profile
viewpoint

hadley/r-internals 216

Documentation for R's internal C API

kevinushey/configure 41

Configure R packages for installation with R.

kevinushey/cronR 17

A simple R package for managing your cron jobs.

kevinushey/2017-rstudio-conf 14

Slides used for my "What's New in the RStudio IDE" talk.

jjallaire/sparklingwater 13

Sparkling Water for R

kevinushey/data.table.extras 8

Auxiliary functions to assist in the usage of data.table

kevinushey/2020-rstudio-conf 6

Slides for my presentation on renv at rstudio::conf(2020L).

kevinushey/aRrgh 2

A newcomer's (angry) guide to data types in R

issue openedrstudio/reticulate

Error: could not find a Python environment for python

I use py_install() but get an error which says no Python environment can be found. I tried installing the latest version from Github but still not work

> py_path = "/Users/wsx/anaconda3/bin/python"
> reticulate::use_python(py_path, required = TRUE)
> reticulate::py_module_available("multiprocess")
[1] FALSE
> reticulate::py_install("multiprocess", pip = TRUE)
Error: could not find a Python environment for /Users/wsx/anaconda3/bin/python

> reticulate::py_config()
python:         /Users/wsx/anaconda3/bin/python
libpython:      /Users/wsx/anaconda3/lib/libpython3.7m.dylib
pythonhome:     /Users/wsx/anaconda3:/Users/wsx/anaconda3
version:        3.7.4 (default, Aug 13 2019, 15:17:50)  [Clang 4.0.1 (tags/RELEASE_401/final)]
numpy:          /Users/wsx/anaconda3/lib/python3.7/site-packages/numpy
numpy_version:  1.19.0

NOTE: Python version was forced by use_python function

created time in 19 minutes

issue commentrstudio/rstudio

Restarting R with .rs.restartR() crashes RStudio after reading file with read_table

I tested it with RStudio-1.4.904. Was the issue already resolved in this version? I did re-install all R packages, but maybe my cpp11 or compiler are out of date.

uqrmaie1

comment created time in 2 hours

issue commentrstudio/rstudio

Support for separate font settings for text and code block in R Markdown

Yes, but I notice that users can't customize font settings in visual mode.

enixam

comment created time in 3 hours

issue openedrstudio/shinydashboard

update documentation Glyphicons link

The documentation shows a Glyphicons link as http://getbootstrap.com/components/#glyphicons I believe this should be update to: https://icons.getbootstrap.com/

Screen Shot 2020-11-30 at 8 48 38 PM

I'm happy to submit a PR if you're open to it and point me to where the documentation lives. Thanks.

created time in 3 hours

fork jcheng5/paws

Paws, a package for Amazon Web Services in R

https://paws-r.github.io

fork in 3 hours

started1Hibiki1/locks-py

started time in 4 hours

issue closedr-lib/testthat

expect_snapshot_value( style = "serialize") should store data binary, but it converts it to ASCII

https://github.com/r-lib/testthat/blob/efaf620af6823d0658cb16e6a1e1676624a790f6/R/snapshot.R#L127-L151

For me this was an unexpected behavior, as the reference states that it will be saved as a binary object. So is this intended?

And if yes, could there be an option included to choose between ASCII & binary?

Because to quote the R manual:

"ASCII saves used to be useful for moving data between platforms but are now mainly of historical interest. They can be more compact than binary saves where compression is not used, but are almost always slower to both read and write: binary saves compress much better than ASCII ones. Further, decimal ASCII saves may not restore double/complex values exactly, and what value is restored may depend on the R platform. " https://stat.ethz.ch/R-manual/R-devel/library/base/html/save.html

And it is indeed very slow in my case.

closed time in 4 hours

srfall

issue commentr-lib/testthat

expect_snapshot_value( style = "serialize") should store data binary, but it converts it to ASCII

Never mind, i just tried to take snapshots that were too big.

srfall

comment created time in 4 hours

push eventrstudio/rstudio

Maria Semple

commit sha a3a1dae557a2d06116c5668edc1e75c1137b0316

add notes about Slurm features and bugfixes * GRES/GPU support * Requeuing support * Slurm supported version update * Resource stream bugfix

view details

push time in 4 hours

push eventrstudio/rstudio

Maria Semple

commit sha 07c90eeb9894793f433d67761cd303d4022eaf32

fix a typo (disabile -> disable)

view details

push time in 5 hours

issue commentrstudio/rstudio

News icon on the Update GUI generates wrong URL

Can you try executing the following:

trace(.rs.rpc.get_package_news_url, quote(print(ls.str())))

and then sharing the output you see after you try to update packages + navigate to that NEWS URL?

This is what I get

trace(.rs.rpc.get_package_news_url, quote(print(ls.str()))) Tracing function ".rs.rpc.get_package_news_url" in package "rstudio" [1] ".rs.rpc.get_package_news_url" Tracing (new("functionWithTrace", .Data = function (packageName, libraryPath) .... on entry libraryPath : chr "/usr/local/lib/R/site-library" packageName : chr "ggeffects"

Warning message: In (new("functionWithTrace", .Data = function (packageName, libraryPath) : Failed to infer appropriate NEWS URL: using 'news/news.html' as best-guess candidate

andresrcs

comment created time in 5 hours

issue openedr-lib/testthat

expect_snapshot_value( style = "serialize") should store data binary, but its converts it to ASCII

https://github.com/r-lib/testthat/blob/efaf620af6823d0658cb16e6a1e1676624a790f6/R/snapshot.R#L127-L151

For me this was an unexpected behavior, as the reference states that it will be saved as a binary object. So is this intended?

And if yes, could there be an option included to choose between ASCII & binary?

Because to quote the R manual:

"ASCII saves used to be useful for moving data between platforms but are now mainly of historical interest. They can be more compact than binary saves where compression is not used, but are almost always slower to both read and write: binary saves compress much better than ASCII ones. Further, decimal ASCII saves may not restore double/complex values exactly, and what value is restored may depend on the R platform. " https://stat.ethz.ch/R-manual/R-devel/library/base/html/save.html

And it is indeed very slow in my case.

created time in 5 hours

issue commentrstudio/rstudio

Intermittent error connecting to R on desktop build

This had resolved for me, but it appeared again today on RDP 1.4.1063-1

jmcphers

comment created time in 6 hours

issue commentrstudio/rstudio

adapt to updated graphics device

Verified; in Windows Desktop v1.4.1029 when I tried to plot with R 4.1 (devel) installed an error was shown in console and the plot was shown in a separate window (not the Plots pane).

With 1.4.1067, the plot showed in the pane, as expected.

kevinushey

comment created time in 7 hours

issue commentrstudio/rstudio

Crashing when building and reinstalling an R package

@mikebessuille v1.3.1093

Still crashing with build and install. With a very similar error message

30 Nov 2020 22:10:03 [rsession-ncsg3] ERROR Unexpected exception: boost: mutex lock failed in pthread_mutex_lock: Invalid argument

BTW, this is a completely different R package I'm working on, so it's not the package.


For info, my error log also contains this

30 Nov 2020 22:08:31 [rsession-ncsg3] ERROR r error 7 (Unexpected data type); OCCURRED AT rstudio::core::Error rstudio::r::sexp::extractFunctionInfo(SEXP, rstudio::core::r_util::FunctionInformation*, bool, bool) src/cpp/r/RSexp.cpp:1744; LOGGED FROM: rstudio::core::r_util::FunctionInformation rstudio::session::modules::rparser::{anonymous}::getInfoAssociatedWithFunctionAtCursor(rstudio::core::r_util::token_cursor::RTokenCursor, rstudio::session::modules::rparser::ParseStatus&) src/cpp/session/modules/SessionRParser.cpp:1158

multiple times

csgillespie

comment created time in 7 hours

issue commentrstudio/r-builds

Duration / maintenance of R builds

We may stop building binaries for unsupported operating systems or R versions at some point (e.g., Ubuntu 14 was dropped in the past). However, we'll still continue to host existing binaries. The only exception might be that some unpatchable security vulnerability forces us to remove an old, existing binary.

For operating systems, CentOS 6 is EOL soon and might be dropped in the next year. I don't think we'll be dropping R versions any time soon though.

JosiahParry

comment created time in 7 hours

issue commentrstudio/rstudio

Certain structure of Rmd file blocks chunk widgets

One more interesting combination: rs-bug--rmd-do-not-run-chunk-2

The output loads but newer gets fully loaded: I waited for 5 minutes and the rprocess did not finish.

GegznaV

comment created time in 7 hours

issue openedrstudio/rstudio

Certain structure of Rmd file blocks chunk widgets

<!-- IMPORTANT: Please fill out this template fully! Failure to do so will result in the issue being closed automatically.

This issue tracker is for bugs and feature requests in the RStudio IDE. If you're having trouble with R itself or an R package, see https://www.r-project.org/help.html, and if you want to ask a question rather than report a bug, go to https://community.rstudio.com/. Finally, if you use RStudio Server Pro, get in touch with our Pro support team at support@rstudio.com.

-->

System details

RStudio Edition : Desktop 
RStudio Version : 1.4.1066
OS Version      : Windows 10
R Version       : 4.0.3

Steps to reproduce the problem

  1. Create a new empty Rmd document.
  2. Paste the code that you find below (without leading spaces). Required conditions: a. Create header as in row 2. b. The header should be above the chunk. c. The first line in the code should be "R" comment, which defines a section (i.e., ends in ------) as in row 5.
  3. Try to run the code in the chunk by clicking on arrow as shonw in the gif.

rs-bug--rmd-do-not-run-chunk

The code:

    # Section 1
    
    ```{r}
    # Comment ----------
    print("The code was evaluated")
    ```

Describe the problem in detail

The widget to run the code does not work if these conditions are present.

Describe the behavior you expected

I expect code to run as usual when the arrow is clicked.

  • [x] I have read the guide for submitting good bug reports.
  • [x] I have installed the latest version of RStudio, and confirmed that the issue still persists.
  • [ ] If I am reporting a RStudio crash, I have included a diagnostics report.
  • [x] I have done my best to include a minimal, self-contained set of instructions for consistently reproducing the issue.

created time in 7 hours

issue commentrstudio/rstudio

Clicking on hyperlinks does nothing in R markdown knitted to HTML from Desktop

Hi, I also ran into that issue with local files. I'm working on Windows OS, and clicking the hyperlink linked to an URL works but nothing happens when it's linked to a local file. Everything works fine if I right click and do "Open in the Browser". Any help/guidance would be much appreciated!

ronblum

comment created time in 7 hours

issue openedrstudio/rstudio

Trying to save a .rmd file while installing/updating R packages makes R/RStudio crash

System details

RStudio Edition : Desktop (preview)
RStudio Version : Version 1.4.1048
OS Version      : macOS 10.15.7 (19H15)
R Version       : 4.0.2 (2020-06-22)

Steps to reproduce the problem

  1. Begin updating or installing packages, either from the install.packages() function, or using the IDE's package update tool.
  2. Make a change to a .rmd file.
  3. Attempt to save the file.
  4. Observe that the package installation freezes, and that the file is not saved.
  5. Wait a couple minutes to receive a "Terminate R" modal with the text:
R is not responding to your request to interrupt processing so to stop the current operation you may need to terminate R entirely.

Terminating R will cause your R session to immediately abort. Active computations will be interrupted and unsaved source file changes and workspace objects will be discarded.

Do you want to terminate R now?

[No] [Yes]

Selecting "Yes" will result in a R fatal error dialog. See screenshot below. image

Describe the problem in detail

While attempting to save a file I was working on during package updates, the R process unexpectedly experienced a fatal error. I am able to reproduce this consistently.

The diagnostics report is attached. diagnostics-report.txt

Describe the behavior you expected

I expected that I could save files in the IDE without crashing the R process.

  • [x] I have read the guide for submitting good bug reports.
  • [ ] I have installed the latest version of RStudio, and confirmed that the issue still persists.
  • [x] If I am reporting a RStudio crash, I have included a diagnostics report.
  • [x] I have done my best to include a minimal, self-contained set of instructions for consistently reproducing the issue.

created time in 8 hours

issue commentrstudio/rstudio

Add syntax highlighting support for languages inside R strings

I've already +1'd this feature request but would also like to offer another solution.

Per @kevinushey, requiring RStudio to parse SQL statements for validity would get expensive. I'm personally forced to work with humongous code files with many thousands of lines, much of which is embedded SQL. So I can see how parsing that for validity would get expensive (though the debug advantage mentioned by @moodymudskipper would be nice).

One solution would be for the R team to add a shorthand, as they did with r"()" for raw strings in 4.0.0. Maybe s"()". But obviously this isn't the proper forum to talk about changes to the R language itself.

So a temporary solution would be for RStudio to format any string found within specific function calls, such as DBI::SQL() and DBI::dbGetQuery(), or any function which will basically only ever be called with SQL strings. I'm no expert

If this is an appealing idea, I'd also ask that the list of functions which get formatted be extendable (even if via a config file), since I personally work with multiple wrapper functions around DBI calls (I can't remember the last time I called DBI::dbGetQuery() directly).

talgalili

comment created time in 8 hours

issue commentrstudio/rstudio

Bundle RStudio Desktop for Linux as a Flatpak, Snap, or AppImage

yes, @eoli3n is correct, the discussion about Faltapak, Snap, or AppImage goes since June, 2018 and there isn't anything on horizon

jmcphers

comment created time in 8 hours

pull request commentr-lib/testthat

skip_if_not(): do not append " is not TRUE" to custom message

A doc fix would be great. There's no need for a skip_on_github_actions() since you can just use skip_on_ci(). The docs probably need to be updated to reflect that we no longer recommend skip_on_travis() or skip_on_appveyor().

dpprdan

comment created time in 8 hours

issue commentrstudio/rstudio

Implement focus for all Source Pane tab types

Verified I can F6 and Shift+F6 into and out of source panes with the 4 indicated types. Used macOS Desktop build 1.4.1066.

melissa-barca

comment created time in 9 hours

push eventrstudio/rsconnect

Jonathan McPherson

commit sha f50110e32ba410008ab1e792948f6b69d6f7ae17

update NEWS

view details

push time in 9 hours

pull request commentrstudio/packrat

Optimise package search tree and insert verbose logger messages

Replaced by https://github.com/rstudio/packrat/pull/615 (now merged).

andrie

comment created time in 9 hours

delete branch rstudio/rsconnect

delete branch : bugfix/available-packages-remove-version-filter

delete time in 9 hours

push eventrstudio/rsconnect

Kevin Ushey

commit sha aa90e168336056af5bb1db2bdf426665cc3e062d

don't use 'R_version' filter when querying available packages

view details

Aron Atkins

commit sha 75bb79c6c282c42a399d1cacd1f9df9183de165d

Merge pull request #467 from rstudio/bugfix/available-packages-remove-version-filter don't use 'R_version' filter when querying available packages

view details

push time in 9 hours

PR merged rstudio/rsconnect

Reviewers
don't use 'R_version' filter when querying available packages

This PR fixes an issue where rsconnect::writeManfiest() could fail if:

  1. The current version of R is older (e.g. R 3.5), but
  2. The application uses a package whose current version on CRAN requires a new version.

An example of this is the foreign package; it currently requires (R >= 4.0.0). This implies that manifest generation will fail, with an error of the form:

> rsconnect::writeManifest(appDir = "usesforeign/")
Warning: Unable to determine the repository for package foreign
Error: 
* Unable to determine the location for some packages. Packages must come from a package
   repository like CRAN or a source control system. Check that options('repos') refers to a
   package repository containing the needed package versions.
+22 -8

4 comments

1 changed file

kevinushey

pr closed time in 9 hours

more