Check a list of URLs for existence of the Google Tag Manager Container code (via Donal Phipps)

gtm_checker.R

#' Check a list of URLs for existence of the Google Tag Manager Container code
#'
#' @param containers
#' A character vector of the container IDs for Google Tag Manager
#' @param urlFile
#' A character vector providing the path to a csv file with 2 columns: url, the full url to the page to be checked; country, the country where the URL is hosted.
#'
#' @param outputdir
#' A character vector providing the path to the desired output directory for the csv of results. If not provided, the csv will be output to the current working directory.
#'
#' @return
#' Writes a csv to outputdir with several columns: url, the url provided in the urlFile csv. country: the country providedd in the urlFile csv. After this, the function will return 2 columns for each GTM container snippet provided as an argument in containers - one column confirming if the snippet is provided in an iFrame, and a second column confirming if the snippet is provided in a script tag.
#' @export
#'
#' @examples
#' check_gtm(containers = c("GTM-TT98938", "GTM-FFD99D"), urlFile = "downloads/urlFile.csv")
check_gtm <- function(containers, urlFile, outputdir = getwd() ) {

  if (!require(httr)) install.packages("httr")
  if (!require(XML)) install.packages("XML")
  if (!require(dplyr)) install.packages("dplyr")

  resultsFrame<-NULL
  urlList <- read.csv(urlFile, stringsAsFactors=FALSE)

  for ( i in seq_along(urlList$url) ) {
    message(paste0("Checking url number ", i, " of ", length(urlList$url)))
    url <- urlList$url[i]
    html2<-GET(urlList[i,1])
    content2<-content(html2, as="text")
    parsedHtml<-htmlParse(content2,asText=TRUE)

    # helper functions
      # Check for tracking ID in iframe
      iframe_checker <- function(tagID) {
      src<-xpathApply(parsedHtml, paste0("//iframe[@src='//www.googletagmanager.com/ns.html?id=",tagID,"']"), xmlGetAttr, "src")
      print(src)
      if ( length(src) == 0 ) {
        iFrameResult <-"Not found"
      } else {
        iFrameResult <-"OK"
      }
      iFrameResult
      }

      # check for tracking ID in script tag
      script_checker <- function(tagID) {
        scriptNode<-xpathApply(parsedHtml, "//script", xmlValue)
        scripChars<-as.character(scriptNode)
        scriptloc<-grep(pattern = tagID, x = scripChars)

        if ( length(scriptloc) == 0 ) {
          scriptResult<-"Not found"
        } else {
          scriptResult<-"OK"
        }
        scriptResult
      }

    # Loop through containers and check for GTM in iFrame or script tags
    checkResults <- data.frame( url = url, country = urlList$country[i])
    for ( container in containers ) {
      iFrameStatus <- iframe_checker(container)
      scriptStatus <- script_checker(container)
      results <- data.frame(cbind(iFrameStatus, scriptStatus))
      names(results) <- c(paste0(container, "-iFrame"), paste0(container, "-script"))
      checkResults <- cbind(checkResults, results)
    }

    resultsFrame<-rbind(resultsFrame,checkResults)

  }

  if ( !file.exists(outputdir) ) {
    dir.create(outputdir)
  }

  write.csv(resultsFrame, file=paste0(outputdir,"/GTM check Full Results-",Sys.Date(),".csv") )
  message(paste0("Form check complete. Results have been written to:\n", outputdir,"/GTM check Full Results-", Sys.Date(),".csv"))
  rm(list=ls())
}