Collect information about YouTube videos: Date created, Author (Channel Name), Author (Channel) URL, Title, Description, Likes, Dislikes, Number of Views, Number of Comments

Get_info_from_YT_videos.R

library(RSelenium)
psPath <-
  "C:/*/phantomjs.exe"
pJS <- phantom(pjs_cmd = psPath)
remDr <- remoteDriver(browserName = "phantomjs")
remDr$open()
# For more details of the initial configuration and 
# some commands used below see 
# https://gist.github.com/Nachtjagdgeschwader/e0775141c2171f769b8939006ba9624b
URLs <-
  read.csv("C:/*/URLs2.csv",
           # it's a .csv file with YouTube videos URLs in the first column
           header = F)$V1
URLs <- as.character(URLs)
datalist = list()
n <- length(URLs)
library(tcltk)
pb <-
  tkProgressBar(
    title = "Collencting videos data",
    min = 0,
    max = n,
    width = 200
  )
for (i in 1:n) {
  tryCatch({
    Sys.sleep(0.001)
    setTkProgressBar(pb, i, label = paste(round(i / n * 100, 1), "% done"))
    remDr$navigate(URLs[i])
    Date <-
      remDr$findElement(using = 'css', "strong.watch-time-text")$getElementText()
    Author <-
      remDr$findElement(using = 'css', "div.yt-user-info")$getElementText()
    AuthorURL <-
      remDr$findElement(using = 'css',
                        "div.yt-user-info a.g-hovercard.yt-uix-sessionlink.spf-link")$getElementAttribute("href")
    URL <-
      remDr$findElement(using = 'css', "a.ytp-title-link.yt-uix-sessionlink")$getElementAttribute("href")
    Title <-
      remDr$findElement(using = 'css', "span#eow-title.watch-title")$getElementText()
    Description <-
      remDr$findElement(using = 'css', "p#eow-description")$getElementText()
    Likes <-
      remDr$findElement(using = 'css',
                        ".like-button-renderer-like-button-unclicked .yt-uix-button-content")$getElementText()
    Dislikes <-
      remDr$findElement(
        using = 'css',
        ".like-button-renderer-dislike-button-unclicked .yt-uix-button-content"
      )$getElementText()
    Views <-
      remDr$findElement(using = 'css', "div.watch-view-count")$getElementText()
    remDr$findElement("css", "body")$sendKeysToElement(list(key = "end"))
    Sys.sleep(1)
    Comments <-
      remDr$findElement(using = 'css', "h2.comment-section-header-renderer")$getElementAttribute("outerHTML")
    
    cbind.fill <- function(...) {
      nm <- list(...)
      nm <- lapply(nm, as.matrix)
      n <- max(sapply(nm, nrow))
      do.call(cbind, lapply(nm, function (x)
        rbind(x, matrix(
          , n - nrow(x), ncol(x)
        ))))
    }
    data <-
      data.frame(
        cbind.fill(
          Date,
          Author,
          AuthorURL,
          URL,
          Title,
          Description,
          Likes,
          Dislikes,
          Views,
          Comments
        )
      )
    colnames(data) <-
      c(
        "Date",
        "Author",
        "AuthorURL",
        "URL",
        "Title",
        "Description",
        "Likes",
        "Dislikes",
        "Views",
        "Comments"
      )
    datalist[[i]] <- data
  }, error = function(e) {
  })
}
big_data = do.call(rbind, datalist)
nrow(big_data)
big_data1 <-
  as.data.frame(big_data, stringsAsFactors = default.stringsAsFactors())
big_data2 <-
  as.data.frame(
    big_data,
    row.names = NULL,
    optional = FALSE,
    cut.names = FALSE,
    col.names = names(
      c(
        "Date",
        "Author",
        "AuthorURL",
        "URL",
        "Title",
        "Description",
        "Likes",
        "Dislikes",
        "Views",
        "Comments"
      )
    ),
    fix.empty.names = TRUE,
    stringsAsFactors = default.stringsAsFactors()
  )
write.csv2(
  big_data,
  "C:/*/Data.csv",
  # destination of the final data save
  col.names = TRUE,
  sep = ";"
)
remDr$close()