since I read a lot similar question on stackoverflow so far, I couldn't find a good solution without updating ggplot2 to the development version.
My problem, I have several scripts which use arrangeGrob to create combined graph out of individual graphs. I save them into a variable and print this variable and/or save it with ggsave. Since a lot of my colleagues update there packages regularly (which is a good thing I think), I always get mails my script no longer work after updating to gridExtra 2.0.0.
I am not sure how to handle this, since the new ggplot2 version where the problem is solved is still under development. I found an article on stack overflow to remove a test if the object to save is a ggplot since the new arrangeGrob function returns a gtable object, but this fails in my case:
library(ggplot2)
library(grid)
library(gridExtra)
a <- data.frame(x=c(1,2,3),
y=c(2,3,4))
p <- ggplot(a, aes(x, y)) + geom_point()
b <- arrangeGrob(p, p)
grid.draw(b)
ggsave('test.pdf', b)
ggsave <- ggplot2::ggsave
body(ggsave) <- body(ggplot2::ggsave)[-2]
ggsave('test.pdf', b)
Some output and error on the console:
d> grid.draw(b)
d> ggsave('test.pdf', b)
Error in ggsave("test.pdf", b) : plot should be a ggplot2 plot
d> ggsave <- ggplot2::ggsave
d> body(ggsave) <- body(ggplot2::ggsave)[-2]
d> ggsave('test.pdf', b)
Saving 10.5 x 10.7 in image
TableGrob (2 x 1) "arrange": 2 grobs
z cells name grob
1 1 (1-1,1-1) arrange gtable[layout]
2 2 (2-2,1-1) arrange gtable[layout]
d>
The test.pdf is created but it is corrupted in any way and can not be opened. Also the gtable object get printed. So I guess something is wrong here.
But, as you can see, I found in the example code, I found the grid.draw function to plot at least my combined graph but I still can not ggsave it after the modification.
I don't want to use the "old" (pdf(file = "test.pdf"); grid.draw(b); dev.off()) device saving functions as suggested in this article, since they are very uncomfortable to use.
In this question someone asked exactly how to save the object, but in the answer they just explain to use grid.darw and he accepted the answer as solving the problem and nobody answered on my comments so far.
So I am pretty lost at the moment, how to provide working scripts for those who have and have not updated to new gridExtra package. The way to remove the test within the ggsave function is I guess the best solution since I can check the gridExtra and ggplot2 version and just overwrite the ggsave function in case there version do not match, but I could not get it to run.
Looking forward to get some help.
EDIT:
maybe the sessionInfo helps
d> sessionInfo()
R version 3.2.0 (2015-04-16)
Platform: x86_64-apple-darwin13.4.0 (64-bit)
Running under: OS X 10.9.5 (Mavericks)
locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
attached base packages:
[1] grid stats graphics grDevices utils datasets methods base
other attached packages:
[1] gridExtra_2.0.0 ggplot2_1.0.1
loaded via a namespace (and not attached):
[1] Rcpp_0.12.1 digest_0.6.8 MASS_7.3-44 plyr_1.8.3 gtable_0.1.2
[6] magrittr_1.5 scales_0.3.0 stringi_1.0-1 reshape2_1.4.1 devtools_1.9.1
[11] proto_0.3-10 tools_3.2.0 stringr_1.0.0 munsell_0.4.2 colorspace_1.2-6
[16] memoise_0.2.1
As an temporary workaround for this unfortunate transition period, you could re-implement the class hack that used to be in gridExtra,
class(b) <- c("arrange","ggplot", class(b))
print.arrange <- function(x) grid.draw(x)
ggsave('test.pdf', b)
Pascal brought me finally to the idea to check for the differences between ggplot 1.0.1 and ggplot 1.0.1.9003, since I don't want to or force the development version of ggplot.
So my idea is a function which will be executed within each script which overwrites the default ggsave function.
I tested it now a little, if there are any bugs or so, please let me know. But the way I do it now it works so far.
repairGgsave <- function() {
ggplot_version <-
compareVersion(as.character(packageVersion('ggplot2')),
'1.0.1.9003')
gridextra_version <-
compareVersion(as.character(packageVersion('gridExtra')),
'0.9.1')
if(gridextra_version > 0) {
if(ggplot_version <= 0) {
ggsave <- function(filename, plot = last_plot(),
device = NULL, path = NULL, scale = 1,
width = NA, height = NA, units = c("in", "cm", "mm"),
dpi = 300, limitsize = TRUE, ...) {
dev <- plot_dev(device, filename, dpi = dpi)
dim <- plot_dim(c(width, height), scale = scale, units = units,
limitsize = limitsize)
if (!is.null(path)) {
filename <- file.path(path, filename)
}
dev(file = filename, width = dim[1], height = dim[2], ...)
on.exit(utils::capture.output(grDevices::dev.off()))
grid.draw(plot)
invisible()
}
assign("ggsave", ggsave, .GlobalEnv)
plot_dim <<- function(dim = c(NA, NA), scale = 1, units = c("in", "cm", "mm"),
limitsize = TRUE) {
units <- match.arg(units)
to_inches <- function(x) x / c(`in` = 1, cm = 2.54, mm = 2.54 * 10)[units]
from_inches <- function(x) x * c(`in` = 1, cm = 2.54, mm = 2.54 * 10)[units]
dim <- to_inches(dim) * scale
if (any(is.na(dim))) {
if (length(grDevices::dev.list()) == 0) {
default_dim <- c(7, 7)
} else {
default_dim <- dev.size() * scale
}
dim[is.na(dim)] <- default_dim[is.na(dim)]
dim_f <- prettyNum(from_inches(dim), digits = 3)
message("Saving ", dim_f[1], " x ", dim_f[2], " ", units, " image")
}
if (limitsize && any(dim >= 50)) {
stop("Dimensions exceed 50 inches (height and width are specified in '",
units, "' not pixels). If you're sure you a plot that big, use ",
"`limitsize = FALSE`.", call. = FALSE)
}
dim
}
plot_dev <<- function(device, filename, dpi = 300) {
if (is.function(device))
return(device)
eps <- function(...) {
grDevices::postscript(..., onefile = FALSE, horizontal = FALSE,
paper = "special")
}
devices <- list(
eps = eps,
ps = eps,
tex = function(...) grDevices::pictex(...),
pdf = function(..., version = "1.4") grDevices::pdf(..., version = version),
svg = function(...) grDevices::svg(...),
emf = function(...) grDevices::win.metafile(...),
wmf = function(...) grDevices::win.metafile(...),
png = function(...) grDevices::png(..., res = dpi, units = "in"),
jpg = function(...) grDevices::jpeg(..., res = dpi, units = "in"),
jpeg = function(...) grDevices::jpeg(..., res = dpi, units = "in"),
bmp = function(...) grDevices::bmp(..., res = dpi, units = "in"),
tiff = function(...) grDevices::tiff(..., res = dpi, units = "in")
)
if (is.null(device)) {
device <- tolower(tools::file_ext(filename))
}
if (!is.character(device) || length(device) != 1) {
stop("`device` must be NULL, a string or a function.", call. = FALSE)
}
dev <<- devices[[device]]
if (is.null(dev)) {
stop("Unknown graphics device '", device, "'", call. = FALSE)
}
dev
}
}
}
}
It basically overwrites the ggsave and creates two new functions from the development version.
After executing the function everything seems to work.
Fix for me was explicitly defining the file:
ggsave(file='test.pdf', b)
Related
I have been using R and RStudio with base and ggplot for a couple of years. Lately, it seems that when some plots are generated, other plots are lost/deleted.
In RStudio, I use the "Plots" pane/tab, and the blue left and right arrows to view the plots that have been created. By disappears, I mean a plot was created and visible here, but if I use the arrows, I can't see it anymore. It was there, and it is not there anymore, it has "disappeared".
After running the code below using "Source" button, which does 3 plots, I can only see 2 plots. If I step through the code using Command-Return, I see the 3 plots generated, but then the middle one gets lost.
After doing more testing, I see for this code it is the call to:
DataExplorer::plot_str(mtcars)
that is the problem. If I comment that line and use another plot from DataExplorer like:
DataExplorer::plot_intro(mtcars)
all the plots are available after the script is complete.
I believe everything is using the latest versions:
mac os - 11.5.2
R - R version 4.1.1 (2021-08-10)
RStudio - RStudio 2021.09.0+351
R libraries - have just updated all the r libraries
Here is a simple base case that seems to recreate the issue. In the example below I am using the DataExplorer library, but I get similar stuff happening in Keras.
Code
########################### Start Initialisation ########################################################################################
# Remove objects from environment
rm(list = ls())
# Clear the R studio console
cat("\014")
# Clear all plots in R studio
try(dev.off(dev.list()["RStudioGD"]),silent<-TRUE)
try(dev.off(),silent<-TRUE)
# Load packages, installing first if not already installed
if (!require(DataExplorer)) {
install.packages("DataExplorer")
library(DataExplorer)
}
##################################
# Use base plotting to plot iris, this works
base::plot(iris, main="iris 1")
# Use DataExplorer::plot_str() to plot mtcars, this works, but then disappears
data("mtcars")
# DataExplorer::plot_str(mtcars) # This causes the problem
DataExplorer::plot_intro(mtcars) # This works
# This plot appears, but seems to cause the one above "mtcars" to disappear
base::plot(iris, main="iris 2")
sessionInfo()
Output:
> # Clear all plots in R studio
> try(dev.off(dev.list()["RStudioGD"]),silent<-TRUE)
null device
1
> try(dev.off(),silent<-TRUE)
> # Load packages, installing first if not already installed
> if (!require(DataExplorer)) {
+ install.packages("DataExplorer")
+ library(DataExpl .... [TRUNCATED]
> ##################################
>
> # Use base plotting to plot iris, this works
> base::plot(iris, main="iris 1")
> # Use DataExplorer::plot_str() to plot mtcars, this works, but then disappears
> data("mtcars")
> DataExplorer::plot_str(mtcars)
> # This plot appears but seems to cause the one above "mtcars" to disappear
> base::plot(iris, main="iris 2")
> sessionInfo()
R version 4.1.1 (2021-08-10)
Platform: x86_64-apple-darwin17.0 (64-bit)
Running under: macOS Big Sur 11.5.2
Matrix products: default
LAPACK: /Library/Frameworks/R.framework/Versions/4.1/Resources/lib/libRlapack.dylib
locale:
[1] en_AU.UTF-8/en_AU.UTF-8/en_AU.UTF-8/C/en_AU.UTF-8/en_AU.UTF-8
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] DataExplorer_0.8.2
loaded via a namespace (and not attached):
[1] pillar_1.6.3 compiler_4.1.1 tools_4.1.1 digest_0.6.28 jsonlite_1.7.2 evaluate_0.14 lifecycle_1.0.1 tibble_3.1.4 gtable_0.3.0 pkgconfig_2.0.3
[11] rlang_0.4.11 igraph_1.2.6 DBI_1.1.1 yaml_2.2.1 parallel_4.1.1 xfun_0.26 fastmap_1.1.0 gridExtra_2.3 dplyr_1.0.7 knitr_1.36
[21] generics_0.1.0 vctrs_0.3.8 htmlwidgets_1.5.4 grid_4.1.1 tidyselect_1.1.1 glue_1.4.2 data.table_1.14.2 R6_2.5.1 fansi_0.5.0 rmarkdown_2.11
[31] ggplot2_3.3.5 purrr_0.3.4 magrittr_2.0.1 scales_1.1.1 ellipsis_0.3.2 htmltools_0.5.2 networkD3_0.4 assertthat_0.2.1 colorspace_2.0-2 utf8_1.2.2
[41] munsell_0.5.0 crayon_1.4.1
I thought perhaps the issue was due to mixing different plotting libraries, but the code below which draws 6 plots, 3 using base:plot() and 3 using ggplot2::ggplot() works and generates 6 plots that I can navigate through in RStudio using the blue left and right arrow buttons on the "Plots" pane/tab.
base::plot(mpg, main="mpg 1 - base")
ggplot2::ggplot(mpg, aes(displ, hwy, colour = class)) +
geom_point() +
ggtitle("mpg 1 - ggplot2")
base::plot(ToothGrowth, main="ToothGrowth 2 - base")
ggplot2::ggplot(ToothGrowth, aes(x=supp, y=len, fill=dose))+
geom_boxplot() +
ggtitle("ToothGrowth 2 - ggplot2")
base::plot(iris, main="iris 3 - base")
ggplot2::ggplot(iris, aes(x=Sepal.Length, y=Sepal.Width, fill=Species))+
geom_boxplot() +
ggtitle("iris 3 - ggplot2")
RStudio is:
RStudio 2021.09.0+351 "Ghost Orchid" Release (077589bcad3467ae79f318afe8641a1899a51606, 2021-09-20) for macOS
Mozilla/5.0 (Macintosh; Intel Mac OS X 11_5_2) AppleWebKit/537.36 (KHTML, like Gecko) QtWebEngine/5.12.10 Chrome/69.0.3497.128 Safari/537.36
Mac os version:
(base) % uname -a
Darwin xxxx.lan 20.6.0 Darwin Kernel Version 20.6.0: Wed Jun 23 00:26:31 PDT 2021; root:xnu-7195.141.2~5/RELEASE_X86_64 x86_64
(base) % sw_vers
ProductName: macOS
ProductVersion: 11.5.2
BuildVersion: 20G95
I think the reason is a simple one: you're not outputting a real plot (static).
DataExplorer::plot_intro(mtcars) is the problematic "plot" and the underlying code for this includes diagonalNetwork towards the end of the function:
DataExplorer
function (data, type = c("diagonal", "radial"), max_level = NULL,
print_network = TRUE, ...)
{
i <- idx <- parent <- NULL
str_output_raw <- capture.output(str(data, vec.len = 0, give.attr = FALSE,
give.length = FALSE, list.len = 2000000000L))
str_output <- unlist(lapply(str_output_raw, function(x) {
gsub(" \\.{2}\\#", "\\$\\#", x)
}))
n <- length(str_output)
base_split <- tstrsplit(str_output[2:n], "\\$")
nest_level <- (nchar(base_split[[1]]) - nchar(gsub(" \\.{2}",
"", base_split[[1]])))/3 + 1
diff_nl <- diff(nest_level)
s4_start_index <- which(diff_nl > 1L) + 1L
if (length(s4_start_index) > 0) {
s4_end_index <- which(diff_nl == -2L)
s4_index_range <- unique(unlist(lapply(s4_start_index,
function(i) {
seq.int(i, s4_end_index[which.min(abs(s4_end_index -
i))])
})))
nest_level[s4_index_range] <- nest_level[s4_index_range] -
1L
}
if (is.null(max_level)) {
max_level <- max(nest_level)
}
else if (max_level <= 0 | max_level > max(nest_level)) {
stop(paste0("max_level should be between 1 and ",
max(nest_level)))
}
else {
max_level <- max_level
}
comp_split <- tstrsplit(base_split[[2]], ":")
comp_root <- gsub(" ", "", comp_split[[1]])
comp_root[which(comp_root == "")] <- make.names(comp_root[which(comp_root ==
"")], unique = TRUE)
if (anyDuplicated(comp_root))
comp_root[which(duplicated(comp_root))] <- make.names(comp_root[which(duplicated(comp_root))],
unique = TRUE)
comp_output <- paste0(comp_root, " (", trimws(gsub("NULL|\\.{3}|\\.{2}",
"", comp_split[[2]])), ")")
str_dt <- data.table(idx = seq_along(nest_level), nest_level,
parent = comp_output)[nest_level <= max_level]
str_dt <- str_dt[str_dt[, list(i = idx, nest_level = nest_level -
1, child = parent)], on = list(nest_level, idx < i),
mult = "last"]
drop_columns(str_dt[is.na(parent), `:=`(parent, paste0("root (",
str_output[1], ")"))], c("idx", "nest_level"))
str_to_list <- function(str_dt, root_name = as.character(str_dt[["parent"]][1])) {
str_list <- list(name = root_name)
children <- str_dt[parent == root_name][["child"]]
if (length(children) > 0) {
str_list[["children"]] <- lapply(children,
str_to_list, str_dt = str_dt)
}
str_list
}
str_list <- str_to_list(str_dt)
if (print_network) {
type <- match.arg(type)
if (type == "diagonal")
print(diagonalNetwork(str_list, ...))
if (type == "radial")
print(radialNetwork(str_list, ...))
}
invisible(str_list)
}
where diagonalNetwork uses htmlwidgets:
function (List, height = NULL, width = NULL, fontSize = 10, fontFamily = "serif",
linkColour = "#ccc", nodeColour = "#fff", nodeStroke = "steelblue",
textColour = "#111", opacity = 0.9, margin = NULL)
{
if (!is.list(List))
stop("List must be a list object.")
root <- List
margin <- margin_handler(margin)
options = list(height = height, width = width, fontSize = fontSize,
fontFamily = fontFamily, linkColour = linkColour, nodeColour = nodeColour,
nodeStroke = nodeStroke, textColour = textColour, margin = margin,
opacity = opacity)
htmlwidgets::createWidget(name = "diagonalNetwork",
x = list(root = root, options = options), width = width,
height = height, htmlwidgets::sizingPolicy(padding = 10,
browser.fill = TRUE), package = "networkD3")
}
htmlwidgets is an interactive "plot" and opens under RStudio's viewer pane and not the static plots pane.
I have set up a fresh R installation in a Windows 10 machine and can't run something as simple as:
data.frame(a = rnorm(100), b = rnorm(100)) |>
ggplot(aes(a, b)) +
ggsave("temp.png")
because I get the following error:
Error: Can't add `ggsave("temp.png")` to a ggplot object.
My session info is:
R version 4.1.0 (2021-05-18)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows >= 8 x64 (build 9200)
Matrix products: default
locale:
[1] LC_COLLATE=Catalan_Spain.1252 LC_CTYPE=Catalan_Spain.1252 LC_MONETARY=Catalan_Spain.1252 LC_NUMERIC=C
[5] LC_TIME=Catalan_Spain.1252
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] ggplot2_3.3.4 dplyr_1.0.6
loaded via a namespace (and not attached):
[1] magrittr_2.0.1 tidyselect_1.1.1 munsell_0.5.0 colorspace_2.0-1 R6_2.5.0 rlang_0.4.11 fansi_0.5.0 tools_4.1.0
[9] grid_4.1.0 data.table_1.14.0 gtable_0.3.0 utf8_1.2.1 withr_2.4.2 ellipsis_0.3.2 digest_0.6.27 tibble_3.1.2
[17] lifecycle_1.0.0 crayon_1.4.1 purrr_0.3.4 farver_2.1.0 vctrs_0.3.8 glue_1.4.2 labeling_0.4.2 compiler_4.1.0
[25] pillar_1.6.1 generics_0.1.0 scales_1.1.1 pkgconfig_2.0.3
I have given permissions to the directory I'm working on and also tried in different directories and run with RScript, RStudio and Pycharm R Console, always with the same issue.
Thanks in advance.
EDIT: this used to work on ggplot2 3.3.3, it's the update to 3.3.4 that breaks things.
You can no longer "add" ggsave to a ggplot addition-pipe.
Edit: this is a recent change in ggplot2-3.3.4. The previous answer is preserved below if you want to work around the new behavior. If you're particular annoyed by it, you might submit a new issue to ggplot2 suggesting that they either (a) undo the breaking change, or (b) better document the change in unintended functionality.
Side note: the day after this answer was posted, commit 389b864 included the following text: "Note that, as a side effect, an unofficial hack <ggplot object> + ggsave() no longer works (#4513)."
(In truth, I don't recall seeing documentation that suggests that + ggsave(.) should work, so the response to a new issue might be that they do not want to preserve an unintended "feature" for the sake of giving up some other elegant completeness.)
The changes from 3.3.3 to 3.3.4 (for save.R) are mostly unrelated to the act of saving the file. However, one functional change is the return value from ggsave:
## -90,5 +98,5 ## ggsave <- function(filename, plot = last_plot(),
grid.draw(plot)
- invisible()
+ invisible(filename)
}
In retrospect, this makes sense: ggplot2's ability to use +-pipes tends to be okay with trying to add NULL-like objects. That is, this works without error:
data.frame(a = rnorm(100), b = rnorm(100)) |>
ggplot(aes(a, b)) +
NULL
Why is NULL relevant here? Because the previous (3.3.3) version of ggsave ends with invisible(), which is invisibly returning NULL. (Internally, ggplot2:::add_ggplot begins with if (is.null(object)) return(p), which explains why that works.)
With the change to invisible(filename) (which, imo, is actually a little better), however, this is effectively the same as
data.frame(a = rnorm(100), b = rnorm(100)) |>
ggplot(aes(a, b)) +
"temp.png"
which does not make sense, so the +-piping fails.
In ggplot2-3.3.3, one can replicate this error with a hack/ugly code:
data.frame(a = rnorm(100), b = rnorm(100)) |>
ggplot(aes(a, b)) +
{ ggsave("temp.png"); "temp.png"; }
# Error: Can't add `{` to a ggplot object.
# * Can't add ` ggsave("temp.png")` to a ggplot object.
# * Can't add ` "temp.png"` to a ggplot object.
# * Can't add `}` to a ggplot object.
which is close enough to the error you saw to (I believe) prove my point: the new-and-improved ggplot2-3.3.4 is returning a string, and that is different enough to break your habit-pattern of adding ggsave to a ggplot2 object.
If you're going to submit a new issue to ggplot2, then I suggest you frame it as a "feature request": if the invisible(filename) is instead a class object for which + works, then the previous behavior can be retained while still supporting the string-return. For example (completely untested):
ggsave <- function(file, ...) {
# .....
class(filename) <- c("ggplot2_string", "character")
invisible(filename)
}
and then extend the +.gg-logic to actually work for strings, perhaps something like
`+.gg` <- function (e1, e2) {
if (missing(e2)) {
abort("Cannot use `+.gg()` with a single argument. Did you accidentally put + on a new line?")
}
if (inherits(e2, "ggplot2_string")) {
e2 <- NULL
e2name <- "NULL"
} else {
e2name <- deparse(substitute(e2))
}
if (is.theme(e1))
add_theme(e1, e2, e2name)
else if (is.ggplot(e1))
add_ggplot(e1, e2, e2name)
else if (is.ggproto(e1)) {
abort("Cannot add ggproto objects together. Did you forget to add this object to a ggplot object?")
}
}
No, I don't think this is the best way, but it is one way, open for discussion.
Four things you can do:
Plot it, then save it. It will be displayed in your graphic device/pane.
data.frame(a = rnorm(100), b = rnorm(100)) |>
ggplot(aes(a, b))
ggsave("temp.png")
Save to an intermediate object without rendering, and save that:
gg <- data.frame(a = rnorm(100), b = rnorm(100)) |>
ggplot(aes(a, b))
ggsave("temp.png", plot = gg)
If R-4.1, pipe it do the plot= argument. While I don't have R-4.1 yet, based on comments I am led to believe that while |> will always pass the previous result as the next call's first argument, you can work around this by naming the file= argument, which means that R-4.1 will pass to the first available argument which (in this case) happens to be plot=, what we need.
data.frame(a = rnorm(100), b = rnorm(100)) |>
ggplot(aes(a, b)) |>
ggsave(file = "temp.png")
If you're using magrittr pipes, then you can do the same thing a little more succinctly:
library(magrittr) # or dplyr, if you're using it for other things
data.frame(a = rnorm(100), b = rnorm(100)) %>% # or |> here
ggplot(aes(a, b)) %>% # but not |> here
ggsave("temp.png", plot = .)
Just remove the plus sign.
data.frame(a = rnorm(100), b = rnorm(100)) |>
ggplot(aes(a, b))
ggsave("temp.png")
ggsave has a default input last_plot()
You could always define your own ggsave() function that restores the old behavior.
library(tidyverse)
ggsave <- function(...) {
ggplot2::ggsave(...)
invisible()
}
data.frame(a = rnorm(100), b = rnorm(100)) |>
ggplot(aes(a, b)) +
ggsave("temp.png")
#> Saving 7 x 5 in image
Created on 2021-06-17 by the reprex package (v1.0.0)
Note: I do not think this is a good idea. Just pointing out that it's possible.
I don't know if my setup differs from others in some way, or if it's just the way my ggsave code happens to be configured, but it still works for me, while also now kicking up the error.
In this, Thomas says:
First, while you’ll get an error in v3.3.4, the plot is actually saved
to a file since the error is thrown after the evaluation of ggsave().
This means that you can “fix” your code by putting the whole
expression in a try() block (please don’t do this though 😬)
But you don't have to do the try block. My code blocks are:
ggplot(stuff) +
ggsave(paste0(today(), "_PlotName", ".png"),
plot = last_plot(),
device = "png", path = "", scale = 3.5, width = 8,
height = 4, units = "in", dpi = 300, limitsize = TRUE)
Presumably specifying plot = last_plot() keeps things working. But don't tell anyone, in case they take it away ;)
I know that the title of this question is a duplicate of this Question and this Question but the solutions over there don't work for me and the error message is (slightly) different:
Error in grid.Call(L_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
polygon edge not found
(note the missing part about the missing font)
I tried all suggestions that I found (updating / reinstalling all loaded graphic packages, ggplot2, GGally, and scales, reinitialising the Fonts on Mac OSX by starting in safe mode, moving the Fonts from /Fonts/ (Disabled) back into /Fonts...) but none of it resolved the problem.
The error seems to occure when I plot a ggplot graph with
scale_y_continuous(label=scientific_10)
where scientific_10 is defined as
scientific_10 <- function(x) {
parse(text = gsub("e", " %*% 10^", scientific_format()(x)))
}
Therefore the I suspect that the scales library has something to do with it.
The most puzzling is that the error only occurs each so-and-so many times, maybe each 3rd or 5th time i try to plot the same graph...
> sessionInfo()
R version 3.2.2 (2015-08-14)
Platform: x86_64-apple-darwin13.4.0 (64-bit)
Running under: OS X 10.9.5 (Mavericks)
locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] gridExtra_2.0.0 scales_0.3.0 broom_0.4.0 tidyr_0.3.1 ggplot2_1.0.1 GGally_0.5.0 dplyr_0.4.3
loaded via a namespace (and not attached):
[1] Rcpp_0.11.5 magrittr_1.5 MASS_7.3-43 mnormt_1.5-1 munsell_0.4.2 colorspace_1.2-6 lattice_0.20-33 R6_2.0.1
[9] stringr_0.6.2 plyr_1.8.1 tools_3.2.2 parallel_3.2.2 grid_3.2.2 gtable_0.1.2 nlme_3.1-121 psych_1.5.8
[17] DBI_0.3.1 htmltools_0.2.6 lazyeval_0.1.10 yaml_2.1.13 assertthat_0.1 digest_0.6.8 reshape2_1.4.1 rmarkdown_0.8.1
[25] labeling_0.3 reshape_0.8.5 proto_0.3-10
traceback()
35: grid.Call(L_textBounds, as.graphicsAnnot(x$label), x$x, x$y,
resolveHJust(x$just, x$hjust), resolveVJust(x$just, x$vjust),
x$rot, 0)
34: widthDetails.text(x)
33: widthDetails(x)
32: (function (x)
{
widthDetails(x)
})(list(label = expression(5 %*% 10^+5, 7.5 %*% 10^+5, 1 %*%
10^+6, 1.25 %*% 10^+6, 1.5 %*% 10^+6), x = 1, y = c(0.0777214770341215,
0.291044141334423, 0.504366805634725, 0.717689469935027, 0.931012134235329
), just = "centre", hjust = 1, vjust = 0.5, rot = 0, check.overlap = FALSE,
name = "axis.text.y.text.8056", gp = list(fontsize = 9.6,
col = "black", fontfamily = "", lineheight = 0.9, font = 1L),
vp = NULL))
31: grid.Call.graphics(L_setviewport, vp, TRUE)
30: push.vp.viewport(X[[i]], ...)
I solved it by installing the library extrafont, installing a set of specific fonts and forcing ggplot to use only these fonts:
require(extrafont)
# need only do this once!
font_import(pattern="[A/a]rial", prompt=FALSE)
require(ggplot2)
# extending the help file example
df <- data.frame(gp = factor(rep(letters[1:3], each = 10)), y = rnorm(30))
ds <- plyr::ddply(df, "gp", plyr::summarise, mean = mean(y), sd = sd(y))
plotobj <- ggplot(df, aes(gp, y)) +
geom_point() +
geom_point(data = ds, aes(y = mean), colour = 'red', size = 3) +
theme(text=element_text(size=16, family="Arial"))
print(plotobj)
I experienced the same issue when trying to plot ggplot/grid output to the graph window in Rstudio. However, plotting to an external graphing device seems to work fine.
The external device of choice depends on your system, but the script below, paraphrased from this blog, works for most systems:
a = switch(tolower(Sys.info()["sysname"]),
"darwin" = "quartz",
"linux" = "x11",
"windows" = "windows")
options("device" = a)
graphics.off()
rm(a)
and to switch back to using the Rstudio plot window:
options("device"="RStudioGD")
graphics.off()
Note that by switching, you lose any existing plots.
A lot of solutions for this particular error direct you to look under the hood of your computer but this error can also be caused by a scripting error in which R expects to match elements from two data structures but cannot.
For me the error was caused by calling a fairly complex graphing function (see below) that read an ordered character vector as well as a matrix whose row names were supposed to each match a value in the ordered character vector. The problem was that some of my values contained dashes in them and R's read.table() function translated those dashes to periods (Ex: "HLA-DOA" became "HLA.DOA").
I was using the ComplexHeatmap package with a call like this:
oncoPrint(mat,
get_type = function(x) strsplit(x, ";")[[1]],
alter_fun_list = alter_fun_list,
col = col,
row_order = my_order,
column_title = "OncoPrint",
heatmap_legend_param = list(title = "Alternations", at = c("AMP", "HOMDEL", "MUT"), labels = c("Amplification", "Deep deletion", "Mutation"))
)
In this call:
mat was a matrix that had dashes swapped out for periods
my_order was a character vector containing the same values as the row names of matexcept the dashes remained
every other argument is essential to the call but irrelevant to this post
To help R find this elusive "polygon edge", I just edited my character vector with:
row_order <- gsub("\\.", "-", row_order)
If you've tried re-installing packages, restarting your computer and re-enabling fonts - maybe check and see if you've got some faulty character matching going on in your call.
i tried to set the font of aes,returned the error info
the added words:
p <- p + theme(text = element_text(family = "宋体"))
when i tried to remove the setting,it's ok then.
Actually, I have the same problem on my MAC and couldn't solve it on a regular base... Since it also happens like every 5th or 10th execution I decided to wrap the whole ggplot command into a trycatch call and execute it until it doesn't fail...
The code would looks like this
error_appeared <- FALSE
repeat{
tryCatch({ # we put everything into a try catch block, because sometimes we get an error
gscat <-
ggplot() # my ggplot command which sometimes fail
ggsave('file.pdf', gscat, width=8,height=8)
plot(gscat)
},
error=function(e) {
print('redo the ratioscatterplot.')
error_appeared <- TRUE
}
)
if(!error_appeared){
break
}
}
Actually I figured out, only the drawing/plotting of the figure gives problems! Saving always works.
Maybe this is helping someone, since I couldn't find a solution which actually solves the whole thing!
Additional:
If somebody wants to play with the problem on a "reproducible example" the code below throws an average of 2 errors out of 20 within the loop.
library(scales)
library(ggplot2)
df <- data.frame(
log2.Ratio.H.L.normalized.rev = c(2.53861265542646, 0.402176424979483, 0.438931541934545, 0.639695233399582, 0.230203013366421,
2.88223218956399, 1.23051046036618, 2.56554843533357, 0.265436896049098,
1.32866415755805, -0.92108963514092, 0.0976107966264223, -0.43048946484291,
-0.558665259531966, 4.13183638727079, 0.904580434921318, -0.0733780789564803,
-0.621932351219966, 1.48594198341242, -0.365611185917855, 1.21088754922081,
-2.3717583289898, 2.95160644380282, 3.71446534016249),
Intensity = c(5951600000, 2.4433e+10, 1.1659e+10, 2273600000, 6.852e+10, 9.8746e+10, 5701600000,
1758500000, 987180000, 3.4167e+11, 1.5718e+10, 6.8888e+10, 5.5936e+10,
8702900000, 1093500000, 4426200000, 1.3681e+11, 7.773e+09, 5860400000,
1.2861e+12, 2017900000, 2061300000, 240520000, 1382700000),
my_label = c("RPL18",
"hCG_2024613", "NOL7", "PRPF4B", "HIST1H2BC", "XRCC1", "C9orf30",
"CABIN1", "MGC3731", "XRCC6", "RPL23", "RPL27", "RPL17", "RPL32",
"XPC", "RPL15", "GNL3", "RPL29", "JOSD3", "PARP1", "DNAPTP6",
"ORC2L", "NCL", "TARDBP"))
unlink("figures", recursive=TRUE)
if(!dir.exists('figures')) dir.create('figures')
for(i in 1:20) {
error_appeared <- FALSE
repeat{
tryCatch({ # we put everything into a try catch block, because sometimes we get an error
gscat <-
ggplot(df, aes_string("log2.Ratio.H.L.normalized.rev", 'Intensity')) +
geom_point(data=df[abs(df[["log2.Ratio.H.L.normalized.rev"]]) < 1,],
color='black', alpha=.3, na.rm=TRUE) +
scale_y_log10(labels = scales::trans_format("log10", scales::math_format()))
ggsave(file.path('figures', paste0('intensity_scatter_', i, '.pdf')),
gscat, width=8, height=8)
plot(gscat)
},
error=function(e) {
# print(e)
print(sprintf('%s redo the ratioscatterplot.', i))
error_appeared <- TRUE
}
)
if(!error_appeared){
break
}
}
}
I am trying to generate plots by looping, here is my code:
n <- unique(wide_data$Product.Code)[1:3]
for (i in n)
{
my.prod2 <- filter(tall_bind, Product.Code == i, Date > ymd("2012/04/01"))
dev.new()
mypath <- file.path("C:","R","SAVEHERE",paste("myplot_", i, ".jpg", sep = ""))
jpeg(file=mypath)
mytitle = paste("Plot for product", i)
p <- qplot(Date, Sold, data = my.prod2, geom = "line", main=mytitle, group = Model, colour = Model) + facet_grid(Model ~ .)
ggsave("myplot_", i, plot=p, device= "jpg" )
}
I get the following error for the above code:
Saving 6.67 x 6.67 in image
Error in ggsave("myplot_", i, plot = p, device = "jpg") : could
not find function "device"
Earlier when I used dev.off() at the end of the loop, I found that though the graphs were being generated they were totally blank.
Could someone please help me understand where is the mistake in my code?
You can leave out the dev.new() and jpg() commands, and also your arguments to ggsave() are incorrect. This should work:
n <- unique(wide_data$Product.Code)[1:3]
for (i in n) {
my.prod2 <- filter(tall_bind, Product.Code == i, Date > ymd("2012/04/01"))
mypath <- file.path("C:","R","SAVEHERE",paste("myplot_", i, ".jpg", sep = ""))
mytitle = paste("Plot for product", i)
p <- qplot(Date, Sold, data = my.prod2, geom = "line", main=mytitle, group = Model, colour = Model) + facet_grid(Model ~ .)
ggsave(filename = mypath, plot = p)
}
What you did was creating a new default graphics device, typically a plotting window, then a jpeg graphics device, i.e. a file. Then you tried to make ggplot2 to plot to directly to file using ggsave, i.e. using its own (jpg) device, and not using either of the two graphics devices you created.
The error, however, was because you gave ggsave the wrong arguments. But even with the right arguments, you would still have ended up with additional unused graphics windows and files through the dev.new() and jpeg() commands. I suggest some extra reading of the help (e.g. type ?ggsave at the r console).
Typically, when using ggplot2 you do not need to worry about dev.new, jpeg and the like. qplot or ggplot and ggsave should do all you need.
I am trying to plot a graph with price and a few technical indicators such as ADX, RSI, and OBV. I cannot figure out why addOBV is giving an error and why addADX not showing at all in the graph lines in the chart?
Here my code:
tmp <- read.csv(paste("ProcessedQuotes/",Nifty[x,],".csv", sep=""),
as.is=TRUE, header=TRUE, row.names=NULL)
tmp$Date<-as.Date(tmp$Date)
ydat = xts(tmp[,-1],tmp$Date)
lineChart(ydat, TA=NULL, name=paste(Nifty[x,]," Technical Graph"))
plot(addSMA(10))
plot(addEMA(10))
plot(addRSI())
plot(addADX())
plot(addOBV())
Error for addOBV is:
Error in try.xts(c(2038282, 1181844, -1114409, 1387404, 3522045, 4951254, :
Error in as.xts.double(x, ..., .RECLASS = TRUE) :
order.by must be either 'names()' or otherwise specified
Below you can see DIn is not shown fully in the graphs.
> class(ydat)
[1] "xts" "zoo"
> head(ydat)
Open High Low Close Volume Trades Sma20 Sma50 DIp DIn DX ADX aroonUp aroonDn oscillator macd signal RSI14
I don't know why that patch doesn't work for you, but you can just create a new function (or you could mask the one from quantmod). Let's just make a new, patched version called addOBV2 which is the code for addOBV except for the one patched line. (x <- as.matrix(lchob#xdata) is replaced with x <- try.xts(lchob#xdata, error=FALSE)).
addOBV2 <- function (..., on = NA, legend = "auto")
{
stopifnot("package:TTR" %in% search() || require("TTR", quietly = TRUE))
lchob <- quantmod:::get.current.chob()
x <- try.xts(lchob#xdata, error=FALSE)
#x <- as.matrix(lchob#xdata)
x <- OBV(price = Cl(x), volume = Vo(x))
yrange <- NULL
chobTA <- new("chobTA")
if (NCOL(x) == 1) {
chobTA#TA.values <- x[lchob#xsubset]
}
else chobTA#TA.values <- x[lchob#xsubset, ]
chobTA#name <- "chartTA"
if (any(is.na(on))) {
chobTA#new <- TRUE
}
else {
chobTA#new <- FALSE
chobTA#on <- on
}
chobTA#call <- match.call()
legend.name <- gsub("^.*[(]", " On Balance Volume (", deparse(match.call()))#,
#extended = TRUE)
gpars <- c(list(...), list(col=4))[unique(names(c(list(col=4), list(...))))]
chobTA#params <- list(xrange = lchob#xrange, yrange = yrange,
colors = lchob#colors, color.vol = lchob#color.vol, multi.col = lchob#multi.col,
spacing = lchob#spacing, width = lchob#width, bp = lchob#bp,
x.labels = lchob#x.labels, time.scale = lchob#time.scale,
isLogical = is.logical(x), legend = legend, legend.name = legend.name,
pars = list(gpars))
if (is.null(sys.call(-1))) {
TA <- lchob#passed.args$TA
lchob#passed.args$TA <- c(TA, chobTA)
lchob#windows <- lchob#windows + ifelse(chobTA#new, 1,
0)
chartSeries.chob <- quantmod:::chartSeries.chob
do.call("chartSeries.chob", list(lchob))
invisible(chobTA)
}
else {
return(chobTA)
}
}
Now it works.
# reproduce your data
ydat <- getSymbols("ZEEL.NS", src="yahoo", from="2012-09-11",
to="2013-01-18", auto.assign=FALSE)
lineChart(ydat, TA=NULL, name=paste("ZEEL Technical Graph"))
plot(addSMA(10))
plot(addEMA(10))
plot(addRSI())
plot(addADX())
plot(addOBV2())
This code reproduces the error:
library(quantmod)
getSymbols("AAPL")
lineChart(AAPL, 'last 6 months')
addOBV()
Session Info:
sessionInfo()
R version 2.15.0 (2012-03-30)
Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit)
locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] quantmod_0.3-17 TTR_0.21-1 xts_0.9-1 zoo_1.7-9 Defaults_1.1-1 rgeos_0.2-11
[7] sp_1.0-5 sos_1.3-5 brew_1.0-6
loaded via a namespace (and not attached):
[1] grid_2.15.0 lattice_0.20-6 tools_2.15.0
Googling around, the error seems to be related to the fact that addOBV converts the data into a matrix, which causes problems with TTR::OBV. A patch has been posted on RForge.