markitect-main/markitect/assets/models.py

"""
Asset model classes for a clean object-oriented interface.

This module provides dataclasses for representing assets with proper
type hints and methods, following the interface expectations from tests.
"""

from dataclasses import dataclass, field
from pathlib import Path
from typing import Optional, Dict, Any, List
from datetime import datetime
from enum import Enum


class ReferenceType(Enum):
    """Types of asset references in markdown."""
    IMAGE = "image"
    LINK = "link"
    EMBED = "embed"
    REFERENCE_STYLE = "reference_style"


@dataclass
class Asset:
    """Represents a managed asset with content-addressable storage."""

    # Core identification
    content_hash: str
    filename: str

    # File properties
    size_bytes: int
    mime_type: str

    # Storage paths
    path: str  # Content-addressable storage path
    original_path: Optional[str] = None

    # Metadata
    created_at: Optional[datetime] = None
    description: Optional[str] = None
    tags: list[str] = field(default_factory=list)

    # Alternative names for compatibility with existing tests
    @property
    def size(self) -> int:
        """Alternative name for size_bytes."""
        return self.size_bytes

    @property
    def checksum(self) -> str:
        """Alternative name for content_hash."""
        return self.content_hash

    @property
    def hash(self) -> str:
        """Alternative name for content_hash."""
        return self.content_hash

    @property
    def storage_path(self) -> Path:
        """Get storage path as Path object."""
        return Path(self.path)

    def get_extension(self) -> str:
        """Get file extension."""
        return Path(self.filename).suffix.lower()

    def is_image(self) -> bool:
        """Check if asset is an image."""
        return self.mime_type.startswith('image/')

    def is_document(self) -> bool:
        """Check if asset is a document."""
        return self.mime_type in ['application/pdf', 'text/markdown', 'text/plain']

    @classmethod
    def from_dict(cls, data: Dict[str, Any]) -> 'Asset':
        """Create Asset from dictionary (for migration from dict-based storage)."""
        # Handle various field name variations
        return cls(
            content_hash=data.get('content_hash', data.get('hash', '')),
            filename=cls._extract_filename_from_path(data.get('path', '')),
            size_bytes=data.get('size_bytes', data.get('size', 0)),
            mime_type=data.get('mime_type', 'application/octet-stream'),
            path=data.get('path', ''),
            original_path=data.get('original_path'),
            created_at=cls._parse_datetime(data.get('created_at')),
            description=data.get('description'),
            tags=data.get('tags', [])
        )

    def to_dict(self) -> Dict[str, Any]:
        """Convert Asset to dictionary (for storage)."""
        return {
            'content_hash': self.content_hash,
            'filename': self.filename,
            'size_bytes': self.size_bytes,
            'mime_type': self.mime_type,
            'path': self.path,
            'original_path': self.original_path,
            'created_at': self.created_at.isoformat() if self.created_at else None,
            'description': self.description,
            'tags': self.tags
        }

    @staticmethod
    def _extract_filename_from_path(path: str) -> str:
        """Extract original filename from storage path when possible."""
        if not path:
            return ""
        storage_path = Path(path)
        # For content-addressable storage, we'll use the hash + extension
        return storage_path.name

    @staticmethod
    def _parse_datetime(dt_str: Optional[str]) -> Optional[datetime]:
        """Parse datetime string."""
        if not dt_str:
            return None
        try:
            return datetime.fromisoformat(dt_str.replace('Z', '+00:00'))
        except (ValueError, AttributeError):
            return None


@dataclass
class AssetReference:
    """Represents a reference to an asset from a markdown file."""

    source_file: Path
    asset_path: str
    reference_type: str  # 'image', 'link', etc.
    line_number: int
    alt_text: str = ""
    title: str = ""
    is_broken: bool = False
    resolved_asset: Optional[Asset] = None


@dataclass
class AssetCollection:
    """Represents a collection of assets with metadata."""

    assets: list[Asset] = field(default_factory=list)
    total_size: int = 0
    created_at: Optional[datetime] = None

    def __post_init__(self):
        """Calculate total size."""
        self.total_size = sum(asset.size_bytes for asset in self.assets)

    def filter_by_type(self, mime_type_prefix: str) -> 'AssetCollection':
        """Filter assets by MIME type prefix."""
        filtered = [asset for asset in self.assets
                   if asset.mime_type.startswith(mime_type_prefix)]
        return AssetCollection(assets=filtered)

    def get_images(self) -> 'AssetCollection':
        """Get only image assets."""
        return self.filter_by_type('image/')

    def get_documents(self) -> 'AssetCollection':
        """Get only document assets."""
        docs = [asset for asset in self.assets if asset.is_document()]
        return AssetCollection(assets=docs)