用pandas过滤出WordPress精品插件.ipynb

用pandas过滤出WordPress精品插件.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 首先引入所有需要用的库\n",
    "\n",
    "#读取jsonl文件的库\n",
    "import jsonlines\n",
    "\n",
    "# 数据分析的库\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "\n",
    "import maya\n",
    "import json\n",
    "\n",
    "import requests\n",
    "\n",
    "import functools\n",
    "\n",
    "import html\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "plugins=[]\n",
    "\n",
    "keepkeys=[\n",
    "        \"slug\",\n",
    "        \"name\",\n",
    "        \"author\",\n",
    "        \"downloaded\",\n",
    "        \"rating\",\n",
    "        \"num_ratings\",\n",
    "        \"added\",\n",
    "        \"last_updated\",\n",
    "        \"tested\",\n",
    "        \"support_threads\",\n",
    "        \"support_threads_resolved\"]\n",
    "\n",
    "#  output.jsonl 由爬虫得来，详见 ：\n",
    "#  用Python爬取WordPress官网所有插件\n",
    "#  https://bestscreenshot.com/scrap-all-plugins-from-wordpress-org/\n",
    "\n",
    "with jsonlines.open('../output.jsonl') as reader:\n",
    "    for obj in reader:        \n",
    "        for k in list(obj.keys()):\n",
    "            if k not in keepkeys:\n",
    "                del obj[k]\n",
    "        plugins.append(obj)\n",
    "        \n",
    "df = pd.DataFrame(plugins)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 把name里的html转义符做一下反转义\n",
    "df['name']=df['name'].apply(lambda x: html.unescape(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 根据当前日期计算出距离上次更新的天数\n",
    "df['last_updated_days']=df['last_updated'].apply(lambda x: (maya.now() -  maya.when(x) ).days  ) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 返回符合所有条件为真的数据\n",
    "def conjunction(*conditions):\n",
    "    return functools.reduce(np.logical_and, conditions)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 所有的过滤的条件\n",
    "last_updated_days_bool = df['last_updated_days'] <= 365\n",
    "\n",
    "downloaded_bool = df['downloaded'] > 1000\n",
    "\n",
    "rating_bool = df['rating'] >= 90 \n",
    "\n",
    "num_ratings_bool = df['num_ratings'] >= 100 \n",
    "\n",
    "support_threads_bool = df['support_threads'] > 0  \n",
    "\n",
    "support_threads_resolved_bool = df[\"support_threads_resolved\"] > 0 \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_filtered = df[conjunction(\n",
    "                    last_updated_days_bool,\n",
    "                    downloaded_bool,\n",
    "                    rating_bool,\n",
    "                    num_ratings_bool,\n",
    "                    support_threads_bool,\n",
    "                    support_threads_resolved_bool,\n",
    "                    )]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "added                       312\n",
       "author                      312\n",
       "downloaded                  312\n",
       "last_updated                312\n",
       "name                        312\n",
       "num_ratings                 312\n",
       "rating                      312\n",
       "slug                        312\n",
       "support_threads             312\n",
       "support_threads_resolved    312\n",
       "tested                      312\n",
       "last_updated_days           312\n",
       "dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_filtered.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 把过滤结果转换成字典\n",
    "data_filtered_dict=data_filtered.set_index('slug').T.to_dict()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 把字典转换成列表\n",
    "data_filtered_list=[]\n",
    "for k,v in data_filtered_dict.items():\n",
    "    x={'slug':k}\n",
    "    y=v\n",
    "    z = {**x, **y}\n",
    "    data_filtered_list.append(z)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'slug': 'accelerated-mobile-pages',\n",
       " 'added': '2016-02-07',\n",
       " 'author': '<a href=\"https://ampforwp.com/\">Ahmed Kaludi, Mohammed Kaludi</a>',\n",
       " 'downloaded': 4137343,\n",
       " 'last_updated': '2019-02-28 12:52pm GMT',\n",
       " 'name': 'AMP for WP – Accelerated Mobile Pages',\n",
       " 'num_ratings': 780,\n",
       " 'rating': 90,\n",
       " 'support_threads': 286,\n",
       " 'support_threads_resolved': 60,\n",
       " 'tested': '5.1',\n",
       " 'last_updated_days': 30}"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_filtered_list[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 把结果列表存为json文件\n",
    "json.dump(data_filtered_list, open(\"data_filtered_result.json\",\"w\"))\n"
   ]
  }
 ],
 "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": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}