R/get_study_fields.R

In AdlCruz/rctapi: rctapi: Download Clinical Trial data from ClinicalTrials.gov

#' Get Study Fields
#'
#' This function takes a search expression and character vector with any number
#' of field names and returns one dataframe with the merged and parsed csv API response.
#' 
#' @param search_expr A string following the querying guidelines at
#' https://clinicaltrials.gov/api/gui/ref/syntax and https://clinicaltrials.gov/api/gui/ref/expr
#' @param fields A character vector with field names. All field names available at
#' https://clinicaltrials.gov/api/info/study_fields_list?fmt=JSON, with "all_fields",
#' or with "get_vector_all_study_fields()"
#' @param max_studies A number indicating how many studies of all the ones that
#' match the search expression will be returned.
#' @param response_content Logical indicating whether the returned object is the unparsed but decoded,
#' content of the API query response. Will only return first 20 fields in fields argument. 
#' Suited for troubleshooting.
#' @keywords get many fields
#' @export
#' @examples
#' get_study_fields(search_expr = 'heart attack',fields = all_fields)

get_study_fields <- function(search_expr,
                             fields,
                             max_studies = 500,
                             response_content = FALSE) {
  # limit of 20 fields per query
  fields_matrix <- split_vector_unique(fields,20)
  
  # decoded but unparsed content of API response
  if (response_content == TRUE) {
    content <- get_response_content(search_expr = search_expr, fields = fields, max_studies = max_studies)
    return(content)
  }
  else {
  
    for (i in names(fields_matrix)) {
      # subset into search list
      fields_character_vector <- (fields_matrix[[i]])
      fields_list <- as.list(na.omit(fields_character_vector))
  
      # if i is name of the first column of the matrix create first df, then continue
      if (i == colnames(fields_matrix[1])) {
        df_merge <- get_fields(search_expr, fields = fields_list, max_studies = max_studies)
      }
      else {
        next_df <- get_fields(search_expr, fields = fields_list, max_studies = max_studies)
        df_merge <- cbind(df_merge,next_df,stringsAsFactors = FALSE)
  
      }
      # precautionary query delay
      q_delay(periods = c(0.2, 0.4))
  
    }
    # no duplicated columns, removes Rank columns
    df_merge <- df_merge[, !duplicated(colnames(df_merge))]
    return(df_merge)
   }
}