google_search.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import getpass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "········\n"
     ]
    }
   ],
   "source": [
    "api_key = getpass.getpass()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "········\n"
     ]
    }
   ],
   "source": [
    "cse_id = getpass.getpass()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from googleapiclient.discovery import build\n",
    "import pprint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def google_search(search_term, api_key, cse_id, **kwargs):\n",
    "    service = build(\"customsearch\", \"v1\", developerKey=api_key)\n",
    "    res = service.cse().list(q=search_term, cx=cse_id, **kwargs).execute()\n",
    "    return res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "results = google_search('stackoverflow site:en.wikipedia.org', api_key, cse_id, num=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "'https://en.wikipedia.org/wiki/Stack_overflow'\n",
      "'https://en.wikipedia.org/wiki/Stack_overflow_(disambiguation)'\n",
      "'https://en.wikipedia.org/wiki/Joel_Spolsky'\n",
      "'https://en.wikipedia.org/wiki/Stack_buffer_overflow'\n",
      "'https://en.wikipedia.org/wiki/Segmentation_fault'\n",
      "'https://en.wikipedia.org/wiki/Fog_Creek_Software'\n",
      "'https://en.wikipedia.org/wiki/Stack_Exchange'\n",
      "'https://en.wikipedia.org/wiki/Jeff_Atwood'\n",
      "'https://en.wikipedia.org/wiki/Buffer_overflow'\n",
      "'https://en.wikipedia.org/wiki/Buffer_overflow_protection'\n"
     ]
    }
   ],
   "source": [
    "for result in results['items']:\n",
    "    pprint.pprint(result['link'])\n",
    "    #print(i['title'] + \": \" + i['url'])\n",
    "#pprint.pprint(results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "google_search_service = build(\"customsearch\", \"v1\", developerKey=api_key)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "results = google_search_service.cse().list(q='\"World Wide Fund for Nature\"', cx=cse_id).execute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'formattedSearchTime': '0.63',\n",
       " 'formattedTotalResults': '142,000',\n",
       " 'searchTime': 0.630174,\n",
       " 'totalResults': '142000'}"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results[\"searchInformation\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [],
   "source": [
    "results = google_search_service.cse().list(q='\"World Wide Fund for Nature\"', cx=cse_id,\n",
    "                                           cr='countryPL').execute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'formattedSearchTime': '0.53',\n",
       " 'formattedTotalResults': '519',\n",
       " 'searchTime': 0.53247,\n",
       " 'totalResults': '519'}"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results[\"searchInformation\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [],
   "source": [
    "result_counts = {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {},
   "outputs": [],
   "source": [
    "country_code = 'BG'\n",
    "result_counts[country_code] = {}\n",
    "for year in range(2000, 2015):\n",
    "    results = google_search_service.cse().list(q='\"World Wide Fund for Nature\"', cx=cse_id,\n",
    "                                               cr='country{}'.format(country_code), sort='date:r:19900101:{}0101'.format(year)).execute()\n",
    "    result_counts[country_code][year] = int(results[\"searchInformation\"]['totalResults'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'BG': {2000: 0,\n",
       "  2001: 0,\n",
       "  2002: 2,\n",
       "  2003: 2,\n",
       "  2004: 4,\n",
       "  2005: 5,\n",
       "  2006: 5,\n",
       "  2007: 9,\n",
       "  2008: 338,\n",
       "  2009: 339,\n",
       "  2010: 305,\n",
       "  2011: 257,\n",
       "  2012: 307,\n",
       "  2013: 265,\n",
       "  2014: 315},\n",
       " 'PL': {2000: 0,\n",
       "  2001: 0,\n",
       "  2002: 4,\n",
       "  2003: 5,\n",
       "  2004: 7,\n",
       "  2005: 8,\n",
       "  2006: 10,\n",
       "  2007: 441,\n",
       "  2008: 440,\n",
       "  2009: 438,\n",
       "  2010: 499,\n",
       "  2011: 499,\n",
       "  2012: 495,\n",
       "  2013: 492,\n",
       "  2014: 409}}"
      ]
     },
     "execution_count": 158,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_counts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [],
   "source": [
    "results = google_search_service.cse().list(q='\"World Wide Fund for Nature\"', cx=cse_id,\n",
    "                                           cr='countryPL', sort='date:r:19900101:20000101').execute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'0'"
      ]
     },
     "execution_count": 152,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "int(results[\"searchInformation\"]['totalResults'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [],
   "source": [
    "results = google_search_service.cse().list(q='\"World Wide Fund for Nature\" daterange:2447893-2451911', cx=cse_id,\n",
    "                                           cr='countryPL').execute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'formattedSearchTime': '0.20',\n",
       " 'formattedTotalResults': '517',\n",
       " 'searchTime': 0.201784,\n",
       " 'totalResults': '517'}"
      ]
     },
     "execution_count": 149,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results[\"searchInformation\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "results = google_search('\"World Wide Fund for Nature\"', api_key, cse_id, num=10, cr='pl', hl='en')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "results = google_search('\"World Wide Fund for Nature\"', api_key, cse_id, num=10, dateRestrict='???', cr='???')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import urllib.parse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'https://www.google.co.jp/search?q=\"World+Wide+Fund+for+Nature\"&lr=&cr=countryPL&hl=en&as_qdr=all&source=lnt&tbs=ctr:countryPL,cdr:1,cd_min:,cd_max:1/1/2014&tbm='"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "urllib.parse.unquote('https://www.google.co.jp/search?q=%22World+Wide+Fund+for+Nature%22&lr=&cr=countryPL&hl=en&as_qdr=all&source=lnt&tbs=ctr%3AcountryPL%2Ccdr%3A1%2Ccd_min%3A%2Ccd_max%3A1%2F1%2F2014&tbm=')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "'https://www.google.com/search?q=\"World+Wide+Fund+for+Nature\"&lr=&cr=countryPL&hl=en&as_qdr=all&tbs=&tbm='"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cd_min:,cd_max:1/1/2014"
   ]
  }
 ],
 "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": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}