In jimrothstein/yt_api: R functions to use Google Youtube API

---- PURPOSE ----

Generic

---

Use one common structure to build up bundle. Except the query will be modified based on $$\code{type}$$

knitr::opts_chunk$set(echo = TRUE,
                      comment="      ##",  
                      error=TRUE, 
                      collapse=TRUE)
load_all()

return structure with all params

## if you need:  
playlists  <- readRDS(playlists, file=here("data","playlists.RDS"))

##  Choose type
    types <- c("playlists",
               "videos",
               "comments",
               "playlist_count",
               "video_count")

# REFERENCE
# curl \
#   'https://youtube.googleapis.com/youtube/v3/playlistItems?part=snippet%2CcontentDetails&playlistId=PLbcglKxZP5PNx1CGmLKT68vKELqrhj4hY&fields=pageInfo&key=[YOUR_API_KEY]' \
#   --header 'Authorization: Bearer [YOUR_ACCESS_TOKEN]' \
#   --header 'Accept: application/json' \
#   --compressed

## practice
    #type  <- "videos"
    type  <- "comments"

## to get all videos, for ONE playlist
global_params <- initialize()
global_params

No S3, construct one generic query.

generic_q  <- function(){

    ## query_defaults (starter values)
    if (!exists("query_defaults")) query_defaults <- get_typical_values()

    ## put it all togehter
    list( 
      ## from global_params
      key = global_params$api$api_key,

    ## for videos:
    base_url = "https://www.googleapis.com/youtube/v3/playlistItems",
    part  =  "snippet",
    ## only change
    fields  = "pageInfo.totalResults",

    #fields=paste(sep=",", "nextPageToken",
    #                  "items(id,snippet(title,description,publishedAt))"),


    ## from query_defaults (only what comments needs)
      maxResults = query_defaults$maxResults,
      channelId = query_defaults$channelId,
      playlistId  = "PLbcglKxZP5PN3RCRdidmIxRrB8luufWIW",  # gloria dehaven

    ## don't forget 
      pageToken = NULL
    )
   } 


query_params  <- generic_q()
query_params

Based on type, we now modify query_params list.

# =============================
    ## assume type = "comments"
# =============================


    query_params$base_url  <- "https://www.googleapis.com/youtube/v3/commentThreads"
    query_params$part   <-   "snippet,replies"

    query_params$fields  <-  paste(sep = ", ", "nextPageToken",
      "items(snippet(topLevelComment(snippet(videoId,textDisplay))))"
    )

    query_params$videoId  = "Mec9sw1cJk8"  # carter

    ## no playlistId
    query_params["playlistId"]  <- NULL
    #query_params$playlistId  =  "PLlXfTHzgMRUIqYrutsFXCOmiqKUgOgGJ5"  # Pavel Grinfeld, Linear Alg 3

    ## and no channelId
    query_params["channelId"]  <- NULL


    query_params

## spares
## query_params$playlistId  <- "PLbcglKxZP5PMU2rNPBpYIzLTgzd5aOHw2"
# playlistId  <- "PLbcglKxZP5PN3RCRdidmIxRrB8luufWIW"  # gloria dehaven
# spare:
# query_params$playlistId =    "PLbcglKxZP5PN3RCRdidmIxRrB8luufWIW"  # gloria dehaven
# playlistId  <-  "PLlXfTHzgMRUIqYrutsFXCOmiqKUgOgGJ5"  # Pavel Grinfeld, Linear Alg 3
# videoId  <- "Mec9sw1cJk8"  # carter
#playlistId  <- "PLbcglKxZP5PMZ7afIT7E2o9NwQIzqTI5l"

Have everything, pack into bundle (3 elements)

bundle  <- list(
                url = query_params$base_url,
                config = global_params$config ,
                query = query_params
                )

bundle$url
bundle$config
bundle$query

get_batch_videos

get_batch   <- function(url = NULL, query = NULL, config = NULL) {
      r  <- httr::GET(url =  url, 
                      query = query, 
                      config = config,
                      httr::verbose())


      ## process resonse, return table
      ##process(x = videos, r=r, videosdb = videosdb)
}

No S3, process batch

# ======================
##  assume type=videos
# ======================

process  <- function(r = NULL, type = NULL, db = NULL){

      ## rbind is done aftert this call; not in this function
      json_content <- get_json(r)

      if (type == "videos") {
          db <- json_content$items$snippet
          nextPageToken <- json_content$nextPageToken
          list(
            nextPageToken = nextPageToken,
            db = db
          )
      }
      if (type == "comments") {
          db  <- json_content$items$snippet$topLevelComment$snippet
          db  <-  tibble::as_tibble(db)
          nextPageToken <- json_content$nextPageToken
          list(
            nextPageToken = nextPageToken,
            db = db
          )
      }

}

sample call

process(r = r, type = "videos", db = )

#### combine 1st batch + additional (if any)
```r

## run 1st time (may refresh stale token)

      db  <- NULL
      r  <- get_batch(url = bundle$url, query=bundle$query, config=bundle$config)
      r  <- process(r=r, type="comments", db = db)
      db <- r$db
      nextPageToken <- r$nextPageToken

## Continue to get batches till null Token
      while (!is.null(nextPageToken)) {
        bundle$query$pageToken <- nextPageToken
        r <- get_batch(url = bundle$url, query=bundle$query, config=bundle$config)
        r  <- process(r = r, type=type, db=db)
        db <- rbind(db, r$db)
        nextPageToken <- r$nextPageToken
      }
## return
  db 
}
##videos  <- get_playlist_videos(playlistId)
head(db, 2)

View(db)

saveRDS(db, file=here::here("data/videosdb.RDS"))

cleanup videos

t  <- videos %>% dplyr::select(title, position) %>% 
    dplyr::arrange(title)
t
str(t)

here()
file <- "rmd/012_get_videos.Rmd"
file  <- basename(file)
dir="rmd"

jimTools::ren_pdf(file,dir)
jimTools::ren_github(file, dir)

jimrothstein/yt_api documentation built on Nov. 5, 2022, 8:05 p.m.