IMO no issues are important enough to not release...

I've also added a small PR #21 to fix the one error, but after that (and installing base64enc from the git source... I do love the change in the compiler on win...) it's all good:

c:\data\external\R
λ build_rpackage.bat --Rdevel --check irdisplay
# Building packages: irdisplay
# Checking: yes
# Building for R-devel
c:\Program Files\R\R-devel
 - Building package irdisplay
* checking for file 'irdisplay/DESCRIPTION' ... OK
* preparing 'IRdisplay':
* checking DESCRIPTION meta-information ... OK
* checking for LF line-endings in source and make files
* checking for empty or unneeded directories
* building 'IRdisplay_0.3.0.9000.tar.gz'

# CHECKING the package
* using log directory 'c:/data/external/R/IRdisplay.Rcheck'
* using R Under development (unstable) (2016-04-25 r70548)
* using platform: x86_64-w64-mingw32 (64-bit)
* using session charset: ISO8859-1
* checking for file 'IRdisplay/DESCRIPTION' ... OK
* this is package 'IRdisplay' version '0.3.0.9000'
* checking package namespace information ... OK
* checking package dependencies ... OK
* checking if this is a source package ... OK
* checking if there is a namespace ... OK
* checking for executable files ... OK
* checking for hidden files and directories ... OK
* checking for portable file names ... OK
* checking whether package 'IRdisplay' can be installed ... OK
* checking installed package size ... OK
* checking package directory ... OK
* checking DESCRIPTION meta-information ... OK
* checking top-level files ... OK
* checking for left-over files ... OK
* checking index information ... OK
* checking package subdirectories ... OK
* checking R files for non-ASCII characters ... OK
* checking R files for syntax errors ... OK
* checking whether the package can be loaded ... OK
* checking whether the package can be loaded with stated dependencies ... OK
* checking whether the package can be unloaded cleanly ... OK
* checking whether the namespace can be loaded with stated dependencies ... OK
* checking whether the namespace can be unloaded cleanly ... OK
* checking loading without being on the library search path ... OK
* checking dependencies in R code ... OK
* checking S3 generic/method consistency ... OK
* checking replacement functions ... OK
* checking foreign function calls ... OK
* checking R code for possible problems ... OK
* checking Rd files ... OK
* checking Rd metadata ... OK
* checking Rd cross-references ... OK
* checking for missing documentation entries ... OK
* checking for code/documentation mismatches ... OK
* checking Rd \usage sections ... OK
* checking Rd contents ... OK
* checking for unstated dependencies in examples ... OK
* checking examples ... NONE
* checking PDF version of manual ... OK
* DONE

Status: OK

* installing to library 'C:/portabel/R/R/win-library/3.4'
* installing *source* package 'IRdisplay' ...
** R
** preparing package for lazy loading
** help
*** installing help indices
** building package indices
** testing if installed package can be loaded
*** arch - i386
*** arch - x64
* MD5 sums
packaged installation of 'IRdisplay' as IRdisplay_0.3.0.9000.zip
* DONE (IRdisplay)
c:\data\external\R

c:\data\external\R
λ

This is similar to this recent question in this stackoverflow: https://stackoverflow.com/questions/50708887/display-html-report-in-jupyter-with-r

in my case, though, I'm trying to display a png file, and this is the error that I'm getting:
`Error in read(file, size): unused argument (size)
Traceback:

1. IRdisplay::display_png(file = "/home/local/ARCS/pw2470/Pictures/Screenshot_from_2018-05-24_14-26-03.png")
2. display_raw("image/png", TRUE, data, file, img_metadata(width,
   . height))
3. prepare_content(isbinary, data, file)
4. read_all(file, isbinary)
   `

The weird thing is that this used to work prior to me upgrading to R v.3.5 and re-installing/upgrading everything. is it a compatibility issue with R v.3.5.0 or is there a recent change that broke everything?

When I run the following code on Jupyter Notebook, the display size changes in accordance with the width option.

library(ggplot2)
g <- qplot(1:5, 1:5)
ggsave("foo.jpeg", g, width=2, height=1)
IRdisplay::display_jpeg(file="foo.jpeg", width=100)
IRdisplay::display_jpeg(file="foo.jpeg", width=300)

But when I run the same code on Jupyter Lab, the display size does not change.
Same behavior for display_png.
Is this a bug or is there any fix to this?

I am using IRkernel_1.0.2 and IRdisplay_0.7.0, and Python3.6 with jupyter 1.0.0 and jupytarlab 2.2.6 on a xubuntu 18.04 laptop.

On jupyter notebook:

On jupyter lab:

Hi, I am using IRdisplay in chrome on a jupyter notebook with conda, and it does not show any output or result in any error.
The lm* are linear regression models combined by stargazer. Stargazer returns a proper html file without any errors.

m <- capture.output(stargazer(lm_enroll,lm_grades,lm_score,
         digits = 3,
          header = FALSE,
          type = "html", 
          se = rob_se,
          title = "Linear Panel Regression Model",
          model.numbers = FALSE)

IRdisplay::display_html(paste("<center>","",m,"","</center>"))

I am unable to install IRdisplay, as it redirects (wrongly) to hadley.

> install_github('irkernel/IRdisplay')
Installing github repo(s) irkernel/IRdisplay/master from hadley
Downloading irkernel/IRdisplay.zip from https://github.com/hadley/irkernel/IRdisplay/archive/master.zip
Error in (function (url, name = NULL, subdir = NULL, config = list(),  :
  client error: (404) Not Found

I am get a segfault error when trying to call display_html from a python environment using rpy2 package. My R notebook is calling a simple a simple python script file that imports rpy2 then imports IRdisplay and then invokes display_html. This is obviously a proof of error script as my use case is calling display_html to render HTML output from a custom python application that I am trying to enable in R. Thanks for any help.

Given:

• Create test_IRdisplay.py file with content:

from rpy2.robjects.packages import importr rdisplay = importr('IRdisplay') rdisplay.display_html('<h1>Hi</h1>')

• R Packages installed are:
  IRkernel 1.2
  reticulate 1.22
  IRdisplay 1.0

• Python Packages installed are:
  rpy2 = "3.4.5"
  notebook = "6.4.3"

Steps to Reproduce:

1. Upload test_IRdisplay.py file to Jupyter
2. Open new R notebook in Jupyter
3. Type py_run_file("test_IRdisplay.py")
4. Execute cell repeatedly until Dead Kernel error (for me the error always occurs on the 8th run - 7 successful runs, 8th is an error)

Logs show:
*** caught segfault *** address 0xb8, cause 'memory not mapped' An irrecoverable exception occurred. R is aborting now ...

I do see this warning after the firs time the cell is executed from step 4:
/usr/local/lib/python3.8/site-packages/rpy2/rinterface_lib/embedded.py:134: UserWarning: R was initialized outside of rpy2 (R_NilValue != NULL). Trying to use it nevertheless. warnings.warn(

This option belongs to the kernel and is only kept here to be compatible with an older IRkernel, which expects it set. Remove it after both projects have a release.

[edit]
Currently, without a change in the frontend, the best we can do is simple send the complete metadata to the frontend in the normal display data: If we manage the styling, we send the styling only once on the first exec of such code. But then the frontend (=Notebook) could reevaluate the cell which currently has the styling and we would no notice this and no send styling, and the notebook would end up without styling information.

See also jupyter/notebook#116 and jupyter/roadmap#9
[/]
It would be nice to be able to specify that something only needs to be displayed once, e.g. the styling for a table or an interactive plot.

The repr side probably only needs to document the way this should happen, the actually implementation needs to be in the display package.

From the old issue comment by @flying-sheep:

yeah, the point is that there needn’t be a new interface: we can simply add attributes to the character()/raw() returned, which avoids changing signatures.

an example where i loop a bunch of things, create their HTML reprs and only do setup once:

repr_html.stuff <- function(x, ...) {
    ...
    structure(html_markup, setup = list(stuff_setup_script = '<script>...</script>')
}

html_setups_seen <- character()
for (thing in things) {
    thing_repr <- repr_html(thing)
    thing_setups <- attr(thing_repr, 'setup')
    new_setups <- stuff_setups[intersect(html_setups_seen, names(thing_setups))]
    html_setups_seen <- c(html_setups_seen, names(new_setups))

    display_html_setups(new_setups)
    display_html(stuff_repr)
}

PS: for the implementation, we should have a list of seen setups, keyed by format and class. this way there will be less potential for name collisions. or we just us the passed code itself as key.

hi @takluyver, i was asked during the submission process by kurt how this and that fits together.

probably you should be “cre”?

see also IRkernel/IRkernel#322

#devtools::install_github("renkun-ken/formattable")
library(formattable)

df <- data.frame(
  id = 1:10,
  name = c("Bob", "Ashley", "James", "David", "Jenny", 
    "Hans", "Leo", "John", "Emily", "Lee"), 
  age = c(28, 27, 30, 28, 29, 29, 27, 27, 31, 30),
  grade = c("C", "A", "A", "C", "B", "B", "B", "A", "C", "C"),
  test1_score = c(8.9, 9.5, 9.6, 8.9, 9.1, 9.3, 9.3, 9.9, 8.5, 8.6),
  test2_score = c(9.1, 9.1, 9.2, 9.1, 8.9, 8.5, 9.2, 9.3, 9.1, 8.8),
  final_score = c(9, 9.3, 9.4, 9, 9, 8.9, 9.25, 9.6, 8.8, 8.7),
  registered = c(TRUE, FALSE, TRUE, FALSE, TRUE, TRUE, TRUE, FALSE, FALSE, FALSE),
  stringsAsFactors = FALSE)

f <- formattable(df, list(
  age = color_tile("white", "orange"),
  grade = formatter("span",
    style = x ~ ifelse(x == "A", style(color = "green", font.weight = "bold"), NA)),
  test1_score = normalize_bar("pink", 0.2),
  test2_score = normalize_bar("pink", 0.2),
  final_score = formatter("span",
    style = x ~ style(color = ifelse(rank(-x) <= 3, "green", "gray")),
    x ~ sprintf("%.2f (rank: %02d)", x, rank(-x))),
  registered = formatter("span", 
    style = x ~ style(color = ifelse(x, "green", "red")),
    x ~ icontext(ifelse(x, "ok", "remove"), ifelse(x, "Yes", "No")))
))

f # does not display anything
class(f) # displays "formattable" "data.frame"
print(f) # nothing

I think this happens when print(f) does not print anything and we don't have anything to make a rich display...

In rstudio, the f already triggers something, so we really should implement the same way rstudio handles this...

x = print(f)

this also triggers a display in rstudio and sets x:

> x = print(f)
> class(x)
[1] "formattable_widget" "htmlwidget" 

So it seems we should implement a repr_html.htmlwidget (and maybe repr_html.formattable? I haven't fully understand the knit_print.*() hierarchy/call chain in formattable yet) and also check that we have the same display mechanism as rstudio (do they have a way to deliver plots to the plot area?).

Regarding things we have no rich printing for: I've no clue how to reliably tell if we miss a repr_xxx.class_type (and tell the user that they should bug us or the package owner for such a method) or that the package author didn't want a visible output. If someone has a clue...

CC: @renkun-ken (owner of the formattable package)

From: https://stackoverflow.com/questions/35361751/embed-html-in-jupyter-with-r-kernel

This should work out of the box:

library(IRdisplay)
library(rCharts)
n1 <- nPlot(Freq ~ V2, group = "V1", data = x, type = "multiBarChart")
n1$save("NAME.html", standalone = TRUE)
display_html(file =  "NAME.html") # must be isolated...

display_html should do the same stuff as the kernel does: IRkernel/IRkernel@f2a153d#diff-6eae7f9cccd978b5b4d819baa7a34b1eR212

Can you tag a release? See also IRkernel/IRkernel#60.

At least on Jupyter notebooks with irkernel 1.1 and irdisplay 0.7, printing a data.frame generated a table with the rows and columns listed in the upper left of the table (eg., A data.frame 14 x 60). For example:

However, in Jupyter lab 3.0.3, the row & column size information is centered in the top-middle of the table. For example:

...so for tables with many rows, the user must now constantly scroll over just to see the data.frame size. Is there a way to move the row & column size information back to the top left of the table?

currently only the options are tested, but there should be tests for each display_*. Probably easiest wit two helper functions:

last_data <- NULL
display_func <- function(data, metadata=NULL){
   last_data <<- list(data, metadata)
}
get_last_data <- function(){
    ret <- last_data
    last_data <<- NULL
    ret
}
options(jupyter.base_display_func=display_func)

# tests ala
display_html(...)
assert_equal(get_last_data(), list(...))

Background

This is when I want to use a package called "table1", which outputs an HTML file with its own CSS-style and then automatically activate a page in the browser to show the output.

When I use the traditional console

All well when using the traditional console

When I use the JupyterLab without IRdisplay

It outputs some codes instead of a visualized table.

When I use the IRdisplay

It shows the table correctly but with the wrong style.

My problem

Umm... I want to know if there is any method for solving this issue? Because the style of the table is very important for an article that without it, I have to draw it by myself in the WORD.
Or is there any means for me to let the package to activate this own page like that it does in the traditional console?
Thank you very much~~~

https://twitter.com/hadleywickham/status/712277632428015616

it looks like you generated the plots with the default 72 dpi, for complicated reasons res = 96 looks much better for ggplot2

Hello, I wanted to remove autoscrolling in jupiterlab and while I get hundreds of hits in python it seems very complicated to find out how to do it in R.
From this post on SO it seems that I should be able to do
IRdisplay::display_javascript('IPython.OutputArea.auto_scroll_threshold = 99999;') however when I do this in jupyterlab I get the error :
Javascript Error: IPython is not defined
Note that this seems to be working in jupyter-notebook
Sorry for the trivial question but I am new to jupyter in R

In IPython I call from IPython.display import clear_output to clear a cell output. How do I achieve the same with IRKernel or IRDisplay. I’m using reticulate to import a python module and call a user defined method which sends Brno back to the client updating its progress. Without clearing the output on each progress change, the results are displayed in succession.

Following the example in the manual:

> display_jpeg(file = url('http://example.com/example.jpg'), width = 100)

I did:

> display_jpeg(file=url('https://upload.wikimedia.org/wikipedia/en/a/a9/Example.jpg'))
Error in file.info(file): invalid filename argument
Traceback:

1. display_jpeg(file = url("https://upload.wikimedia.org/wikipedia/en/a/a9/Example.jpg"))
2. display_raw("image/jpeg", TRUE, data, file, img_metadata(width, 
 .     height))
3. prepare_content(isbinary, data, file)
4. file.info(file)

Just url:

> url('https://upload.wikimedia.org/wikipedia/en/a/a9/Example.jpg')
A connection with                                                                        
description "https://upload.wikimedia.org/wikipedia/en/a/a9/Example.jpg"
class       "url-libcurl"                                               
mode        "r"                                                         
text        "text"                                                      
opened      "closed"                                                    
can read    "yes"                                                       
can write   "no"                                                        

Verison info:

> sessionInfo()
R version 3.4.1 (2017-06-30)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 16.04.4 LTS

Matrix products: default
BLAS: /opt/conda/lib/R/lib/libRblas.so
LAPACK: /opt/conda/lib/R/lib/libRlapack.so

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] IRdisplay_0.4.4 bit64_0.9-5     bit_1.1-12     httr_1.3.1     

loaded via a namespace (and not attached):
 [1] Rcpp_0.12.15      digest_0.6.15     crayon_1.3.4      R6_2.2.2         
 [5] repr_0.13         jsonlite_1.5      magrittr_1.5      evaluate_0.10.1  
 [9] stringi_1.1.7     curl_3.2          uuid_0.1-2        data.table_1.11.0
[13] IRkernel_0.8.11   tools_3.4.1       stringr_1.3.0     compiler_3.4.1   
[17] base64enc_0.1-3   htmltools_0.3.6   pbdZMQ_0.3-2      openssl_1.0.1    

Hi, I just opened an issue on IRkernel, IRkernel/IRkernel#392 but I realized that perhaps it is related to IRdisplay and not the kernel itself. should I close the issue and move it here instead?

#5 totally derailed and IRkernel/IRkernel#97 makes this urgent so let me get this done:

We want the following functions:

1. high level display creating a mimebundle by calling all repr_* functions on an object, then calling base_display on that mimebundle
2. low level display_mimebundle and display1 functions directly accepting mimebundle-like lists
3. high level display_* convenience functions for a single object or file.

more questions:

1. should we call base_display only with base64 encoded text strings or also with raw vectors and rely on simplejson automatically base64’ing those?

2. i like the ... API instead of a dedicated list: which should we do:

   • display_mimebundle(list('application/png' = bar,
                             'image/png' = foo),
                        metadata = list(
                             'image/png' = list(width = 400)))

   • display_mimebundle('application/png' = bar,
                        'image/png' = foo, 
                        metadata = list(
                              'image/png' = list(width = 400)))

   • display_mimebundle('application/png' = bar,
                        'image/png' = list(foo, width = 400))

   the latter of course would also accept a dedicated metadata parameter.

change in IRkernel: jupyter.result_mimetypes → jupyter.display_mimetypes

…which we then use here in the new display function to determine which reprs to call

since IRkernel sets the option (if not also set) and this project doesn’t depend on IRkernel: should this project also set a default for that option?

I am trying to display pdf in the Google Colaboratory environment.

When I ran:

library(IRdisplay)
display_pdf(file='test.pdf')

The R notebook returned nothing.

I had correct pdf format file in the file system, and all other functions display_png, display_jpeg, display_svg worked for me.

I'm wondering if there's a recognised way of displaying an audio media player as part of cell output in a Jupyter / IRkernel notebook?

What I'd like to be able to do is something like:

display('myFile.wav')
display('myMidi.mid')
display(myAudioObject)

There doesn't seem to be a display function for it - is the best route to try to wrap something as an HTMLwidget and then render that?

Hello!

I am trying to execute the following lines of codes (from here: https://plot.ly/ggplot2/), but then it yields an error message, saying: "HTML widgets cannot be represented in plain text (need html)". Why is this and how can this be fixed?

require(ggplot2); require(plotly)

set.seed(100)
d <- diamonds[sample(nrow(diamonds), 1000), ]

p <- ggplot(data = d, aes(x = carat, y = price)) +
geom_point(aes(text = paste("Clarity:", clarity)), size = 4) +
geom_smooth(aes(colour = cut, fill = cut)) + facet_wrap(~ cut)

(gg <- ggplotly(p))

IMO this should also get a test that this function is used by the display functions, so display(something) should also raise that warning.

-> #32 (comment)

This is my code in R

trend1 <- gtrends (c(paste0( "House for sale in London", "House for sale in Birmingham")), geo = "GB", time = "2004-01-01 2016-12-30")

And the error is

Error: widget$status_code == 200 is not TRUE

It doesn't seem to be a common error. I don't know what happens to it because I don't think the code is wrong though. Please help!

This StackOverflow question: https://stackoverflow.com/q/72698223 illustrates what looks like a bug in the display of graphics. This script:

temp <- mtcars[, 1:3]
colnames(temp) <- c("x1", "z", "y1")

plotting_function <- function(data_to_vis) {
  
  p1 <- ggplot(data_to_vis, aes(x=x1, y=y1)) +
    geom_point(size = 3)       
  options(repr.plot.width = 5, repr.plot.height = 5)     
  print(p1)
  p2 <- ggplot(data_to_vis, aes(x=x1, y=y1)) +
    geom_point() +
    facet_wrap(~z) +
    theme(aspect.ratio = 1)
  options(repr.plot.width = 20, repr.plot.height = 20) 
  print(p2)
}

library(ggplot2)
plotting_function(temp)

uses the 2nd options() setting to set the width and height of both plots. To get them to appear as intended, the first setting needs to be moved after the print(p1) statement.

This was tested on the R kernel at https://jupyter.org/try, which reports "The version of the notebook server is: 6.4.10", "Current kernel information: R version 4.1.2 (2021-11-01)".

Compared to ipython, we are currently missing:

• display_markdown -> that's interesting because pander, which displays mostly every thing in R in markdown
• display_jpeg -> to complement the *_png and *_svg one?
• display_latex -> also lots of stuff in R which can transform to a latex representation
• display_json -> that's the one I want -> jupyter/jupyter_client#6 (comment) :-)
• display_pdf -> nice for plotting, which should be included in latex/pdf
• display_javascript -> in python used to implement magics, which influences the behaviour of the notebook... Not sure if we need that...

#devtools::install_github("renkun-ken/formattable")
df <- data.frame(
  id = 1:10,
  name = c("Bob", "Ashley", "James", "David", "Jenny", 
    "Hans", "Leo", "John", "Emily", "Lee"), 
  age = c(28, 27, 30, 28, 29, 29, 27, 27, 31, 30),
  grade = c("C", "A", "A", "C", "B", "B", "B", "A", "C", "C"),
  test1_score = c(8.9, 9.5, 9.6, 8.9, 9.1, 9.3, 9.3, 9.9, 8.5, 8.6),
  test2_score = c(9.1, 9.1, 9.2, 9.1, 8.9, 8.5, 9.2, 9.3, 9.1, 8.8),
  final_score = c(9, 9.3, 9.4, 9, 9, 8.9, 9.25, 9.6, 8.8, 8.7),
  registered = c(TRUE, FALSE, TRUE, FALSE, TRUE, TRUE, TRUE, FALSE, FALSE, FALSE),
  stringsAsFactors = FALSE)
library(formattable)

f <- formattable(df, list(
  age = color_tile("white", "orange"),
  grade = formatter("span",
    style = x ~ ifelse(x == "A", style(color = "green", font.weight = "bold"), NA)),
  test1_score = normalize_bar("pink", 0.2),
  test2_score = normalize_bar("pink", 0.2),
  final_score = formatter("span",
    style = x ~ style(color = ifelse(rank(-x) <= 3, "green", "gray")),
    x ~ sprintf("%.2f (rank: %02d)", x, rank(-x))),
  registered = formatter("span", 
    style = x ~ style(color = ifelse(x, "green", "red")),
    x ~ icontext(ifelse(x, "ok", "remove"), ifelse(x, "Yes", "No")))
))
w <- as.htmlwidget(f)
h <- htmlwidgets:::toHTML(w, standalone = FALSE, knitrOptions = NULL)
as.character(h)

Results in

Error: is.vector(vec) || is.factor(vec) is not TRUE
Traceback:

1. tryCatch({
 .     for (mime in getOption("jupyter.display_mimetypes")) {
 .         r <- mime2repr[[mime]](obj)
 .         if (!is.null(r)) 
 .             data[[mime]] <- r
 .     }
 . }, error = handle_error)

"

I remove the svg display options like this:

old = getOption("jupyter.display_mimetypes")
options(jupyter.display_mimetypes = old[1:length(old)-1]) # the last element is svg
print(getOption("jupyter.display_mimetypes"))
[1] "text/plain"             "text/html"              "text/markdown"         
[4] "text/latex"             "application/json"       "application/javascript"
[7] "application/pdf"        "image/png"              "image/jpeg" 

But the notebook still contains svg and I can't right click on the image and save the image :-(

We currently define displayas:

#' Display an object using any available reprs
#'
#' @param obj  The object to be displayed
#' @importFrom repr mime2repr repr_text
#' @export
display <- function (obj) {
    data <- namedlist()
    if (getOption('jupyter.rich_display')) {
        for (mime in getOption('jupyter.display_mimetypes')) {
            r <- mime2repr[[mime]](obj)
            if (!is.null(r)) data[[mime]] <- r
        }
    } else {
        data[['text/plain']] <- repr_text(obj)
    }
    publish_mimebundle(data)
}

I would like to shorten that to

#' Display an object using any available reprs
#'
#' @param obj  The object to be displayed
#' @importFrom repr mime2repr repr_text
#' @export
display <- function (obj) {
    data <- namedlist()
    for (mime in getOption('jupyter.display_mimetypes')) {
        r <- mime2repr[[mime]](obj)
        if (!is.null(r)) data[[mime]] <- r
    }
    publish_mimebundle(data)
}

This would mean that I can remove the auto rich display output (jupyter.rich_display = FALSE) and could still use display(something) and get all available output. That would in line with display_XXX which do not honor this option (and rightly so). It would also be more in line of the current doc string: "using any available reprs".

This would also mean that jupyter.rich_display would be an option of the kernel alone and not of the IRdisplay package. IMO it could then be renamed to jupyter.auto_rich_display.

svg is simply not ready to be used for bigger plots :-(

This plot makes scrolling while the plot is visible a pain. Igraph also has some problems with it's svg based plotting. IMO it should be removed in irkernel (following the way it is done by the ipython kernel) or demoted in the notebook behind the bitmap based graphic formats.

library(magrittr)
library(ggplot2)
#options(jupyter.plot_mimetypes = c("text/plain", "image/png" )) # no more svg...
opt = theme(legend.position  = "none",
panel.background = element_rect(fill="violetred4"),
axis.ticks       = element_blank(),
panel.grid       = element_blank(),
axis.title       = element_blank(),
axis.text        = element_blank())
seq(from=-10, to=10, by = 0.05) %>%
expand.grid(x=., y=.) %>%
ggplot(aes(x=x, y=y)) +
geom_point(alpha=.1, shape=20, size=1, color="white", stroke=0) + opt

Not sure where this comes from:

library(grid)
library(gridExtra)
d <- head(iris, 3)
g <- tableGrob(d)
grid.newpage()
grid.draw(g)

produces a plot which has lots of whitespace around it :-(

Happens with svg and svg removed