plot_str Error: C stack usage 7970280 is too close to the limit

[boxuancui]

Identified by @holgerbrandl from Revolutions blog post.

plot_str just gives:
Error: C stack usage 7970280 is too close to the limit

@holgerbrandl Do you mind sharing a subset of your data, so that I can re-produce?

[boxuancui]

@holgerbrandl On another note, thanks for your suggestions on the theme issue. @amrrs just created #56 and maybe I can do something there.

[holgerbrandl]

Interestintly, the error changes when running it twice

> plot_str(dds)
Error: evaluation nested too deeply: infinite recursion / options(expressions=)?
> plot_str(dds)
Error: C stack usage  7970312 is too close to the limit
> devtools::session_info()
Session info ------------------------------------------------------------------
 setting  value                       
 version  R version 3.4.0 (2017-04-21)
 system   x86_64, darwin15.6.0        
 ui       X11                         
 language (EN)                        
 collate  en_US.UTF-8                 
 tz       <NA>                        
 date     2018-02-14                  

To reproduce it, try

dds = local(get(load(url("https://cloud.mpi-cbg.de/index.php/s/1FVDEUnWhm8N7cJ/download"))))
DataExplorer::plot_str(dds)

The dds is a data structure created with https://bioconductor.org/packages/release/bioc/html/DESeq2.html

Error is the same on MacOS and linux for me.

[boxuancui]

Note to myself:

• I don't think it fully support S4 objects yet.
• Figure out what went wrong in the recursion.
• Same error on glmnet object:

  test <- glmnet(matrix(rnorm(100 * 20), 100, 20), rnorm(100))
  plot_str(test)
  # Error: evaluation nested too deeply: infinite recursion / options(expressions=)?
  # Error during wrapup: evaluation nested too deeply: infinite recursion / options(expressions=)?
  plot_str(test, max_level = 4)
  # Error: C stack usage  7969624 is too close to the limit
  

  • Setting max_level to 1, 2 and 3 produces the same chart, very likely due to the S4 methods.

[boxuancui]

@holgerbrandl When I run the following command, it is giving me an error. Since plot_str is based on the output of str, it will fail as well. Do you have the same problem?

> dds <- local(get(load(url("https://cloud.mpi-cbg.de/index.php/s/1FVDEUnWhm8N7cJ/download"))))
> str(dds)
Formal class 'DESeqDataSet' [package "DESeq2"] with 8 slots
  ..@ design            :Class 'formula'  language ~sample
  .. .. ..- attr(*, ".Environment")=<environment: R_GlobalEnv> 
  ..@ dispersionFunction:function (q)  
  .. ..- attr(*, "coefficients")= Named num [1:2] 0.0839 2.3722
  .. .. ..- attr(*, "names")= chr [1:2] "asymptDisp" "extraPois"
  .. ..- attr(*, "fitType")= chr "parametric"
  .. ..- attr(*, "varLogDispEsts")= num 0.566
  .. ..- attr(*, "dispPriorVar")= num 0.25
  ..@ rowRanges         :Formal class 'GRangesList' [package "GenomicRanges"] with 5 slots
  .. .. ..@ unlistData     :Error in FUN(X[[i]], ...) : 
  no slot of name "elementType" for this object of class "GRanges"

However, the infinite recursion and C stack overflow are mostly due to the lack of S4 support, which should be fixed (in develop) now.

[boxuancui]

I am closing this for v0.6 release. As I said in the comment above, the infinite recursion and C stack overflow errors were caused by S4 objects, and it should be fixed. Please create a separate issue if anything else is not working.