Client for the YouTube Data API

Search public YouTube data and retrieve channels, videos, playlists, comments, captions, live broadcasts, and reference data. Authenticated methods support common uploads, playlist changes, comment moderation, and media updates. See the YouTube Data API documentation at < https://developers.google.com/youtube/v3/>.

tuber

tuber is an R client for the YouTube Data API and selected YouTube Live Streaming API endpoints. It searches public YouTube data, retrieves channels, videos, playlists, comments, and captions, and supports common authenticated operations such as uploads, comment moderation, and playlist changes.

The package does not wrap every YouTube endpoint. Run vignette("api-conventions", package = "tuber") for the supported resources and known omissions.

Installation

Install the CRAN release:

install.packages("tuber")

Install the development version from GitHub:

# install.packages("pak")
pak::pak("gojiplus/tuber")

Authentication

Public read functions use an API key by default. Create a key in Google Cloud, enable the YouTube Data API v3 for that project, and set the key for the current R session:

yt_set_key("YOUR_YOUTUBE_API_KEY")

OAuth is required for private account data and every operation that changes YouTube data. Create a desktop OAuth client in Google Cloud, then authenticate in the system browser:

yt_oauth("YOUR_CLIENT_ID", "YOUR_CLIENT_SECRET")

yt_oauth() stores its token in the user's R cache directory by default. Pass an explicit token path if you need a different location.

Examples

Search for videos:

videos <- yt_search("Barack Obama", max_results = 25)

Retrieve fixed, snake-case video statistics:

stats <- get_video_stats("N708P-A45D0")

List a channel's playlists and the videos in one playlist:

playlists <- list_playlists(channel_id = "UCMtFAi84ehTSYSE9XoHefig")
items <- list_playlist_items(
  playlist_id = playlists$playlist_id[[1]],
  max_results = 100
)

Collect top-level comments and all replies:

comments <- get_all_comments("a-UQz7fqR3w", max_results = 500)

List and download caption tracks for a video owned by the authenticated account:

tracks <- list_captions("yJXTXN4xrI8")
caption <- download_caption(tracks$caption_id[[1]], as_raw = FALSE)

API conventions

• list_*() functions mirror YouTube list endpoints and return a data frame by default. Use simplify = FALSE to keep the collected API response.
• get_*() functions return a derived, enriched, or singular result.
• Plural ID arguments accept vectors. Singular ID arguments accept one value.
• max_results is the total result limit, even when the function makes several paginated requests.
• Fixed simplified schemas use snake-case column names. The columns returned by get_video_details() depend on part and retain YouTube's field names.
• Public reads default to auth = "key". Functions that can use either form of authentication accept auth = "key" or auth = "token". OAuth-only functions do not expose an auth argument.

License

tuber is released under the MIT License.