2021q2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib\n",
    "matplotlib.style.use('ggplot')\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from sqlalchemy import create_engine, text, MetaData\n",
    "from sqlalchemy.exc import IntegrityError\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "engine = create_engine('postgresql:///piwheels')\n",
    "conn = engine.connect()\n",
    "meta = MetaData()\n",
    "meta.bind = conn\n",
    "meta.reflect(views=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "start_date = '2021-04-01'\n",
    "start_time = '2021-04-01 00:00:00'\n",
    "end_date = '2021-06-30'\n",
    "end_time = '2021-07-01 00:00:00'\n",
    "one_year_ago = '2020-07-01'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>downloads_in_period</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8471941</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   downloads_in_period\n",
       "0              8471941"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_sql(f\"\"\"\n",
    "    SELECT COUNT(*) AS downloads_in_period\n",
    "    FROM downloads\n",
    "    WHERE accessed_at::date BETWEEN '{start_date}' AND '{end_date}'\n",
    "\"\"\", conn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>downloads_to_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>48609840</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   downloads_to_date\n",
       "0           48609840"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_sql(f\"\"\"\n",
    "    SELECT COUNT(*) AS downloads_to_date\n",
    "    FROM downloads\n",
    "    WHERE accessed_at::date < '{end_date}'\n",
    "\"\"\", conn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>project_page_hits</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>764031</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   project_page_hits\n",
       "0             764031"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_sql(f\"\"\"\n",
    "    SELECT COUNT(*) AS project_page_hits\n",
    "    FROM project_page_hits\n",
    "    WHERE accessed_at::date BETWEEN '{start_date}' AND '{end_date}'\n",
    "\"\"\", conn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>project_json_downloads</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>116486</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   project_json_downloads\n",
       "0                  116486"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_sql(f\"\"\"\n",
    "    SELECT COUNT(*) AS project_json_downloads\n",
    "    FROM project_json_downloads\n",
    "    WHERE accessed_at::date BETWEEN '{start_date}' AND '{end_date}'\n",
    "\"\"\", conn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>web_page_hits</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>386472</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   web_page_hits\n",
       "0         386472"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_sql(f\"\"\"\n",
    "    SELECT COUNT(*) AS web_page_hits\n",
    "    FROM web_page_hits\n",
    "    WHERE accessed_at::date BETWEEN '{start_date}' AND '{end_date}'\n",
    "\"\"\", conn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>m</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2020-07-01</td>\n",
       "      <td>1102139</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2020-08-01</td>\n",
       "      <td>1737186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2020-09-01</td>\n",
       "      <td>1653471</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2020-10-01</td>\n",
       "      <td>1850781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2020-11-01</td>\n",
       "      <td>1240661</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2020-12-01</td>\n",
       "      <td>1552508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>1960137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2021-02-01</td>\n",
       "      <td>3424032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2021-03-01</td>\n",
       "      <td>3878585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2021-04-01</td>\n",
       "      <td>2956219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2021-05-01</td>\n",
       "      <td>2752983</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2021-06-01</td>\n",
       "      <td>2762739</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            m        c\n",
       "0  2020-07-01  1102139\n",
       "1  2020-08-01  1737186\n",
       "2  2020-09-01  1653471\n",
       "3  2020-10-01  1850781\n",
       "4  2020-11-01  1240661\n",
       "5  2020-12-01  1552508\n",
       "6  2021-01-01  1960137\n",
       "7  2021-02-01  3424032\n",
       "8  2021-03-01  3878585\n",
       "9  2021-04-01  2956219\n",
       "10 2021-05-01  2752983\n",
       "11 2021-06-01  2762739"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "downloads_by_month = pd.read_sql(f\"\"\"\n",
    "    SELECT date_trunc('month', accessed_at) m, COUNT(*) AS c\n",
    "    FROM downloads\n",
    "    WHERE accessed_at::date BETWEEN '{one_year_ago}' AND '{end_date}'\n",
    "    GROUP BY m\n",
    "    ORDER BY m\n",
    "\"\"\", conn)\n",
    "\n",
    "downloads_by_month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlMAAAF2CAYAAABKyB3wAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Z1A+gAAAACXBIWXMAAAsTAAALEwEAmpwYAAAvAUlEQVR4nO3deXRUZYL+8aeyExJokhAgYVGCShOaLZFVMJA0ZMBWx9GgEQYUlxY3mkURbYN6kACiDpvKIraACq2Otq04EhGRIEgMiIBsA7hAMCYBg4kh2/v7gx81lhAq8KYqFfv7Ocdz6t771q2nyqrKw93KYYwxAgAAwAXxq+8AAAAADRllCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCkCdcDgcWr58uccfZ926dXI4HPruu+8u6P6HDh2Sw+HQhg0b6jiZbxo9erRSUlLqOwbwm0aZAhqA0aNHy+FwyOFwKDAwUFFRUbriiis0c+ZMlZSU1Hc8+IANGzbI4XDo0KFD9R0F+JdDmQIaiP79+ysvL09ff/21PvroI918882aN2+eevTooe+//76+4wHAvyzKFNBABAUFqWXLloqJidEf/vAH3XXXXfr000/1ww8/aPLkyc5xFRUVmjx5smJjYxUUFKROnTrplVdecS7/61//qn79+jmnP/roIzkcDj3yyCPOeQ8//LD69Okj6f92q61Zs0YDBgxQaGioOnXqpNWrV58zb15enm688Ub97ne/U6NGjZSUlKScnBzncmOMbr/9dsXFxalRo0Zq3769pkyZopMnT7qsZ+7cuWrdurVCQ0M1ZMgQffPNNy7Li4uLdcstt6hly5YKDg5WmzZtNH78eLev56FDh5ScnOx87Ndee825LCkpSXfccYfLeGOM4uLi9MQTT9S4TofDoblz52r48OFq3Lix2rZtq9dff10//vijbr75ZoWHh6t9+/Z64403XO63Z88eDRs2TGFhYQoLC9Of/vQn7d+/37n8pZdeUkBAgLKzs9WjRw+FhoYqISFBW7ZscT6X/v37S5IuvvhiORwOJSUluTzGwoUL1a5dOzVp0kRXX301BRyoSwaAzxs1apRJTk4+67J77rnHNGnSxFRVVRljjJk4caKJiIgwq1atMnv27DHTpk0zDofDZGVlGWOM+fDDD01gYKA5ceKEMcaYRx55xDRv3tz06dPHuc7evXubKVOmGGOM+eijj4wk06VLF7N69Wqzd+9eM3r0aBMeHm6Kioqc95Fkli1bZowxprq62vTs2dN07drVfPLJJ2b79u0mLS3N/O53vzM//PCDMcaYqqoqM2XKFLNp0yZz8OBB8/bbb5uWLVuaRx991LnOt956y/j7+5vZs2ebPXv2mMWLF5vo6GgjyXz77bfGGGPuvfde06VLF7Np0ybz9ddfm+zsbLNw4cIaX8uDBw8aSaZVq1Zm+fLlZvfu3ebhhx82fn5+Jjc31xhjzCuvvGLCwsKcr5ExxmRlZRl/f39z+PDhGtctybRo0cK89NJLZt++feauu+4yISEhJjU11SxdutTs27fP3HPPPSY0NNQUFBQYY4wpLS01bdu2NYMGDTI5OTkmJyfHJCUlmbi4OHPy5EljjDFLly41DofD9O/f36xfv9589dVXJjU11Vx00UWmoqLCVFZWmrfffttIMp999pnJy8szhYWFzvdOkyZNzI033mi+/PJLs3HjRnPRRReZESNG1Pg8AJwfyhTQAJyrTD333HNGkvn+++9NSUmJCQoKMvPnz3cZc+2115qBAwcaY4z5+eefTXBwsHn33XeNMcb07dvXzJo1y1mwiouLTUBAgLN8nS5Tb7zxhnN9R48eNZLM+++/75z3yzKVlZVlJJmdO3c6l5eVlZmWLVuaxx57rMbn+fTTT5sOHTo4p/v162fS09NdxkyYMMGlTF199dVm1KhRNa7z106XqUceecRlfp8+fZwFo6yszERFRZlFixY5l994443m6quvPue6JZn777/fOZ2fn28kmXvuucc5r6ioyEgy77zzjjHGmMWLF5tGjRo5S6Yxp17fkJAQ87e//c0Yc6pMSTKff/65c8ymTZuMJLN7925jjDGffPKJkWQOHjzokmnUqFGmefPmpqyszDkvMzPTtGzZ8pzPBUDt1etuvgULFui2227ThAkTajV+48aN+stf/qLx48frv/7rvzycDmgYzP//rXKHw6H9+/ervLxcAwYMcBlz5ZVXaufOnZKkkJAQ9enTR2vXrtVPP/2kLVu2KD09XZdcconWr1+v9evXy9/f32VXoCR169bNebtFixby9/evcVfRzp07FRkZqU6dOjnnBQcHq1evXs4ckrRo0SL16tVLLVq0UFhYmB566CF9/fXXzuW7du1S3759XdZ9xRVXuEyPHTtWr7/+ujp37qz7779fq1evVnV1tbuXzbkb87R+/fo5swUHB2v06NFatGiRJKmwsFD//d//rdtvv93tert27eq83bx5c/n7+6tLly7Oec2aNVNQUJDy8/MlnXqtOnXqpKioKOeYFi1a6LLLLnN5rRwOh8u6Y2JiJKlWu+s6duyo4OBgl/uymw+oO/VappKSkjRlypRajc3Ly9Nbb72lJ554Qk8//bRGjx7t2XBAA7Fz5041bdpUkZGRtb7PoEGD9OGHH+qTTz5R+/btFRMTo0GDBmnt2rVau3at+vTpo5CQEJf7BAUFnbGe2pSWmvz973/X3XffreHDh+u9997T1q1b9eijj6qiouK81nP6OKqHH35YZWVlGjFihAYNGqSqqqoLziZJd955p7Zs2aLt27dr2bJlat68uf7t3/7N7f0CAwPdznM4HOf92vn5+cnf399lHVLt/h/8+v+dw+FwlnAA9uq1THXq1ElhYWEu844ePapp06bpwQcf1KOPPqrDhw9Lkj788EMNGTLEOb5p06Zezwv4msOHD2vFihW67rrr5Ofnpw4dOig4OFjr1693Gffxxx+rc+fOzumBAwfqiy++0N///nclJydLkkuZGjRokFWu+Ph4FRYWateuXc55J0+e1ObNm5051q9fr+7du2v8+PFKSEjQJZdccsZp/Z06ddLGjRtd5mVnZ5/xeBEREbrpppv0wgsv6N1339XHH3/s8thns2nTJpfpjRs3umxJ69ChgwYNGqRFixZp8eLFuvXWW13KTF2Jj4/Xrl27VFBQ4Jz3/fffa8+ePS7/z9w5XZhsSySA8xdQ3wF+beHChbr99tvVqlUr7du3T4sXL1ZGRoaOHDki6dSZSNXV1brhhhtcdjsAv3Xl5eU6evSoqqurVVhYqA0bNmj69OmKjo7W9OnTJUmhoaG677779Ne//lXNmzdX165d9frrr+vtt9/WmjVrnOvq1auXQkNDtWzZMudZbElJSUpLS1NVVZXmzZtnlXXQoEHq2bOn0tPTNX/+fDVt2lRPPPGEysrKdNddd0mSLrvsMi1ZskRvv/22OnfurH/+85968803XdYzYcIE3XDDDerZs6eGDh2qDRs2aNmyZS5jHn74YSUkJCg+Pl5+fn5asWKFwsLC1LZt23NmXLJkiTp27KjExEQtX75cn376qebOnesy5s4779SIESNUWVmp2267zeo1qUl6eroef/xxDR8+XLNmzZIxRhMnTlRsbKyGDx9e6/W0a9dOfn5+eu+99zR8+HAFBwfzj07AS3zq0ghlZWXas2ePnn76aU2aNEkLFy7U8ePHJZ3alJ2Xl6eMjAzdf//9euGFF7hYIf6lfPLJJ2rVqpXatm2rpKQkrVixQvfcc49yc3PVokUL57hp06bp9ttv17hx49S5c2ctX75cy5cvd26Bkk7tdurXr5+qqqo0cOBASaeO5enSpYsaN26snj17WmV1OBx666231LFjRw0bNkyXX365jh49qjVr1jiPDbrzzjs1cuRI3XLLLerevbs2b96sqVOnuqzn3//93zV79mzNnDlTXbp00YoVKzRjxgyXMSEhIXr00UeVkJCgxMREbd++XatXr3ZbJDIzM7Vw4UJ16dJFy5Yt0/Lly9WjRw+XMddee62aNm2q1NRUtWnTxuo1qUmjRo30wQcfKDg4WAMGDNCVV16pxo0b6/333z/rrtWatGjRQtOnT1dmZqZatWqla665xiN5AZzJYep5x3l+fr5mzJih2bNnq7S0VOPGjdPChQvPGLdw4UJdcsklzi/+xx9/XOnp6erQoYO3IwP4F1FYWKjWrVvrtddeo5wAqJFPbZkKDQ1VdHS0Pv30U0mnzlI6fQxFz549nWe2FBcXKy8vz+Vf4wBQVyoqKnT06FE9/PDDio2N1Z/+9Kf6jgTAh9Xrlqlnn31Wu3bt0okTJ9S0aVOlpaWpc+fOWrRokY4fP67Kykr169dP119/vYwxevnll7Vt2zb5+fnpuuuuO+PUbQCoC+vWrdPAgQN18cUXa9myZXzXADinet/NBwAA0JD51G4+AACAhoYyBQAAYIEyBQAAYKFeL9p5+kKcvi4qKsrl6sS+jKyeQVbPIKtnkNUzyOoZDSXr6d/DPJtal6nq6mpNnjxZERERmjx5ssuyiooKzZs3TwcOHFB4eLjGjRun6OjoC08MAADQQNR6N997772n2NjYsy5bu3atGjdurLlz52rYsGFasWJFnQUEAADwZbUqU4WFhcrNzXX5OYpfysnJUVJSkiSpd+/e2rFjB79IDgAA/iXUajffSy+9pBEjRujnn38+6/KioiJFRkZKkvz9/RUaGqoTJ06oSZMmLuOysrKUlZUl6dTvYp3+jS5fFxAQQFYPIKtnkNUzyOoZZPUMsnqX2zL1+eefq2nTpmrfvr3z51wuVEpKilJSUpzTDeGAM6nhHBwnkdVTyOoZZPUMsnoGWT2joWS1OgB9z549ysnJ0datW1VeXq6ff/5Zc+bM0X333eccExERocLCQkVGRqqqqkqlpaUKDw+vm/QAAAA+zG2ZSk9PV3p6uiRp586deuedd1yKlCQlJCRo3bp1uvTSS7Vp0ybFx8fL4XB4JjEAAIAPueCLdq5cuVI5OTmSpEGDBumnn37Svffeq3/+85+6+eab6ywgAACALzuvi3bGx8crPj5ekjR8+HDn/KCgII0fP75ukwEAADQA/JwMAACABcoUAACABcoUAACABcoUAACAhfM6AB0AAEmKja35Aobnz25dhw8fqaMcwIVhyxQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAIAFyhQAAICFAHcDysvLlZGRocrKSlVVVal3795KS0tzGbNu3TotW7ZMERERkqTU1FQlJyd7JjEAAIAPcVumAgMDlZGRoZCQEFVWVurRRx9Vt27ddOmll7qM69u3r8aMGeOxoAAAAL7I7W4+h8OhkJAQSVJVVZWqqqrkcDg8HgwAAKAhcBhjjLtB1dXVevDBB3X06FENGTJEI0aMcFm+bt06vfLKK2rSpIlatWqlUaNGKSoq6oz1ZGVlKSsrS5KUmZmp8vLyOnoanhUQEKDKysr6jlErZPUMsnoGWT3DG1mDg4M8uv7zcfKkd/6W8B7wjIaSNSio5vd8rcrUaSUlJXrqqad0yy23qG3bts75J06cUEhIiAIDA7VmzRpt3LhRGRkZbtd35MiR2j50vYqKilJBQUF9x6gVsnoGWT2DrJ7hjayxsTEeXf/5OHzYO39LeA94RkPJGhNT83v+vM7ma9y4seLj47Vt2zaX+eHh4QoMDJQkJScn68CBA+efEgAAoAFyW6aKi4tVUlIi6dSZfdu3b1dsbKzLmGPHjjlv5+TkqHXr1nUcEwAAwDe5PZvv2LFjmj9/vqqrq2WMUZ8+fZSQkKCVK1cqLi5OiYmJWr16tXJycuTv76+wsDCNHTvWG9kBAADqndsy1a5dO82cOfOM+cOHD3feTk9PV3p6et0mAwAAaAC4AjoAAIAFt1umAADeUbdnyNmty1tnyAG/BWyZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsECZAgAAsBDgbkB5ebkyMjJUWVmpqqoq9e7dW2lpaS5jKioqNG/ePB04cEDh4eEaN26coqOjPRYaAADAV7jdMhUYGKiMjAzNmjVLM2fO1LZt27R3716XMWvXrlXjxo01d+5cDRs2TCtWrPBYYAAAAF/itkw5HA6FhIRIkqqqqlRVVSWHw+EyJicnR0lJSZKk3r17a8eOHTLG1H1aAAAAH+N2N58kVVdX68EHH9TRo0c1ZMgQXXLJJS7Li4qKFBkZKUny9/dXaGioTpw4oSZNmtR9YgAAAB9SqzLl5+enWbNmqaSkRE899ZS++eYbtW3b9rwfLCsrS1lZWZKkzMxMRUVFnfc66kNAQABZPYCsnkFWz2hIWetCQ3qu3srakN4DZPWuWpWp0xo3bqz4+Hht27bNpUxFRESosLBQkZGRqqqqUmlpqcLDw8+4f0pKilJSUpzTBQUFFtG9JyoqiqweQFbPIKtneCdrjIfXX3vun2tDylo3eL96RkPJGhNT83ve7TFTxcXFKikpkXTqzL7t27crNjbWZUxCQoLWrVsnSdq0aZPi4+PPOK4KAADgt8jtlqljx45p/vz5qq6uljFGffr0UUJCglauXKm4uDglJiZq0KBBmjdvnu69916FhYVp3LhxXogOAIB7Mb/aAGC1Lsv7Hzl8uE5ywLe4LVPt2rXTzJkzz5g/fPhw5+2goCCNHz++bpMBAAA0AFwBHQAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwEJAfQcAAACnxMTG1t26LO9/5PDhc6+/AWX1NLZMAQAAWKBMAQAAWKBMAQAAWKBMAQAAWKBMAQAAWKBMAQAAWKBMAQAAWKBMAQAAWKBMAQAAWKBMAQAAWKBMAQAAWKBMAQAAWKBMAQAAWKBMAQAAWKBMAQAAWKBMAQAAWAhwN6CgoEDz58/X8ePH5XA4lJKSoqFDh7qM2blzp2bOnKno6GhJUq9evXT99dd7JjEAAIAPcVum/P39NXLkSLVv314///yzJk+erC5duqh169Yu437/+99r8uTJHgsKAADgi9zu5mvWrJnat28vSWrUqJFiY2NVVFTk8WAAAAANgdstU7+Un5+vgwcPqkOHDmcs27t3ryZNmqRmzZpp5MiRatOmTZ2FBAAA8FUOY4ypzcCysjJlZGTouuuuU69evVyWlZaWys/PTyEhIcrNzdVLL72kOXPmnLGOrKwsZWVlSZIyMzNVXl5eB0/B8wICAlRZWVnfMWqFrJ5BVs8gq6vg4CCPrv98nDx57u/nhpQ1KDjYS0ncKz958pzLyXph3GWtC0FBNb/na7VlqrKyUrNnz1b//v3PKFKSFBoa6rzdo0cPLVmyRMXFxWrSpInLuJSUFKWkpDinCwoKavPw9S4qKoqsHkBWzyCrZ3gna4yH11977p9rw8nqO0nJ6ine+B6Jian5Gbs9ZsoYo+eff16xsbG66qqrzjrm+PHjOr2Ba//+/aqurlZ4ePgFxgUAAGg43G6Z2rNnj9avX6+2bdtq0qRJkqSbbrrJ2QIHDx6sTZs26YMPPpC/v7+CgoI0btw4ORwOzyYHAADwAW7LVMeOHbVq1apzjklNTVVqamqdhQIAAGgouAI6AACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACABcoUAACAhQB3AwoKCjR//nwdP35cDodDKSkpGjp0qMsYY4yWLl2qrVu3Kjg4WGPHjlX79u09FhoAAMBXuC1T/v7+GjlypNq3b6+ff/5ZkydPVpcuXdS6dWvnmK1bt+ro0aOaM2eO9u3bp8WLF+vJJ5/0aHAAAABf4HY3X7NmzZxbmRo1aqTY2FgVFRW5jMnJydGAAQPkcDh06aWXqqSkRMeOHfNMYgAAAB9yXsdM5efn6+DBg+rQoYPL/KKiIkVFRTmnIyMjzyhcAAAAv0Vud/OdVlZWptmzZ2v06NEKDQ29oAfLyspSVlaWJCkzM9OlgPmygIAAsnoAWT2DrJ7RkLLWhYb0XMnqGWStvVqVqcrKSs2ePVv9+/dXr169zlgeERGhgoIC53RhYaEiIiLOGJeSkqKUlBTn9C/v48uioqLI6gFk9QyyeoZ3ssZ4eP215/65NpysvpOUrJ7ije+RmJian7Hb3XzGGD3//POKjY3VVVddddYxiYmJWr9+vYwx2rt3r0JDQ9WsWbMLTwwAANBAuN0ytWfPHq1fv15t27bVpEmTJEk33XSTswUOHjxY3bt3V25uru677z4FBQVp7Nixnk0NALUUG1uX/362W9fhw0fqKAcAX+K2THXs2FGrVq065xiHw6HbbrutzkIBAAA0FFwBHQAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwEKtf5sPaKjq7qKNXLARAHAmtkwBAABYoEwBAABYoEwBAABYoEwBAABYoEwBAABYoEwBAABYoEwBAABYoEwBAABY4KKdPoSLSwIA0PCwZQoAAMACZQoAAMACZQoAAMACZQoAAMACZQoAAMACZ/PhgnDmIQAAp7BlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwALXmQJwQbjWGACc4rZMLViwQLm5uWratKlmz559xvKdO3dq5syZio6OliT16tVL119/fd0nBQAA8EFuy1RSUpJSU1M1f/78Gsf8/ve/1+TJk+s0GAAAQEPg9pipTp06KSwszBtZAAAAGpw6OWZq7969mjRpkpo1a6aRI0eqTZs2dbFaAAAAn2ddpi6++GItWLBAISEhys3N1axZszRnzpyzjs3KylJWVpYkKTMzU1FRUbYP7xUBAQENJmtdaEjPlaxn4v3qu8jqGWT1DLLWnnWZCg0Ndd7u0aOHlixZouLiYjVp0uSMsSkpKUpJSXFOFxQU2D68V0RFRXkpa12dHWWnds+VrOfLW+933q+/5hs5JbJ6irusvpOUrJ7ije+8mJian7H1daaOHz8uY4wkaf/+/aqurlZ4eLjtagEAABoEt1umnn32We3atUsnTpzQn//8Z6WlpamyslKSNHjwYG3atEkffPCB/P39FRQUpHHjxsnhcHg8OAAAgC9wW6bGjRt3zuWpqalKTU2tqzwAAAANCj8nAwAAYIEyBQAAYIEyBQAAYIEyBQAAYIEyBQAAYIEyBQAAYIEyBQAAYIEyBQAAYIEyBQAAYIEyBQAAYIEyBQAAYIEyBQAAYIEyBQAAYIEyBQAAYIEyBQAAYCGgvgMA+D+xsTF1tCa79Rw+fKSOcgDAbx9bpgAAACxQpgAAACxQpgAAACxQpgAAACxQpgAAACxQpgAAACxQpgAAACxQpgAAACxQpgAAACxQpgAAACxQpgAAACz85n+bj986AwAAnsSWKQAAAAuUKQAAAAuUKQAAAAtuj5lasGCBcnNz1bRpU82ePfuM5cYYLV26VFu3blVwcLDGjh2r9u3beyQsAACAr3G7ZSopKUlTpkypcfnWrVt19OhRzZkzR3fccYcWL15cpwEBAAB8mdsy1alTJ4WFhdW4PCcnRwMGDJDD4dCll16qkpISHTt2rE5DAgAA+CrrY6aKiooUFRXlnI6MjFRRUZHtagEAABoEr15nKisrS1lZWZKkzMxMlxL2W9eQnitZPYOsnkFWzyCrZ5DVM+o7q3WZioiIUEFBgXO6sLBQERERZx2bkpKilJQU5/Qv7+c5dXXRTju1e65kPV9k9YzfVlbfyCmR1VPcZfWdpGT1FG/0iZiYmp+x9W6+xMRErV+/XsYY7d27V6GhoWrWrJntagEAABoEt1umnn32We3atUsnTpzQn//8Z6WlpamyslKSNHjwYHXv3l25ubm67777FBQUpLFjx3o8NAAAgK9wW6bGjRt3zuUOh0O33XZbXeUBAABoULgCOgAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgAXKFAAAgIWA2gzatm2bli5dqurqaiUnJ+vaa691Wb5u3TotW7ZMERERkqTU1FQlJyfXeVgAAABf47ZMVVdXa8mSJXrkkUcUGRmphx56SImJiWrdurXLuL59+2rMmDEeCwoAAOCL3O7m279/v1q2bKkWLVooICBAffv21ZYtW7yRDQAAwOe53TJVVFSkyMhI53RkZKT27dt3xrjNmzfrq6++UqtWrTRq1ChFRUXVbVIAAAAfVKtjptxJSEhQv379FBgYqDVr1mj+/PnKyMg4Y1xWVpaysrIkSZmZmf9ShashPVeyegZZPYOsnkFWzyCrZ9R3VrdlKiIiQoWFhc7pwsJC54Hmp4WHhztvJycna/ny5WddV0pKilJSUpzTBQUF5x34/MV44THcq91zJev5Iqtn/Lay+kZOiaye4i6r7yQlq6d4o0/ExNT8jN0eMxUXF6e8vDzl5+ersrJSGzduVGJiosuYY8eOOW/n5OSccXA6AADAb5XbLVP+/v669dZbNW3aNFVXV2vgwIFq06aNVq5cqbi4OCUmJmr16tXKycmRv7+/wsLCNHbsWG9kBwAAqHe1OmaqR48e6tGjh8u84cOHO2+np6crPT29bpMBAAA0AFwBHQAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwAJlCgAAwEJAbQZt27ZNS5cuVXV1tZKTk3Xttde6LK+oqNC8efN04MABhYeHa9y4cYqOjvZEXgAAAJ/idstUdXW1lixZoilTpuiZZ55Rdna2vvvuO5cxa9euVePGjTV37lwNGzZMK1as8FhgAAAAX+K2TO3fv18tW7ZUixYtFBAQoL59+2rLli0uY3JycpSUlCRJ6t27t3bs2CFjjEcCAwAA+BK3ZaqoqEiRkZHO6cjISBUVFdU4xt/fX6GhoTpx4kQdRwUAAPA9tTpmqq5kZWUpKytLkpSZmamYmBiPP6bvbCBz/1zJeiHI6hm/nay+k1Miq6e4eb/6UFi3nyyyXhDPt4lzc7tlKiIiQoWFhc7pwsJCRURE1DimqqpKpaWlCg8PP2NdKSkpyszMVGZmpm1ur5o8eXJ9R6g1snoGWT2DrJ5BVs8gq2c0pKw1cVum4uLilJeXp/z8fFVWVmrjxo1KTEx0GZOQkKB169ZJkjZt2qT4+Hg5HA6PBAYAAPAlbnfz+fv769Zbb9W0adNUXV2tgQMHqk2bNlq5cqXi4uKUmJioQYMGad68ebr33nsVFhamcePGeSE6AABA/avVMVM9evRQjx49XOYNHz7ceTsoKEjjx4+v22Q+JCUlpb4j1BpZPYOsnkFWzyCrZ5DVMxpS1po4DNcwAAAAuGD8nAwAAIAFytSvjBw58pzLp06dqv/93//1UpozffbZZ0pLS9Phw4frLUNtvfnmmxo/frwmTpyoSZMmad++ffUd6ZwKCws1c+ZM3Xfffbr33nu1dOlSVVZW1jj+3Xff1cmTJ72YUEpLS9PLL7/snP7HP/6hVatWeTVDbQ0fPlyTJk3S+PHjNWnSJL3zzjuqrq6u71huufsO8AWnX9vT/+Xn59c4tj6/s9LS0jRnzhzndFVVlcaMGePTZ3T7+ndsQ3xNpYbxubLh1etMwV52drY6duyo7OxspaWl1XecGu3du1eff/65ZsyYocDAQBUXF5+zmNQ3Y4yeeuopDR48WA888ICqq6v1wgsv6NVXX63xS+C9995T//79FRwc7LWcgYGB2rx5s6699lo1adLEa497IYKCgjRr1ixJ0o8//qg5c+bo559/9un3bUPxy9fWlwUHB+vbb79VeXm5goKCtH379jMureNOVVWV/P39PZTwTBf6HVtdXS0/P89vn6iL1xR1jzJ1Fjt37tQ777zjvPbFkiVLFBcX5/zJnPpSVlam3bt3KyMjQzNmzFBaWto5s+bm5urll19WcHCwLrvsMuXn53vteh7Hjh1TeHi4AgMDJcn5h//AgQP629/+prKyMjVp0kRjx45Vs2bNNHXqVLVr1067du1SdXW17rrrLnXo0MErWSVpx44dCgoK0sCBAyVJfn5+GjVqlO655x7dcMMNWrVqlb744gs5HA4lJyfLGKOioiI99thjatKkiTIyMryS08/PTykpKXr33Xd10003uSzLz8/Xc889pxMnTjhf29DQUE2cOFHz5s2Tn5+fysrK9Je//EVz585VQID3Pv5NmzbVHXfcoYceekg33HCDjDFasWKFdu3apYqKCg0ZMkR//OMfJUlvvfWWPvnkE/n5+albt266+eabvZbztLKyMs2cOVMlJSWqrKzUjTfeqMsvv1z5+fmaPn26LrvsMu3du1cRERF64IEHFBQU5PWMv1bTZ0uS1q9fr+eff75ePlvdu3dXbm6uevfurezsbPXr10+7d++WdOrnypYuXaqKigoFBQVp7NixiomJ0bp167R582aVlZWpurpajz32mFey1vQdu2rVKoWEhOjo0aOKj4/XbbfdJj8/P40cOVJ//OMf9eWXX2rMmDHq2LGjV3JeyGuakZGhW265RRdddJEk6a9//avGjBnjnPaGc/29uvvuu3XllVfq888/V2VlpcaPH6/Y2FivZbPFbr4GZMuWLerWrZtiYmIUHh6uAwcO1Di2vLxcixYt0pQpUzRjxgwVFxd7ManUtWtXFRYW6v7779fixYu1a9cuVVZW6sUXX9SECRM0Y8YMDRw4UK+++qrzPidPntSsWbM0ZswYPffcc17N++233+riiy92mRcaGqqoqCitXbtWP/zwg2bOnKmnnnpK/fv319ChQxUREaGMjAyvFanThgwZog0bNqi0tNRl/osvvqgrr7xSTz31lK644gq9+OKLCg0N1UUXXaRdu3ZJknJzc9W1a1evFqnTWrRooerqav34449au3atQkNDNX36dE2fPl0ffvih8vPztXXrVuXk5OjJJ5/UrFmzdM0113g9p3RqC+DEiRM1Y8YMZWRk6OWXX3b+3mheXp5SU1P19NNPKzQ0VJs2bfJ6vvLycucuvlmzZvn0Z6tfv37Kzs5WeXm5vv76a11yySXOZTExMXr88cc1c+ZMpaWl6ZVXXnEuO3jwoCZMmOC1IiXV/B27f/9+3XrrrXrmmWf0/fff67PPPpN06nXt0KGDZs2a5bUiJV3Yazpw4EDn9SCPHDmiiooKrxap2ggPD9eMGTM0ePBgvfPOO/Ud57ywZaoByc7O1tChQyVJffv21YYNG5SQkHDWsUeOHFF0dLSio6MlSVdccYXzp3y8ISQkRDNmzNBXX32lnTt36plnntF//Md/6Ntvv9UTTzwh6dRm8dP/cj6dUZI6deqk0tJSlZSUqHHjxl7LXJOdO3dqyJAhzl0NYWFh9ZonNDRUAwYM0HvvveeyRWTfvn2aOHGiJGnAgAFasWKFpFPvlY0bN6pz587Kzs7WkCFD6iX3L33xxRf65ptvnEWktLRUeXl5+vLLL5WUlOTcdVpfr7UxRq+++qq++uorORwOFRUV6ccff5QkRUdHO/8ItW/fXj/88IPX8/16N98333zjs5+tdu3a6YcfflB2dra6d+/usqy0tFTz58/X0aNHJZ3apXdaly5dvP7/v6bv2A4dOqhFixaS5NwK1Lt3b/n5+al3795ezShd2Gvap08fvfHGGxoxYoQ++uijet/Tcja9evWSdOpzdbqwNhSUqbPw9/fXL68YUVFRUY9pTvnpp5+0Y8cOffPNN3I4HM4DeS+//HKfy3qan5+f4uPjFR8fr7Zt2+p//ud/1Lp1a02bNq1W9/fmVfRbt26tzZs3u8wrLS1VQUGBmjdv7rUctTVs2DA9+OCDtfpCTExM1KuvvqqffvpJBw4cUOfOnT0f8Cy+//57+fn5qWnTpjLG6JZbblG3bt1cxnzxxRf1ku3XNmzYoOLiYmVmZiogIEB33323ysvLJcm561o69R4/Pb+++epnSzr1Hly2bJmmTp2qEydOOOevXLlS8fHxzoPof7kVypvHIko1f8f++hqLvxQYGOiV46TO5nxf0+DgYHXp0kU5OTn69NNP6+WAdXd/W09vMffz83Mp1g0Bu/nOIioqSt99950qKipUUlKiL7/8sr4jadOmTRowYIAWLFig+fPn67nnnlN0dLSMMWfNGhMTo/z8fOdZPhs3bvRq3iNHjigvL885fejQIcXGxqq4uFh79+6VJFVWVurbb791jjmdcffu3QoNDVVoaKjX8v7hD3/QyZMn9fHHH0s69S/7l19+WUlJSeratavWrFnj/HD/9NNPkk5tfSsrK/Naxl8KCwtTnz59tHbtWue8Sy+91PkabtiwwbnbISQkRHFxcVq6dKkSEhLq5cu/uLhYixYtUmpqqhwOh7p166YPPvjAeVLCkSNHVFZWpi5dumjdunXOsyRPv9beVlpaqqZNmyogIEA7duyol61P5yMmJsZnP1vSqV1M119/vdq2besyv7S01Hnw9OldUPWlpu/Y3bt3a//+/crPz1d1dbU+/fRTr+7Sq8mFvKbJyclaunSp4uLi6mWrry/+ba0rbJn6haqqKgUGBioqKkp9+vTRhAkTFB0dfcaxNPUhOzv7jONHevXqpezs7LNmDQoK0pgxY/Tkk08qODhYcXFxXs1bVlamF198USUlJfL391fLli11xx13KCUlRUuXLlVpaamqqqo0dOhQtWnTxpn5gQceUFVVle666y6v5nU4HJo4caIWL16sN954Q8YYde/eXTfddJP8/PyUl5eniRMnKiAgQMnJyUpNTVVKSoqmTZvmPHbK26666iq9//77zulbb71VCxYs0D/+8Q/nAcin9e3bV08//bSmTp3qtXynj+s5fTZW//79ddVVV0mSBg0apPz8fD344IOSTp2gMGnSJHXr1k2HDh3S5MmTFRAQoO7duys9Pd1rmU9/B1xxxRWaMWOGJkyYoLi4OJ8/EDYgIEATJkzwyc+WJEVGRjp3n/3SNddco/nz5+vNN9885xYgb6jpO3bNmjXq0KGDlixZ4jwAvWfPnvWU8v9cyGvavn17NWrUyHmijbf48t/WusIV0H/h0KFDeuGFFzR9+vT6jlInysrKFBISImOMlixZopYtWzr/mPmaqVOnauTIkV4vfcAv/da+A2Dv12egNWSnz0B+5plnvLqF+l/hc8WWqf/vgw8+0OrVqzV69Oj6jlJnsrKy9PHHH6uyslIXX3yx89RzAGf6LX4HAKd9/PHHeu211/Sf//mfXi1S/yqfK7ZMAQAAWOAAdAAAAAuUKQAAAAuUKQAAAAuUKQAAAAuUKQAAAAuUKQAAAAv/D6i8ul9qC+nmAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = (10, 6)\n",
    "\n",
    "months = [m.month_name()[:3] for m in downloads_by_month['m']]\n",
    "downloads = [c for c in downloads_by_month['c']]\n",
    "\n",
    "colors = ['blue']*9 + ['red']*3\n",
    "plt.bar(range(12), downloads, color=colors)\n",
    "plt.title('Downloads by month')\n",
    "plt.xticks(range(12), months)\n",
    "\n",
    "plt.savefig('downloads-by-month.png')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>d</th>\n",
       "      <th>downloads</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2021-04-01</td>\n",
       "      <td>134775</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2021-04-02</td>\n",
       "      <td>168082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2021-04-03</td>\n",
       "      <td>107416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2021-04-04</td>\n",
       "      <td>108983</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2021-04-05</td>\n",
       "      <td>93094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86</th>\n",
       "      <td>2021-06-26</td>\n",
       "      <td>69684</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>87</th>\n",
       "      <td>2021-06-27</td>\n",
       "      <td>74094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88</th>\n",
       "      <td>2021-06-28</td>\n",
       "      <td>104068</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>89</th>\n",
       "      <td>2021-06-29</td>\n",
       "      <td>131575</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90</th>\n",
       "      <td>2021-06-30</td>\n",
       "      <td>95098</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>91 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             d  downloads\n",
       "0   2021-04-01     134775\n",
       "1   2021-04-02     168082\n",
       "2   2021-04-03     107416\n",
       "3   2021-04-04     108983\n",
       "4   2021-04-05      93094\n",
       "..         ...        ...\n",
       "86  2021-06-26      69684\n",
       "87  2021-06-27      74094\n",
       "88  2021-06-28     104068\n",
       "89  2021-06-29     131575\n",
       "90  2021-06-30      95098\n",
       "\n",
       "[91 rows x 2 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "downloads_by_day = pd.read_sql(f\"\"\"\n",
    "    SELECT accessed_at::date AS d, COUNT(*) AS downloads\n",
    "    FROM downloads\n",
    "    WHERE accessed_at::date BETWEEN '{start_date}' AND '{end_date}'\n",
    "    GROUP BY d\n",
    "    ORDER BY d\n",
    "\"\"\", conn)\n",
    "\n",
    "downloads_by_day"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p = downloads_by_day.plot.bar(title='Downloads by day')\n",
    "p.figure.savefig('downloads-by-day.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "56200613"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_sql(f\"\"\"\n",
    "    SELECT COUNT(*) AS searches_in_period\n",
    "    FROM searches\n",
    "    WHERE accessed_at >= '{start_time}' AND accessed_at < '{end_time}'\n",
    "\"\"\", conn)\n",
    "\n",
    "searches = df.values[0][0]\n",
    "searches"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>month</th>\n",
       "      <th>searches</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2021-04-01</td>\n",
       "      <td>19101324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2021-05-01</td>\n",
       "      <td>18305260</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2021-06-01</td>\n",
       "      <td>18794029</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       month  searches\n",
       "0 2021-04-01  19101324\n",
       "1 2021-05-01  18305260\n",
       "2 2021-06-01  18794029"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_sql(f\"\"\"\n",
    "    SELECT date_trunc('month', accessed_at) AS month, COUNT(*) AS searches\n",
    "    FROM searches\n",
    "    WHERE accessed_at >= '{start_time}' AND accessed_at < '{end_time}'\n",
    "    GROUP BY month\n",
    "    ORDER BY month\n",
    "\"\"\", conn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>day</th>\n",
       "      <th>searches</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2021-04-01</td>\n",
       "      <td>889337</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2021-04-02</td>\n",
       "      <td>895889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2021-04-03</td>\n",
       "      <td>780547</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2021-04-04</td>\n",
       "      <td>770063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2021-04-05</td>\n",
       "      <td>639760</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86</th>\n",
       "      <td>2021-06-26</td>\n",
       "      <td>568262</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>87</th>\n",
       "      <td>2021-06-27</td>\n",
       "      <td>528281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88</th>\n",
       "      <td>2021-06-28</td>\n",
       "      <td>643457</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>89</th>\n",
       "      <td>2021-06-29</td>\n",
       "      <td>688179</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90</th>\n",
       "      <td>2021-06-30</td>\n",
       "      <td>610818</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>91 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           day  searches\n",
       "0   2021-04-01    889337\n",
       "1   2021-04-02    895889\n",
       "2   2021-04-03    780547\n",
       "3   2021-04-04    770063\n",
       "4   2021-04-05    639760\n",
       "..         ...       ...\n",
       "86  2021-06-26    568262\n",
       "87  2021-06-27    528281\n",
       "88  2021-06-28    643457\n",
       "89  2021-06-29    688179\n",
       "90  2021-06-30    610818\n",
       "\n",
       "[91 rows x 2 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "searches_by_day = pd.read_sql(f\"\"\"\n",
    "    SELECT accessed_at::date AS day, COUNT(*) AS searches\n",
    "    FROM searches\n",
    "    WHERE accessed_at::date BETWEEN '{start_date}' AND '{end_date}'\n",
    "    GROUP BY day\n",
    "    ORDER BY day\n",
    "\"\"\", conn)\n",
    "\n",
    "searches_by_day"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p = searches_by_day.plot.bar(title='Searches by day')\n",
    "p.figure.savefig('searches-by-day.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>total_time_saved_in_quarter</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>13937 days 00:11:17.998327</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  total_time_saved_in_quarter\n",
       "0  13937 days 00:11:17.998327"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_sql(f\"\"\"\n",
    "    SELECT JUSTIFY_INTERVAL(SUM(\n",
    "        CASE f.platform_tag\n",
    "            WHEN 'linux_armv7l' THEN 1                           \n",
    "            WHEN 'linux_armv6l' THEN 6                                                       \n",
    "            ELSE 0           \n",
    "        END *\n",
    "        CASE                        \n",
    "            WHEN b.duration > INTERVAL '1 day' THEN INTERVAL '0'\n",
    "            WHEN b.duration > INTERVAL '6.7 seconds' THEN b.duration - INTERVAL '6.7 seconds'\n",
    "            ELSE INTERVAL '0'               \n",
    "        END                  \n",
    "        )) AS total_time_saved_in_quarter                        \n",
    "    FROM downloads d\n",
    "    JOIN files f ON d.filename = f.filename\n",
    "    JOIN builds b ON b.build_id = f.build_id\n",
    "    WHERE f.abi_tag <> 'none'\n",
    "    AND d.accessed_at::date between '{start_date}' and '{end_date}'\n",
    "\"\"\", conn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>month</th>\n",
       "      <th>time_saved_in_month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2020-07-01</td>\n",
       "      <td>8803 days 06:58:40.185313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2020-08-01</td>\n",
       "      <td>11274 days 21:23:53.931867</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2020-09-01</td>\n",
       "      <td>9542 days 09:45:47.040065</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2020-10-01</td>\n",
       "      <td>8074 days 17:26:14.875176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2020-11-01</td>\n",
       "      <td>6374 days 06:58:39.848346</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2020-12-01</td>\n",
       "      <td>3878 days 06:59:28.147684</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>3727 days 17:44:07.111225</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2021-02-01</td>\n",
       "      <td>5639 days 08:25:11.612328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2021-03-01</td>\n",
       "      <td>6756 days 17:05:34.494183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2021-04-01</td>\n",
       "      <td>5148 days 08:09:03.090226</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2021-05-01</td>\n",
       "      <td>4542 days 20:14:38.898707</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2021-06-01</td>\n",
       "      <td>4240 days 19:47:36.009394</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        month        time_saved_in_month\n",
       "0  2020-07-01  8803 days 06:58:40.185313\n",
       "1  2020-08-01 11274 days 21:23:53.931867\n",
       "2  2020-09-01  9542 days 09:45:47.040065\n",
       "3  2020-10-01  8074 days 17:26:14.875176\n",
       "4  2020-11-01  6374 days 06:58:39.848346\n",
       "5  2020-12-01  3878 days 06:59:28.147684\n",
       "6  2021-01-01  3727 days 17:44:07.111225\n",
       "7  2021-02-01  5639 days 08:25:11.612328\n",
       "8  2021-03-01  6756 days 17:05:34.494183\n",
       "9  2021-04-01  5148 days 08:09:03.090226\n",
       "10 2021-05-01  4542 days 20:14:38.898707\n",
       "11 2021-06-01  4240 days 19:47:36.009394"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "time_saved_by_month = pd.read_sql(f\"\"\"\n",
    "    SELECT date_trunc('month', d.accessed_at) as month, JUSTIFY_INTERVAL(SUM(\n",
    "        CASE f.platform_tag\n",
    "            WHEN 'linux_armv7l' THEN 1\n",
    "            WHEN 'linux_armv6l' THEN 6\n",
    "            ELSE 0\n",
    "        END *\n",
    "        CASE\n",
    "            WHEN b.duration > INTERVAL '1 day' THEN INTERVAL '0'\n",
    "            WHEN b.duration > INTERVAL '6.7 seconds' THEN b.duration - INTERVAL '6.7 seconds'\n",
    "            ELSE INTERVAL '0'\n",
    "        END\n",
    "            )) AS time_saved_in_month\n",
    "    FROM downloads d\n",
    "    JOIN files f ON d.filename = f.filename\n",
    "    JOIN builds b ON b.build_id = f.build_id\n",
    "    WHERE f.abi_tag <> 'none'\n",
    "    AND d.accessed_at::date between '{one_year_ago}' and '{end_date}'\n",
    "    GROUP BY month\n",
    "    ORDER BY month\n",
    "\"\"\", conn)\n",
    "\n",
    "time_saved_by_month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "months = [m.month_name()[:3] for m in time_saved_by_month['month']]\n",
    "days_saved = [ts.days for ts in time_saved_by_month['time_saved_in_month']]\n",
    "\n",
    "colors = ['blue']*9 + ['red']*3\n",
    "plt.bar(range(12), days_saved, color=colors)\n",
    "plt.title('Time saved by month')\n",
    "plt.xticks(range(12), months)\n",
    "\n",
    "plt.savefig('time-saved-by-month.png')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>package</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>requests</td>\n",
       "      <td>153175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>chardet</td>\n",
       "      <td>150367</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>numpy</td>\n",
       "      <td>140902</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>importlib-metadata</td>\n",
       "      <td>131558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>octoprint-pisupport</td>\n",
       "      <td>124520</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>pip</td>\n",
       "      <td>118897</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>idna</td>\n",
       "      <td>109821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>h11</td>\n",
       "      <td>104934</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>urllib3</td>\n",
       "      <td>99970</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>typing-extensions</td>\n",
       "      <td>99920</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               package       c\n",
       "0             requests  153175\n",
       "1              chardet  150367\n",
       "2                numpy  140902\n",
       "3   importlib-metadata  131558\n",
       "4  octoprint-pisupport  124520\n",
       "5                  pip  118897\n",
       "6                 idna  109821\n",
       "7                  h11  104934\n",
       "8              urllib3   99970\n",
       "9    typing-extensions   99920"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_10_downloads = pd.read_sql(f\"\"\"\n",
    "    select b.package, count(*) c\n",
    "    from downloads d\n",
    "    join files f on f.filename = d.filename\n",
    "    join builds b on b.build_id = f.build_id\n",
    "    where accessed_at::date between '{start_date}' and '{end_date}'\n",
    "    group by b.package\n",
    "    order by c desc\n",
    "    limit 10;\n",
    "\"\"\", conn)\n",
    "\n",
    "top_10_downloads"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<ol>\n",
      "<li><strong><a href='https://www.piwheels.org/project/requests'>requests</a></strong> (153,175)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/chardet'>chardet</a></strong> (150,367)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/numpy'>numpy</a></strong> (140,902)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/importlib-metadata'>importlib-metadata</a></strong> (131,558)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/octoprint-pisupport'>octoprint-pisupport</a></strong> (124,520)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/pip'>pip</a></strong> (118,897)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/idna'>idna</a></strong> (109,821)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/h11'>h11</a></strong> (104,934)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/urllib3'>urllib3</a></strong> (99,970)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/typing-extensions'>typing-extensions</a></strong> (99,920)</li>\n",
      "</ol>\n"
     ]
    }
   ],
   "source": [
    "print(\"<ol>\")\n",
    "for i, row in top_10_downloads.iterrows():\n",
    "    print(f\"<li><strong><a href='https://www.piwheels.org/project/{row['package']}'>{row['package']}</a></strong> ({row['c']:,})</li>\")\n",
    "print(\"</ol>\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>package</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>pip</td>\n",
       "      <td>8729381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>wheel</td>\n",
       "      <td>5982230</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>setuptools</td>\n",
       "      <td>3978520</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>pyrogram</td>\n",
       "      <td>1502564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>future</td>\n",
       "      <td>751603</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>youtube-dl</td>\n",
       "      <td>749239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>aiohttp</td>\n",
       "      <td>696422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>pyngrok</td>\n",
       "      <td>661208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>aiodns</td>\n",
       "      <td>649962</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>six</td>\n",
       "      <td>547745</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      package        c\n",
       "0         pip  8729381\n",
       "1       wheel  5982230\n",
       "2  setuptools  3978520\n",
       "3    pyrogram  1502564\n",
       "4      future   751603\n",
       "5  youtube-dl   749239\n",
       "6     aiohttp   696422\n",
       "7     pyngrok   661208\n",
       "8      aiodns   649962\n",
       "9         six   547745"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_10_searches = pd.read_sql(f\"\"\"\n",
    "    select package, count(*) c\n",
    "    from searches\n",
    "    where accessed_at >= '{start_time}' AND accessed_at < '{end_time}'\n",
    "    group by package\n",
    "    order by c desc\n",
    "    limit 10;\n",
    "\"\"\", conn)\n",
    "\n",
    "top_10_searches"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<ol>\n",
      "<li><strong><a href='https://www.piwheels.org/project/pip'>pip</a></strong> (8,729,381)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/wheel'>wheel</a></strong> (5,982,230)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/setuptools'>setuptools</a></strong> (3,978,520)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/pyrogram'>pyrogram</a></strong> (1,502,564)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/future'>future</a></strong> (751,603)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/youtube-dl'>youtube-dl</a></strong> (749,239)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/aiohttp'>aiohttp</a></strong> (696,422)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/pyngrok'>pyngrok</a></strong> (661,208)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/aiodns'>aiodns</a></strong> (649,962)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/six'>six</a></strong> (547,745)</li>\n",
      "</ol>\n"
     ]
    }
   ],
   "source": [
    "print(\"<ol>\")\n",
    "for pkg, c in top_10_searches.values:\n",
    "    print(f\"<li><strong><a href='https://www.piwheels.org/project/{pkg}'>{pkg}</a></strong> ({c:,})</li>\")\n",
    "print(\"</ol>\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>package</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>opencv-python</td>\n",
       "      <td>1210</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>opencv-contrib-python</td>\n",
       "      <td>1108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>numpy</td>\n",
       "      <td>981</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>scipy</td>\n",
       "      <td>700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>tensorflow</td>\n",
       "      <td>686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>qhub-jupyterhub-theme</td>\n",
       "      <td>493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>pillow</td>\n",
       "      <td>445</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>cryptography</td>\n",
       "      <td>429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>pandas</td>\n",
       "      <td>381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>pip</td>\n",
       "      <td>335</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 package     c\n",
       "0          opencv-python  1210\n",
       "1  opencv-contrib-python  1108\n",
       "2                  numpy   981\n",
       "3                  scipy   700\n",
       "4             tensorflow   686\n",
       "5  qhub-jupyterhub-theme   493\n",
       "6                 pillow   445\n",
       "7           cryptography   429\n",
       "8                 pandas   381\n",
       "9                    pip   335"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_10_project_pages = pd.read_sql(f\"\"\"\n",
    "    select package, count(*) c\n",
    "    from project_page_hits\n",
    "    where accessed_at::date between '{start_date}' and '{end_date}'\n",
    "    group by package\n",
    "    order by c desc\n",
    "    limit 10;\n",
    "\"\"\", conn)\n",
    "\n",
    "top_10_project_pages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<ol>\n",
      "<li><strong><a href='https://www.piwheels.org/project/opencv-python'>opencv-python</a></strong> (1,210)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/opencv-contrib-python'>opencv-contrib-python</a></strong> (1,108)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/numpy'>numpy</a></strong> (981)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/scipy'>scipy</a></strong> (700)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/tensorflow'>tensorflow</a></strong> (686)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/qhub-jupyterhub-theme'>qhub-jupyterhub-theme</a></strong> (493)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/pillow'>pillow</a></strong> (445)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/cryptography'>cryptography</a></strong> (429)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/pandas'>pandas</a></strong> (381)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/pip'>pip</a></strong> (335)</li>\n",
      "</ol>\n"
     ]
    }
   ],
   "source": [
    "print(\"<ol>\")\n",
    "for pkg, c in top_10_project_pages.values:\n",
    "    print(f\"<li><strong><a href='https://www.piwheels.org/project/{pkg}'>{pkg}</a></strong> ({c:,})</li>\")\n",
    "print(\"</ol>\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>package</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>awscrt</td>\n",
       "      <td>12815</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>numpy</td>\n",
       "      <td>220</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>requirements-parser</td>\n",
       "      <td>91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>scipy</td>\n",
       "      <td>62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>opencv-python</td>\n",
       "      <td>58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>requests</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>notion2alfred-snippet</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>pyyaml</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>devjohn06test</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>opencv-contrib-python</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 package      c\n",
       "0                 awscrt  12815\n",
       "1                  numpy    220\n",
       "2    requirements-parser     91\n",
       "3                  scipy     62\n",
       "4          opencv-python     58\n",
       "5               requests     46\n",
       "6  notion2alfred-snippet     43\n",
       "7                 pyyaml     40\n",
       "8          devjohn06test     38\n",
       "9  opencv-contrib-python     37"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_10_project_json = pd.read_sql(f\"\"\"\n",
    "    select package, count(*) c\n",
    "    from project_json_downloads\n",
    "    where accessed_at::date between '{start_date}' and '{end_date}'\n",
    "    group by package\n",
    "    order by c desc\n",
    "    limit 10;\n",
    "\"\"\", conn)\n",
    "\n",
    "top_10_project_json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<ol>\n",
      "<li><strong><a href='https://www.piwheels.org/project/awscrt'>awscrt</a></strong> (12,815)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/numpy'>numpy</a></strong> (220)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/requirements-parser'>requirements-parser</a></strong> (91)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/scipy'>scipy</a></strong> (62)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/opencv-python'>opencv-python</a></strong> (58)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/requests'>requests</a></strong> (46)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/notion2alfred-snippet'>notion2alfred-snippet</a></strong> (43)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/pyyaml'>pyyaml</a></strong> (40)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/devjohn06test'>devjohn06test</a></strong> (38)</li>\n",
      "<li><strong><a href='https://www.piwheels.org/project/opencv-contrib-python'>opencv-contrib-python</a></strong> (37)</li>\n",
      "</ol>\n"
     ]
    }
   ],
   "source": [
    "print(\"<ol>\")\n",
    "for pkg, c in top_10_project_json.values:\n",
    "    print(f\"<li><strong><a href='https://www.piwheels.org/project/{pkg}'>{pkg}</a></strong> ({c:,})</li>\")\n",
    "print(\"</ol>\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>82603</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       c\n",
       "0  82603"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_sql(f\"\"\"\n",
    "    select count(distinct(package)) c\n",
    "    from project_json_downloads\n",
    "    where accessed_at::date between '{start_date}' and '{end_date}'\n",
    "\"\"\", conn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_agent</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mozilla</td>\n",
       "      <td>102782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>python-urllib</td>\n",
       "      <td>12943</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ccbot</td>\n",
       "      <td>395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>node-fetch</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>python-requests</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>goodbot</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>geturl</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>lanaibot please contact contactlanaibot@gmail....</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>linguee bot (http:</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>lanaibotapr1</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>tinytestbot</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>curl</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>java</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>mauibot (crawler.feedback+wc@gmail.com)</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>wordpress</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>zoominfobot (zoominfobot at zoominfo dot com)</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>lanaibotapr8 runid=lanaibotapr8</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>netestate ne crawler (+http:</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>opera</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>pip</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>python-httpx</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>yisouspider</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>apache-httpclient</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           user_agent       c\n",
       "0                                             mozilla  102782\n",
       "1                                       python-urllib   12943\n",
       "2                                               ccbot     395\n",
       "3                                          node-fetch     123\n",
       "4                                     python-requests      85\n",
       "5                                             goodbot      40\n",
       "6                                              geturl      34\n",
       "7   lanaibot please contact contactlanaibot@gmail....      25\n",
       "8                                  linguee bot (http:      15\n",
       "9                                        lanaibotapr1       8\n",
       "10                                        tinytestbot       7\n",
       "11                                               curl       7\n",
       "12                                               java       7\n",
       "13            mauibot (crawler.feedback+wc@gmail.com)       4\n",
       "14                                          wordpress       3\n",
       "15      zoominfobot (zoominfobot at zoominfo dot com)       1\n",
       "16                    lanaibotapr8 runid=lanaibotapr8       1\n",
       "17                       netestate ne crawler (+http:       1\n",
       "18                                              opera       1\n",
       "19                                                pip       1\n",
       "20                                       python-httpx       1\n",
       "21                                        yisouspider       1\n",
       "22                                  apache-httpclient       1"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_sql(f\"\"\"\n",
    "    select user_agent, count(*) c\n",
    "    from project_json_downloads\n",
    "    where accessed_at::date between '{start_date}' and '{end_date}'\n",
    "    group by user_agent\n",
    "    order by c desc\n",
    "\"\"\", conn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>os_name</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Linux</td>\n",
       "      <td>56167991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Darwin</td>\n",
       "      <td>18367</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Windows</td>\n",
       "      <td>11247</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>FreeBSD</td>\n",
       "      <td>2394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>OpenBSD</td>\n",
       "      <td>569</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>None</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Salman Rahwidean Janotama</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Dewandiaksa SM</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Ajie Kurniawan S</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     os_name         c\n",
       "0                      Linux  56167991\n",
       "1                     Darwin     18367\n",
       "2                    Windows     11247\n",
       "3                    FreeBSD      2394\n",
       "4                    OpenBSD       569\n",
       "5                       None        41\n",
       "6  Salman Rahwidean Janotama         2\n",
       "7             Dewandiaksa SM         1\n",
       "8           Ajie Kurniawan S         1"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "os_usage = pd.read_sql(f\"\"\"\n",
    "    select os_name, count(*) c\n",
    "    from searches\n",
    "    where accessed_at >= '{start_time}' AND accessed_at < '{end_time}'\n",
    "    group by os_name\n",
    "    order by c desc\n",
    "\"\"\", conn)\n",
    "\n",
    "os_usage"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99.94% linux\n"
     ]
    }
   ],
   "source": [
    "linux_usage_pc = 100 * (os_usage.values[0][1] / searches)\n",
    "print(f'{linux_usage_pc:2.2f}% linux')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>distro_name</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Raspbian GNU/Linux</td>\n",
       "      <td>49808463</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Debian GNU/Linux</td>\n",
       "      <td>5735305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Ubuntu</td>\n",
       "      <td>517307</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Alpine Linux</td>\n",
       "      <td>29273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Red Hat Enterprise Linux Server</td>\n",
       "      <td>27245</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       distro_name         c\n",
       "0               Raspbian GNU/Linux  49808463\n",
       "1                 Debian GNU/Linux   5735305\n",
       "2                           Ubuntu    517307\n",
       "3                     Alpine Linux     29273\n",
       "4  Red Hat Enterprise Linux Server     27245"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "distro_usage = pd.read_sql(f\"\"\"\n",
    "    select distro_name, count(*) c\n",
    "    from searches\n",
    "    where accessed_at >= '{start_time}' AND accessed_at < '{end_time}'\n",
    "    and os_name = 'Linux'\n",
    "    group by distro_name\n",
    "    order by c desc\n",
    "    limit 5\n",
    "\"\"\", conn)\n",
    "\n",
    "distro_usage"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>55543768</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          c\n",
       "0  55543768"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "debian_usage = pd.read_sql(f\"\"\"\n",
    "    select count(*) c\n",
    "    from searches\n",
    "    where accessed_at >= '{start_time}' AND accessed_at < '{end_time}'\n",
    "    and os_name = 'Linux'\n",
    "    and distro_name in ('Raspbian GNU/Linux', 'Debian GNU/Linux')\n",
    "    order by c desc\n",
    "    limit 5\n",
    "\"\"\", conn)\n",
    "\n",
    "debian_usage"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "98.83% debian\n"
     ]
    }
   ],
   "source": [
    "debian_usage_pc = 100 * (debian_usage.c[0] / searches)\n",
    "print(f'{debian_usage_pc:2.2f}% debian')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.92% ubuntu\n"
     ]
    }
   ],
   "source": [
    "ubuntu_usage = pd.read_sql(f\"\"\"\n",
    "    select count(*) c\n",
    "    from searches\n",
    "    where accessed_at >= '{start_time}' AND accessed_at < '{end_time}'\n",
    "    and os_name = 'Linux'\n",
    "    and distro_name = 'Ubuntu'\n",
    "    order by c desc\n",
    "    limit 5\n",
    "\"\"\", conn)\n",
    "\n",
    "ubuntu_usage_pc = 100 * (ubuntu_usage.c[0] / searches)\n",
    "print(f'{ubuntu_usage_pc:2.2f}% ubuntu')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Debian / Raspberry Pi OS': 55543768, 'Others': 656845}"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "debian = debian_usage.values[0][0]\n",
    "others = searches - debian\n",
    "\n",
    "distro_usage = {\n",
    "    'Debian / Raspberry Pi OS': debian,\n",
    "    'Others': others,\n",
    "}\n",
    "\n",
    "distro_usage"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.pie(distro_usage.values(), labels=distro_usage.keys())\n",
    "plt.title('Linux distro usage')\n",
    "\n",
    "plt.savefig('distro-usage.png')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>distro_version</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>52049617</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9</td>\n",
       "      <td>3415782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8</td>\n",
       "      <td>1278</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  distro_version         c\n",
       "0             10  52049617\n",
       "1              9   3415782\n",
       "2              8      1278"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "distro_versions = pd.read_sql(f\"\"\"\n",
    "    select distro_version, count(*) c\n",
    "    from searches\n",
    "    where accessed_at >= '{start_time}' AND accessed_at < '{end_time}'\n",
    "    and os_name = 'Linux'\n",
    "    and distro_name in ('Raspbian GNU/Linux', 'Debian GNU/Linux')\n",
    "    and distro_version in ('8', '9', '10')\n",
    "    group by distro_version\n",
    "    order by c desc\n",
    "\"\"\", conn)\n",
    "\n",
    "distro_versions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10 93.8%\n",
      "9 6.2%\n",
      "8 0.0%\n"
     ]
    }
   ],
   "source": [
    "total = sum(distro_versions.c)\n",
    "\n",
    "for v in distro_versions.values:\n",
    "    print(v[0], f'{(100 * v[1] / total):.1f}%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "version_names = {'10': 'Buster', '9': 'Stretch', '8': 'Jessie'}\n",
    "plt.pie(distro_versions.c, labels=[version_names[v] for v in distro_versions.distro_version])\n",
    "plt.title('Raspbian/Debian version usage')\n",
    "\n",
    "plt.savefig('debian-usage.png')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>arch</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>armv7l</td>\n",
       "      <td>41318543</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>aarch64</td>\n",
       "      <td>8775655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>x86_64</td>\n",
       "      <td>4157557</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>armv6l</td>\n",
       "      <td>1876199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>i686</td>\n",
       "      <td>40499</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>AMD64</td>\n",
       "      <td>11247</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>armv8l</td>\n",
       "      <td>10101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>arm64</td>\n",
       "      <td>7640</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>amd64</td>\n",
       "      <td>2963</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>ppc64le</td>\n",
       "      <td>135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>None</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>armv5tel</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>mips</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        arch         c\n",
       "0     armv7l  41318543\n",
       "1    aarch64   8775655\n",
       "2     x86_64   4157557\n",
       "3     armv6l   1876199\n",
       "4       i686     40499\n",
       "5      AMD64     11247\n",
       "6     armv8l     10101\n",
       "7      arm64      7640\n",
       "8      amd64      2963\n",
       "9    ppc64le       135\n",
       "10      None        41\n",
       "11  armv5tel        22\n",
       "12      mips        11"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "arch = pd.read_sql(f\"\"\"\n",
    "    select arch, count(*) c\n",
    "    from searches\n",
    "    where accessed_at >= '{start_time}' AND accessed_at < '{end_time}'\n",
    "    group by arch\n",
    "    order by c desc\n",
    "\"\"\", conn)\n",
    "\n",
    "arch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.pie(arch.c[:4], labels=arch.arch[:4])\n",
    "plt.title('Architecture usage')\n",
    "\n",
    "plt.savefig('arch.png')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "armv7l 79.6%\n",
      "aarch64 16.8%\n",
      "armv6l 3.6%\n"
     ]
    }
   ],
   "source": [
    "arm_arch = pd.read_sql(f\"\"\"\n",
    "    select arch, count(*) c\n",
    "    from searches\n",
    "    where accessed_at >= '{start_time}' AND accessed_at < '{end_time}'\n",
    "    and os_name = 'Linux'\n",
    "    and distro_name in ('Raspbian GNU/Linux', 'Debian GNU/Linux')\n",
    "    and arch in ('armv6l', 'armv7l', 'aarch64')\n",
    "    group by arch\n",
    "    order by c desc\n",
    "\"\"\", conn)\n",
    "\n",
    "total = sum(arm_arch.c)\n",
    "\n",
    "for v in arm_arch.values:\n",
    "    print(v[0], f'{(100 * v[1] / total):.1f}%') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.pie(arm_arch.c, labels=arm_arch.arch)\n",
    "plt.title('Raspbian/Debian Arm architecture usage')\n",
    "\n",
    "plt.savefig('debian-arch.png')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pyver</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3.7</td>\n",
       "      <td>36773442</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.7</td>\n",
       "      <td>7941645</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3.8</td>\n",
       "      <td>3509013</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3.9</td>\n",
       "      <td>1664067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3.5</td>\n",
       "      <td>1562918</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3.6</td>\n",
       "      <td>367980</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  pyver         c\n",
       "0   3.7  36773442\n",
       "1   2.7   7941645\n",
       "2   3.8   3509013\n",
       "3   3.9   1664067\n",
       "4   3.5   1562918\n",
       "5   3.6    367980"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "py_vers = pd.read_sql(f\"\"\"\n",
    "    select substring(py_version from 1 for 3) pyver, count(*) c\n",
    "    from searches\n",
    "    where accessed_at >= '{start_time}' AND accessed_at < '{end_time}'\n",
    "    and os_name = 'Linux'\n",
    "    and distro_name in ('Raspbian GNU/Linux', 'Debian GNU/Linux')\n",
    "    and arch in ('armv6l', 'armv7l', 'aarch64')\n",
    "    and py_name = 'CPython'\n",
    "    group by pyver\n",
    "    order by c desc\n",
    "    limit 6\n",
    "\"\"\", conn)\n",
    "\n",
    "py_vers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.pie(py_vers.c, labels=py_vers.pyver)\n",
    "plt.title('Python version usage')\n",
    "\n",
    "plt.savefig('py-vers.png')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.7 70.97%\n",
      "2.7 15.33%\n",
      "3.8 6.77%\n",
      "3.9 3.21%\n",
      "3.5 3.02%\n",
      "3.6 0.71%\n"
     ]
    }
   ],
   "source": [
    "total = sum(row[1] for row in py_vers.values)\n",
    "\n",
    "for ver, c in py_vers.values:\n",
    "    print(ver, f'{100 * (c / total):.2f}%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>installer_version</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>21.1.2</td>\n",
       "      <td>13562312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>18.1</td>\n",
       "      <td>12414762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>21.0.1</td>\n",
       "      <td>9830117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>21.1.1</td>\n",
       "      <td>8027994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>21.1</td>\n",
       "      <td>1748748</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>19.0.1</td>\n",
       "      <td>1501309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>9.0.1</td>\n",
       "      <td>1436057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>21.1.3</td>\n",
       "      <td>1435949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>20.3.4</td>\n",
       "      <td>1187729</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>20.3.3</td>\n",
       "      <td>985534</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  installer_version         c\n",
       "0            21.1.2  13562312\n",
       "1              18.1  12414762\n",
       "2            21.0.1   9830117\n",
       "3            21.1.1   8027994\n",
       "4              21.1   1748748\n",
       "5            19.0.1   1501309\n",
       "6             9.0.1   1436057\n",
       "7            21.1.3   1435949\n",
       "8            20.3.4   1187729\n",
       "9            20.3.3    985534"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pip_versions = pd.read_sql(f\"\"\"\n",
    "    select installer_version, count(*) c\n",
    "    from searches\n",
    "    where accessed_at >= '{start_time}' AND accessed_at < '{end_time}'\n",
    "    and installer_name = 'pip'\n",
    "    and installer_version is not null\n",
    "    group by installer_version\n",
    "    order by c desc\n",
    "\"\"\", conn)\n",
    "\n",
    "pip_versions[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "total = sum(v[1] for v in pip_versions.values)\n",
    "labels = [v[0] if v[1] / total > 0.01 else '' for v in pip_versions.values]\n",
    "\n",
    "plt.pie(pip_versions.c, labels=labels)\n",
    "plt.title('pip version usage')\n",
    "\n",
    "plt.savefig('pip-vers.png')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>setuptools_version</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>40.8.0</td>\n",
       "      <td>26933766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>57.0.0</td>\n",
       "      <td>7295780</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>56.0.0</td>\n",
       "      <td>3236865</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>41.2.0</td>\n",
       "      <td>2763925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>56.2.0</td>\n",
       "      <td>2148493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>40.7.1</td>\n",
       "      <td>1493317</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>53.1.0</td>\n",
       "      <td>1440726</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>49.2.1</td>\n",
       "      <td>1228322</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>54.2.0</td>\n",
       "      <td>938802</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>51.1.2</td>\n",
       "      <td>763037</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  setuptools_version         c\n",
       "0             40.8.0  26933766\n",
       "1             57.0.0   7295780\n",
       "2             56.0.0   3236865\n",
       "3             41.2.0   2763925\n",
       "4             56.2.0   2148493\n",
       "5             40.7.1   1493317\n",
       "6             53.1.0   1440726\n",
       "7             49.2.1   1228322\n",
       "8             54.2.0    938802\n",
       "9             51.1.2    763037"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "setuptools_versions = pd.read_sql(f\"\"\"\n",
    "    select setuptools_version, count(*) c\n",
    "    from searches\n",
    "    where accessed_at >= '{start_time}' AND accessed_at < '{end_time}'\n",
    "    and installer_name = 'pip'\n",
    "    and setuptools_version is not null\n",
    "    group by setuptools_version\n",
    "    order by c desc\n",
    "\"\"\", conn)\n",
    "\n",
    "setuptools_versions[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "total = sum(v[1] for v in setuptools_versions.values)\n",
    "labels = [v[0] if v[1] / total > 0.03 else '' for v in setuptools_versions.values]\n",
    "\n",
    "plt.pie(setuptools_versions.c, labels=labels)\n",
    "plt.title('setuptools version usage')\n",
    "\n",
    "plt.savefig('setuptools-vers.png')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "import csv\n",
    "from glob import glob\n",
    "from collections import defaultdict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "ONE_TB = 1e12\n",
    "months = ('202007', '202008', '202009', '202010', '202011', '202012', '202101', '202102', '202103', '202104', '202105', '202106')\n",
    "\n",
    "bandwidth = {}\n",
    "bandwidth_totals = defaultdict(int)\n",
    "for month in months:\n",
    "    bandwidth[month] = defaultdict(int)\n",
    "    files = glob('/home/ben/piwheels-logs/' + month + '/*-sizes.csv')\n",
    "    for file in files:\n",
    "        with open(file) as f:\n",
    "            r = csv.reader(f)\n",
    "            for log_type, count in r:\n",
    "                bandwidth[month][log_type] += int(count)\n",
    "                bandwidth_totals[month] += int(count) / ONE_TB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "202007 3.92 TB\n",
      "202008 6.38 TB\n",
      "202009 6.27 TB\n",
      "202010 6.58 TB\n",
      "202011 4.43 TB\n",
      "202012 3.32 TB\n",
      "202101 3.79 TB\n",
      "202102 5.87 TB\n",
      "202103 7.07 TB\n",
      "202104 5.35 TB\n",
      "202105 4.82 TB\n",
      "202106 4.72 TB\n"
     ]
    }
   ],
   "source": [
    "for month, data in bandwidth_totals.items():\n",
    "    print(month, f'{data:.2f} TB')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'pip downloads': 12508507792700,\n",
       " 'Misc': 1939319281327,\n",
       " 'Searches': 435480540129,\n",
       " '': 10516781945}"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m1, m2, m3 = months[-3:]\n",
    "\n",
    "bandwidth_in_period = {\n",
    "    'pip downloads': bandwidth[m1]['LOGDOWNLOAD'] + bandwidth[m2]['LOGDOWNLOAD'] + bandwidth[m3]['LOGDOWNLOAD'],\n",
    "    'Misc': bandwidth[m1]['OTHER'] + bandwidth[m2]['OTHER'] + bandwidth[m3]['OTHER'],\n",
    "    'Searches': bandwidth[m1]['LOGSEARCH'] + bandwidth[m2]['LOGSEARCH'] + bandwidth[m3]['LOGSEARCH'],\n",
    "    '': (\n",
    "        bandwidth[m1]['LOGPROJECT'] + bandwidth[m2]['LOGPROJECT'] + bandwidth[m3]['LOGPROJECT'] +\n",
    "        bandwidth[m1]['LOGJSON'] + bandwidth[m2]['LOGJSON'] + bandwidth[m3]['LOGJSON'] +\n",
    "        bandwidth[m1]['LOGPAGE'] + bandwidth[m2]['LOGPAGE'] + bandwidth[m3]['LOGPAGE']\n",
    "    ),\n",
    "}\n",
    "\n",
    "bandwidth_in_period"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.pie(bandwidth_in_period.values(), labels=bandwidth_in_period.keys())\n",
    "plt.title('Bandwidth breakdown')\n",
    "\n",
    "plt.savefig('bandwidth.png')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "pip downloads 83.98%\n",
      "Misc 13.02%\n",
      "Searches 2.92%\n",
      " 0.07%\n"
     ]
    }
   ],
   "source": [
    "for k, v in bandwidth_in_period.items():\n",
    "    pc = 100 * (v / sum(bandwidth_in_period.values()))\n",
    "    print(k, f'{pc:.2f}%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "bandwidth_by_month = {\n",
    "    'Jul': bandwidth_totals['202007'],\n",
    "    'Aug': bandwidth_totals['202008'],\n",
    "    'Sep': bandwidth_totals['202009'],\n",
    "    'Oct': bandwidth_totals['202010'],\n",
    "    'Nov': bandwidth_totals['202011'],\n",
    "    'Dec': bandwidth_totals['202012'],\n",
    "    'Jan': bandwidth_totals['202101'],\n",
    "    'Feb': bandwidth_totals['202102'],\n",
    "    'Mar': bandwidth_totals['202103'],\n",
    "    'Apr': bandwidth_totals['202104'],\n",
    "    'May': bandwidth_totals['202105'],\n",
    "    'Jun': bandwidth_totals['202106'],\n",
    "}\n",
    "\n",
    "colors = ['blue']*9 + ['red']*3\n",
    "plt.bar(range(len(bandwidth_by_month)), bandwidth_by_month.values(), color=colors)\n",
    "plt.title('Bandwidth by month (TB)')\n",
    "plt.xticks(range(len(bandwidth_by_month)), list(bandwidth_by_month.keys()))\n",
    "\n",
    "plt.savefig('bandwidth-by-month.png')\n",
    "\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}