bgg_api_test.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "ddc8fe73",
   "metadata": {},
   "outputs": [],
   "source": [
    "from IPython.display import clear_output\n",
    "from bs4 import BeautifulSoup as bs\n",
    "from airtable import Airtable\n",
    "import requests\n",
    "\n",
    "\n",
    "airtable_api = '' # Sequência de 16 dígitos. Vá em Account > API para pegar o seu.\n",
    "table_id = 'appEc9XLglmx5igrn' # É o que vem após a primeira \"/\" da url.\n",
    "airtable = Airtable(table_id, 'acervo', airtable_api)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "bb08d806",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Functions\n",
    "\n",
    "# Essa função aceita um texto (tipo \"Puerto Rico\") e traz uma lista de dicionários \n",
    "# com informações desses jogos, para que seja fácil descobrir o \"bgg_id\" dele.\n",
    "def search(query):\n",
    "    url = f'https://www.boardgamegeek.com/xmlapi/search?search={query}'\n",
    "    r = requests.get(url)\n",
    "    soup = bs(r.content, \"lxml\")\n",
    "    data = []\n",
    "    for i, bg in enumerate(soup.find_all(\"boardgame\")):\n",
    "        try:\n",
    "            bgg_id = int(bg['objectid'])\n",
    "            name = bg.find(\"name\").text\n",
    "            year = bg.find(\"yearpublished\").text\n",
    "            result = i\n",
    "            link = f'https://boardgamegeek.com/boardgame/{bgg_id}'\n",
    "            obj = {\"bgg_id\": bgg_id, \"name\": name, \"year\": year, \"result\": result, \"link\": link}\n",
    "            data.append(obj)\n",
    "        except AttributeError:\n",
    "            continue\n",
    "    return data\n",
    "\n",
    "# Dado um bgg_id, puxa alguns dados. É possível pegar vários outros\n",
    "# Ver mais em: https://boardgamegeek.com/wiki/page/BGG_XML_API2\n",
    "def game_info(bgg_id):\n",
    "    url = f'https://www.boardgamegeek.com/xmlapi2/thing?id={bgg_id}&stats=1'\n",
    "    r = requests.get(url)\n",
    "    soup = bs(r.content, \"lxml\")\n",
    "    nome = soup.find(\"name\", {\"type\": \"primary\"})['value']\n",
    "    rank = int(soup.find(\"ranks\").find(\"rank\", {\"name\": \"boardgame\"})['value'])\n",
    "    nota = float(soup.find(\"ratings\").find(\"average\")['value'])\n",
    "    age_poll = soup.find(\"poll\", {\"name\": \"suggested_playerage\"})\n",
    "    age_poll_results = [(int(x['numvotes']), x['value']) for x in age_poll.find_all(\"result\")]\n",
    "    idade = max(age_poll_results)[1]\n",
    "    author_list = soup.find_all(\"link\", {\"type\": \"boardgamedesigner\"})\n",
    "    autores = ', '.join([x['value'] for x in author_list])\n",
    "    thumb_url = soup.find(\"thumbnail\").text\n",
    "    obj = {\"jogo\": nome, \"rank\": rank, \"nota_bgg\": nota, \"idade\": idade,\n",
    "           \"autores\": autores, \"thumbnail\": [{\"url\": thumb_url}]}\n",
    "    return obj\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f85f9cf1",
   "metadata": {},
   "source": [
    "### Atualizando table pela primeira vez\n",
    "\n",
    "Subimos um csv criado manualmente para o Airtable, chamamos de \"acervo\" e trabalhamos daí "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "79a4a17f",
   "metadata": {},
   "outputs": [],
   "source": [
    "def first_dump()\n",
    "    acervo = airtable.get_all()\n",
    "\n",
    "    for jogo in acervo:\n",
    "        if not jogo['fields'].get(\"bgg_id\"):\n",
    "            query = jogo['fields']['search_query']\n",
    "            print(query)\n",
    "            pesquisa = search(query)\n",
    "            if not pesquisa:\n",
    "                continue\n",
    "            for r in pesquisa:\n",
    "                print(f'{r[\"result\"]} {r[\"name\"]} ({r[\"year\"]})')\n",
    "                print(r['link'])\n",
    "                print('\\n')\n",
    "            print('\\n\\n')\n",
    "            correto = input(\"which one?\")\n",
    "            bgg_id = pesquisa[int(correto)]['bgg_id']\n",
    "            airtable.update(jogo['id'], {\"bgg_id\": bgg_id})\n",
    "            clear_output(wait=True)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b201d62",
   "metadata": {},
   "source": [
    "### Usando dados do BGG na tabela em questão:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fb9de5a1",
   "metadata": {},
   "outputs": [],
   "source": [
    "acervo = airtable.get_all()\n",
    "for jogo in acervo:\n",
    "    if not 'rank' in jogo['fields']:\n",
    "        print(jogo)\n",
    "        game = game_info(jogo['fields']['bgg_id'])\n",
    "        airtable.update(jogo['id'], game)"
   ]
  }
 ],
 "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}