{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Yahoo知恵袋のスクレイピングのPythonサンプルプログラム\n",
    "\n",
    "seleniumをつかったyahoo知恵袋のスクレイピングのサンプルプログラムです。<br>\n",
    "質問の検索結果まで出力できます。<br>\n",
    "出力結果は、csvファイルに書き出します。<br>\n",
    "\n",
    "\n",
    "参考にしたコード<br>\n",
    "【Python×Selenium】超簡単にWebサイトをスクレイピングしてみる<br>\n",
    "https://miyanetdev.com/archives/327"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from selenium import webdriver\n",
    "from time import sleep\n",
    "import urllib\n",
    "import re\n",
    "import pandas as pd\n",
    "\n",
    "PAGE_LIMIT = 20 #ページ遷移の最大の回数\n",
    "SEARCH_QUERY = \"プログラミング\"\n",
    "SQRAPING_URL = \"https://chiebukuro.yahoo.co.jp/\"\n",
    "\n",
    "#出力結果を格納数csvファイル\n",
    "csv_file_name = SEARCH_QUERY + \".csv\"\n",
    "\n",
    "#ドライバーを設定する\n",
    "#linuxなどGUIがない環境で動かす場合は、ヘッドレスモードを入れておく\n",
    "#options = webdriver.ChromeOptions()\n",
    "#options.add_argument('--headless')\n",
    "\n",
    "\n",
    "#driver = webdriver.Chrome('./chromedriver', options)\n",
    "driver = webdriver.Chrome('./chromedriver')\n",
    "\n",
    "#知恵袋ページを読み込む\n",
    "driver.get(SQRAPING_URL)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#該当ページを解析する\n",
    "def analysis_action():\n",
    "\n",
    "    elems = driver.find_elements_by_xpath('//*[@id=\"sr\"]/ul/li[*]')\n",
    "    # 取得した要素を1つずつ表示\n",
    "\n",
    "    out_puts = []\n",
    "\n",
    "    if(len(elems) == 0):\n",
    "        print(\"ページは存在しないよ〜\")\n",
    "    else:\n",
    "        for elem in elems:\n",
    "            out_dic ={}\n",
    "            out_dic['query_key'] = SEARCH_QUERY\n",
    "            out_dic['rs_title'] = elem.find_elements_by_xpath('h3/a')[0].text\n",
    "            out_dic['rs_link']  = elem.find_elements_by_xpath('h3/a')[0].get_attribute('href')\n",
    "            out_dic['rs_summary'] = elem.find_elements_by_xpath('p[1]')[0].text\n",
    "            #print(out_dic)\n",
    "            out_puts.append(out_dic)\n",
    "            #print(\"*\" * 60)\n",
    "            \n",
    "    return out_puts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def next_page_action():\n",
    "    \"\"\"\n",
    "    現在のページから次のページを読み込むアクションを実行する\n",
    "    \"\"\"\n",
    "    rtn = False\n",
    "    \n",
    "    #次へボタンのクリック\n",
    "    elems = driver.find_elements_by_xpath('//*[@id=\"pg_low\"]/div/a[*]')\n",
    "\n",
    "    #現在のページ\n",
    "    print(\"ページ遷移前のurl:\")\n",
    "    print(driver.current_url)\n",
    "    if(len(elems) == 0):\n",
    "        print(\"次のページは存在しないよ〜\")\n",
    "    else:\n",
    "        for elem in elems:\n",
    "            #print(elem.text)\n",
    "            if(elem.text != \"次へ\"):\n",
    "                continue\n",
    "            url = elem.get_attribute('href')\n",
    "            driver.get(url)\n",
    "            rtn = True\n",
    "            break\n",
    "\n",
    "    return rtn\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 最初の検索を実行する\n",
    "search_box = driver.find_element_by_css_selector('input.txtKeyword')\n",
    "search_box.send_keys(SEARCH_QUERY)\n",
    "search_button_container = driver.find_element_by_css_selector('p.btnSearch')\n",
    "search_button = search_button_container.find_element_by_css_selector('input')\n",
    "search_button.click()\n",
    "sleep(2)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#知恵袋の検索結果の一覧をpandasに格納してcsvに書き出す\n",
    "#csvには、途中で止まっても良いように、1ページ終わったら書き出すようにしている\n",
    "\n",
    "d = analysis_action()\n",
    "df=pd.DataFrame(d)  \n",
    "df.to_csv(csv_file_name, encoding=\"utf_8_sig\")\n",
    "\n",
    "analysis_list = []\n",
    "analysis_list.extend(d)\n",
    "\n",
    "for page in range(PAGE_LIMIT):\n",
    "    \n",
    "    print(\"ページ %dを実行中\" % page)\n",
    "    sleep(5)\n",
    "    \n",
    "    #次のページに遷移する\n",
    "    rtn = next_page_action()\n",
    "    if(rtn == False):\n",
    "        break\n",
    "        \n",
    "    #知恵袋の質問リストを格納する\n",
    "    d = analysis_action()\n",
    "    if(len(d) > 0):\n",
    "        analysis_list.extend(d)\n",
    "        df=pd.DataFrame(analysis_list)  \n",
    "        df.to_csv(csv_file_name, encoding=\"utf_8_sig\")\n",
    "        \n",
    "driver.close()\n",
    "driver.quit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}