do we really need to attach package maps to use `map_data` ?

[moodymudskipper]

The package maps is loaded silently when using ggplot2::map_data.

It does message about the conflicts, but it's easy to miss it and is especially annoying because it masks purr::map.

Is there a good reason why it needs to be attached ? In the current situation I think I'd rather have it fail and tell me to attach maps myself, so I can add the library call before attaching purrr and control what is happening.

This happens with version 3.0, apologies if it has been solved since, I'm not able to test it on the last version.

[clauswilke]

The line that is responsible here is this one:

ggplot2/R/fortify-map.r

Line 78 in 9108d9b

try_require("maps", "map_data")

@hadley I've seen the try_require() pattern in several places in the ggplot2 code base. Is there a reason we attach the respective packages instead of simply testing whether they exist? I think it would generally be preferable to not attach packages from inside ggplot2.

[clauswilke]

In other words, would anything terrible happen if we removed line 47 here:

ggplot2/R/utilities.r

Lines 45 to 49 in e9d4e5d

	try_require <- function(package, fun) {
	if (requireNamespace(package, quietly = TRUE)) {
	library(package, character.only = TRUE)
	return(invisible())
	}

[hadley]

I think you'd need to review on a case-by-case basis. I agree that it's undesirable to load packages, but I'm almost certain it was necessary for maps (but a recent release may have fixed the problem). I also have vague recollection that stat_smooth() may have some issues for mgcv — i.e. something like y ~ mgcv::s(x) doesn't work?

[clauswilke]

How about adding an argument attach = FALSE to try_require(), so we need to explicitly turn it on when we're sure we need attaching. Initially, all calls to try_require() would be changed so they say attach = TRUE, and then we can one-by-one remove those statements.

I'm pretty certain mgcv is fine without attaching now. I just used it in my ungeviz package.

[hadley]

There are only 5 uses of try_require() so I think someone could just remove the library() call, and then carefully check the code. I think StatBinhex and StatSummaryHex will need some ::. Looking at the code I now wonder if it was actually quantreg and/or MatrixModels in stat_quantile() caused problems. It's also quite possible the problems have been resolved since I wrote the code, or I may have been confused in the first place (I think much of this was written before I understood namespaces).

[clauswilke]

I'll take a look. The problem is that for some of the cases where we're using try_require() I have no idea what kind of code example would trigger the respective lines. That's why I think I'll have to follow the step-wise process of eliminating one case at a time.

@moodymudskipper Could you provide an example with map_data, ideally one that currently breaks something because the maps package gets attached?

[hadley]

I can probably dredge my memory if you can point me to specific cases.

[clauswilke]

I just opened a PR. I have resolved hexbin, and I assume maps will be straightforward once @moodymudskipper provides a reprex. mgcv is imported, so we don't need to worry about it. This means the only two remaining cases are quantreg and MatrixModels in stat_quantile() here:

ggplot2/R/stat-quantile.r

Lines 53 to 60 in c84d9a0

	try_require("quantreg", "stat_quantile")
	if (is.null(formula)) {
	if (method == "rqss") {
	try_require("MatrixModels", "stat_quantile")
	formula <- eval(substitute(y ~ qss(x, lambda = lambda)),
	list(lambda = lambda))
	} else {

I suspect I can resolve them with the examples provided in geom_quantile().

[clauswilke]

The problem seems to be that we can't use match.fun() to get a function from a package that is not attached. Is there an alternative?

match.fun("rq")
#> Error in get(as.character(FUN), mode = "function", envir = envir): object 'rq' of mode 'function' was not found

match.fun("quantreg::rq")
#> Error in get(as.character(FUN), mode = "function", envir = envir): object 'quantreg::rq' of mode 'function' was not found

require(quantreg)
#> Loading required package: quantreg
#> Loading required package: SparseM
#> 
#> Attaching package: 'SparseM'
#> The following object is masked from 'package:base':
#> 
#>     backsolve
match.fun("rq")
#> function (formula, tau = 0.5, data, subset, weights, na.action, 
#>     method = "br", model = TRUE, contrasts = NULL, ...) 
#> {
#>     call <- match.call()
#>     mf <- match.call(expand.dots = FALSE)
#>     m <- match(c("formula", "data", "subset", "weights", "na.action"), 
#>         names(mf), 0)
#>     mf <- mf[c(1, m)]
#>     mf$drop.unused.levels <- TRUE
#>     mf[[1]] <- as.name("model.frame")
#>     mf <- eval.parent(mf)
#>     if (method == "model.frame") 
#>         return(mf)
#>     mt <- attr(mf, "terms")
#>     weights <- as.vector(model.weights(mf))
#>     Y <- model.response(mf)
#>     if (method == "sfn") {
#>         if (requireNamespace("MatrixModels") && requireNamespace("Matrix")) {
#>             X <- MatrixModels::model.Matrix(mt, data, sparse = TRUE)
#>             vnames <- dimnames(X)[[2]]
#>             X <- as(X, "matrix.csr")
#>             mf$x <- X
#>         }
#>     }
#>     else X <- model.matrix(mt, mf, contrasts)
#>     eps <- .Machine$double.eps^(2/3)
#>     Rho <- function(u, tau) u * (tau - (u < 0))
#>     if (length(tau) > 1) {
#>         if (any(tau < 0) || any(tau > 1)) 
#>             stop("invalid tau:  taus should be >= 0 and <= 1")
#>         if (any(tau == 0)) 
#>             tau[tau == 0] <- eps
#>         if (any(tau == 1)) 
#>             tau[tau == 1] <- 1 - eps
#>         coef <- matrix(0, ncol(X), length(tau))
#>         rho <- rep(0, length(tau))
#>         fitted <- resid <- matrix(0, nrow(X), length(tau))
#>         for (i in 1:length(tau)) {
#>             z <- {
#>                 if (length(weights)) 
#>                   rq.wfit(X, Y, tau = tau[i], weights, method, 
#>                     ...)
#>                 else rq.fit(X, Y, tau = tau[i], method, ...)
#>             }
#>             coef[, i] <- z$coefficients
#>             resid[, i] <- z$residuals
#>             rho[i] <- sum(Rho(z$residuals, tau[i]))
#>             fitted[, i] <- Y - z$residuals
#>         }
#>         taulabs <- paste("tau=", format(round(tau, 3)))
#>         dimnames(coef) <- list(dimnames(X)[[2]], taulabs)
#>         dimnames(resid) <- list(dimnames(X)[[1]], taulabs)
#>         fit <- z
#>         fit$coefficients <- coef
#>         fit$residuals <- resid
#>         fit$fitted.values <- fitted
#>         if (method == "lasso") 
#>             class(fit) <- c("lassorqs", "rqs")
#>         else if (method == "scad") 
#>             class(fit) <- c("scadrqs", "rqs")
#>         else class(fit) <- "rqs"
#>     }
#>     else {
#>         process <- (tau < 0 || tau > 1)
#>         if (tau == 0) 
#>             tau <- eps
#>         if (tau == 1) 
#>             tau <- 1 - eps
#>         fit <- {
#>             if (length(weights)) 
#>                 rq.wfit(X, Y, tau = tau, weights, method, ...)
#>             else rq.fit(X, Y, tau = tau, method, ...)
#>         }
#>         if (process) 
#>             rho <- list(x = fit$sol[1, ], y = fit$sol[3, ])
#>         else {
#>             if (length(dim(fit$residuals))) 
#>                 dimnames(fit$residuals) <- list(dimnames(X)[[1]], 
#>                   NULL)
#>             rho <- sum(Rho(fit$residuals, tau))
#>         }
#>         if (method == "lasso") 
#>             class(fit) <- c("lassorq", "rq")
#>         else if (method == "scad") 
#>             class(fit) <- c("scadrq", "rq")
#>         else class(fit) <- ifelse(process, "rq.process", "rq")
#>     }
#>     fit$na.action <- attr(mf, "na.action")
#>     fit$formula <- formula
#>     fit$terms <- mt
#>     fit$xlevels <- .getXlevels(mt, mf)
#>     fit$call <- call
#>     fit$tau <- tau
#>     fit$weights <- weights
#>     fit$residuals <- drop(fit$residuals)
#>     fit$rho <- rho
#>     fit$method <- method
#>     fit$fitted.values <- drop(fit$fitted.values)
#>     attr(fit, "na.message") <- attr(m, "na.message")
#>     if (model) 
#>         fit$model <- mf
#>     fit
#> }
#> <bytecode: 0x7fcfd8023d58>
#> <environment: namespace:quantreg>

^{Created on 2019-02-12 by the reprex package (v0.2.1)}

[hadley]

Why we using match.fun() on strings? Can we just pass function objects instead?

[clauswilke]

That's how it's currently written, with the string coming in from the function arguments:

ggplot2/R/stat-quantile.r

Lines 13 to 22 in 43dcd63

	stat_quantile <- function(mapping = NULL, data = NULL,
	geom = "quantile", position = "identity",
	...,
	quantiles = c(0.25, 0.5, 0.75),
	formula = NULL,
	method = "rq",
	method.args = list(),
	na.rm = FALSE,
	show.legend = NA,
	inherit.aes = TRUE) {

Would something like this work instead of match.fun()?

if (identical(method, "rq")) method <- quantreg::rq
else if (identical(method, "rqss")) method <- quantreg::rqss
else stop("Unknown method")

[moodymudskipper]

Here's a reprex :

library(tidyverse)
monuments <- tribble(
  ~region, ~monument, ~LON, ~LAT,
  "Europe","Eiffel Tower", 2.29398, 48.8586,
  "Europe","Big Ben",  -0.124661, 51.5008,
  "Americas","Statue of Liberty", -74.04454, 40.68923,
  "Americas","Christ the Redeemer", -43.21118, -22.951871
)

world <- map_data("world")
#> 
#> Attachement du package : 'maps'
#> The following object is masked from 'package:purrr':
#> 
#>     map
ggplot(world,aes(long, lat,group=group)) +
  geom_polygon(fill="lightgrey",color="darkgrey") +
  coord_map("ortho",orientation = c(0,0,0)) + 
  theme_minimal() +
  geom_point(aes(LON,LAT,color= region, group = 1),data=monuments, size=6)

map(monuments, class)
#> Warning: Unknown or uninitialised column: 'maptype'.
#> Error in paste("(^", regions, ")", sep = "", collapse = "|"): cannot coerce type 'builtin' to vector of type 'character'

Created on 2019-02-13 by the reprex package (v0.2.0).

[clauswilke]

Looks like my PR fixes this.

library(tidyverse)
devtools::load_all("~/github/ggplot2")
#> Loading ggplot2
#> unloadNamespace("ggplot2") not successful, probably because another loaded package depends on it.Forcing unload. If you encounter problems, please restart R.
monuments <- tribble(
  ~region, ~monument, ~LON, ~LAT,
  "Europe","Eiffel Tower", 2.29398, 48.8586,
  "Europe","Big Ben",  -0.124661, 51.5008,
  "Americas","Statue of Liberty", -74.04454, 40.68923,
  "Americas","Christ the Redeemer", -43.21118, -22.951871
)

world <- map_data("world")
ggplot(world,aes(long, lat,group=group)) +
  geom_polygon(fill="lightgrey",color="darkgrey") +
  coord_map("ortho",orientation = c(0,0,0)) + 
  theme_minimal() +
  geom_point(aes(LON,LAT,color= region, group = 1),data=monuments, size=6)

map(monuments, class)
#> $region
#> [1] "character"
#> 
#> $monument
#> [1] "character"
#> 
#> $LON
#> [1] "numeric"
#> 
#> $LAT
#> [1] "numeric"

^{Created on 2019-02-12 by the reprex package (v0.2.1)}