dl-organizer/tests/test_reports_integration.py

"""Integration tests for report generation with analysis engine.

Tests the complete workflow from analysis to report generation.
"""

import json
from pathlib import Path
from datetime import datetime, timezone
from vlm.models import SeriesIdentity, VideoFile, MovieIdentity
from vlm.analysis import analyze_series_completeness, detect_duplicates
from vlm.reports import generate_completeness_report, generate_duplicate_report, generate_summary_report


class TestReportsIntegration:
    """Test report generation integrated with analysis engine."""
    
    def test_completeness_workflow(self):
        """Test complete workflow from series analysis to completeness report."""
        # Create test episodes with gaps
        episodes = [
            SeriesIdentity("Breaking Bad", 1, [1], 0.9, False, "Breaking.Bad.S01E01.mkv"),
            SeriesIdentity("Breaking Bad", 1, [2], 0.9, False, "Breaking.Bad.S01E02.mkv"),
            SeriesIdentity("Breaking Bad", 1, [4], 0.9, False, "Breaking.Bad.S01E04.mkv"),
            SeriesIdentity("The Wire", 1, [1], 0.9, False, "The.Wire.S01E01.mkv"),
            SeriesIdentity("The Wire", 1, [3], 0.9, False, "The.Wire.S01E03.mkv"),
        ]
        
        # Analyze completeness
        analysis = analyze_series_completeness(episodes)
        
        # Generate text report
        library_root = Path("/mnt/nas/videos")
        text_report = generate_completeness_report(analysis, "text", library_root)
        
        # Verify report contains expected information
        assert "Breaking Bad" in text_report
        assert "The Wire" in text_report
        assert "Episodes missing:" in text_report
        
        # Generate JSON report
        json_report = generate_completeness_report(analysis, "json", library_root)
        data = json.loads(json_report)
        
        # Verify JSON structure
        assert data["metadata"]["series_count"] == 2
        assert len(data["series"]) == 2
    
    def test_duplicate_workflow(self):
        """Test complete workflow from duplicate detection to duplicate report."""
        # Create test identities and files
        identities = [
            MovieIdentity("The Matrix", 1999, 0.9, False, "The.Matrix.1999.1080p.mkv"),
            MovieIdentity("The Matrix", 1999, 0.9, False, "The.Matrix.1999.720p.mkv"),
            MovieIdentity("Inception", 2010, 0.9, False, "Inception.2010.mkv"),
        ]
        
        files = [
            VideoFile(
                Path("/movies/The.Matrix.1999.1080p.mkv"),
                "The.Matrix.1999.1080p.mkv",
                2000000000,
                datetime.now(timezone.utc),
                "movie",
                resolution="1920x1080",
                codec="h264"
            ),
            VideoFile(
                Path("/movies/The.Matrix.1999.720p.mkv"),
                "The.Matrix.1999.720p.mkv",
                1000000000,
                datetime.now(timezone.utc),
                "movie",
                resolution="1280x720",
                codec="h264"
            ),
            VideoFile(
                Path("/movies/Inception.2010.mkv"),
                "Inception.2010.mkv",
                1500000000,
                datetime.now(timezone.utc),
                "movie"
            ),
        ]
        
        # Detect duplicates
        duplicates = detect_duplicates(list(zip(identities, files)))
        
        # Generate text report
        library_root = Path("/mnt/nas/videos")
        text_report = generate_duplicate_report(duplicates, "text", library_root)
        
        # Verify report contains expected information
        assert "The Matrix (1999)" in text_report
        assert "1920x1080" in text_report
        assert "1280x720" in text_report
        
        # Generate JSON report
        json_report = generate_duplicate_report(duplicates, "json", library_root)
        data = json.loads(json_report)
        
        # Verify JSON structure
        assert data["metadata"]["duplicate_groups"] == 1
        assert len(data["duplicates"]) == 1
        assert data["duplicates"][0]["file_count"] == 2
    
    def test_summary_workflow(self):
        """Test summary report generation with mixed file types."""
        # Create test files
        files = [
            VideoFile(Path("/movies/Movie1.mkv"), "Movie1.mkv", 2000000000, datetime.now(timezone.utc), "movie"),
            VideoFile(Path("/movies/Movie2.mkv"), "Movie2.mkv", 1500000000, datetime.now(timezone.utc), "movie"),
            VideoFile(Path("/series/Show.S01E01.mkv"), "Show.S01E01.mkv", 1000000000, datetime.now(timezone.utc), "series"),
            VideoFile(Path("/series/Show.S01E02.mkv"), "Show.S01E02.mkv", 1000000000, datetime.now(timezone.utc), "series"),
            VideoFile(Path("/anime/Anime1.mkv"), "Anime1.mkv", 800000000, datetime.now(timezone.utc), "anime"),
        ]
        
        # Generate summary report
        library_root = Path("/mnt/nas/videos")
        report = generate_summary_report(files, library_root)
        
        # Verify report contains expected information
        assert "Total Files: 5" in report
        assert "Movie:" in report
        assert "Series:" in report
        assert "Anime:" in report
        assert "Category Breakdown:" in report
    
    def test_all_reports_include_metadata(self):
        """Test that all reports include generation timestamp and library root."""
        library_root = Path("/mnt/nas/videos")
        
        # Test completeness report
        completeness_report = generate_completeness_report([], "text", library_root)
        assert "Generated:" in completeness_report
        assert str(library_root) in completeness_report
        
        # Test duplicate report
        duplicate_report = generate_duplicate_report([], "text", library_root)
        assert "Generated:" in duplicate_report
        assert str(library_root) in duplicate_report
        
        # Test summary report
        summary_report = generate_summary_report([], library_root)
        assert "Generated:" in summary_report
        assert str(library_root) in summary_report