import os
import re
import json
import csv
from pathlib import Path
from datetime import datetime
from tqdm import tqdm

# Pattern pour extraire les infos du nom de fichier
FILENAME_PATTERN = re.compile(r'auto_.*?_b(\d+)_.*?_(\d{10})\.h5$', re.IGNORECASE)

DATA_ROOTS = [Path("/mnt/kingston"), Path("/mnt/data_sdb1")]
POSITIONS_CSV = Path("/mnt/kingston/Copie de SETE_AUV_DARFV4-Copier(1).csv")
OUTPUT_INDEX = Path("/mnt/kingston/seismic_webapp/data/index.json")

def load_node_positions(csv_path):
    positions = {}
    if not csv_path.exists(): return positions
    with open(csv_path, 'r', encoding='utf-8', errors='replace') as f:
        lines = f.readlines()
        if len(lines) < 5: return positions
        headers = lines[3].strip().split(',')
        try:
            node_code_idx = headers.index('NodeCode')
            easting_idx = headers.index('Aslaid Easting') if 'Aslaid Easting' in headers else headers.index('Preplot Easting')
            northing_idx = headers.index('Aslaid Northing') if 'Aslaid Northing' in headers else headers.index('Preplot Northing')
        except: return positions
        for line in lines[4:]:
            parts = line.strip().split(',')
            try:
                nid = parts[node_code_idx].strip()
                positions[nid] = {
                    'easting': float(parts[easting_idx]),
                    'northing': float(parts[northing_idx]),
                    'depth': float(parts[headers.index('Aslaid Depth')]) if 'Aslaid Depth' in headers else 0.0
                }
            except: continue
    return positions

def scan_all():
    index = {}
    pos = load_node_positions(POSITIONS_CSV)
    print(f"Positions chargées: {len(pos)}")
    
    file_count = 0
    for root in DATA_ROOTS:
        print(f"Scan de {root}...")
        for h5_file in root.rglob("*.h5"):
            # Extraction ID node et timestamp
            match = re.search(r'_b(\d+)_.*?(\d{10})\.h5$', h5_file.name)
            if not match: continue
            
            node_id = match.group(1)
            ts = int(match.group(2))
            date_str = datetime.fromtimestamp(ts).strftime('%Y-%m-%d')
            
            if node_id not in index:
                index[node_id] = {'id': node_id, 'position': pos.get(node_id), 'dates': {}, 'hasDates': True}
            
            if date_str not in index[node_id]['dates']:
                index[node_id]['dates'][date_str] = []
            
            index[node_id]['dates'][date_str].append({
                'path': str(h5_file),
                'timestamp': ts,
                'channels': ['ch0', 'ch1', 'ch2', 'ch3'],
                'size_bytes': h5_file.stat().st_size
            })
            file_count += 1
            
    # Ajouter les nodes sans fichiers mais avec position
    for nid, p in pos.items():
        if nid not in index:
            index[nid] = {'id': nid, 'position': p, 'dates': {}, 'hasDates': False}

    full_index = {
        'generated_at': datetime.now().isoformat(),
        'sample_rate_hz': 200,
        'nodes': index,
        'dates': sorted(list(set(d for n in index.values() for d in n['dates'].keys())))
    }
    
    with open(OUTPUT_INDEX, 'w') as f: json.dump(full_index, f, indent=2)
    print(f"Index total généré: {file_count} fichiers, {len(index)} nodes.")

if __name__ == '__main__': scan_all()