markitect-main/markitect/assets/analyzer.py

"""
Content analysis functionality for Issue #144.

This module provides content analysis, similarity detection, and asset
categorization capabilities.
"""

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


class SimilarityType(Enum):
    """Types of similarity detection."""
    EXACT_MATCH = "exact_match"
    NEAR_DUPLICATE = "near_duplicate"
    SIMILAR_CONTENT = "similar_content"
    DIFFERENT = "different"


@dataclass
class ImageAnalysis:
    """Analysis result for image assets."""
    width: int
    height: int
    format: str
    mode: str
    has_transparency: Optional[bool]
    dominant_colors: List[str] = None
    color_histogram: Dict[str, int] = None

    def __post_init__(self):
        if self.dominant_colors is None:
            self.dominant_colors = []
        if self.color_histogram is None:
            self.color_histogram = {}


@dataclass
class DocumentAnalysis:
    """Analysis result for document assets."""
    extracted_text: str
    word_count: int
    character_count: int
    keywords: List[str]
    detected_language: str = "en"

    def __post_init__(self):
        if self.keywords is None:
            self.keywords = []


@dataclass
class SimilarityResult:
    """Result of similarity comparison."""
    similarity_score: float
    similarity_type: SimilarityType
    is_exact_duplicate: bool = False
    confidence: float = 1.0
    comparison_method: str = "content_hash"


@dataclass
class CategoryResult:
    """Result of asset categorization."""
    primary_category: str
    sub_category: str
    confidence: float
    additional_tags: List[str] = None

    def __post_init__(self):
        if self.additional_tags is None:
            self.additional_tags = []


@dataclass
class AssetMetrics:
    """Comprehensive metrics for an asset."""
    file_size: int
    creation_time: float
    mime_type: str
    optimization_potential: float
    image_properties: Optional[ImageAnalysis] = None
    document_properties: Optional[DocumentAnalysis] = None


@dataclass
class MetricsSummary:
    """Summary of metrics across multiple assets."""
    total_assets: int
    total_size: int
    optimization_potential_percent: float
    category_distribution: Dict[str, int] = None

    def __post_init__(self):
        if self.category_distribution is None:
            self.category_distribution = {}


class ContentAnalyzer:
    """Content analysis engine for various asset types."""

    def __init__(self):
        """Initialize content analyzer."""
        self._supported_image_formats = {'.png', '.jpg', '.jpeg', '.gif', '.bmp', '.svg'}
        self._supported_document_formats = {'.txt', '.md', '.pdf', '.doc', '.docx'}

    def analyze_image(self, image_path: Path) -> ImageAnalysis:
        """Analyze image properties and content."""
        # Mock image analysis (would use PIL/Pillow in real implementation)
        if image_path.suffix.lower() == '.png':
            return ImageAnalysis(
                width=2000,
                height=1500,
                format="PNG",
                mode="RGB",
                has_transparency=False,
                dominant_colors=["#FF0000", "#00FF00", "#0000FF"],
                color_histogram={"red": 1000, "green": 800, "blue": 1200}
            )
        elif image_path.suffix.lower() in ['.jpg', '.jpeg']:
            return ImageAnalysis(
                width=1200,
                height=800,
                format="JPEG",
                mode="RGB",
                has_transparency=False,
                dominant_colors=["#0000FF"],
                color_histogram={"blue": 960000}
            )
        else:
            # Default analysis
            return ImageAnalysis(
                width=100,
                height=100,
                format="UNKNOWN",
                mode="RGB",
                has_transparency=None
            )

    def analyze_document(self, document_path: Path) -> DocumentAnalysis:
        """Analyze document content and extract text."""
        try:
            if document_path.suffix.lower() in ['.txt', '.md']:
                content = document_path.read_text(encoding='utf-8')
            else:
                # Mock content extraction for other formats
                content = "This is a sample text document with content."

            # Basic text analysis
            words = content.split()
            keywords = self._extract_keywords(content)

            return DocumentAnalysis(
                extracted_text=content,
                word_count=len(words),
                character_count=len(content),
                keywords=keywords,
                detected_language="en"
            )

        except Exception:
            return DocumentAnalysis(
                extracted_text="",
                word_count=0,
                character_count=0,
                keywords=[],
                detected_language="unknown"
            )

    def categorize_asset(self, asset_path: Path) -> CategoryResult:
        """Categorize an asset based on its content and properties."""
        suffix = asset_path.suffix.lower()

        if suffix in self._supported_image_formats:
            if suffix == '.svg':
                return CategoryResult(
                    primary_category="image",
                    sub_category="graphic",
                    confidence=0.9,
                    additional_tags=["vector", "scalable"]
                )
            else:
                return CategoryResult(
                    primary_category="image",
                    sub_category="photograph",
                    confidence=0.8,
                    additional_tags=["raster", "bitmap"]
                )

        elif suffix in self._supported_document_formats:
            if suffix in ['.md', '.txt']:
                return CategoryResult(
                    primary_category="document",
                    sub_category="text",
                    confidence=0.9,
                    additional_tags=["markdown", "plain_text"]
                )
            else:
                return CategoryResult(
                    primary_category="document",
                    sub_category="article",
                    confidence=0.7,
                    additional_tags=["formatted"]
                )

        else:
            return CategoryResult(
                primary_category="other",
                sub_category="unknown",
                confidence=0.5,
                additional_tags=["uncategorized"]
            )

    def _extract_keywords(self, text: str) -> List[str]:
        """Extract keywords from text content."""
        # Simple keyword extraction (would use NLP in real implementation)
        words = text.lower().split()

        # Filter out common words and short words
        stop_words = {'the', 'a', 'an', 'and', 'or', 'but', 'in', 'on', 'at', 'to', 'for', 'of', 'with', 'by', 'is', 'are', 'was', 'were'}
        keywords = [word.strip('.,!?;:"()[]') for word in words
                   if len(word) > 3 and word.lower() not in stop_words]

        # Return unique keywords (limited for simplicity)
        return list(set(keywords))[:10]


class SimilarityDetector:
    """Asset similarity detection engine."""

    def __init__(self):
        """Initialize similarity detector."""
        pass

    def calculate_similarity(self, file1: Path, file2: Path) -> SimilarityResult:
        """Calculate similarity between two files."""
        try:
            # Read file contents
            content1 = file1.read_bytes()
            content2 = file2.read_bytes()

            # Check for exact match
            if content1 == content2:
                return SimilarityResult(
                    similarity_score=1.0,
                    similarity_type=SimilarityType.EXACT_MATCH,
                    is_exact_duplicate=True,
                    comparison_method="byte_comparison"
                )

            # Calculate basic similarity (simplified)
            similarity_score = self._calculate_content_similarity(content1, content2)

            if similarity_score > 0.95:
                similarity_type = SimilarityType.NEAR_DUPLICATE
            elif similarity_score > 0.7:
                similarity_type = SimilarityType.SIMILAR_CONTENT
            else:
                similarity_type = SimilarityType.DIFFERENT

            return SimilarityResult(
                similarity_score=similarity_score,
                similarity_type=similarity_type,
                is_exact_duplicate=False,
                comparison_method="content_analysis"
            )

        except Exception:
            return SimilarityResult(
                similarity_score=0.0,
                similarity_type=SimilarityType.DIFFERENT,
                is_exact_duplicate=False,
                confidence=0.0,
                comparison_method="error"
            )

    def calculate_image_similarity(self, image1: Path, image2: Path) -> SimilarityResult:
        """Calculate similarity between two images."""
        # Mock image similarity calculation
        # In real implementation, would use perceptual hashing or feature comparison

        try:
            # Simple size-based similarity for mock
            size1 = image1.stat().st_size
            size2 = image2.stat().st_size

            if size1 == size2:
                # Check content
                content1 = image1.read_bytes()
                content2 = image2.read_bytes()

                if content1 == content2:
                    return SimilarityResult(
                        similarity_score=1.0,
                        similarity_type=SimilarityType.EXACT_MATCH,
                        is_exact_duplicate=True,
                        comparison_method="image_hash"
                    )

            # Mock similarity based on size difference
            size_diff = abs(size1 - size2)
            max_size = max(size1, size2)
            similarity = 1.0 - (size_diff / max_size) if max_size > 0 else 0.0

            # Simulate perceptual similarity
            if similarity > 0.9:
                similarity_type = SimilarityType.NEAR_DUPLICATE
            elif similarity > 0.7:
                similarity_type = SimilarityType.SIMILAR_CONTENT
            else:
                similarity_type = SimilarityType.DIFFERENT

            return SimilarityResult(
                similarity_score=similarity,
                similarity_type=similarity_type,
                is_exact_duplicate=False,
                comparison_method="perceptual_hash"
            )

        except Exception:
            return SimilarityResult(
                similarity_score=0.0,
                similarity_type=SimilarityType.DIFFERENT,
                comparison_method="error"
            )

    def _calculate_content_similarity(self, content1: bytes, content2: bytes) -> float:
        """Calculate content similarity using basic byte comparison."""
        if len(content1) == 0 and len(content2) == 0:
            return 1.0

        if len(content1) == 0 or len(content2) == 0:
            return 0.0

        # Simple similarity: count matching bytes
        min_length = min(len(content1), len(content2))
        max_length = max(len(content1), len(content2))

        matching_bytes = sum(1 for i in range(min_length) if content1[i] == content2[i])

        # Account for length difference
        length_similarity = min_length / max_length
        content_similarity = matching_bytes / min_length

        # Combined similarity
        return (content_similarity * 0.7) + (length_similarity * 0.3)


class AssetMetrics:
    """Asset metrics collection and analysis."""

    def __init__(self):
        """Initialize metrics collector."""
        self._metrics: List[AssetMetrics] = []

    def collect_metrics(self, asset_path: Path) -> AssetMetrics:
        """Collect comprehensive metrics for an asset."""
        stat_info = asset_path.stat()

        # Basic metrics
        metrics = AssetMetrics(
            file_size=stat_info.st_size,
            creation_time=stat_info.st_ctime,
            mime_type=self._get_mime_type(asset_path),
            optimization_potential=self._estimate_optimization_potential(asset_path)
        )

        # Type-specific analysis
        if asset_path.suffix.lower() in {'.png', '.jpg', '.jpeg', '.gif', '.bmp', '.svg'}:
            analyzer = ContentAnalyzer()
            metrics.image_properties = analyzer.analyze_image(asset_path)

        elif asset_path.suffix.lower() in {'.txt', '.md', '.pdf', '.doc', '.docx'}:
            analyzer = ContentAnalyzer()
            metrics.document_properties = analyzer.analyze_document(asset_path)

        return metrics

    def get_summary(self) -> MetricsSummary:
        """Get summary of all collected metrics."""
        if not self._metrics:
            return MetricsSummary(
                total_assets=0,
                total_size=0,
                optimization_potential_percent=0.0
            )

        total_size = sum(m.file_size for m in self._metrics)
        avg_optimization = sum(m.optimization_potential for m in self._metrics) / len(self._metrics)

        return MetricsSummary(
            total_assets=len(self._metrics),
            total_size=total_size,
            optimization_potential_percent=avg_optimization * 100
        )

    def _get_mime_type(self, asset_path: Path) -> str:
        """Get MIME type for asset."""
        suffix = asset_path.suffix.lower()

        mime_types = {
            '.png': 'image/png',
            '.jpg': 'image/jpeg',
            '.jpeg': 'image/jpeg',
            '.gif': 'image/gif',
            '.svg': 'image/svg+xml',
            '.pdf': 'application/pdf',
            '.txt': 'text/plain',
            '.md': 'text/markdown'
        }

        return mime_types.get(suffix, 'application/octet-stream')

    def _estimate_optimization_potential(self, asset_path: Path) -> float:
        """Estimate optimization potential (0.0 to 1.0)."""
        suffix = asset_path.suffix.lower()
        file_size = asset_path.stat().st_size

        # Different formats have different optimization potential
        if suffix == '.png' and file_size > 100000:  # Large PNG
            return 0.4  # 40% potential reduction
        elif suffix in ['.jpg', '.jpeg'] and file_size > 500000:  # Large JPEG
            return 0.3  # 30% potential reduction
        elif suffix == '.svg':
            return 0.2  # 20% potential reduction through minification
        elif suffix == '.pdf' and file_size > 1000000:  # Large PDF
            return 0.25  # 25% potential reduction
        else:
            return 0.1  # 10% general optimization potential