Dplyr: Dplyr crashes R 3.3.0 32-bit when using mutate or summarize when code should return an error

[daltonhance]

I can confirm that this happens only with 3.3.0 32 bit. It does not occur with 64 bit 3.3.0 or with 32-bit 3.2.4. I am not able to confirm the exact circumstances under which it happens.

library(dplyr)

tst <- data.frame(
  grp = c(rep("A", 16), rep("B", 16), rep("C", 16)),
  x   = rnorm(16*3),
  y   = rnorm(16*3)
)

tst %>% 
  mutate(corr = cor(x, y, na.rm =  T))

Using the 64-bit 3.3.0, the second chuck of code correctly returns the helpful error:
Error: unused argument (na.rm = TRUE). Under 32-bit 3.3.0, R instead crashes.

[hadley]

@pkoki @daltonhance I delete the irrelevant stuff to make it easier to see the actual issue

But it's unlikely I'll have the time to fix given that it doesn't reproduce on my computer.

[pradeepthiyyagura]

I agree with the bug.

Summarize does not work in the new 3.3 version

R.version
_
platform x86_64-w64-mingw32
arch x86_64
os mingw32
system x86_64, mingw32
status
major 3
minor 3.0
year 2016
month 05
day 03
svn rev 70573
language R
version.string R version 3.3.0 (2016-05-03)
nickname Supposedly Educational
iris %>% group_by(Species) %>% summarize(avg =mean(Sepal.Width)) %>% arrange(avg)
avg
1 3.057333

R.version
_
platform x86_64-pc-linux-gnu
arch x86_64
os linux-gnu
system x86_64, linux-gnu
status
major 3
minor 1.2
year 2014
month 10
day 31
svn rev 66913
language R
version.string R version 3.1.2 (2014-10-31)
nickname Pumpkin Helmet
iris %>% group_by(Species) %>% summarize(avg =mean(Sepal.Width)) %>% arrange(avg)
Source: local data frame [3 x 2]

 Species   avg
  (fctr) (dbl)

1 versicolor 2.770
2 virginica 2.974
3 setosa 3.428

[kevinushey]

I was initially able to reproduce this crash on 32bit R, but it went away after reinstalling both Rcpp and dplyr.

[naofumi-fujii]

I faced this problem on

• Windows7
• 32bit
• RStudio 0.98.1103, 0.99.902
• R 3.3.0, 3.3.1

and this is my code that causes crash.

iris %>%
  group_by(Sepal.Length, Sepal.Width, some_not_existing_colname) %>%
  summarise(sum(Petal.Width))

this code works(with proper columns)

iris %>%
  group_by(Sepal.Length, Sepal.Width) %>%
  summarise(sum(Petal.Width))

hope this helps.

[mikedolanfliss]

Same here on Win7, 32bit, R3.3.0. dplyr errors crash R entirely. Tried a reinstall of dplyr and Rcpp, but no dice yet.

[kevinushey]

This is indeed a bad interaction between 32bit versions of R 3.3.x and RStudio. Hopefully, this will be fixed on the RStudio side soon.

[CBSti]

Had the same problem and same OS/software conditions (Windows 7, 32 bit, R 3.3.1 ...). I installed the preview version of RStudio from
https://www.rstudio.com/products/rstudio/download/preview/
as suggested, now it works again. I get proper error messages instead of the R Session and/or RStudio crashing. Just some feedback, if this was already known, feel free to delete this comment.

[kevinushey]

Thanks for the confirmation @CBSti -- indeed, this should now be resolved in the RStudio preview release.

To anyone stumbling upon this issue, please install the preview release of RStudio to resolve these crashes.

[ZedLeb]

I experienced these fatal error crashes today and would like to try the preview release but I'm using RStudio Portable (v 0.99.902).

Does anyone know if (and how) I can install?

[kevinushey]

Both installers and portable versions of the preview release of RStudio are available at https://www.rstudio.com/products/rstudio/download/preview/.

[ZedLeb]

Many thanks for rapid response!
On 7 Sep 2016 10:57 p.m., Kevin Ushey [email protected] wrote:Both installers and portable versions of the preview release of RStudio are available at https://www.rstudio.com/products/rstudio/download/preview/.

—You are receiving this because you commented.Reply to this email directly, view it on GitHub, or mute the thread.

[ZedLeb]

Installed the preview version today and I'm not crashing any more. Thank you!
I noticed this new version runs the whole of a dplyr piped chain too - that caught me off guard :)

[pushkar-patil]

Me too was facing the same issue. While using mutate() in dplyr, the rstudio crashed completely.
Installed the preview version and now its working. Thanks!!!

[krlmlr]

Seems to be resolved by the RStudio preview release. Can we close this?

[kevinushey]

Yes indeed (the fixes are also part of the current RStudio release, v1.0.44).

[tandrewsimpson]

Instead of reinstalling RStudio, try removing all your packges. That worked for me! I used this code:

ip <- installed.packages()
pkgs.to.remove <- ip[!(ip[,"Priority"] %in% c("base", "recommended")), 1]
sapply(pkgs.to.remove, remove.packages)