import csv

def check_lengths(filename, label):
    lengths = []
    failed = 0
    total = 0
    with open(filename, 'r') as f:
        reader = csv.reader(f)
        next(reader)
        for row in reader:
            if not row or row[0].startswith('#'):
                continue
            total += 1
            try:
                length = int(row[2])
                lengths.append(length)
                if length == 0:
                    failed += 1
            except:
                pass
    
    print(f"\n=== {label} ===")
    print(f"Total paths: {total}")
    print(f"Failed paths (length=0): {failed} ({100*failed/total if total else 0:.1f}%)")
    print(f"Paths with length > 0: {total - failed}")
    
    if lengths:
        avg = sum(lengths) / len(lengths)
        print(f"Avg length (all): {avg:.1f}")
        nonzero = [l for l in lengths if l > 0]
        if nonzero:
            print(f"Avg length (non-zero): {sum(nonzero)/len(nonzero):.1f}")
            print(f"Max length: {max(lengths)}")
    
    # Length buckets
    print("\nLength distribution:")
    buckets = [(0, 0), (1, 5), (6, 10), (11, 50), (51, 100), (100, 1000)]
    for lo, hi in buckets:
        if lo == hi == 0:
            count = sum(1 for l in lengths if l == 0)
        else:
            count = sum(1 for l in lengths if lo <= l <= hi)
        pct = 100*count/len(lengths) if lengths else 0
        print(f"  {lo}-{hi if hi < 1000 else 'inf'} nodes: {count} ({pct:.1f}%)")

check_lengths("/home/popertots/bench/pathfinding_benchmark_baseline_release.csv", "BASELINE")
check_lengths("pathfinding_benchmark_current.csv", "CURRENT")