#!/usr/bin/env python3
# -*- coding: utf-8 -*-
# Konwersja .wmt do JSON i csv
# 22.03.2026 LK
# python3 reader2.py '/Users/lklich/Desktop/Adam Błachowicz' -a -f -d <-wszystkie, +podfoldery, usuwa stare wmt  

from __future__ import annotations
import argparse
import pathlib
import struct
import json
import pandas as pd
from datetime import datetime
from typing import Tuple, Dict, Any

HEADER_FMT = "<3sHHIII"   # 19 B
HEADER_SIZE = struct.calcsize(HEADER_FMT)

SAMPLE_FMT  = "<IBhhhB"   # 12 B
SAMPLE_SIZE = struct.calcsize(SAMPLE_FMT)

def read_wmt_one(path: pathlib.Path) -> Tuple[pd.DataFrame, Dict[str, Any]]:
    blob = path.read_bytes()

    if len(blob) < HEADER_SIZE:
        raise ValueError(f"{path.name}: plik zbyt krótki ({len(blob)} B).")

    magic, version, headerSize, sampleSize, start_unix, reccount = struct.unpack_from(HEADER_FMT, blob, 0)

    if magic != b"WMT":
        raise ValueError(f"{path.name}: niepoprawny magic {magic!r}.")
    
    dt_obj = datetime.fromtimestamp(start_unix)
    start_time_pl = dt_obj.strftime("%d.%m.%Y %H:%M:%S")
    start_time_iso = dt_obj.isoformat()

    data = blob[headerSize:]
    nrec = len(data) // SAMPLE_SIZE

    meta = {
        "filename": path.name,
        "full_path": str(path.absolute()),
        "wmt_version": version,
        "start_unix": start_unix,
        "start_time_iso": start_time_iso,
        "start_time_pl": start_time_pl,
        "declared_reccount": reccount,
        "actual_reccount": nrec,
        "is_incomplete": len(data) % SAMPLE_SIZE != 0
    }

    rows = []
    off = 0
    for _ in range(nrec):
        rec = data[off:off+SAMPLE_SIZE]
        offset_us, sensor_id, x, y, z, ready = struct.unpack(SAMPLE_FMT, rec)
        unix_ts = start_unix + (offset_us / 1_000_000.0)
        rows.append((start_unix, offset_us, unix_ts, sensor_id, x, y, z, int(ready)))
        off += SAMPLE_SIZE

    df = pd.DataFrame(rows, columns=["start_unix", "offset_us", "unix_ts", "sensor_id", "x", "y", "z", "ready"])
    df = df.astype({"sensor_id": "int32", "x": "int32", "y": "int32", "z": "int32", "ready": "int32"})
    
    return df, meta

def save_outputs(df: pd.DataFrame, base_path: pathlib.Path, meta: Dict[str, Any], no_header: bool, suffix: str):
    csv_path = base_path.with_suffix(suffix)
    json_path = base_path.with_suffix(".json")

    with open(csv_path, 'w', encoding='utf-8') as f:
        f.write(f"# Plik: {meta.get('filename', 'N/A')}\n")
        f.write(f"# Wersja WMT: {meta.get('wmt_version', 'N/A')}\n")
        f.write(f"# Start: {meta.get('start_time_pl', 'N/A')} (Unix: {meta.get('start_unix', 0)})\n")
        f.write(f"# Rekordy: {meta.get('actual_reccount', 0)}\n")
        df.to_csv(f, index=False, header=not no_header)

    with open(json_path, 'w', encoding='utf-8') as f:
        json.dump(meta, f, indent=4, ensure_ascii=False)

def main():
    ap = argparse.ArgumentParser(description="Konwersja .wmt -> CSV + JSON.")
    ap.add_argument("input", help="Ścieżka do pliku lub katalogu")
    ap.add_argument("-a", "--all", action="store_true", help="Przetwarzaj wszystkie pliki .wmt w katalogu")
    ap.add_argument("-f", "--recursive", action="store_true", help="Przeszukuj podfoldery (wymaga -a)")
    ap.add_argument("-o", "--overwrite", action="store_true", help="Nadpisuj istniejące pliki wyjściowe")
    ap.add_argument("-d", "--delete", action="store_true", help="USUŃ plik .wmt po udanej konwersji")
    ap.add_argument("--out-suffix", default=".csv", help="Sufiks CSV")
    ap.add_argument("--concat", action="store_true", help="Scal dane do jednego pliku")
    ap.add_argument("--no-header", action="store_true", help="CSV bez nagłówka kolumn")
    args = ap.parse_args()

    input_path = pathlib.Path(args.input)
    files_to_process = []

    if args.all:
        if not input_path.is_dir():
            print(f"[ERROR] Ścieżka {input_path} nie jest katalogiem!")
            return
        files_to_process = sorted(list(input_path.rglob("*.wmt") if args.recursive else input_path.glob("*.wmt")))
    else:
        if input_path.is_dir():
            print(f"[ERROR] Podano katalog, ale brakuje -a.")
            return
        files_to_process = [input_path]

    if not files_to_process:
        print(f"[INFO] Brak plików do przetworzenia.")
        return

    dfs = []
    combined_meta = []
    successfully_processed = []

    for p in files_to_process:
        csv_check = p.with_suffix(args.out_suffix)
        json_check = p.with_suffix(".json")

        if not args.overwrite and not args.concat:
            if csv_check.exists() or json_check.exists():
                print(f"[SKIP] Pominięto {p.name} - pliki wyjściowe już istnieją.")
                continue

        try:
            df, meta = read_wmt_one(p)
            
            if args.concat:
                df["_source_path"] = str(p.relative_to(input_path))
                dfs.append(df)
                combined_meta.append(meta)
                successfully_processed.append(p)
            else:
                save_outputs(df, p, meta, args.no_header, args.out_suffix)
                print(f"Przetworzono: {p.name}")
                if args.delete:
                    p.unlink()
                    print(f"  [DEL] Usunięto plik źródłowy: {p.name}")
                    
        except Exception as e:
            print(f"[ERROR] Błąd w pliku {p.name}: {e}")

    # Obsługa zapisu zbiorczego (concat)
    if args.concat and dfs:
        out_base = input_path / input_path.name if input_path.is_dir() else files_to_process[0]
        big_df = pd.concat(dfs, ignore_index=True)
        meta_summary = {
            "description": "Scalony zestaw danych",
            "files_count": len(combined_meta),
            "sources": combined_meta
        }
        
        try:
            save_outputs(big_df, out_base, meta_summary, args.no_header, args.out_suffix)
            print(f"\nZapisano scalone dane do {out_base.with_suffix(args.out_suffix)}")
            
            # Usuwamy pliki źródłowe dopiero po udanym scaleniu
            if args.delete:
                for p in successfully_processed:
                    p.unlink()
                print(f"  [DEL] Usunięto {len(successfully_processed)} plików źródłowych po scaleniu.")
        except Exception as e:
            print(f"[ERROR] Błąd zapisu pliku scalonego: {e}")

if __name__ == "__main__":
    main()