A generalization of Bert's suggestion is plotSubset <- function (data, subset, qsubset = substitute(subset)) { sdata <- data[eval(qsubset, data), ] with(sdata, plot(x, y, main = subsetToPlotmath(expr = qsubset))) }
subsetToPlotmath <- function(expr) { # Argument 'expr': an expression used as subset argument to subset() # Return: an expression used by plotmath that is more readable to non-programming people if (is.call(expr)) { for(i in seq_along(expr)) { expr[[i]] <- subsetToPlotmath(expr[[i]]) } if (is.name(funcName <- expr[[1]]) && !is.null(func <- env.subsetToPlotmath[[as.character(funcName)]])) { expr <- do.call(func, as.list(expr[-1])) } } expr } env.subsetToPlotmath <- new.env() env.subsetToPlotmath[["&"]] <- function(x, y) substitute(x ~ italic(and) ~ y) env.subsetToPlotmath[["|"]] <- function(x, y) substitute((x) ~ italic(or) ~ (y)) # internal parens not always needed env.subsetToPlotmath[["log10"]] <- function(x) substitute(italic(log)[10](x)) env.subsetToPlotmath[["exp"]] <- function(x) substitute(italic(e)^x) You can add more conversions to the environment env.subsetToPlotmath. Try it with > df <- data.frame(x=1:5, y=1:5) > plotSubset(df, x<1.5 | y>3.5) # see title "(x < 1.5) or (y > 3.5)" and pts at x=1,4,5. It doesn't get right the parentheses needed to enforce the order of evaluation: it always puts parentheses around the arguments to | and never puts them around the arguments to &. Bill Dunlap TIBCO Software wdunlap tibco.com On Mon, Aug 20, 2018 at 4:14 PM, MacQueen, Don via R-help < r-help@r-project.org> wrote: > Thanks Bert! > > It certainly works for the example (and shows a much deeper understanding > of eval, substitute, etc. than I have). But it doesn't appear to generalize > very well in the way I need (which of course I didn't think of mentioning > until after I sent the email -- sorry). > > Suppose subs is any expression that would be valid for the subset argument > of base::subset, for a given data frame. Then I can extract that subset of > the data frame by using > mydf[ with(mydf, eval(subs)) , ] > (or similar). > > Then, having plotted some aspect of that subset, I want to annotate the > plot with the subset specifications. > > I've used this approach to set up a system that helps me to interactively > review various subsets of a large set of data. I save the final selected > subsetting expressions in some sort of data structure, for later use in > preparing a report using rmarkdown. > > I was hoping to use plotmath to improve the appearance of the annotations > -- but I now think it's not worth this kind of effort. I think I'm going to > settle for mtext( as.character(subs) ). > > -Don > > -- > Don MacQueen > Lawrence Livermore National Laboratory > 7000 East Ave., L-627 > Livermore, CA 94550 > 925-423-1062 > Lab cell 925-724-7509 > > > > From: Bert Gunter <bgunter.4...@gmail.com> > Date: Monday, August 20, 2018 at 3:38 PM > To: "MacQueen, Don" <macque...@llnl.gov> > Cc: array R-help <r-help@r-project.org> > Subject: Re: [R] plotmath and logical operators? > > This is clumsy and probably subject to considerable improvement, but does > it work for you: > > left <- quote(x >= 3) > right <- quote(y <= 3) ## these can be anything > > ## the plot: > plot(1) > eval(substitute(mtext(expression(paste(left, " & ",right))), list(left = > left, right = right))) > > ## Expression evaluation > eval(substitute(with(df,left & right), list(left = left, right = right))) > Cheers, > Bert > > > Bert Gunter > > "The trouble with having an open mind is that people keep coming along and > sticking things into it." > -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) > > > On Mon, Aug 20, 2018 at 2:00 PM MacQueen, Don via R-help < > r-help@r-project.org<mailto:r-help@r-project.org>> wrote: > I would like to use plotmath to annotate a plot with an expression that > includes a logical operator. > > ## works well > tmp <- expression(x >= 3) > plot(1) > mtext(tmp) > > ## not so well > tmp <- expression(x >= 3 & y <= 3) > plot(1) > mtext(tmp) > > Although the text that's displayed makes sense, it won't be obvious to my > non-mathematical audience. > > I'd appreciate suggestions. > > > I've found a work-around that gets the annotation to look right > tmpw <- expression(paste( x >= 3, " & ", y <= 3) ) > plot(1) > mtext(tmpw) > > > But it breaks my original purpose, illustrated by this example: > > df <- data.frame(x=1:5, y=1:5) > tmp <- expression(x >= 3 & y <= 3) > tmpw <- expression(paste( x >= 3, " & ", y <= 3) ) > with(df, eval(tmp)) > [1] FALSE FALSE TRUE FALSE FALSE > with(df, eval(tmpw)) > [1] "FALSE & TRUE" "FALSE & TRUE" "TRUE & TRUE" "TRUE & FALSE" > "TRUE & FALSE" > > Thanks > -Don > > -- > Don MacQueen > Lawrence Livermore National Laboratory > 7000 East Ave., L-627 > Livermore, CA 94550 > 925-423-1062 > Lab cell 925-724-7509 > > > > ______________________________________________ > R-help@r-project.org<mailto:R-help@r-project.org> mailing list -- To > UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/ > posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/ > posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > [[alternative HTML version deleted]] ______________________________________________ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.