Youtube API v3 and python to generate list of views/likes on own youtube videos

Question

I'm trying to get a RaspberryPi3 Python based birdhouse reading the popularity of its own uploaded videos (which enables it to deduct which ones should be deleted and avoid hundreds of uploaded files). I thought the best way to read #views/#likes, was to use yt_analytics_report.py

When I input it always returns 0 values:

When I input:

$python yt_analytics_report.py --filters="video==bb_o--nP1mg"

or

$python yt_analytics_report.py --filters="channel==UCGXIq7h2UeFh9RhsSEsFsiA"

The output is:

$ python yt_analytics_report.py --filters="video==bb_o--nP1mg" {'metrics': 'views,estimatedMinutesWatched,averageViewDuration', 'filters': 'video==bb_o--nP1mg', 'ids': 'channel==MINE', 'end_date': '2018-01-12', 'start_date': '2018-01-06'}
Please visit this URL to authorize this application: [note: here was url with authe sequence etc. which I ackced] followed by the result:
views estimatedMinutesWatched averageViewDuration 0.0 0.0 0.0

I'm new to this; The last 3 days I've been testing a variety of filters, but the result is always the same. I guess I do something severely wrong. The (auto sensor triggered) video uploads work excellent so I presume the root cause is related to the way I'm using the yt-analytics example. Any suggestions on rootcause or alternative methods to retrieve #views/#likes of self uploaded youtubes are appreciated.

Answer 1

After a few days trying I have found a solution how to generate with Python and the Youtube API v3 a list which contains views, likes etc of uploaded videos of my own Youtube channel. I would like to share the complete code just in case anyone is having the same challenge. The code contains remarks and referrals to additional information. Please be aware that using the API consumes API-credits... This implies that (when you run this script continuesly or often) you can run out of your daily maximum numbers of API-credits set by Google.

# This python 2.7.14 example shows how to retrieve with Youtube API v3 a list of uploaded Youtube videos in a channel and also
# shows additional statistics of each individual youtube video such as number of views, likes etc.
# Please notice that YOU HAVE TO change API_KEY and Youtube channelID
# Have a look at the referred examples to get to understand how the API works
#
# The code consists of two parts:
# - The first part queries the videos in a channel and stores it in a list
# - The second part queries in detail each individual video
#
# Credits to the Coding 101 team, the guy previously guiding me to a query and Google API explorer who/which got me on track.
#
# RESULTING EXAMPLE OUTPUT: The output of the will look a bit like this:
#
# https://www.youtube.com/watch?v=T3U2oz_Y8T0
# Upload date:        2018-01-13T09:43:27.000Z
# Number of views:    8
# Number of likes:    2
# Number of dislikes: 0
# Number of favorites:0
# Number of comments: 0
#
# https://www.youtube.com/watch?v=EFyC8nhusR8
# Upload date:        2018-01-06T14:24:34.000Z
# Number of views:    6
# Number of likes:    2
# Number of dislikes: 0
# Number of favorites:0
# Number of comments: 0
#
#
import urllib #importing to use its urlencode function
import urllib2 #for making http requests
import json #for decoding a JSON response
#
API_KEY = 'PLACE YOUR OWN YOUTUBE API HERE'                 # What? How? Learn here: https://www.youtube.com/watch?v=JbWnRhHfTDA 
ChannelIdentifier = 'PLACE YOUR OWN YOUTUBE channelID HERE' # What? How? Learn here: https://www.youtube.com/watch?v=tf42K4pPWkM
#
# This first part will query the list of videos uploaded of a specific channel
# The identification is done through the ChannelIdentifier hwich you have defined as a variable
# The results from this first part will be stored in the list videoMetadata. This will be used in the second part of the code below.
#
# This code is based on the a very good example from Coding 101 which you can find here:https://www.youtube.com/watch?v=_M_wle0Iq9M
#
url = 'https://www.googleapis.com/youtube/v3/search?part=snippet&channelId='+ChannelIdentifier+'&maxResults=50&type=video&key='+API_KEY
response = urllib2.urlopen(url) #makes the call to YouTube
videos = json.load(response) #decodes the response so we can work with it
videoMetadata = [] #declaring our list
for video in videos['items']:
  if video['id']['kind'] == 'youtube#video':
      videoMetadata.append(video['id']['videoId']) #Appends each videoID and link to our list
#
# In this second part, a loop will run through the listvideoMetadata
# During each step the details a specific video are retrieved and displayed
# The structure of the API-return can be tested with the API explorer (which you can excecute without OAuth):
# https://developers.google.com/apis-explorer/#p/youtube/v3/youtube.videos.list?part=snippet%252CcontentDetails%252Cstatistics&id=Ks-_Mh1QhMc&_h=1&
#
for metadata in videoMetadata:
  print "https://www.youtube.com/watch?v="+metadata  # Here the videoID is printed
  SpecificVideoID = metadata
  SpecificVideoUrl = 'https://www.googleapis.com/youtube/v3/videos?part=snippet%2CcontentDetails%2Cstatistics&id='+SpecificVideoID+'&key='+API_KEY
  response = urllib2.urlopen(SpecificVideoUrl) #makes the call to a specific YouTube
  videos = json.load(response) #decodes the response so we can work with it
  videoMetadata = [] #declaring our list
  for video in videos['items']: 
    if video['kind'] == 'youtube#video':
        print "Upload date:        "+video['snippet']['publishedAt']    # Here the upload date of the specific video is listed
        print "Number of views:    "+video['statistics']['viewCount']   # Here the number of views of the specific video is listed
        print "Number of likes:    "+video['statistics']['likeCount']   # etc
        print "Number of dislikes: "+video['statistics']['dislikeCount']
        print "Number of favorites:"+video['statistics']['favoriteCount']
        print "Number of comments: "+video['statistics']['commentCount']
        print "\n"

Answer 2

Building on Sefo's answer above, I was able to clean up the outputs a bit.

The first function creates a list of relevant videos (you could replace this however you want), and the second iterates through this list and grabs the statistics and basic text data associated with each individual video.

The output is a list of dictionaries, perfect for conversion into a pandas DataFrame.

def youtube_search_list(q, max_results=10):
  # Call the search.list method to retrieve results matching the specified
  # query term.
    youtube = build(YOUTUBE_API_SERVICE_NAME, YOUTUBE_API_VERSION,
    developerKey=DEVELOPER_KEY)

  # Call the search.list method to retrieve results matching the specified
  # query term.
    search_response = youtube.search().list(
        q=q,
        part='id,snippet',
        maxResults=max_results,
        order='viewCount'
      ).execute()

    return search_response

def youtube_search_video(q='spinners', max_results=10):
    max_results = max_results
    order = "viewCount"
    token = None
    location = None
    location_radius = None
    youtube = build(YOUTUBE_API_SERVICE_NAME, YOUTUBE_API_VERSION,         
    developerKey=DEVELOPER_KEY)
    q=q
    #Return list of matching records up to max_search
    search_result = youtube_search_list(q, max_results)

    videos_list = []
    for search_result in search_result.get("items", []):

        if search_result["id"]["kind"] == 'youtube#video':
            temp_dict_ = {}
            #Available from initial search
            temp_dict_['title'] = search_result['snippet']['title']  
            temp_dict_['vidId'] = search_result['id']['videoId']  

            #Secondary call to find statistics results for individual video
            response = youtube.videos().list(
                part='statistics, snippet', 
                id=search_result['id']['videoId']
                    ).execute()  
            response_statistics = response['items'][0]['statistics']
            response_snippet = response['items'][0]['snippet']


            snippet_list = ['publishedAt','channelId', 'description', 
                            'channelTitle', 'tags', 'categoryId', 
                            'liveBroadcastContent', 'defaultLanguage', ]
            for val in snippet_list:
                try:
                    temp_dict_[val] = response_snippet[val]
                except:
                    #Not stored if not present
                    temp_dict_[val] = 'xxNoneFoundxx'    

            stats_list = ['favoriteCount', 'viewCount', 'likeCount', 
                          'dislikeCount', 'commentCount']
            for val in stats_list:
                try:
                    temp_dict_[val] = response_statistics[val]
                except:
                    #Not stored if not present
                    temp_dict_[val] = 'xxNoneFoundxx'

            #add back to main list
            videos_list.append(temp_dict_)

    return videos_list

Answer 3

this code will help you a ton, i was struggling with this for a long time, just provide the API key and the youtube channel name and the channel id in the search list from googleapiclient.discovery import build

DEVELOPER_KEY = "paste your API key here"
YOUTUBE_API_SERVICE_NAME = "youtube"
YOUTUBE_API_VERSION = "v3"
title = []
channelId = []
channelTitle = []
categoryId = []
videoId = []
viewCount = []
likeCount = []
dislikeCount = []
commentCount = []
favoriteCount = []
category = []
tags = []
videos = []
tags = []
max_search = 50
order = "relevance"
token = None
location = None
location_radius = None
youtube = build(YOUTUBE_API_SERVICE_NAME, YOUTUBE_API_VERSION,         
developerKey=DEVELOPER_KEY)
search_result = youtube.search().list(q="put the channel name here", type="video",     
pageToken=token, order=order, part="id,snippet",
                                  maxResults=max_search, location=location,
                                  locationRadius=location_radius, channelId='put the     
channel ID here').execute()  # this line is to get the videos of the channel by the 
name of it
for search_result in search_result.get("items", []):
if search_result["id"]["kind"] == 'youtube#video':
    title.append(search_result['snippet']['title'])  # the title of the video
    videoId.append(search_result['id']['videoId'])  # the ID of the video
    response = youtube.videos().list(part='statistics, snippet',     
id=search_result['id'][
        'videoId']).execute()  # this is the other request because the statistics and     
snippet require this because of the API
    channelId.append(response['items'][0]['snippet']['channelId'])  # channel ID, 
which is constant here
    channelTitle.append(response['items'][0]['snippet']['channelTitle'])    # channel 
title, also constant
    categoryId.append(response['items'][0]['snippet']['categoryId'])        # stores 
the categories of the videos
    favoriteCount.append(response['items'][0]['statistics']['favoriteCount'])   # 
stores the favourite count of the videos
    viewCount.append(response['items'][0]['statistics']['viewCount'])           # 
stores the view counts
    """  

    the likes and dislikes had a bug all along, which required the if else instead of 
just behaving like the viewCount"""
    if 'likeCount' in response['items'][0]['statistics'].keys():             # checks 
for likes count then restores it, if no likes stores 0
        likeCount.append(response['items'][0]['statistics']['likeCount'])
    else:
        likeCount.append('0')
    if 'dislikeCount' in response['items'][0]['statistics'].keys():             # checks for dislikes count then stores it, if no dislikes found returns 0
        dislikeCount.append(response['items'][0]['statistics']['dislikeCount'])
    else:
        likeCount.append('0')
    if 'commentCount' in response['items'][0]['statistics'].keys():             # checks for comment count then stores it, if no comment found stores 0
        commentCount.append(response['items'][0]['statistics']['commentCount'])
    else:
        commentCount.append('0')
    if 'tags' in response['items'][0]['snippet'].keys():                        # checks for tags count then stores it, if none found stores 0
        tags.append(response['items'][0]['snippet']['tags'])
    else:
        tags.append('0')
youtube_dict = {
'tags': tags,
'channelId': channelId,
'channelTitle': channelTitle,
'categoryId': categoryId,
'title': title,
'videoId': videoId,
'viewCount': viewCount,
'likeCount': likeCount,
'dislikeCount': dislikeCount,
'commentCount': commentCount,
'favoriteCount': favoriteCount
}
for x in youtube_dict:
print(x)
for y in youtube_dict[x]:
    print(y)

please re indent the code because the site ruined the indentation of python to make the code in the code section not in the words section. good luck