{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "0b6e3da3-d921-4445-927c-b8a66b3a2428",
   "metadata": {},
   "outputs": [],
   "source": [
    "from github import Github\n",
    "import requests"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "1b5b14e4-3b9d-41c0-9df6-4f63d76e05da",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_diff(url):\n",
    "    return requests.get(url).content.decode()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "id": "ebe058f9-f62d-4978-890a-3e8200999779",
   "metadata": {},
   "outputs": [],
   "source": [
    "tok = \"MYGITHUBTOKEN\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "id": "19d37444-e62e-4e31-bce8-6cd1ee63e133",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tqdm.notebook import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "id": "fd25eafa-fb29-4051-ae0a-f71a0dbc92db",
   "metadata": {},
   "outputs": [],
   "source": [
    "gh = Github(\"MYGITNAME\", tok)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "id": "f3b9c5f6-31a0-49b2-834b-982952c6b7e5",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "id": "c60588b7-56de-457f-93a6-3e8b3b8965c8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 out of 17\n",
      "Data length: 29\n",
      "1 out of 17\n",
      "Data length: 53\n",
      "2 out of 17\n",
      "Data length: 79\n",
      "3 out of 17\n",
      "Data length: 107\n",
      "4 out of 17\n",
      "Data length: 134\n",
      "5 out of 17\n",
      "Data length: 159\n",
      "6 out of 17\n",
      "Data length: 187\n",
      "7 out of 17\n",
      "Data length: 211\n",
      "8 out of 17\n",
      "Data length: 239\n",
      "9 out of 17\n",
      "Data length: 266\n",
      "10 out of 17\n",
      "Data length: 295\n",
      "11 out of 17\n",
      "Data length: 323\n",
      "12 out of 17\n",
      "Data length: 353\n",
      "13 out of 17\n",
      "Data length: 380\n",
      "14 out of 17\n",
      "Data length: 409\n",
      "15 out of 17\n",
      "Data length: 439\n",
      "16 out of 17\n",
      "Data length: 454\n"
     ]
    }
   ],
   "source": [
    "repo = gh.get_repo(\"huggingface/accelerate\")\n",
    "pulls = repo.get_pulls(state=\"closed\")\n",
    "total_pages = (pulls.totalCount // 30) + 1\n",
    "total = pulls.totalCount\n",
    "for page_id in range(total_pages):\n",
    "    page = pulls.get_page(page_id)\n",
    "    for pull in page:\n",
    "        pull_id = pull.id\n",
    "        pull_diff = get_diff(pull.diff_url)\n",
    "        pull_user = pull.user.login\n",
    "        if pull.is_merged():\n",
    "            code_comments = []\n",
    "            conversation = []\n",
    "            # Code comments\n",
    "            for comment in pull.get_comments():\n",
    "                code_comments.append(\n",
    "                    {\n",
    "                        \"body\":comment.body,\n",
    "                        \"diff_hunk\":comment.diff_hunk,\n",
    "                        \"from_author\": comment.user.login == pull_user\n",
    "                    }\n",
    "                )\n",
    "            for comment in pull.get_issue_comments():\n",
    "                conversation.append(\n",
    "                    dict(\n",
    "                        body=comment.body,\n",
    "                        from_author=comment.user.login == pull_user\n",
    "                    )\n",
    "                )\n",
    "            data[pull_id] = {\n",
    "                \"diff\":pull_diff, \n",
    "                \"code_comments\":code_comments, \n",
    "                \"context\":conversation,\n",
    "                \"url\":pull.url\n",
    "            }     \n",
    "    print(f\"{page_id} out of {total_pages}\")\n",
    "    print(f'Data length: {len(data)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "id": "16140ee6-88e7-4cdf-b215-2282f1034dcf",
   "metadata": {},
   "outputs": [],
   "source": [
    "new_code = []\n",
    "for item in list(data.keys()):\n",
    "    d = data[item]\n",
    "    d[\"pr_id\"] = item\n",
    "    if len(d[\"code_comments\"]) > 1:\n",
    "        new_code.append(d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "id": "1149e33e-75bf-4b0b-a94c-fe5ee2a291bc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "134"
      ]
     },
     "execution_count": 174,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(new_code)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "id": "7763a1aa-727c-41c9-b348-a4adaaca8445",
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "with open(\"data.json\", \"w+\") as f:\n",
    "    json.dump(new_code, f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "71e93f8e-35a3-4a8c-a706-4dc88d38f3fc",
   "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}