#!/usr/bin/env python3
"""OCR semua gambar di folder pakai RapidOCR."""
import os, sys, json
from rapidocr_onnxruntime import RapidOCR

src = sys.argv[1]
out = sys.argv[2] if len(sys.argv) > 2 else "/home/mantapman/.hermes/profiles/atun/workspace/tsc_stock_ocr.json"
engine = RapidOCR()
res_all = {}
files = sorted([f for f in os.listdir(src) if f.lower().endswith((".png", ".jpg", ".jpeg"))],
               key=lambda x: int(''.join(c for c in x if c.isdigit()) or 0))
for f in files:
    p = os.path.join(src, f)
    r, _ = engine(p)
    lines = []
    if r:
        # urutkan atas→bawah
        for box, txt, score in sorted(r, key=lambda x: min(pt[1] for pt in x[0])):
            lines.append(txt.strip())
    res_all[f] = lines
    print(f"\n=== {f} ({os.path.getsize(p)//1024} KB) ===")
    for l in lines:
        print("  ", l)
json.dump(res_all, open(out, "w"), indent=1, ensure_ascii=False)
print("\nsaved:", out)