powerhour_build.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# -*- coding: utf-8 -*-\n",
    "\"\"\"\n",
    "Created on Sat Apr 18 20:18:15 2020\n",
    "\n",
    "@author: jeffr\n",
    "\"\"\"\n",
    "\n",
    "# packages needed\n",
    "import os\n",
    "import pandas as pd\n",
    "\n",
    "from pytube import YouTube\n",
    "from moviepy.video.io.ffmpeg_tools import ffmpeg_extract_subclip\n",
    "from moviepy.editor import *\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   number        clipid  year          event  \\\n",
      "0       1  HorseRacing1  2015  Triple crown    \n",
      "1       2         Golf1  2019        Masters   \n",
      "2       3         Golf7  2017           Open   \n",
      "3       4         Golf9  2016      Ryder Cup   \n",
      "4       5        Golf10  2016        Masters   \n",
      "\n",
      "                                                desc  \\\n",
      "0                                   American Pharaoh   \n",
      "1  Tiger wins 2019 masters (water balls by compet...   \n",
      "2                            Spieth Open go get that   \n",
      "3                   Reed-Rory Ryder cup singles 2016   \n",
      "4                                    Spieth collapse   \n",
      "\n",
      "                                                link  duration  \n",
      "0  https://www.youtube.com/watch?v=WyDcXtVb8_U&t=101        60  \n",
      "1    https://www.youtube.com/watch?v=pE-aOBGN8UM&t=0        70  \n",
      "2    https://www.youtube.com/watch?v=-0q3doVGUuE&t=0        80  \n",
      "3    https://www.youtube.com/watch?v=3fKB0UJxhe4&t=0        68  \n",
      "4  https://www.youtube.com/watch?v=bMf0JpfpZu8&t=209        60  \n"
     ]
    }
   ],
   "source": [
    "# import dataframe of powerhour videos\n",
    "df = pd.read_csv('./powerhour.csv')\n",
    "\n",
    "print(df.head(5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "https://www.youtube.com/watch?v=WyDcXtVb8_U&t=101\n"
     ]
    }
   ],
   "source": [
    "# create iterables for video loops\n",
    "itrlink = df.link\n",
    "itrduration = df.duration\n",
    "\n",
    "#df = df[0:4]\n",
    "\n",
    "print(df.iloc[0].link)\n",
    "\n",
    "VIDEO_STAGE_PATH = './video_staging/'\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "pass\n",
      "Task Completed!\n",
      "pass\n",
      "Task Completed!\n",
      "pass\n",
      "Task Completed!\n",
      "pass\n",
      "Task Completed!\n",
      "pass\n",
      "Task Completed!\n",
      "pass\n",
      "Task Completed!\n",
      "pass\n",
      "Task Completed!\n",
      "pass\n",
      "Task Completed!\n",
      "pass\n",
      "Task Completed!\n",
      "pass\n",
      "Task Completed!\n",
      "pass\n",
      "Task Completed!\n",
      "cant cast\n",
      "pass\n",
      "Task Completed!\n",
      "cant cast\n",
      "pass\n",
      "Task Completed!\n",
      "cant cast\n",
      "pass\n",
      "Task Completed!\n",
      "pass\n",
      "Task Completed!\n",
      "cant cast\n",
      "pass\n",
      "Task Completed!\n",
      "pass\n",
      "Task Completed!\n",
      "cant cast\n",
      "pass\n",
      "Task Completed!\n",
      "cant cast\n",
      "pass\n",
      "Task Completed!\n",
      "cant cast\n",
      "pass\n",
      "Task Completed!\n",
      "cant cast\n",
      "pass\n",
      "Task Completed!\n",
      "cant cast\n",
      "pass\n",
      "Task Completed!\n",
      "cant cast\n",
      "pass\n",
      "Task Completed!\n",
      "cant cast\n",
      "pass\n",
      "Task Completed!\n",
      "cant cast\n",
      "pass\n",
      "Task Completed!\n",
      "pass\n",
      "Task Completed!\n"
     ]
    }
   ],
   "source": [
    "# initialize empty dataframe to put metadata in\n",
    "video_length_logging = pd.DataFrame(columns=['video_title', 'video_start', 'video_finish'])\n",
    "\n",
    "#  download youtube videos\n",
    "for clip in range(0,len(df)):\n",
    "        \n",
    "    video_link = df.iloc[clip].link\n",
    "    clipid = df.iloc[clip].clipid\n",
    "    yt = YouTube(video_link)\n",
    "\n",
    "    # grab only video (mp4) files\n",
    "    mp4files = yt.streams.filter('mp4')\n",
    "    #\n",
    "    d_video = yt.streams.first()#(mp4files[-1].extension,mp4files[-1].resolution)\n",
    "    \n",
    "    # build some metadata for slicing videos later\n",
    "    video_title = d_video.title\n",
    "    try: \n",
    "        video_start = video_link.split(\"&t=\",1)[1]\n",
    "        float(video_start)\n",
    "    except:\n",
    "        print(\"cant cast\")\n",
    "        video_start = 0\n",
    "    video_finish = float(video_start) + df.iloc[clip].duration\n",
    "    video_length_logging = video_length_logging.append(\n",
    "        {'video_title': clipid, \n",
    "         'video_start': video_start, \n",
    "         'video_finish': video_finish}\n",
    "        , ignore_index=True)\n",
    "    \n",
    "    try: \n",
    "        #downloading the video \n",
    "        d_video.download(VIDEO_STAGE_PATH, filename=clipid, skip_existing=True) \n",
    "    except: \n",
    "        print(\"Some Error!\") \n",
    "    print('Task Completed!')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# clip videos to relevant time periods\n",
    "\n",
    "for video_clip_name in os.listdir('video_staging/'):\n",
    "    \n",
    "    video_metadata = video_length_logging[video_length_logging['video_title']==video_clip_name.split(\".mp4\",1)[0]]\n",
    "    \n",
    "    test_clip = \"video_staging/\" + video_clip_name\n",
    "    trimmed_clip = ffmpeg_extract_subclip(test_clip,\n",
    "                                          float(video_metadata['video_start']),\n",
    "                                          float(video_metadata['video_finish']),\n",
    "                                          targetname='./video_clipping/' + video_clip_name)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Intialize empty set for videos\n",
    "L =[]\n",
    "\n",
    "for root, dirs, files in os.walk(\"video_clipping/\"):\n",
    "\n",
    "    files.sort(reverse=True)\n",
    "\n",
    "    for file in files:\n",
    "        if os.path.splitext(file)[1] == '.mp4':\n",
    "            filePath = os.path.join(root, file)\n",
    "            video = VideoFileClip(filePath)\n",
    "            L.append(video)\n",
    "\n",
    "final_clip = concatenate_videoclips(L)\n",
    "final_clip.to_videofile(\"outputs/output.mp4\", fps=24, remove_temp=False)"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Edit 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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}