{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "574d024a-3ffc-40d7-98f4-e74744e65435",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "unziped\n",
      "Total number of images: 30340\n",
      "copy from eshooshoo2019\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Processing images:   1%|          | 320/30340 [01:38<2:31:15,  3.31it/s]"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import zipfile\n",
    "import zlib\n",
    "import shutil\n",
    "from pathlib import Path\n",
    "from PIL import Image\n",
    "from tqdm import tqdm \n",
    "\n",
    "maxsize = 1280\n",
    "\n",
    "def unzip_files_in_directory(directory):\n",
    "    for root, _, files in os.walk(directory):\n",
    "        for file in files:\n",
    "            if file.endswith('.zip'):\n",
    "                zip_path = os.path.join(root, file)\n",
    "                # Create a new subdirectory for the extracted files\n",
    "                extract_dir = os.path.join(root, os.path.splitext(file)[0])\n",
    "                os.makedirs(extract_dir, exist_ok=True)\n",
    "                try:\n",
    "                    with zipfile.ZipFile(zip_path, 'r') as zip_ref:\n",
    "                        zip_ref.extractall(extract_dir)\n",
    "                except (zipfile.BadZipFile, FileNotFoundError, OSError) as e:\n",
    "                    print(f\"Skipping corrupted or unreadable file: {zip_path}. Error: {e}\")\n",
    "\n",
    "                # Optionally, remove the zip file after extraction\n",
    "                os.remove(zip_path)\n",
    "\n",
    "def list_image_files(directory, image_extensions=('.png', '.jpg', '.jpeg', '.gif', '.bmp', '.tiff')):\n",
    "    image_files = []\n",
    "    for root, _, files in os.walk(directory):\n",
    "        for file in files:\n",
    "            if file.lower().endswith(image_extensions):\n",
    "                image_files.append(os.path.join(root, file))\n",
    "    return image_files\n",
    "\n",
    "def compute_crc32(file_path):\n",
    "    crc_value = 0\n",
    "    with open(file_path, 'rb') as file:\n",
    "        for chunk in iter(lambda: file.read(4096), b''):\n",
    "            crc_value = zlib.crc32(chunk, crc_value)\n",
    "    return format(crc_value & 0xFFFFFFFF, '08x')\n",
    "\n",
    "\n",
    "def copy_images_with_crc32(image_files, output_directory, base_folder_name):\n",
    "    output_path = Path(output_directory) / base_folder_name\n",
    "    output_path.mkdir(parents=True, exist_ok=True)\n",
    "\n",
    "    for image_file in tqdm(image_files, desc=\"Processing images\"):\n",
    "        try:\n",
    "            path = Path(image_file)\n",
    "            ext = path.suffix.lower()\n",
    "            crc = compute_crc32(image_file)\n",
    "\n",
    "            with Image.open(image_file) as img:\n",
    "                needs_processing = (\n",
    "                    (ext !=\".jpg\") or\n",
    "                    (img.mode != \"RGB\") or\n",
    "                    (max(img.size) > maxsize)\n",
    "                )\n",
    "\n",
    "                if needs_processing:\n",
    "                    # Convert to RGB (handle transparency)\n",
    "                    if img.mode in (\"RGBA\", \"LA\", \"P\"):\n",
    "                        bg = Image.new(\"RGB\", img.size, (255, 255, 255))\n",
    "                        if img.mode == \"P\":\n",
    "                            img = img.convert(\"RGBA\")\n",
    "                        if img.mode == \"RGBA\":\n",
    "                            bg.paste(img, mask=img.split()[-1])\n",
    "                        img = bg\n",
    "                    elif img.mode != \"RGB\":\n",
    "                        img = img.convert(\"RGB\")\n",
    "\n",
    "                    # Resize if too big\n",
    "                    if max(img.size) > maxsize:\n",
    "                        img.thumbnail((maxsize, maxsize), Image.BICUBIC)\n",
    "\n",
    "                    # Save as PNG\n",
    "                    out_name = f\"{base_folder_name}_{crc}.png\"\n",
    "                    img.save(output_path / out_name, \"PNG\")#, optimize=True)\n",
    "                else:\n",
    "                    # Copy as-is\n",
    "                    out_name = f\"{base_folder_name}_{crc}{ext}\"\n",
    "                    shutil.move(image_file, output_path / out_name)\n",
    "\n",
    "            # Copy .txt if exists\n",
    "            txt_path = path.with_suffix(\".txt\")\n",
    "            if txt_path.exists():\n",
    "                shutil.move(txt_path, output_path / Path(out_name).with_suffix(\".txt\"))\n",
    "\n",
    "        except Exception as e:\n",
    "            print(f\"Skipping {image_file}: {e}\")\n",
    "\n",
    "def main(directory, output_directory):\n",
    "    base_folder_name = os.path.basename(directory.rstrip('/').rstrip('\\\\'))\n",
    "\n",
    "    # Unzip all zip files in the directory\n",
    "    unzip_files_in_directory(directory)\n",
    "    print('unziped')\n",
    "\n",
    "    # List all image files\n",
    "    image_files = list_image_files(directory)\n",
    "    print(f\"Total number of images: {len(image_files)}\")\n",
    "\n",
    "    # Copy images with CRC32 hash in their names\n",
    "    print('copy from',base_folder_name)\n",
    "    copy_images_with_crc32(image_files, output_directory, base_folder_name)\n",
    "    print('ok')\n",
    "\n",
    "# Example usage\n",
    "directory_path = '/workspace/ds/eshooshoo2019'\n",
    "output_directory_path = '/workspace/ds/2015-19'\n",
    "main(directory_path, output_directory_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e047aa2a-f7cc-4332-91cf-17c373f61ea2",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}