R/reqParse.R

In MarkEdmondson1234/searchConsoleR: Google Search Console R Client

make_empty_df <- function(){
  warning("No data found for supplied dates - returning NA", call. = FALSE)
  the_df <- data.frame(matrix(NA, nrow = 1, ncol = 4))
  names(the_df) <- c('clicks','impressions','ctr','position')
  
  the_df
}

#' Parsing function for \code{\link{search_analytics}}
#'
#' @param x req$content from API response
#' @param dim the dimensions passed from search_analytics
#' @param pn PrettyNames passed from search_analytics
#'
#' @keywords internal
#' @family parsing functions
parse_search_analytics <- function(x, dim, prettyNames=TRUE){

  the_data <- x$rows
  
  if(is.null(dim)){
    if(is.null(the_data)){
      the_df <- make_empty_df()
    } else {
      the_df <- the_data
      attr(the_df, "aggregationType") <- x$responseAggregationType
      attr(the_df, "totalRows") <- nrow(x$rows)      
    }
    
    return(the_df)
  }


  # a bit of jiggery pokery (data processing)
  dimensionCols <- data.frame(
    Reduce(rbind, lapply(the_data$keys, function(x) rbind(x))),
    row.names=NULL, 
    stringsAsFactors = FALSE)
  
  ## if no rows, get out of here.
  if(!nrow(dimensionCols) > 0) {
    empty_df <- make_empty_df()
    if('date' %in% dim) empty_df$date <- as.Date(NA)
    
    return(empty_df)
  }
  
  names(dimensionCols ) <- dim
  dimensionCols <- lapply(dimensionCols, unname)
  
  if('date' %in% dim){
    dimensionCols$date <- as.Date(dimensionCols$date)
  }
  
  if(all('country' %in% dim, prettyNames)){
    dimensionCols$countryName <- sapply(dimensionCols$country, lookupCountryCode)
  }
  
  metricCols <- the_data[setdiff(names(the_data), 'keys')]
  
  metricCols$ctr <- as.numeric(metricCols$ctr)
  metricCols$position <- as.numeric(metricCols$position)
  
  the_df <- data.frame(dimensionCols, 
                       metricCols, 
                       stringsAsFactors = FALSE, 
                       row.names = NULL)
  
  attr(the_df, "aggregationType") <- x$responseAggregationType
  attr(the_df, "totalRows") <- nrow(x$rows)
  
  the_df

}

#' Parsing function for \code{\link{list_sitemaps}}
#'
#' @param x req$content from API response
#'
#' @keywords internal
#' @family parsing functions
parse_sitemaps <- function(x){
  list(sitemap = x$sitemap[, setdiff(names(x$sitemap), "contents")],
       contents = x$sitemap$contents)

}

#' Parsing function for \code{\link{crawl_errors}}
#'
#' @param x req$content from API response
#'
#' @keywords internal
#' @family parsing functions
parse_crawlerrors <- function(x){
  cpt <- x$countPerTypes
  ## data parsing gymnastics
  ldf <- Reduce(rbind,
                apply(cpt, 1, function(row) {
                  data.frame(platform = row['platform'],
                             category = row['category'],
                             count = row$entries$count,
                             timecount = row$entries$timestamp )
                })
  )

  ## transform to something useable
  ldf$platform <- as.character(ldf$platform)
  ldf$category <- as.character(ldf$category)
  ldf$count <- as.numeric(as.character(ldf$count))
  ldf$timecount <- RFC_convert(ldf$timecount, drop_time = TRUE)

  ldf

}

#' Parsing function for \code{\link{list_crawl_error_samples}}
#'
#' @param x req$content from API response
#'
#' @keywords internal
#' @family parsing functions
parse_crawlerror_sample <- function(x){
  errs <- x$urlCrawlErrorSample

  if(!is.null(errs)){
    errs$last_crawled <- RFC_convert(errs$last_crawled)
    errs$first_detected <- RFC_convert(errs$first_detected)

    errs      
  } else {
    message("No error samples available.")
    NULL
  }

}

#' Parsing function for \code{\link{error_sample_url}}
#'
#' @param x req$content from API response
#'
#' @keywords internal
#' @family parsing functions
parse_errorsample_url <- function(x){
  raw_details <- x

  if(all(c('last_crawled', 'first_detected') %in% names(raw_details))){
    raw_details$last_crawled <- RFC_convert(raw_details$last_crawled)
    raw_details$first_detected <- RFC_convert(raw_details$first_detected)
    inner_details <- Reduce(rbind, raw_details$urlDetails)

    data.frame(linkedFrom=inner_details,
               last_crawled=raw_details$last_crawled,
               first_detected=raw_details$first_detected,
               pageUrl=raw_details$pageUrl)

  }

}