of Reasoning")
        for item in metadata['reasoning_type']:
            md.append(f"- {item}")
        md.append("")
    
    # Bias
    if metadata.get('bias'):
        md.append("### Identified Bias")
        for item in metadata['bias']:
            md.append(f"- {item}")
        md.append("")
        
        if metadata.get('cause_of_bias'):
            md.append("**Cause(s) of Bias:**")
            for item in metadata['cause_of_bias']:
                md.append(f"- {item}")
            md.append("")
    
    # Error
    if metadata.get('error'):
        md.append("### Error")
        md.append(f"{metadata['error']}\n")
        
        if metadata.get('cause_of_error'):
            md.append("**Cause(s) of Error:**")
            for item in metadata['cause_of_error']:
                md.append(f"- {item}")
            md.append("")
    
    md.append("---\n")
    
    # Top Level Elements (Figure 6)
    md.append("## Top Level Elements (DF MC 0)\n")
    md.append("*Based on Figure 6 - Top Level Elements*\n")
    
    applicable_top_level = {
        k: v for k, v in top_level.items() 
        if v.get("applicable") and v.get("description")
    }
    
    if applicable_top_level:
        for key, value in applicable_top_level.items():
            title = key.replace("_", " ").title()
            md.append(f"### {title}")
            md.append(f"{value['description']}\n")
    else:
        md.append("*No top-level elements specified.*\n")
    
    md.append("---\n")
    
    # Data & Processes (Figure 7)
    md.append("## Data Types and Analytical Processes (DF MC 1)\n")
    md.append("*Based on Figure 7 - DF Data types and analytical processes (Hargreaves et al., 2024)*\n")
    
    applicable_processes = {
        k: v for k, v in processes.items()
        if v.get("applicable") and v.get("description")
    }
    
    if applicable_processes:
        for key, value in applicable_processes.items():
            title = key.replace("_", " ").title()
            md.append(f"### {title}")
            md.append(f"{value['description']}\n")
    else:
        md.append("*No data processes specified.*\n")
    
    md.append("---\n")
    
    # References
    md.append("## References\n")
    md.append("This model card is based on the following works:\n")
    md.append("1. **Di Maio, P.** (2024). Towards Open Standards for Systemic Complexity in Digital Forensics. https://papers.cool/arxiv/2512.12970")
    md.append("2. **Hargreaves, C., Nelson, A., & Casey, E.** (2024). An abstract model for digital forensic analysis tools—A foundation for systematic error mitigation analysis. *Forensic Science International: Digital Investigation*, 48.\n")
    
    # Footer
    md.append("---\n")
    md.append(f"*Generated by [Digital Forensics Model Card Generator](https://huggingface.co/spaces/forensic-model-card-generator) v{generator_version}*")
    
    return "\n".join(md)