When using set() "Internal error, please report" #2996
Comments
|
please be sure to setDT() after reading from .rda
…On Thu, Aug 9, 2018, 11:36 PM Joseph M. Burling ***@***.***> wrote:
Get an internal error with set() when loading an object to an environment
and getting then using object from same environment.
*Example:*
library(data.table)
print(sessionInfo())
load_tmp_rda <- function(v, folder, env) {
obj_name <- paste0("tmp", v)
filename <- file.path(folder, paste0(obj_name, ".rda"))
load(filename, env)
obj <- get(obj_name, env)
message(obj_name, " has ", nrow(obj), " rows")
set(obj, NULL, "version", v)
}
# doesn't work
local({
tmp1 <- data.table(runif(10000))
tmp_dir <- tempdir()
tmp_file <- file.path(tmp_dir, "tmp1.rda")
save(tmp1, file=tmp_file)
this <- environment()
load_tmp_rda(1L, tmp_dir, this)
})
# works
local({
tmp1 <- data.table(runif(10000))
obj_name <- paste0("tmp", 1L)
this <- environment()
obj <- get(obj_name, this)
set(obj, NULL, "version", 1L)
})
*Output:*
R version 3.5.1 (2018-07-02)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows >= 8 x64 (build 9200)
Matrix products: default
locale:
[1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 LC_MONETARY=English_United States.1252 LC_NUMERIC=C
[5] LC_TIME=English_United States.1252
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] data.table_1.11.4
loaded via a namespace (and not attached):
[1] Rcpp_0.12.18 pillar_1.3.0 compiler_3.5.1 plyr_1.8.4 bindr_0.1.1 tools_3.5.1 digest_0.6.15 packrat_0.4.9-3 evaluate_0.11
[10] tibble_1.4.2 gtable_0.2.0 pkgconfig_2.0.1 rlang_0.2.1 rstudioapi_0.7 bindrcpp_0.2.2 gridExtra_2.3 stringr_1.3.1 dplyr_0.7.6
[19] styler_1.0.2 knitr_1.20 rprojroot_1.3-2 grid_3.5.1 tidyselect_0.2.4 glue_1.3.0 R6_2.2.2 rmarkdown_1.10 ggplot2_3.0.0
[28] purrr_0.2.5 mejr_1.2.0.9000 magrittr_1.5 rematch2_2.0.1 backports_1.1.2 scales_0.5.0 htmltools_0.3.6 rfmt_1.0 assertthat_0.2.0
[37] colorspace_1.3-2 stringi_1.1.7 lazyeval_0.2.1 munsell_0.5.0 crayon_1.3.4
tmp1 has 10000 rows
Error in set(obj, NULL, "version", v) :
Internal error, please report (including result of sessionInfo()) to data.table issue tracker: oldtncol (0) < oldncol (1) but tl of class is marked.
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#2996>, or mute the thread
<https://github.com/notifications/unsubscribe-auth/AHQQdTr2LGGttcAEzq8a9PtMJa_uH5e6ks5uPP-zgaJpZM4V3iVv>
.
|
|
using library(data.table)
load_tmp_rda <- function(v, folder, env) {
obj_name <- paste0("tmp", v)
filename <- file.path(folder, paste0(obj_name, ".rda"))
load(filename, env)
obj <- get(obj_name, env)
setDT(obj) ## added this line
message(obj_name, " has ", nrow(obj), " rows")
set(obj, NULL, "version", v)
}
local({
tmp1 <- data.table(runif(10000))
tmp_dir <- tempdir()
tmp_file <- file.path(tmp_dir, "tmp1.rda")
save(tmp1, file=tmp_file)
this <- environment()
load_tmp_rda(1L, tmp_dir, this)
})
#tmp1 has 10000 rowserror is gone. |
Probably we can improve the error message, it sounds like you did something really scary to see Can have a branch if |
|
Is there not an FAQ about this issue? Or something in the Wiki? Would like to refer to something in the error message... |
In the FAQ, under "Reading data.table from RDS or RData file" , it says to use alloc.col. I always use setDT. If that's recommended or just as good as alloc.col, maybe the FAQ can be updated to match. |
|
error message has to point to |
Get an internal error with set() when loading an object to an environment and getting then using object from same environment.
Example:
Output:
The text was updated successfully, but these errors were encountered: