Segfault when writing dataframes with two or more factor columns

[LTLA]

set.seed(0)
library(rhdf5)
Y <- data.frame(
    X=sample(LETTERS, 20000, replace=TRUE),
    Y=sample(LETTERS, 20000, replace=TRUE)
)
unlink('stuff.h5')
h5write(Y, file="stuff.h5", name='data', DataFrameAsCompound=FALSE)

Running it locally on my mac indicates that memory usage climbs inexorably - as of time of writing, I'm at 3 GB usage in an R session that's stuck on h5write. Running it on Unix on an interactive job with an 8GB memory limit eventually causes a segfault, presumably because all memory is allocated. valgrind indicates that the error occurs in _H5Dwrite.

No problems are observed if there is only one factor column (i.e., either column is converted to a character or integer), the command above completes instantly.

Session information

R version 3.6.1 Patched (2019-10-31 r77366)
Platform: x86_64-apple-darwin17.7.0 (64-bit)
Running under: macOS High Sierra 10.13.6

Matrix products: default
BLAS:   /Users/luna/Software/R/R-3-6-branch/lib/libRblas.dylib
LAPACK: /Users/luna/Software/R/R-3-6-branch/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

other attached packages:
[1] rhdf5_2.30.0

loaded via a namespace (and not attached):
[1] compiler_3.6.1 Rhdf5lib_1.8.0

[grimbough]

Thanks for the report, can confirm I see the same issue on my machine. It also happens if both are converted to character, which is what is happening prior to the write anyway. - I forgot the default for data.frame() is stringsAsFactors = TRUE.

[grimbough]

This should now be fixed in both release (2.30.1) and devel (2.31.1). Let me know if it's still causing issues.