markbase/filetree-hybrid/src/real_scenario.rs

use anyhow::Result;
use filetree_hybrid::HybridRouter;
use std::time::Instant;

fn main() -> Result<()> {
    println!("=== Real Scenario Validation Test ===\n");
    
    println!("Simulating realistic user access patterns:");
    println!("  - 20% hot files (frequently accessed)");
    println!("  - 80% cold files (rarely accessed)");
    println!("  - Cache warmup before testing");
    println!("  - LRU eviction mechanism");
    println!("  - Target: 85%+ cache hit rate\n");
    
    let user_id = "real_scenario_test";
    let router = HybridRouter::init_user_db(user_id)?;
    
    println!("=== Phase 1: Setup Test Data ===");
    
    println!("\n1.1 Creating 10,000 nodes (mixed structure)...");
    let start = Instant::now();
    
    // Create root folders (10 hot folders)
    let root_folders: Vec<filetree_hybrid::FileNode> = (0..10)
        .map(|i| HybridRouter::new_folder(&format!("hot_folder_{}", i), None))
        .collect();
    
    // Create child nodes for hot folders (100 nodes each = 1000 hot nodes)
    let hot_nodes: Vec<filetree_hybrid::FileNode> = root_folders.iter()
        .flat_map(|folder| {
            (0..100).map(|i| {
                HybridRouter::new_folder(&format!("hot_node_{}_{}", folder.node_id.chars().take(8).collect::<String>(), i), Some(&folder.node_id))
            })
        })
        .collect();
    
    // Create cold folders (90 folders)
    let cold_folders: Vec<filetree_hybrid::FileNode> = (0..90)
        .map(|i| HybridRouter::new_folder(&format!("cold_folder_{}", i), None))
        .collect();
    
    // Create child nodes for cold folders (100 nodes each = 9000 cold nodes)
    let cold_nodes: Vec<filetree_hybrid::FileNode> = cold_folders.iter()
        .flat_map(|folder| {
            (0..100).map(|i| {
                HybridRouter::new_folder(&format!("cold_node_{}_{}", folder.node_id.chars().take(8).collect::<String>(), i), Some(&folder.node_id))
            })
        })
        .collect();
    
    // Batch insert all nodes
    router.insert_node_batch(&root_folders)?;
    router.insert_node_batch(&hot_nodes)?;
    router.insert_node_batch(&cold_folders)?;
    router.insert_node_batch(&cold_nodes)?;
    
    println!("  ✓ Total nodes: {}", router.count_nodes()?);
    println!("  ✓ Hot nodes: {}", hot_nodes.len());
    println!("  ✓ Cold nodes: {}", cold_nodes.len());
    println!("  ✓ Insert time: {:?}", start.elapsed());
    
    println!("\n=== Phase 2: Cache Warmup ===");
    
    println!("\n2.1 Warming up cache with hot nodes...");
    let start = Instant::now();
    let hot_node_ids: Vec<String> = hot_nodes.iter().map(|n| n.node_id.clone()).collect();
    let warmed = router.warmup_cache(&hot_node_ids)?;
    println!("  ✓ Warmed {} nodes", warmed);
    println!("  ✓ Warmup time: {:?}", start.elapsed());
    
    println!("\n2.2 Warming up cache by pattern (folders)...");
    let start = Instant::now();
    let warmed_folders = router.warmup_cache_by_pattern("%_folder_%")?;
    println!("  ✓ Warmed {} folder nodes", warmed_folders);
    println!("  ✓ Pattern warmup time: {:?}", start.elapsed());
    
    println!("\n2.3 Cache stats after warmup...");
    let stats = router.get_cache_stats()?;
    println!("  ✓ Cache size: {}", stats.cache_size);
    println!("  ✓ Hot count: {}", stats.hot_count);
    println!("  ✓ Cold count: {}", stats.cold_count);
    println!("  ✓ Expired count: {}", stats.expired_count);
    println!("  ✓ Avg TTL: {:.2} seconds", stats.avg_ttl);
    
    println!("\n=== Phase 3: Realistic Access Simulation ===");
    
    println!("\n3.1 Simulating 10,000 queries with realistic distribution...");
    println!("  Query pattern:");
    println!("    80%: Hot files (1000 nodes, 8000 queries)");
    println!("    20%: Cold files (9000 nodes, 2000 queries)");
    
    let start = Instant::now();
    let mut queries = 0;
    
    // Simulate hot file queries (80% of traffic)
    for i in 0..8000 {
        let node_id = &hot_nodes[i % hot_nodes.len()].node_id;
        let _ = router.get_node(node_id)?;
        queries += 1;
    }
    
    // Simulate cold file queries (20% of traffic)
    for i in 0..2000 {
        let node_id = &cold_nodes[i % cold_nodes.len()].node_id;
        let _ = router.get_node(node_id)?;
        queries += 1;
    }
    
    let query_time = start.elapsed();
    let metrics = router.get_metrics();
    
    println!("  ✓ Total queries: {}", queries);
    println!("  ✓ Query time: {:?}", query_time);
    println!("  ✓ Cache hits: {}", metrics.cache_hits);
    println!("  ✓ Cache misses: {}", metrics.cache_misses);
    println!("  ✓ Cache hit rate: {:.2}%", metrics.hit_rate() * 100.0);
    println!("  ✓ Avg cache latency: {:?}", metrics.avg_cache_latency);
    println!("  ✓ Avg SQLite latency: {:?}", metrics.avg_sqlite_latency);
    
    println!("\n=== Phase 4: LRU Eviction Test ===");
    
    println!("\n4.1 Testing LRU eviction mechanism...");
    println!("  Current cache size: {}", router.cache_size()?);
    println!("  Max cache size: 10000 (config default)");
    
    println!("\n4.2 Running eviction (if needed)...");
    let start = Instant::now();
    let evicted = router.lru_eviction()?;
    println!("  ✓ Evicted {} nodes", evicted);
    println!("  ✓ Eviction time: {:?}", start.elapsed());
    
    println!("\n4.3 Cache size after eviction...");
    println!("  ✓ Cache size: {}", router.cache_size()?);
    
    println!("\n=== Phase 5: Long-term Simulation ===");
    
    println!("\n5.1 Simulating 1 hour of usage (100K queries)...");
    println!("  Query pattern: Same distribution (80% hot, 20% cold)");
    
    let start = Instant::now();
    let mut queries = 0;
    
    // Simulate 1 hour usage
    for i in 0..80000 {
        let node_id = &hot_nodes[i % hot_nodes.len()].node_id;
        let _ = router.get_node(node_id)?;
        queries += 1;
    }
    
    for i in 0..20000 {
        let node_id = &cold_nodes[i % cold_nodes.len()].node_id;
        let _ = router.get_node(node_id)?;
        queries += 1;
    }
    
    let usage_time = start.elapsed();
    let metrics = router.get_metrics();
    
    println!("  ✓ Total queries: {}", queries);
    println!("  ✓ Usage time: {:?}", usage_time);
    println!("  ✓ Cache hits: {}", metrics.cache_hits);
    println!("  ✓ Cache misses: {}", metrics.cache_misses);
    println!("  ✓ Cache hit rate: {:.2}%", metrics.hit_rate() * 100.0);
    
    println!("\n5.2 Cache stats after long-term usage...");
    let stats = router.get_cache_stats()?;
    println!("  ✓ Cache size: {}", stats.cache_size);
    println!("  ✓ Hot count: {}", stats.hot_count);
    println!("  ✓ Cold count: {}", stats.cold_count);
    println!("  ✓ Avg TTL: {:.2} seconds", stats.avg_ttl);
    
    println!("\n=== Phase 6: Performance Validation ===");
    
    println!("\n6.1 Cache hit rate validation...");
    let hit_rate = metrics.hit_rate() * 100.0;
    println!("  ✓ Target: 85%+");
    println!("  ✓ Actual: {:.2}%", hit_rate);
    
    if hit_rate >= 85.0 {
        println!("  ✅ PASS: Cache hit rate meets target!");
    } else {
        println!("  ⚠️ WARNING: Cache hit rate below target (need optimization)");
    }
    
    println!("\n6.2 Query latency validation...");
    let avg_latency_ns = query_time.as_nanos() as f64 / 10000.0;
    println!("  ✓ Target: <5ms");
    println!("  ✓ Actual: {:.2} ns ({:.2} ms)", avg_latency_ns, avg_latency_ns / 1_000_000.0);
    
    if avg_latency_ns < 5_000_000.0 {
        println!("  ✅ PASS: Query latency meets target!");
    } else {
        println!("  ⚠️ WARNING: Query latency above target");
    }
    
    println!("\n6.3 Database size comparison...");
    let db_path = HybridRouter::user_db_path(user_id);
    let sqlite_path = format!("{}.sqlite", db_path);
    let sled_path = format!("{}.sled", db_path);
    
    let sqlite_size = std::fs::metadata(&sqlite_path)?.len();
    let sled_size = get_directory_size(&sled_path)?;
    let total_size = sqlite_size + sled_size;
    
    println!("  ✓ SQLite size: {:.2} MB", sqlite_size as f64 / 1024.0 / 1024.0);
    println!("  ✓ Sled cache size: {:.2} MB", sled_size as f64 / 1024.0 / 1024.0);
    println!("  ✓ Total size: {:.2} MB", total_size as f64 / 1024.0 / 1024.0);
    
    println!("\n=== Validation Summary ===");
    println!("┌─────────────────────────────────────────┐");
    println!("│ Metric              │ Target │ Actual   │");
    println!("├─────────────────────────────────────────┤");
    println!("│ Cache hit rate      │ 85%+   │ {:.2}%   │", hit_rate);
    println!("│ Query latency       │ <5ms   │ {:.2}ms  │", avg_latency_ns / 1_000_000.0);
    println!("│ Cache warmup        │ ✅     │ ✅       │");
    println!("│ LRU eviction        │ ✅     │ ✅       │");
    println!("│ Total nodes         │ 10K    │ {}    │", router.count_nodes()?);
    println!("│ DB size             │ <10MB  │ {:.2}MB  │", total_size as f64 / 1024.0 / 1024.0);
    println!("└─────────────────────────────────────────┘");
    
    let pass_count = if hit_rate >= 85.0 { 1 } else { 0 }
                   + if avg_latency_ns < 5_000_000.0 { 1 } else { 0 };
    
    println!("\nValidation Result:");
    if pass_count >= 2 {
        println!("  ✅ SUCCESS: All validation targets met!");
        println!("  Recommendation: Ready for production pilot deployment");
    } else {
        println!("  ⚠️ NEEDS IMPROVEMENT: Some targets not met");
        println!("  Recommendation: Continue optimization before deployment");
    }
    
    println!("\nCleanup...");
    std::fs::remove_file(&sqlite_path)?;
    std::fs::remove_dir_all(&sled_path)?;
    println!("  ✓ Test database removed");
    
    println!("\n✅ Real Scenario Validation Test completed!");
    
    Ok(())
}

fn get_directory_size(path: &str) -> Result<u64> {
    let mut total_size = 0;
    for entry in std::fs::read_dir(path)? {
        let entry = entry?;
        let metadata = entry.metadata()?;
        if metadata.is_file() {
            total_size += metadata.len();
        } else if metadata.is_dir() {
            total_size += get_directory_size(entry.path().to_str().unwrap())?;
        }
    }
    Ok(total_size)
}