[Discussion] dtplyr dependency

[artemklevtsov]

Hi.

What is reason to use the dtplyr package in Imports? Really needed only data.table::rbindlist. Let's see a dtplyr dependencies:

R> miniCRAN::pkgDep("dtplyr", availPkgs = available.packages(), suggests = FALSE)
 [1] "dtplyr"     "dplyr"      "data.table" "lazyeval"   "assertthat" "R6"         "Rcpp"       "tibble"    
 [9] "magrittr"   "DBI"        "BH"

but data.table haven't any dependncies.

In fact dtplyr::tbl_dt only sets c("tbl_dt", "tbl", "data.table", "data.frame") as class attribute.

[hrbrmstr]

I despise data.table hieroglyphs syntax (they are the antithesis of literate programming) and hold out hope that various volunteers will take up the maintenance of the dtplyr pkg. I'm also a fan of the tidyverse and those pkgs come for the ride there. It is open source so you are free to fork and do what you like. The next CRAN release will also enable selecting between returning a tbl_dt or tbl_df so said dependencies aren't going away anyway. On Tue, Feb 21, 2017 at 20:14 Artem Klevtsov notifications@github.com wrote:

Hi.

What is reason to use the dtplyr package in Imports? Really needed only data.table::rbindlist. Let's see a dtplyr dependencies:

R> miniCRAN::pkgDep("dtplyr", availPkgs = available.packages(), suggests = FALSE) [1] "dtplyr" "dplyr" "data.table" "lazyeval" "assertthat" "R6" "Rcpp" "tibble" [9] "magrittr" "DBI" "BH"

but data.table haven't any dependncies.

In fact dtplyr::tbl_dt only sets c("tbl_dt", "tbl", "data.table", "data.frame") as class attribute.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/hrbrmstr/ndjson/issues/1, or mute the thread https://github.com/notifications/unsubscribe-auth/AAfHtvH3pUP9RaIdLzq2hKJrW-iNtyFmks5re4vigaJpZM4MIFJ- .