R/ds.mxData.R

In datashield/dsMxClient: Datashield package for advanced structural equation modelling

#' 
#' @title Creates a new MxData object
#' @description This function is similar to OpenMx function \code{mxData}. 
#' @details The same output as the one returned by 'mxData' in the package \code{OpenMx} is generated
#' but an empty (all values = NA) dataframe is returned if the output dataframe is not valid according
#' to the DataSHIELD criteria (i.e. if the dataframe has less than the allowed number of observations/rows.
#' If the output dataframe has 0 observations (number of rows = 0), a data fram with one empty row is returned.
#' @param observed  a string, the name of a matrix or data.frame which provides data to the MxData object.
#' @param type A character string defining the type of data in the 'observed' argument. Must be one of "raw", 
#' "cov", "cor", or "sscp".
#' @param means An optional vector of means for use when 'type' is "cov", or "cor".
#' @param numObs	The number of observations in the data supplied in the 'observed' argument. Required unless 
#' 'type' equals "raw".
#' @param newobj the name of the new variable. By default the name of the new object is "new_mxData".
#' @param datasources a list of opal object(s) obtained after login in to opal servers;
#' these objects hold also the data assign to R, as \code{dataframe}, from opal datasources.
#' By default an internal function looks for 'opal' objects in the environment and sets this parameter. 
#' @return a new mxData object.
#' @author Gaye, A.
#' @export
#' @examples {
#'
#' }
#' @references   Steven M. Boker, Michael C. Neale, Hermine H. Maes, Michael J. Wilde, Michael Spiegel, Timothy R. Brick, 
#' Jeffrey Spies, Ryne Estabrook, Sarah Kenny, Timothy C., Bates, Paras Mehta, and John Fox. (2011) 
#' OpenMx: An Open Source Extended Structural Equation Modeling Framework. Psychometrika. 
#' Steven M. Boker, Michael C. Neale, Hermine H. Maes, Michael J. Wilde, Michael Spiegel, Timothy R. Brick, Ryne Estabrook, 
#' Timothy C. Bates, Paras Mehta, Timo von Oertzen, Ross J. Gore, Michael D. Hunter, Daniel C. Hackett, Julian Karch and 
#' Andreas M. Brandmaier. (2012) OpenMx 1.3 User Guide.
#' 
ds.mxData <- function(observed=NULL, type=NULL, means=NA, numObs=NA, newobj='new_mxData', datasources=NULL){
  
  # if no opal login details were provided look for 'opal' objects in the environment
  if(is.null(datasources)){
    findLogin <- getOpals()
    if(findLogin$flag == 1){
      datasources <- findLogin$opals
    }else{
      if(findLogin$flag == 0){
        stop(" Are yout logged in to any server? Please provide a valid opal login object! ", call.=FALSE)
      }else{
        message(paste0("More than one list of opal login object were found: '", paste(findLogin$opals,collapse="', '"), "'!"))
        userInput <- readline("Please enter the name of the login object you want to use: ")
        datasources <- eval(parse(text=userInput))
        if(class(datasources[[1]]) != 'opal'){
          stop("End of process: you failed to enter a valid login object", call.=FALSE)
        }
      }
    }
  }
  
  # Throw an error message if required arguments are not set
  if(is.null(observed)){
    stop(" Please provie the name of a matrix or data.frame for the argument 'observed'! ", call.=FALSE)
  }
  if(is.null(type)){
    stop(" Please provide a string defining the 'type' of data in the 'observed' argument! ", call.=FALSE)
  }
  
  # call the server side function that does the job
  cally <- paste0("mxData(", observed, ", '", type, "', ", means, ", ", numObs, ")")
  datashield.assign(datasources, newobj, as.symbol(cally))
  
  # check that the new object has been created and display a message accordingly
  finalcheck <- isAssigned(datasources, newobj)
  
}