R/print_error.R

In Marcogarofalo/Rose_vecchie: Cose (Things) with R

#' return a string like 1.32(11)
#'
#' This function allows you read a file in gnuplot style,
#'  Each block is a chunk of data separated by an empty line
#' @param mean mean value
#' @param err error 
#' @details  Important it allows only for 20 columns 
#' @export
#' @examples mean_print(1.32,0.11)

mean_print<-function(ave,err){
  if( !is.numeric(ave) | is.na(ave)  ){
    s=sprintf("NA")
    return(s)
  }
  if(!is.numeric(err) | is.na(err) | err==0 ){
    s=sprintf("%g",ave)
    return(s)
  }  
  
  if(err<0){
    s=sprintf("%g(%g)",ave,err)
    return(s)
  }
  if(ave<0)
    ave1<- -ave
  else
    ave1<-ave
  
  a<-(log10(ave1))
  e<-(log10(err+err/1000))
  if ( a<4  & a>-4){
    we<-(err/10^(as.integer(e-2)))
    if(e<0){
      e<-as.integer(e)
      format=sprintf("%%.%df(%%.0f)",-e+2)
      s=sprintf(format,ave1,we);
      
    }
    else if(e>1){
      s=sprintf("%.0f(%.0f)",ave1,err);
      
    }
    else{
      s=sprintf("%.1f(%.1f)",ave1,err);
      
    }
    
  }
  else{
    a<-as.integer(a)
    e<-as.integer(e)
    
    wm<-( ave1/10^(e))
    we<-(err/10^(e))
    s=sprintf("%.1f(%.1f)e%+-d",wm,we,e);
  }
  if(ave<0)
    s=sprintf("-%s",s);
  
  return(s)
  
}
#ave=1.4
#err=0.1*1000/1001
#mean_print(ave,err)