rytstat - R interface for work with YouTube APIs

rytstat package is an R interface for working with the following YouTube APIs:

• YouTube Reporting API
• YouTube Analytics API
• YouTube Data API

That is, the rytstat allows you to request any data available in the YouTube Creator Studio for further analysis and visualization using the R language.

Privacy Policy

The rytstat package for authorization uses the gargle package, the credentials obtained during authorization are stored exclusively on your local PC, you can find out the folder into which the credentials are cached using the ryt_auth_cache_path() function.

For loading data from your YouTube channel rytstat needs next scopes:

• View monetary and non-monetary YouTube Analytics reports for your YouTube content
• View your YouTube account
• View and manage your assets and associated content on YouTube
• View YouTube Analytics reports for your YouTube content
• Manage your YouTube account

For more details see Official YouTube API documentation.

The package does not transfer your credentials or data obtained from your advertising accounts to third parties, however, the responsibility for information leakage remains on the side of the package user. The author does not bear any responsibility for their safety, be careful when transferring cached credentials to third parties.

For more details, I recommend that you read the following articles from the official documentation of the gargle package:

• Stewarding the cache of user tokens
• Auth when using R in the browser
• How gargle gets tokens

Authorization process

You run gads_auth('me@gmail.com') and start OAuth Dance in the browser:

Upon success, you see this message in the browser:

Authentication complete. Please close this page and return to R.

And you credentials cached locally on your PC in the form of RDS files.

Key points

• By default, gargle caches user tokens centrally, at the user level, and their keys or labels also convey which Google identity is associated with each token.
• Token storage relies on serialized R objects. That is, tokens are stored locally on your PC in the form of RDS files.

Use own OAuth client

You can use own OAuth app:

app <- httr::oauth_app(appname = "app name", key = "app id", secret = "app secret")
ryt_auth_configure(app = app)

# or from json file 
ryt_auth_configure(path = 'D:/ga_auth/app.json')

# run authorization
ryt_auth('me@gmail.com')

Install

You can install rytstat from CRAN or GitHub:

install.packages("rytstat")

или GitHub:

devtools::install_github('selesnow/rytstat')

Examples

Auth

library(rytstat)
library(httr)

# auth app
app <- oauth_app(
    appname = 'my app',
    key = 'app id', 
    secret = 'app secret')

ryt_auth_configure(app = app)

ryt_auth(email = 'me@gmail.com')

Using YouTube Data API

# load channel data
channel <- ryt_get_channels()

# load videos
videos <- ryt_get_video_list()
video_details <- ryt_get_video_details(video_id = videos$id_video_id)

# load playlists
pl <- ryt_get_playlists()
pl_items <- ryt_get_playlist_items(
     playlist_id = pl$id[1],
     part = c('contentDetails', 'snippet'),
     fields = 'items(id,snippet/channelId,snippet/title,contentDetails/videoId)'
)

# search channels, playlists or videos
search_res_videos <- ryt_search(
    type            = 'video',
    q               = 'r language tutorial',
    published_after = '2022-03-01T00:00:00Z',
    published_before = '2022-06-01T00:00:00Z',
    max_results     = 10 
)

search_res_playlists <- ryt_search(
    type             = 'playlist',
    q                = 'r language tutorial',
    published_after  = '2022-03-01T00:00:00Z',
    published_before = '2022-06-01T00:00:00Z',
    max_results      = 50
)

search_res_channels <- ryt_search(
    type             = 'channel',
    q                = 'r language tutorial',
    published_after  = '2022-03-01T00:00:00Z',
    published_before = '2022-06-01T00:00:00Z',
    max_results      = 50
)

Using YouTube Analytics API

library(rytstat)

# get list of videos
videos <- ryt_get_video_list()

# get statistics by day and videos
# you can specify no more than 500 videos at a time
video_stat <- ryt_get_analytics(
  start_date = '2021-01-01', 
  end_date = '2021-09-01',
  dimensions = c('day', 'video'),
  metrics = c('views', 
              'likes', 
              'dislikes', 
              'comments', 
              'shares'),
  filters = str_glue('video=={str_c(head(videos$id_video_id, 500), collapse=",")}')
)

Using YouTube Reporting API

# auth
ryt_auth('me@gmail.com')

# get reporting data
## create job
ryt_create_job('channel_basic_a2')

## get job list
jobs2 <- ryt_get_job_list()

## get job report list
reports <- ryt_get_report_list(
  job_id = jobs$id[1],
  created_after = '2021-10-20T15:01:23.045678Z'
)

# get report data
data <- ryt_get_report(
  download_url = reports$downloadUrl[1]
)

# delete job
ryt_delete_job(jobs$id[1])

Author

Alexey Seleznev, Head of analytics dept. at Netpeak Telegram Channel: R4marketing YouTube Channel: R4marketing email: selesnow@gmail.com facebook: facebook.com/selesnow blog: alexeyseleznev.wordpress.com