burchill / catchr Goto Github PK
View Code? Open in Web Editor NEWcatchr: Flexible, useful tools for dealing with conditions in R, for new users and veterans
License: Other
catchr: Flexible, useful tools for dealing with conditions in R, for new users and veterans
License: Other
I recently updated rlang
to v 0.4.11.9000 and am now encountering an error with make_plans
and make_catch_fn
that has to do with the function catchr:::approx_arg_name
and how expr_deparse
now deals with ...
.
Error trace:
`...` is not empty.
i These dots only exist to allow future extensions and should be empty.
x We detected these problematic arguments:
* `..1`
i Did you misspecify an argument?
Backtrace:
x
1. +-base::source("~/R/Quant/JobsScripts/TuneHP.R", echo = TRUE)
2. | +-base::withVisible(eval(ei, envir))
3. | \-base::eval(ei, envir)
4. | \-base::eval(ei, envir)
5. +-qf::start_cluster(outfile = "~/R/Quant/cl_log.log") ~/R/Quant/JobsScripts/TuneHP.R:12:0
6. | +-base::eval(...)
7. | | \-base::eval(...)
8. | \-catchr::make_catch_fn(...)
9. | \-catchr::make_plans(..., .opts = .opts)
10. | \-catchr:::check_and_clean_input(..., spec_names = special_terms)
11. | \-catchr:::clean_input(akw$kwargs, spec_names)
12. | \-`%>%`(...)
13. +-catchr:::add_back_arg_pos(., qs)
14. | \-purrr::map2(...)
15. +-purrr::map(., ~classify_arg(., spec_names))
16. | \-catchr:::.f(.x[[i]], ...)
17. | \-catchr:::classify_arg(., spec_names)
18. | \-catchr:::approx_arg_name(!!arg)
19. | \-get_expr(enquo(x)) %>% expr_deparse(999) %>% paste(collapse = "")
20. +-base::paste(., collapse = "")
21. \-rlang::expr_deparse(., 999)
22. \-rlang::check_dots_empty0(...)
23. \-rlang::check_dots_empty()
24. \-rlang:::action_dots(...)
SessionInfo:
R version 4.0.2 (2020-06-22)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 19042)
Matrix products: default
locale:
[1] LC_COLLATE=English_United States.1252
[2] LC_CTYPE=English_United States.1252
[3] LC_MONETARY=English_United States.1252
[4] LC_NUMERIC=C
[5] LC_TIME=English_United States.1252
attached base packages:
[1] stats graphics grDevices utils datasets methods
[7] base
other attached packages:
[1] parsnip_0.1.5 RevoUtils_11.0.2 RevoUtilsMath_11.0.0
loaded via a namespace (and not attached):
[1] colorspace_2.0-0 slider_0.1.5 ellipsis_0.3.1
[4] class_7.3-17 timetk_2.6.0 fs_1.5.0
[7] logspline_2.1.16 rstudioapi_0.13 mice_3.13.0
[10] listenv_0.8.0 furrr_0.2.1 dials_0.0.9.9000
[13] bit64_4.0.5 prodlim_2019.11.13 fansi_0.4.2
[16] mvtnorm_1.1-1 lubridate_1.7.9.2 codetools_0.2-16
[19] splines_4.0.2 knitr_1.31 jsonlite_1.7.2
[22] workflows_0.2.2 pROC_1.16.2 broom_0.7.6
[25] anytime_0.3.9 yardstick_0.0.8 tune_0.1.5
[28] compiler_4.0.2 httr_1.4.2 tictoc_1.0
[31] backports_1.2.1 assertthat_0.2.1 Matrix_1.2-18
[34] cli_2.5.0 htmltools_0.5.1.1 tools_4.0.2
[37] gtable_0.3.0 glue_1.4.2 dplyr_1.0.5
[40] Rcpp_1.0.6 DiceDesign_1.8-1 vctrs_0.3.8.9000
[43] debugme_1.1.0 iterators_1.0.13 timeDate_3043.102
[46] gower_0.2.2 xfun_0.22 stringr_1.4.0
[49] globals_0.14.0 lifecycle_1.0.0 qf_0.0.0.9999
[52] future_1.20.1 tsibble_0.9.3 MASS_7.3-51.6
[55] zoo_1.8-8 scales_1.1.1 ipred_0.9-9
[58] parallel_4.0.2 expm_0.999-4 yaml_2.2.1
[61] quantmod_0.4.17 curl_4.3 Exact_2.0
[64] memoise_1.1.0 ggplot2_3.3.3 rpart_4.1-15
[67] stringi_1.5.3 RSQLite_2.2.2 foreach_1.5.1
[70] TTR_0.23-6 lhs_1.1.1 warp_0.2.0
[73] boot_1.3-25 lava_1.6.8.1 rlang_0.4.11.9000
[76] pkgconfig_2.0.3 rsample_0.0.9 evaluate_0.14
[79] lattice_0.20-41 purrr_0.3.4 recipes_0.1.16
[82] bit_4.0.4 tidyselect_1.1.0 parallelly_1.21.0
[85] plyr_1.8.6 magrittr_2.0.1 R6_2.5.0
[88] DescTools_0.99.37 generics_0.1.0 DBI_1.1.1
[91] pillar_1.5.1 withr_2.3.0 xts_0.12.1
[94] RPushbullet_0.3.3.1 survival_3.1-12 nnet_7.3-14
[97] tibble_3.1.0 crayon_1.4.1 catchr_0.2.3
[100] utf8_1.2.1 AlpacaforR_1.0.0 rmarkdown_2.7.1
[103] workflowsets_0.0.2 grid_4.0.2 data.table_1.13.2
[106] blob_1.2.1 digest_0.6.27 tidyr_1.1.3
[109] GPfit_1.0-8 munsell_0.5.0
Hello,
I'm seeing several errors with dev rlang. We plan to release it in 2 to 4 weeks. With R CMD check I see these two:
check_and_clean_input(d1 = base::acosh, spec_names = "acosh")
#> Error in `call_name()`: `call` must be a simple call.
#> ℹ Calls to `::` or `:::` are not simple calls.
#> ℹ See `?is_call_simple`.
opts <- catchr_opts(
default_plan = c(collect, muffle),
drop_empty_conds = FALSE,
bare_if_possible = FALSE
)
plans <- make_plans(warning, message, error, .opts = opts)
res <- catch_expr(condition_thrower(), plans)
res2 <- catch_expr(dispense_collected(res), plans)
waldo::compare(res, res2)
#> `old$error[[1]]` is length 2
#> `new$error[[1]]` is length 3
#>
#> `names(old$error[[1]])`: "message" "call"
#> `names(new$error[[1]])`: "message" "call" "trace"
#>
#> `old$error[[1]]$trace` is absent
#> `new$error[[1]]$trace` is an S3 object of class <rlang_trace/rlib_trace/tbl/data.frame>, a list
The call_name()
error is a planned breakage, it now needs to be paired with is_call_simple()
.
I see more issues with interactive devtools::test()
but haven't investigated. Could you take a look and let me know if you think there is any bug in rlang that causes failures please?
Hi @burchill,
I've been thoroughly making use of catchr
thanks to the info you provided in the issue over on furrr
, thank you for cluing me into this useful package and for your dedicated effort in creating and maintaining it!
I've just encountered a somewhat bizarre error when using catchr
to surface message from background processes using the implementation you provided. The bang bang !!
operator just doesn't seem to work when catch
is used!
Here's a reprex to illustrate:
write_to_file <- function(cond) {
cond_class <- class(cond)[1]
msg <- paste(cond_class, ":", cond$message)
write(msg, file = "outlog", append=TRUE)
}
catch <- catchr::make_catch_fn(
warning = c(write_to_file, muffle),
message = c(write_to_file, muffle),
error = c(write_to_file, catchr::exit_with("Returned error!"))
)
cl <- parallel::makeCluster(2)
future::plan(future::cluster, workers = cl)
.d <- data.frame(time = rnorm(5), t2 = rnorm(5))
`!!` <- rlang::`!!`
`%>%` <- magrittr::`%>%`
# With normal furrr it's fine
furrr::future_imap(1:2, ~{
.t <- "time"
.ts <- rlang::sym(.t)
.d %>%
dplyr::select(!!.ts)
})
# But with catchr the bang bang causes an error
furrr::future_imap(1:2, ~catch({
.t <- "time"
.ts <- rlang::sym(.t)
.d %>%
dplyr::select(!!.ts)
}))
parallel::stopCluster(cl)
Error in enquo(expr) : object '.ts' not found
I should also mention that apparently there's a new immediateMessage
class in future
functionality that allows messages to be surfaced to the console from background processes. You might already know about it, but if you haven't I just recently tagged you again on the issue where you informed me about catchr
where Davis Vaughan and Henrik Bengtsson shared it with me. I feel like it's info that is likely useful you and this package.
Here's my sessionInfo JIC you need it:
R version 3.5.3 (2019-03-11)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 18362)
Matrix products: default
locale:
[1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252
[3] 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] RevoUtils_11.0.3 AlpacaforR_1.0.0 printr_0.1 dplyr_1.0.0 magrittr_1.5
[6] RevoUtilsMath_11.0.0
loaded via a namespace (and not attached):
[1] Rcpp_1.0.5.2 HDA_0.0.0.9000 catchr_0.2.2 pillar_1.4.6.9000 compiler_3.5.3
[6] tools_3.5.3 digest_0.6.25.1 packrat_0.5.0 bit_1.1-15.2 lubridate_1.7.9
[11] jsonlite_1.7.0 RSQLite_2.2.0.9000 memoise_1.1.0 lifecycle_0.2.0.9000 tibble_3.0.3.9000
[16] pkgconfig_2.0.3 rlang_0.4.7.9000 DBI_1.1.0 rstudioapi_0.11 parallel_3.5.3
[21] xfun_0.15.1 furrr_0.1.0.9002 httr_1.4.1 stringr_1.4.0 knitr_1.29.3
[26] globals_0.12.5 generics_0.0.2 vctrs_0.3.2 bit64_0.9-7 websocket_1.1.0
[31] tidyselect_1.1.0 glue_1.4.1.9000 listenv_0.8.0 R6_2.4.1 tidyr_1.1.0
[36] purrr_0.3.4.9000 blob_1.2.1 codetools_0.2-16 ellipsis_0.3.1 future_1.18.0
[41] stringi_1.4.7 crayon_1.3.4
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.