Potential bug using permutest?

[ecologyjh]

I am not an R wizard yet but I just spent 3 hours trying to figure out why my permutest code stopped working. Maybe there's a way to fix this to save other people spending 3 hours on it in the future! Not sure if this would be reproduced on other machines as have not been able to test.

I ran the following rstan code to increase computing power:

rstan_options(auto_write = TRUE)
options(mc.cores = parallel::detectCores())

I then ran betadisper followed by permutest (the code from the example in the permutest.betadisper R documentation returns the same result for me - which can be used to test this) to be returned with the error:

Pstats [,1]:incorrect number of dimensions

When I restarted without running the cores code, there were no issues again. I read somewhere that this was supposed to have been fixed in a previous version - but I still had problems! Hope this is useful.

[gavinsimpson]

I can't reproduce this with:

library('rstan')
rstan_options(auto_write = TRUE)
options(mc.cores = parallel::detectCores())
library('vegan')
example('permutest.betadisper')

on

session info

─ Session info ───────────────────────────────────────────────────────────────
 setting  value                          
 version  R version 3.6.3 (2020-02-29)   
 os       Fedora 32 (Workstation Edition)
 system   x86_64, linux-gnu              
 ui       X11                            
 language (EN)                           
 collate  en_US.utf8                     
 ctype    en_US.utf8                     
 tz       America/Regina                 
 date     2020-06-29                     

─ Packages ───────────────────────────────────────────────────────────────────
 package     * version  date       lib source        
 assertthat    0.2.1    2019-03-21 [1] CRAN (R 3.6.3)
 backports     1.1.8    2020-06-17 [1] CRAN (R 3.6.3)
 callr         3.4.3    2020-03-28 [1] CRAN (R 3.6.3)
 cli           2.0.2    2020-02-28 [1] CRAN (R 3.6.3)
 cluster       2.1.0    2019-06-19 [2] CRAN (R 3.6.3)
 colorspace    1.4-1    2019-03-18 [1] CRAN (R 3.6.3)
 crayon        1.3.4    2017-09-16 [1] CRAN (R 3.6.3)
 desc          1.2.0    2018-05-01 [1] CRAN (R 3.6.3)
 devtools      2.3.0    2020-04-10 [1] CRAN (R 3.6.3)
 digest        0.6.25   2020-02-23 [1] CRAN (R 3.6.3)
 dplyr         1.0.0    2020-05-29 [1] CRAN (R 3.6.3)
 ellipsis      0.3.1    2020-05-15 [1] CRAN (R 3.6.3)
 fansi         0.4.1    2020-01-08 [1] CRAN (R 3.6.3)
 fs            1.4.1    2020-04-04 [1] CRAN (R 3.6.3)
 generics      0.0.2    2018-11-29 [1] CRAN (R 3.6.3)
 ggplot2     * 3.3.1    2020-05-28 [1] CRAN (R 3.6.3)
 glue          1.4.1    2020-05-13 [1] CRAN (R 3.6.3)
 gridExtra     2.3      2017-09-09 [1] CRAN (R 3.6.3)
 gtable        0.3.0    2019-03-25 [1] CRAN (R 3.6.3)
 inline        0.3.15   2018-05-18 [1] CRAN (R 3.6.3)
 lattice     * 0.20-41  2020-04-02 [1] CRAN (R 3.6.3)
 lifecycle     0.2.0    2020-03-06 [1] CRAN (R 3.6.3)
 loo           2.2.0    2019-12-19 [1] CRAN (R 3.6.3)
 magrittr      1.5      2014-11-22 [1] CRAN (R 3.6.3)
 MASS          7.3-51.6 2020-04-26 [1] CRAN (R 3.6.3)
 Matrix        1.2-18   2019-11-27 [2] CRAN (R 3.6.3)
 matrixStats   0.56.0   2020-03-13 [1] CRAN (R 3.6.3)
 memoise       1.1.0    2017-04-21 [1] CRAN (R 3.6.3)
 mgcv          1.8-31   2019-11-09 [2] CRAN (R 3.6.3)
 munsell       0.5.0    2018-06-12 [1] CRAN (R 3.6.3)
 nlme          3.1-148  2020-05-24 [1] CRAN (R 3.6.3)
 permute     * 0.9-5    2019-03-12 [1] CRAN (R 3.6.3)
 pillar        1.4.4    2020-05-05 [1] CRAN (R 3.6.3)
 pkgbuild      1.0.8    2020-05-07 [1] CRAN (R 3.6.3)
 pkgconfig     2.0.3    2019-09-22 [1] CRAN (R 3.6.3)
 pkgload       1.1.0    2020-05-29 [1] CRAN (R 3.6.3)
 prettyunits   1.1.1    2020-01-24 [1] CRAN (R 3.6.3)
 processx      3.4.2    2020-02-09 [1] CRAN (R 3.6.3)
 ps            1.3.3    2020-05-08 [1] CRAN (R 3.6.3)
 purrr         0.3.4    2020-04-17 [1] CRAN (R 3.6.3)
 R6            2.4.1    2019-11-12 [1] CRAN (R 3.6.3)
 Rcpp          1.0.4.6  2020-04-09 [1] CRAN (R 3.6.3)
 remotes       2.1.1    2020-02-15 [1] CRAN (R 3.6.3)
 rlang         0.4.6    2020-05-02 [1] CRAN (R 3.6.3)
 rprojroot     1.3-2    2018-01-03 [1] CRAN (R 3.6.3)
 rstan       * 2.19.3   2020-02-11 [1] CRAN (R 3.6.3)
 scales        1.1.1    2020-05-11 [1] CRAN (R 3.6.3)
 sessioninfo   1.1.1    2018-11-05 [1] CRAN (R 3.6.3)
 StanHeaders * 2.21.0-3 2020-05-28 [1] CRAN (R 3.6.3)
 testthat      2.3.2    2020-03-02 [1] CRAN (R 3.6.3)
 tibble        3.0.1    2020-04-20 [1] CRAN (R 3.6.3)
 tidyselect    1.1.0    2020-05-11 [1] CRAN (R 3.6.3)
 usethis       1.6.1    2020-04-29 [1] CRAN (R 3.6.3)
 vctrs         0.3.1    2020-06-05 [1] CRAN (R 3.6.3)
 vegan       * 2.5-6    2019-09-01 [1] CRAN (R 3.6.3)
 withr         2.2.0    2020-04-20 [1] CRAN (R 3.6.3)

[1] /home/gavin/R/x86_64-redhat-linux-gnu-library/3.6
[2] /usr/lib64/R/library
[3] /usr/share/R/library

[jarioksa]

I can't reproduce this either. Need more info.

[jarioksa]

Confirmed: this seems to happen when we use socket clusters in parallel evaluation (parallel::makeCluster). This happens usually only in Windows and therefore @gavinsimpson and I failed to reproduce the error (I tested in two operating systems, but neither was Windows). We can reproduce this by using socket clusters also in non-Windows-systems:

library(vegan)
data(varespec)
dis <- vegdist(varespec)
groups <- factor(c(rep(1,16), rep(2,8)), labels = c("grazed","ungrazed"))
mod <- betadisper(dis, groups)
cl <- parallel::makeCluster(2) ## to force the use of socket cluster in non-Windows systems
permutest(mod, parallel=cl) ## Error in Pstats[, 1] : incorrect number of dimensions

The error condition emerges because we use parRapply with socket clusters, and this drops dimensions of the result, and when we later try to drop the dimension again, we get an error of not having them, but only a a vector.

[gavinsimpson]

Thanks @jarioksa (shows the importance of a fully reproducible example). I'll take a look.

[jarioksa]

@gavinsimpson : an obvious solution is to use drop(Pstats) instead of Pstats[,1], but that would fail if Pstats has more than one column (I don't know if that can happen). Then there are dirty ifs (but they may be more robust).

Yes, and perhaps we should find a Windows test bed as well. The rstan was not an issue.

It seems that we have had this bug in vegan since release 2.2-0 (Nov 2014), but no Windows-user had reported problems.

[jarioksa]

We are forced to have a new release of vegan (see #382), and this is a release critical bug. How to fix this @gavinsimpson ?

[gavinsimpson]

@jarioksa were you given a timeline for submitting to CRAN? I'll look at this by the weekend unless we need a real ease sooner?

[jarioksa]

@gavinsimpson no deadline. This is a problem in the future R (now R-devel) and may not be very urgent yet. However, I was working for a release, but upon checking issues, noticed this one that I had forgot, and think should be fixed for the release. So you can quite well postpone this to the weekend (and my best greetings to the family).