{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ⚠️ For now, do not use this script as there is one minor problem! ⚠️\n",
    "### Instead just run `create_meta_data_csv_md.py` once and copy & paste the entire content of `meta_data.md` between `### Catalogue` and `### Example`.\n",
    "\n",
    "This is a jupyter notebook with two cells/scripts. However, both have the same flaw: `\\n` new line handling is messy as the table does contain entries like `...| test | \\nArticle | test data | ...` that are parsed as line breaks destroying the markdown table. \n",
    "\n",
    "If you read this and have a quick solution, please share it! I'm currently working on other features and don't have time to investigate further."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1) Read the meta_data file and update readme"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "\n",
    "def update_readme_with_metadata(readme_path='README.md', metadata_path='meta_data.md'):\n",
    "    # Step 1: Read the readme.md file\n",
    "    with open(readme_path, 'r') as file:\n",
    "        readme_content = file.read()\n",
    "    \n",
    "    # Step 2: Identify the section to replace\n",
    "    pattern = r\"### Catalogue(.*?)### Example\"\n",
    "    match = re.search(pattern, readme_content, re.DOTALL)\n",
    "    \n",
    "    if match:\n",
    "        # Step 3: Read the metadata.md file\n",
    "        with open(metadata_path, 'r') as file:\n",
    "            metadata_content = file.read()\n",
    "        \n",
    "        # Step 4: Replace the section\n",
    "        updated_content = re.sub(pattern, f\"### Catalogue\\n{metadata_content}\\n### Example\", readme_content, flags=re.DOTALL)\n",
    "        \n",
    "        # Step 5: Write the updated content back to readme.md\n",
    "        with open(readme_path, 'w') as file:\n",
    "            file.write(updated_content)\n",
    "    else:\n",
    "        print(\"The specified section was not found in the readme.md file.\")\n",
    "\n",
    "# Call the function\n",
    "update_readme_with_metadata()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2) Do everything in one go: parse all json.gz files, create the meta_data file and update readme. Note that this cell might not run in Jupyter due to multiprocessing issues with ipykernel. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "import gzip\n",
    "import csv\n",
    "from multiprocessing import Pool, cpu_count\n",
    "import time\n",
    "\n",
    "def process_json_file(file_path):\n",
    "    with gzip.open(file_path, 'rt', encoding='utf-8') as gz_file:\n",
    "        data = json.load(gz_file)\n",
    "        return data.get('meta', {})\n",
    "\n",
    "def get_file_size_mb(file_path):\n",
    "    return round(os.path.getsize(file_path) / (1024 * 1024), 2)\n",
    "\n",
    "def write_to_csv_and_md(output_csv, output_md, headers, data):\n",
    "    with open(output_csv, 'w', newline='', encoding='utf-8') as csv_file:\n",
    "        writer = csv.DictWriter(csv_file, fieldnames=headers)\n",
    "        writer.writeheader()\n",
    "        writer.writerows(data)\n",
    "\n",
    "    with open(output_md, 'w', encoding='utf-8') as md_file:\n",
    "        md_file.write(\"| \" + \" | \".join(headers) + \" |\\n\")\n",
    "        md_file.write(\"|\" + \"|\".join([\" --- \" for _ in headers]) + \"|\\n\")\n",
    "\n",
    "        for row in data:\n",
    "            md_file.write(\"| \" + \" | \".join([str(row[header]) for header in headers]) + \" |\\n\")\n",
    "\n",
    "def process_file(file_name, input_directory, base_url):\n",
    "    file_path = os.path.join(input_directory, file_name)\n",
    "    meta_data = process_json_file(file_path)\n",
    "    file_size_mb = get_file_size_mb(file_path)\n",
    "\n",
    "    row_data = {\n",
    "        \"filesize\": file_size_mb,\n",
    "        \"filename\": file_name,\n",
    "        \"URL\": f\"{base_url}{file_name.replace('.json.gz', '')}\",\n",
    "        **meta_data\n",
    "    }\n",
    "\n",
    "    return row_data\n",
    "\n",
    "def replace_section_in_readme(readme_path, meta_data_path):\n",
    "    # Read README.md content\n",
    "    with open(readme_path, 'r', encoding='utf-8') as readme_file:\n",
    "        readme_content = readme_file.read()\n",
    "    \n",
    "    # Read meta_data.md content\n",
    "    with open(meta_data_path, 'r', encoding='utf-8') as meta_data_file:\n",
    "        meta_data_content = meta_data_file.read()\n",
    "    \n",
    "    # Pattern to identify the section to replace\n",
    "    pattern = r\"### Catalogue(.*?)### Example\"\n",
    "    # Use re.sub to replace the section with meta_data_content\n",
    "    updated_readme_content = re.sub(pattern, f\"### Catalogue{meta_data_content}### Example\", readme_content, flags=re.DOTALL)\n",
    "    \n",
    "    # Write the updated content back to README.md\n",
    "    with open(readme_path, 'w', encoding='utf-8') as readme_file:\n",
    "        readme_file.write(updated_readme_content)\n",
    "\n",
    "def main(input_directory, output_csv, output_md, base_url=\"https://do-me.github.io/SemanticFinder/?hf=\"):\n",
    "    headers = [\n",
    "        \"filesize\", \"textTitle\", \"textAuthor\", \"textYear\", \"textLanguage\", \"URL\",\n",
    "        \"modelName\", \"quantized\", \"splitParam\", \"splitType\", \"characters\", \"chunks\",\n",
    "        \"wordsToAvoidAll\", \"wordsToCheckAll\", \"wordsToAvoidAny\", \"wordsToCheckAny\",\n",
    "        \"exportDecimals\", \"lines\", \"textNotes\", \"textSourceURL\", \"filename\"\n",
    "    ]\n",
    "\n",
    "    all_data = []\n",
    "    \n",
    "    start_time = time.time()\n",
    "\n",
    "    file_list = [file_name for file_name in os.listdir(input_directory) if file_name.endswith('.json.gz')]\n",
    "\n",
    "    with Pool(cpu_count()) as pool:\n",
    "        all_data = pool.starmap(process_file, [(file_name, input_directory, base_url) for file_name in file_list])\n",
    "\n",
    "    write_to_csv_and_md(output_csv, output_md, headers, all_data)\n",
    "    \n",
    "    replace_section_in_readme(os.path.join(input_directory, 'README.md'), os.path.join(input_directory, output_md))\n",
    "\n",
    "    end_time = time.time()\n",
    "    processing_time = end_time - start_time\n",
    "    print(f\"Processing time: {round(processing_time, 2)} seconds\")\n",
    "\n",
    "if __name__ == \"__main__\":\n",
    "    input_directory = \".\"\n",
    "    output_csv = \"meta_data.csv\"\n",
    "    output_md = \"meta_data.md\"\n",
    "\n",
    "    main(input_directory, output_csv, output_md)\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "py3.11",
   "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": "3.11.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}