{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "7a478c86",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-07-31T01:55:12.877222Z",
     "start_time": "2023-07-31T01:55:12.874203Z"
    }
   },
   "outputs": [],
   "source": [
    "import requests\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "61571547",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-07-31T01:57:23.295679Z",
     "start_time": "2023-07-31T01:57:23.292514Z"
    }
   },
   "outputs": [],
   "source": [
    "t = 2314"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "5d7eda84",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-07-31T01:56:12.033568Z",
     "start_time": "2023-07-31T01:56:12.025127Z"
    }
   },
   "outputs": [],
   "source": [
    "def download_java_files(url, path):\n",
    "    global t\n",
    "    # Send GET request to retrieve the folder contents\n",
    "    response = requests.get(url, params={\"ref\": branch})\n",
    "    if response.status_code == 200:\n",
    "        # Parse the response JSON\n",
    "        contents = response.json()\n",
    "\n",
    "        for item in contents:\n",
    "            if item[\"type\"] == \"file\" and item[\"name\"].endswith(\".java\"):\n",
    "                # Download Java file\n",
    "                download_url = item[\"download_url\"]\n",
    "                file_name = \"train_\"+str(t)\n",
    "                output_file_path = f\"Desktop/MITACS/Dataset/inp-txt/{file_name}\"\n",
    "\n",
    "                # Send GET request to download the file content\n",
    "                file_content = requests.get(download_url).text\n",
    "\n",
    "                # Save the file to the output directory\n",
    "                with open(output_file_path, \"w\") as output_file:\n",
    "                    output_file.write(file_content)\n",
    "\n",
    "                print(f\"Downloaded: {file_name}\")\n",
    "                t+=1\n",
    "            elif item[\"type\"] == \"dir\":\n",
    "                # Recursively navigate into subfolders\n",
    "                subfolder_url = item[\"url\"]\n",
    "                subfolder_path = f\"{path}/{item['name']}\"\n",
    "                download_java_files(subfolder_url, subfolder_path)\n",
    "    else:\n",
    "        print(\"Failed to retrieve folder contents.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "dafd2b43",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-07-31T01:59:48.822445Z",
     "start_time": "2023-07-31T01:59:48.667420Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Failed to retrieve folder contents.\n",
      "2314\n"
     ]
    }
   ],
   "source": [
    "# GitHub repository details\n",
    "repo = \"leakcanary\"\n",
    "owner = \"square\"\n",
    "branch = \"main\"  # Specify the branch you want to download from\n",
    "output_directory = \"DesktopDataset/inp-txt\"  # Specify the directory to save the downloaded files\n",
    "# API endpoint to get the list of files in the repository\n",
    "url = f\"https://api.github.com/repos/{owner}/{repo}/contents\"\n",
    "download_java_files(url, output_directory)\n",
    "print(t)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f51e193d",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.10 (tensorflown)",
   "language": "python",
   "name": "tensorflown"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}