# Map R functions to a GUI window Function map a large class of functions onto a set of `tcltk` widgets that allows the input of its parameter. ```r mapFuncToWidget(f,frm,bttLabel="OK",STORE="ff",callSubst="mini GUI call") ``` ## Arguments - `f`: Function to map. - `frm`: `tcltk` frame to place the GUI window. - `bttLabel`: execution button label. - `STORE`: A string. Name of the place where to store details needed by the GUI to perform the execution. - `callSubst`: string to set `call` attribute/slot in some of the R computations results. ## Details This function returns a frame which contains pairs of `tcltk` labels and text entry (or any other tktcl widget that allows to input values) and a button. In this way, this functions maps an \ function `f` into a GUI window that allows its computation. Therefore, it provides a map from the a set of \ function onto some class of GUI windows. Usually, the way function is executed provides with nasty and long `call` attributes, `call` parameter substitute these allowing a much more comfortable output. The string provided by `STORE` is used to store the function arguments in the list `miniGUIEnvir$miniGUIData`, enabling in this way the computation of the function. ## Returns This function returns a `tcltk` frame (an object created with `tkframe`). ## Author(s) Jorge Luis Ojeda Cabrera (jojeda@unizar.es ). ## See Also `miniGUI`, `makeWidgetCmd`, `tcltk`. ## Examples ```r require(tcltk) ## ## a window for lm ## ## create some data(in the global environment) n <- 100 d <- data.frame(x=runif(n)) d$z <- 0.5 * rnorm(n) d$y <- 2 * d$x + d$z ## create a tcltk frame and give it a title frm <- tktoplevel() tkwm.title(frm,"mapFuncToWidget for lm") ## create the GUI window map of lm mapFuncToWidget(lm,frm) ## ...you may close the window ## ## a window for T tests ## myTtest <- function(x,y,mu=0) return( t.test(x=x,y=y,mu=mu) ) ## create a tcltk frame and give it a title frm <- tktoplevel() tkwm.title(frm,"mapFuncToWidget for T tests") ## create the GUI window map of lm mapFuncToWidget(myTtest,frm) ## ...you may close the window ## ## a simple example ## g <- function(a=1,b=rnorm) {cat("--g--");paste("g(a,b)=",a+b(a))} h <- function(a=1,b=3,c=3) {cat("--h--");paste("h(a,b,c)=",a+b+c)} ## create a tcltk frame and give it a title frm <- tktoplevel() tkwm.title(frm,"mapFuncToWidget for g") ## create the GUI window map of g mapFuncToWidget(g,frm) ## ...you may close the window ## ## the ue of STORE ## frm <- tktoplevel() tkwm.title(frm,"another map") mapFuncToWidget(g,frm,bttLabel="press me !!!") ## ... and if you do not close the window mapFuncToWidget(h,frm,bttLabel="exec h !!!",STORE="fff") ## ...STORE should be added because g and h shares parameter ## names a and b ## now you may close the window ## ## a more involved example(see miniGUI widgets) ## ## some kernels kernels <- list( gaussK=function(x) dnorm(x,0,1), SqK=function(x) ifelse( abs(x) <= 1 , 1 , 0 ), EpaK=function(x) ifelse( abs(x) <= 1 , 3/(4*sqrt(5))*(1 - x^2/5) , 0 ), TrianK=function(x) ifelse( abs(x) <= 1 , (1 - abs(x)) , 0 ) ) ## how to compute the density at x prDensEst <- function(x,dat,h,K) mean( K((x-dat)/h) ) / h ## the fucntion to appear in the frame prDensCurvEst <- function(datos, bandwidth=miniGUIscale(from=.05,to=3,by=.05), Kernel=miniGUImenusel(c('"gaussK"','"SqK"','"EpaK"','"TrianK"','"QuartK"')) ) { n <- length(datos) Kernel <- kernels[[Kernel]] f <- function(x) sapply(x,function(x) prDensEst(x,datos,bandwidth,Kernel)) xeval <- seq(min(datos),max(datos),len=100) ##plot pts in x axis plot(datos,rep(0,n),pch="+",ylim=c(0,1.25*max(f(xeval))), ylab="dens.",main="Density") curve(f,add=T) return(f) } formals(prDensCurvEst)$bandwidth <- quote(miniGUIscale(from=.05,to=3,by=.05)) formals(prDensCurvEst)$Kernel <- quote( miniGUImenusel(c('"gaussK"','"SqK"','"EpaK"','"TrianK"','"QuartK"')) ) frm <- tktoplevel() tkwm.title(frm,"mapFuncToWidget for f") aaa <- mapFuncToWidget(prDensCurvEst,frm) ```

mapFuncToWidget function