mediaproc/tools/__main__.py

"""
Modelgen - Generic Model Generation Tool

Generates typed models from various sources to various formats.

Input sources:
- from-config: Configuration files (soleprint config.json style)
- from-schema: Python dataclasses in schema/ folder
- extract: Existing codebases (Django, SQLAlchemy, Prisma)

Output formats:
- pydantic: Pydantic BaseModel classes
- django: Django ORM models
- typescript: TypeScript interfaces
- protobuf: Protocol Buffer definitions
- prisma: Prisma schema

Usage:
    python -m soleprint.station.tools.modelgen --help
    python -m soleprint.station.tools.modelgen from-config -c config.json -o models.py
    python -m soleprint.station.tools.modelgen from-schema -o models/ --targets pydantic,typescript
    python -m soleprint.station.tools.modelgen extract --source /path/to/django --targets pydantic
"""

import argparse
import sys
from pathlib import Path

from .generator import GENERATORS


def cmd_from_config(args):
    """Generate models from a configuration file (soleprint config.json style)."""
    from .loader import load_config
    from .model_generator import ModelGenerator

    config_path = Path(args.config)
    if not config_path.exists():
        print(f"Error: Config file not found: {config_path}", file=sys.stderr)
        sys.exit(1)

    output_path = Path(args.output)

    print(f"Loading config: {config_path}")
    config = load_config(config_path)

    print(f"Generating {args.format} models to: {output_path}")
    generator = ModelGenerator(
        config=config,
        output_path=output_path,
        output_format=args.format,
    )
    result_path = generator.generate()

    print(f"Models generated: {result_path}")


def cmd_from_schema(args):
    """Generate models from Python dataclasses in schema/ folder."""
    from .loader import load_schema
    from .writer import write_file

    # Determine schema path
    schema_path = Path(args.schema) if args.schema else Path.cwd() / "schema"

    if not schema_path.exists():
        print(f"Error: Schema folder not found: {schema_path}", file=sys.stderr)
        print(
            "Create a schema/ folder with Python dataclasses and an __init__.py",
            file=sys.stderr,
        )
        print("that exports DATACLASSES and ENUMS lists.", file=sys.stderr)
        sys.exit(1)

    print(f"Loading schema: {schema_path}")
    schema = load_schema(schema_path)

    print(f"Found {len(schema.models)} models, {len(schema.enums)} enums")

    # Parse targets
    targets = [t.strip() for t in args.targets.split(",")]
    output_dir = Path(args.output)

    for target in targets:
        if target not in GENERATORS:
            print(f"Warning: Unknown target '{target}', skipping", file=sys.stderr)
            continue

        generator = GENERATORS[target]()
        ext = generator.file_extension()

        # Determine output filename (use target name to avoid overwrites)
        if len(targets) == 1 and args.output.endswith(ext):
            output_file = output_dir
        else:
            output_file = output_dir / f"models_{target}{ext}"

        print(f"Generating {target} to: {output_file}")
        generator.generate(schema, output_file)

    print("Done!")


def cmd_extract(args):
    """Extract models from existing codebase."""
    from .loader.extract import EXTRACTORS

    source_path = Path(args.source)
    if not source_path.exists():
        print(f"Error: Source path not found: {source_path}", file=sys.stderr)
        sys.exit(1)

    # Auto-detect or use specified framework
    framework = args.framework
    extractor = None

    if framework == "auto":
        for name, extractor_cls in EXTRACTORS.items():
            ext = extractor_cls(source_path)
            if ext.detect():
                framework = name
                extractor = ext
                print(f"Detected framework: {framework}")
                break

        if not extractor:
            print("Error: Could not auto-detect framework", file=sys.stderr)
            print(f"Available frameworks: {list(EXTRACTORS.keys())}", file=sys.stderr)
            sys.exit(1)
    else:
        if framework not in EXTRACTORS:
            print(f"Error: Unknown framework: {framework}", file=sys.stderr)
            print(f"Available: {list(EXTRACTORS.keys())}", file=sys.stderr)
            sys.exit(1)
        extractor = EXTRACTORS[framework](source_path)

    print(f"Extracting from: {source_path}")
    models, enums = extractor.extract()

    print(f"Extracted {len(models)} models, {len(enums)} enums")

    # Parse targets
    targets = [t.strip() for t in args.targets.split(",")]
    output_dir = Path(args.output)

    for target in targets:
        if target not in GENERATORS:
            print(f"Warning: Unknown target '{target}', skipping", file=sys.stderr)
            continue

        generator = GENERATORS[target]()
        ext = generator.file_extension()

        # Determine output filename (use target name to avoid overwrites)
        if len(targets) == 1 and args.output.endswith(ext):
            output_file = output_dir
        else:
            output_file = output_dir / f"models_{target}{ext}"

        print(f"Generating {target} to: {output_file}")
        generator.generate((models, enums), output_file)

    print("Done!")


def cmd_list_formats(args):
    """List available output formats."""
    print("Available output formats:")
    for fmt in GENERATORS.keys():
        print(f"  - {fmt}")


def main():
    parser = argparse.ArgumentParser(
        description="Modelgen - Generic Model Generation Tool",
        formatter_class=argparse.RawDescriptionHelpFormatter,
    )

    subparsers = parser.add_subparsers(dest="command", required=True)

    # Available formats for help text
    formats = list(GENERATORS.keys())
    formats_str = ", ".join(formats)

    # from-config command
    config_parser = subparsers.add_parser(
        "from-config",
        help="Generate models from soleprint configuration file",
    )
    config_parser.add_argument(
        "--config",
        "-c",
        type=str,
        required=True,
        help="Path to configuration file (e.g., config.json)",
    )
    config_parser.add_argument(
        "--output",
        "-o",
        type=str,
        required=True,
        help="Output path (file or directory)",
    )
    config_parser.add_argument(
        "--format",
        "-f",
        type=str,
        default="pydantic",
        choices=["pydantic"],  # Only pydantic for config mode
        help="Output format (default: pydantic)",
    )
    config_parser.set_defaults(func=cmd_from_config)

    # from-schema command
    schema_parser = subparsers.add_parser(
        "from-schema",
        help="Generate models from Python dataclasses in schema/ folder",
    )
    schema_parser.add_argument(
        "--schema",
        "-s",
        type=str,
        default=None,
        help="Path to schema folder (default: ./schema)",
    )
    schema_parser.add_argument(
        "--output",
        "-o",
        type=str,
        required=True,
        help="Output path (file or directory)",
    )
    schema_parser.add_argument(
        "--targets",
        "-t",
        type=str,
        default="pydantic",
        help=f"Comma-separated output targets ({formats_str})",
    )
    schema_parser.set_defaults(func=cmd_from_schema)

    # extract command
    extract_parser = subparsers.add_parser(
        "extract",
        help="Extract models from existing codebase",
    )
    extract_parser.add_argument(
        "--source",
        "-s",
        type=str,
        required=True,
        help="Path to source codebase",
    )
    extract_parser.add_argument(
        "--framework",
        "-f",
        type=str,
        choices=["django", "sqlalchemy", "prisma", "auto"],
        default="auto",
        help="Source framework (default: auto-detect)",
    )
    extract_parser.add_argument(
        "--output",
        "-o",
        type=str,
        required=True,
        help="Output path (file or directory)",
    )
    extract_parser.add_argument(
        "--targets",
        "-t",
        type=str,
        default="pydantic",
        help=f"Comma-separated output targets ({formats_str})",
    )
    extract_parser.set_defaults(func=cmd_extract)

    # list-formats command
    formats_parser = subparsers.add_parser(
        "list-formats",
        help="List available output formats",
    )
    formats_parser.set_defaults(func=cmd_list_formats)

    args = parser.parse_args()
    args.func(args)


if __name__ == "__main__":
    main()