socialmixr error on upgrade data.table

[mattdowle]

socialmixr_0.1.6.tar.gz
./socialmixr.Rcheck/00check.log:* using log directory ‘/home/mdowle/build/revdeplib/socialmixr.Rcheck’
./socialmixr.Rcheck/00check.log:* using R Under development (unstable) (2020-07-14 r78854)
./socialmixr.Rcheck/00check.log:* using platform: x86_64-pc-linux-gnu (64-bit)
./socialmixr.Rcheck/00check.log:* using session charset: UTF-8
./socialmixr.Rcheck/00check.log:* checking for file ‘socialmixr/DESCRIPTION’ ... OK
./socialmixr.Rcheck/00check.log:* this is package ‘socialmixr’ version ‘0.1.6’
./socialmixr.Rcheck/00check.log:* package encoding: UTF-8
./socialmixr.Rcheck/00check.log:* checking package namespace information ... OK
./socialmixr.Rcheck/00check.log:* checking package dependencies ... OK
./socialmixr.Rcheck/00check.log:* checking if this is a source package ... OK
./socialmixr.Rcheck/00check.log:* checking if there is a namespace ... OK
./socialmixr.Rcheck/00check.log:* checking for executable files ... OK
./socialmixr.Rcheck/00check.log:* checking for hidden files and directories ... OK
./socialmixr.Rcheck/00check.log:* checking for portable file names ... OK
./socialmixr.Rcheck/00check.log:* checking for sufficient/correct file permissions ... OK
./socialmixr.Rcheck/00check.log:* checking whether package ‘socialmixr’ can be installed ... OK
./socialmixr.Rcheck/00check.log:* checking installed package size ... OK
./socialmixr.Rcheck/00check.log:* checking package directory ... OK
./socialmixr.Rcheck/00check.log:* checking ‘build’ directory ... OK
./socialmixr.Rcheck/00check.log:* checking DESCRIPTION meta-information ... OK
./socialmixr.Rcheck/00check.log:* checking top-level files ... OK
./socialmixr.Rcheck/00check.log:* checking for left-over files ... OK
./socialmixr.Rcheck/00check.log:* checking index information ... OK
./socialmixr.Rcheck/00check.log:* checking package subdirectories ... OK
./socialmixr.Rcheck/00check.log:* checking R files for non-ASCII characters ... OK
./socialmixr.Rcheck/00check.log:* checking R files for syntax errors ... OK
./socialmixr.Rcheck/00check.log:* checking whether the package can be loaded ... OK
./socialmixr.Rcheck/00check.log:* checking whether the package can be loaded with stated dependencies ... OK
./socialmixr.Rcheck/00check.log:* checking whether the package can be unloaded cleanly ... OK
./socialmixr.Rcheck/00check.log:* checking whether the namespace can be loaded with stated dependencies ... OK
./socialmixr.Rcheck/00check.log:* checking whether the namespace can be unloaded cleanly ... OK
./socialmixr.Rcheck/00check.log:* checking loading without being on the library search path ... OK
./socialmixr.Rcheck/00check.log:* checking dependencies in R code ... OK
./socialmixr.Rcheck/00check.log:* checking S3 generic/method consistency ... OK
./socialmixr.Rcheck/00check.log:* checking replacement functions ... OK
./socialmixr.Rcheck/00check.log:* checking foreign function calls ... OK
./socialmixr.Rcheck/00check.log:* checking R code for possible problems ... OK
./socialmixr.Rcheck/00check.log:* checking Rd files ... OK
./socialmixr.Rcheck/00check.log:* checking Rd metadata ... OK
./socialmixr.Rcheck/00check.log:* checking Rd cross-references ... OK
./socialmixr.Rcheck/00check.log:* checking for missing documentation entries ... OK
./socialmixr.Rcheck/00check.log:* checking for code/documentation mismatches ... OK
./socialmixr.Rcheck/00check.log:* checking Rd \usage sections ... OK
./socialmixr.Rcheck/00check.log:* checking Rd contents ... OK
./socialmixr.Rcheck/00check.log:* checking for unstated dependencies in examples ... OK
./socialmixr.Rcheck/00check.log:* checking contents of ‘data’ directory ... OK
./socialmixr.Rcheck/00check.log:* checking data for non-ASCII characters ... OK
./socialmixr.Rcheck/00check.log:* checking data for ASCII and uncompressed saves ... OK
./socialmixr.Rcheck/00check.log:* checking installed files from ‘inst/doc’ ... OK
./socialmixr.Rcheck/00check.log:* checking files in ‘vignettes’ ... OK
./socialmixr.Rcheck/00check.log:* checking examples ... OK
./socialmixr.Rcheck/00check.log:* checking for unstated dependencies in ‘tests’ ... OK
./socialmixr.Rcheck/00check.log:* checking tests ... ERROR
./socialmixr.Rcheck/00check.log:  Running ‘testthat.R’
./socialmixr.Rcheck/00check.log:Running the tests in ‘tests/testthat.R’ failed.
./socialmixr.Rcheck/00check.log:Complete output:
./socialmixr.Rcheck/00check.log:  > library(testthat)
./socialmixr.Rcheck/00check.log:  > library(socialmixr)
./socialmixr.Rcheck/00check.log:  
./socialmixr.Rcheck/00check.log:  Attaching package: 'socialmixr'
./socialmixr.Rcheck/00check.log:  
./socialmixr.Rcheck/00check.log:  The following object is masked from 'package:utils':
./socialmixr.Rcheck/00check.log:  
./socialmixr.Rcheck/00check.log:      cite
./socialmixr.Rcheck/00check.log:  
./socialmixr.Rcheck/00check.log:  > 
./socialmixr.Rcheck/00check.log:  > test_check("socialmixr")
./socialmixr.Rcheck/00check.log:  Error: C stack usage  7976260 is too close to the limit
./socialmixr.Rcheck/00check.log:  Execution halted
./socialmixr.Rcheck/00check.log:* checking for unstated dependencies in vignettes ... OK
./socialmixr.Rcheck/00check.log:* checking package vignettes in ‘inst/doc’ ... OK
./socialmixr.Rcheck/00check.log:* checking running R code from vignettes ... NONE
./socialmixr.Rcheck/00check.log:  ‘introduction.Rmd’ using ‘UTF-8’... OK
./socialmixr.Rcheck/00check.log:* checking re-building of vignette outputs ... OK
./socialmixr.Rcheck/00check.log:* checking PDF version of manual ... OK
./socialmixr.Rcheck/00check.log:* DONE
./socialmixr.Rcheck/00check.log:Status: 1 ERROR

[MichaelChirico]

Error is coming from copy. MRE:

library(socialmixr)
erroneous_survey <- survey(polymod$participants, polymod$contacts, polymod$reference)
erroneous_type1 <- copy(erroneous_survey)

Investigating...

[MichaelChirico]

I think this is a downstream bug.

Here's a smaller example that reproduces the bug:

library(socialmixr)
library(data.table)

y = structure(list(list(given = "Joel", family = "Mossong", role = NULL, email = NULL, comment = NULL)), class = "person")
copy(y)

The issue is that copy is recursively trying to reallocate any objects in the structure of y that are data.tables, however, y[[1L]] is the same as y:

identical(y[[1L]], y)

Therefore lapply fails to make any progress "exploring the tree" of the object & we are stuck in a loop.

This looks like a bug in [[.person from utils to me -- the method is behaving like a [.person method would.

I can't think offhand of a way to fix this in copy besides e.g. putting a "max depth" limit on the reallocate behavior?