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

FILENAME_PATTERN = re.compile(r'_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_pos():
    positions = {}
    if not POSITIONS_CSV.exists(): return {}
    with open(POSITIONS_CSV, 'r', encoding='utf-8', errors='replace') as f:
        lines = f.readlines()
        if len(lines) < 5: return {}
        headers = lines[3].strip().split(',')
        try:
            ni = headers.index('NodeCode')
            ei = headers.index('Aslaid Easting') if 'Aslaid Easting' in headers else headers.index('Preplot Easting')
            oi = headers.index('Aslaid Northing') if 'Aslaid Northing' in headers else headers.index('Preplot Northing')
            di = headers.index('Aslaid Depth') if 'Aslaid Depth' in headers else -1
        except: return {}
        for line in lines[4:]:
            parts = line.strip().split(',')
            try:
                nid = parts[ni].strip()
                positions[nid] = {'easting': float(parts[ei]), 'northing': float(parts[oi]), 'depth': float(parts[di]) if di != -1 else 0.0}
            except: continue
    return positions

def scan():
    pos = load_pos()
    index = {}
    file_count = 0
    for root in DATA_ROOTS:
        print(f"Scanning {root}...")
        for h5_file in root.rglob("*.h5"):
            match = FILENAME_PATTERN.search(h5_file.name)
            if not match: continue
            nid, ts = match.group(1), int(match.group(2))
            # Utilisation de la date du dossier parent si possible, sinon du timestamp
            date_str = datetime.fromtimestamp(ts).strftime('%Y-%m-%d')
            # Forcer la date du dossier (plus fiable pour l'utilisateur)
            for p in h5_file.parents:
                if re.match(r'2020-09-\d{2}', p.name):
                    date_str = p.name
                    break
            
            if nid not in index:
                index[nid] = {'id': nid, 'position': pos.get(nid), 'dates': {}, 'hasDates': True}
            if date_str not in index[nid]['dates']:
                index[nid]['dates'][date_str] = []
            index[nid]['dates'][date_str].append({'path': str(h5_file), 'timestamp': ts, 'channels': ['ch0', 'ch1', 'ch2', 'ch3']})
            file_count += 1
            
    for nid, p in pos.items():
        if nid not in index: index[nid] = {'id': nid, 'position': p, 'dates': {}, 'hasDates': False}

    full = {
        '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, f, indent=2)
    print(f"Index: {file_count} files, {len(index)} nodes, {len(full['dates'])} dates.")

if __name__ == '__main__': scan()