Bump R dependency to 3.4.0 (#6975)

* Bump R dependency to 3.4.0

* rm declaration of routine

* remove registration too

Author: MichaelChirico

.ci/README.md

@@ -12,7 +12,7 @@ Test jobs:
 - `test-lin-rel-cran` - `--as-cran` on Linux, strict test for final status of `R CMD check`.
 - `test-lin-dev-gcc-strict-cran` - `--as-cran` on Linux, `r-devel` built with `-enable-strict-barrier --disable-long-double`, test for compilation warnings, test for new NOTEs/WARNINGs from `R CMD check`.
 - `test-lin-dev-clang-cran` - same as `gcc-strict` job but R built with `clang` and  no `--enable-strict-barrier --disable-long-double` flags.
-- `test-lin-ancient-cran` - Stated R dependency version (currently 3.3.0) on Linux.
+- `test-lin-ancient-cran` - Stated R dependency version (currently 3.4.0) on Linux.
 - `test-lin-dev-san` - `r-devel` on Linux built with `clang -fsanitize=address,undefined` (including LeakSanitizer), test for sanitizer output in tests and examples.
 - `test-win-rel` - `r-release` on Windows.
 - `test-win-dev` - `r-devel` on Windows.

.dev/CRAN_Release.cmd

@@ -235,23 +235,23 @@ system.time(test.data.table(script="*.Rraw"))  # apx 8h = froll 3h + nafill 1m +
 
 
 ###############################################
-#  R 3.3.0 (stated dependency)
+#  R 3.4.0 (stated dependency)
 ###############################################
 
 ### ONE TIME BUILD
 sudo apt-get -y build-dep r-base
 cd ~/build
-wget http://cran.stat.ucla.edu/src/base/R-3/R-3.3.0.tar.gz
-tar xvf R-3.3.0.tar.gz
-cd R-3.3.0
+wget http://cran.stat.ucla.edu/src/base/R-3/R-3.4.0.tar.gz
+tar xvf R-3.4.0.tar.gz
+cd R-3.4.0
 CFLAGS="-fcommon" FFLAGS="-fallow-argument-mismatch" ./configure --without-recommended-packages
 make
-alias R330=~/build/R-3.3.0/bin/R
+alias R340=~/build/R-3.4.0/bin/R
 ### END ONE TIME BUILD
 
 cd ~/GitHub/data.table
-R330 CMD INSTALL ./data.table_1.16.99.tar.gz
-R330
+R340 CMD INSTALL ./data.table_1.16.99.tar.gz
+R340
 require(data.table)
 test.data.table(script="*.Rraw")
 

.devcontainer/r-ancient-gcc/Dockerfile

@@ -1,4 +1,4 @@
-FROM registry.gitlab.com/jangorecki/dockerfiles/r-3.3.0
+FROM registry.gitlab.com/jangorecki/dockerfiles/r-3.4.0
 
 RUN apt-get -qq update \
   && apt-get install -y --no-install-recommends git

.github/workflows/R-CMD-check-occasional.yaml

@@ -15,7 +15,7 @@ jobs:
       fail-fast: false
       matrix:
         os: [macOS-latest, windows-latest, ubuntu-latest]
-        r: ['devel', 'release', '3.3', '3.4', '3.5', '3.6', '4.0', '4.1', '4.2', '4.3']
+        r: ['devel', 'release', '3.4', '3.5', '3.6', '4.0', '4.1', '4.2', '4.3']
         locale: ['en_US.utf8', 'zh_CN.utf8', 'lv_LV.utf8'] # Chinese for translations, Latvian for collate order (#3502)
         exclude:
           # only run non-English locale CI on Ubuntu
@@ -28,8 +28,6 @@ jobs:
           - os: windows-latest
             locale: 'lv_LV.utf8'
           # macOS/arm64 only available for R>=4.1.0
-          - os: macOS-latest
-            r: '3.3'
           - os: macOS-latest
             r: '3.4'
           - os: macOS-latest

.gitlab-ci.yml

@@ -203,13 +203,13 @@ test-lin-dev-clang-cran:
 # stated dependency on R
 test-lin-ancient-cran:
   <<: *test-lin
-  image: registry.gitlab.com/rdatatable/dockerfiles/r-3.3.0
+  image: registry.gitlab.com/rdatatable/dockerfiles/r-3.4.0
   variables:
     _R_CHECK_FORCE_SUGGESTS_: "FALSE" # can be removed if all dependencies are available (knitr, xts, etc.)
   script:
     - *install-deps
     # knitr requires evaluate, which requires R 3.6.0.
-    # Restore checking vignettes if upgrading our R dependency means knitr can be installed.
+    # Restore checking vignettes if upgrading our R dependency means knitr can be installed, or when we switch to litedown.
     - R CMD check --no-manual --no-build-vignettes --ignore-vignettes $(ls -1t data.table_*.tar.gz | head -n 1)
 
 # run the main checks with Address(+Leak),UBSanitizer enabled

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: data.table
 Version: 1.17.99
 Title: Extension of `data.frame`
-Depends: R (>= 3.3.0)
+Depends: R (>= 3.4.0)
 Imports: methods
 Suggests: bit64 (>= 4.0.0), bit (>= 4.0.4), R.utils, xts, zoo (>= 1.8-1), yaml, knitr, markdown
 Description: Fast aggregation of large data (e.g. 100GB in RAM), fast ordered joins, fast add/modify/delete of columns by group using no copies at all, list columns, friendly and fast character-separated-value read/write. Offers a natural and flexible syntax, for faster development.

NEWS.md

@@ -41,6 +41,8 @@
    + Argument `in.place` to `droplevels` has been removed.
    + It's now an error to set `datatable.nomatch`, which has been warning since 1.15.0.
 
+3. {data.table} now depends on R 3.4.0 (2017).
+
 # data.table [v1.17.0](https://github.com/Rdatatable/data.table/milestone/34)  (20 Feb 2025)
 
 ## POTENTIALLY BREAKING CHANGES

R/onAttach.R

@@ -39,10 +39,6 @@
       else
         packageStartupMessagef("This is %s. This warning should not normally occur on Windows or Linux where OpenMP is turned on by data.table's configure script by passing -fopenmp to the compiler. If you see this warning on Windows or Linux, please file a GitHub issue.\n**********", Sys.info()["sysname"])
     }
-    if (.Call(CbeforeR340)) {
-      # not base::getRversion()<"3.4.0" in case the user upgrades R but does not reinstall data.table; a reasonable mistake since data.table would seem to be the latest version
-      packageStartupMessagef("**********\nThis data.table installation was compiled for R < 3.4.0 (Apr 2017) and is known to leak memory. Please upgrade R and reinstall data.table to fix the leak. Maintaining and testing code branches to support very old versions increases development time so please do upgrade R. We intend to bump data.table's dependency from 8 year old R 3.1.0 (Apr 2014) to 5 year old R 3.4.0 (Apr 2017).\n**********")
-    }
   }
 }
 

R/print.data.table.R

@@ -253,7 +253,6 @@ char.trunc = function(x, trunc.char = getOption("datatable.prettyprint.char")) {
   nchar_chars = nchar(x, 'char')
   is_full_width = nchar_width > nchar_chars
   idx = !is.na(x) & pmin(nchar_width, nchar_chars) > trunc.char
-  if (!any(idx)) return(x) # strtrim() errors for width=integer() on R 3.3.0
   x[idx] = paste0(strtrim(x[idx], trunc.char * fifelse(is_full_width[idx], 2L, 1L)), "...")
   x
 }

inst/tests/tests.Rraw

@@ -200,16 +200,6 @@ base_messages = list(
   NULL
 )
 
-# Ensure an operation uses C-locale sorting (#3502). For test set-ups/comparisons that use base operations, which are
-#   susceptible to locale-specific sorting issues, but shouldn't be needed for data.table code, which always uses C sorting.
-# TODO(R>=3.3.0): use order(method="radix") as a way to avoid needing this helper
-with_c_collate = function(expr) {
-  old = Sys.getlocale("LC_COLLATE")
-  on.exit(Sys.setlocale("LC_COLLATE", old))
-  Sys.setlocale("LC_COLLATE", "C")
-  expr
-}
-
 ##########################
 .do_not_rm = ls()  # objects that exist at this point should not be removed by rm_all(); e.g. test_*, base_messages, Ctest_dt_win_snprintf, prevtest, etc
 ##########################
@@ -1841,10 +1831,10 @@ test(609, chorder(character()), base::order(character()))
 test(610, chorder(""), base::order(""))
 # Extra tests of chorder and chgroup
 x = sample(LETTERS)
-test(610.1, chorder(x), with_c_collate(base::order(x)))
+test(610.1, chorder(x), base::order(x, method="radix"))
 test(610.2, chgroup(x), seq_along(x))
 x = sample(LETTERS,1000,replace=TRUE)
-test(610.3, chorder(x), with_c_collate(base::order(x)))
+test(610.3, chorder(x), base::order(x, method="radix"))
 test(610.4, unique(x[chgroup(x)]), unique(x))
 
 # := by group
@@ -4516,7 +4506,9 @@ for (nvars in seq_along(names(DT))) {
   combn(names(DT), nvars, simplify=FALSE, function(x) {  # simplify=FALSE needed for R 3.1.0
     for (i in seq_len(nrow(signs))) {
       test_no <<- test_no + 1L
-      ll = as.call(c(as.name("order"),
+      ll = as.call(c(
+        as.name("order"),
+        method="radix",
         lapply(seq_along(x), function(j) {
           if (signs[i,j] == 1L)
             as.name(x[j])
@@ -4528,7 +4520,7 @@ for (nvars in seq_along(names(DT))) {
           }
         })
       ))
-      test(1223.0 + test_no*0.001, forderv(DT, by=x, order=signs[i,]), with_c_collate(with(DT, eval(ll))))
+      test(1223.0 + test_no*0.001, forderv(DT, by=x, order=signs[i,]), with(DT, eval(ll)))
     }
     integer()
   })
@@ -4609,18 +4601,8 @@ test(1240.2, ans2, ans1)
 # i.e. a three-way match.
 base_order <- function(..., na.last=TRUE, method=c("shell","radix")) {
   ans1 = base::order(..., na.last=na.last, method="shell")
-  if (!is.na(na.last) || base::getRversion()>"3.3.3") {
-    ans2 = base::order(..., na.last=na.last, method="radix")
-    if (!identical(ans1,ans2)) stop("Base R's order(,method='shell') != order(,method='radix')")
-  } else {
-    # Only when na.last=NA in just R 3.3.0-3.3.3 we don't check shell==radix
-    # because there was a problem in base R's port of data.table code then when :
-    #     1) 2 or more vectors were passed to base::order(,method="radix")
-    # AND 2) na.last=NA
-    # AND 3) there is a subgroup of size exactly 2
-    # AND 4) one of those 2 items in the subgroup is NA and the other is not NA
-    # See tests 1728.3 and 1728.13.
-  }
+  ans2 = base::order(..., na.last=na.last, method="radix")
+  if (!identical(ans1, ans2)) stop("Base R's order(,method='shell') != order(,method='radix')")
   ans1
 }
 
@@ -4780,7 +4762,9 @@ for (i in seq_along(names(DT))) {
   ans[[i]] = combn(names(DT), i, function(x) {
     tmp = apply(cj, 1, function(y) {
       test_no <<- test_no + 1L
-      ll = as.call(c(as.name("base_order"),
+      ll = as.call(c(
+        as.name("base_order"),
+        method = "radix",
         lapply(seq_along(x), function(j) {
           if (y[j] == 1L)
             as.name(x[j])
@@ -4793,11 +4777,11 @@ for (i in seq_along(names(DT))) {
         })
       ))
       ans1 = forderv(DT, by=x, order=y, na.last=TRUE)         # adding tests for both nalast=TRUE and nalast=NA
-      test(1252.0 + test_no*0.001, ans1, with_c_collate(with(DT, eval(ll))))
+      test(1252.0 + test_no*0.001, ans1, with(DT, eval(ll)))
       test_no <<- test_no + 1L
       ll <- as.call(c(as.list(ll), na.last=NA))
       ans1 = forderv(DT, by=x, order=y, na.last=NA)           # nalast=NA here.
-      test(1252.0 + test_no*0.001, ans1[ans1 != 0], with_c_collate(with(DT, eval(ll))))
+      test(1252.0 + test_no*0.001, ans1[ans1 != 0], with(DT, eval(ll)))
     })
     dim(tmp)=NULL
     list(tmp)
@@ -12402,10 +12386,7 @@ DT1 = data.table(x = mixed_strings, y = c(latin1_strings, utf8_strings), z = 1:1
 DT2 = copy(DT1)
 setkey(DT1, x)
 setkey(DT2, y)
-# the ans is generated by `sort(c(utf8_strings, utf8_strings), method = "radix")`
-# but we should not use radix sort in the test because it's introduced after R3.3.0
-ans = c("El. pa\u00c5\u00a1tas", "El. pa\u00c5\u00a1tas", "fa\u00e7ile", "fa\u00e7ile",
-"\u00a1tas", "\u00a1tas", "\u00de", "\u00de", "\u00e7ile", "\u00e7ile")
+ans = sort(c(utf8_strings, utf8_strings), method = "radix")
 test(1864.1, DT1$x, ans)
 test(1864.2, DT2$y, ans)
 ans = c(1L, 6L, 2L, 7L, 3L, 8L, 4L, 9L, 5L, 10L)
@@ -17789,12 +17770,9 @@ x = factor(letters[1:10])
 DT = data.table(a = x)[1:5]
 test(2214.01, fdroplevels(factor()), droplevels(factor()))
 test(2214.02, fdroplevels(x[1:5]), droplevels(x[1:5]))
-if (base::getRversion() >= "3.4.0") {
-  # bug fix in R 3.4.0: "droplevels(f) now keeps <NA> levels when present."
-  test(2214.03, fdroplevels(x[1:5], c("b", "d")), droplevels(x[1:5], c("b", "d")))
-  test(2214.04, fdroplevels(x[1:5], letters[1:5]), droplevels(x[1:5], letters[1:5]))
-  test(2214.05, droplevels(DT, exclude=c("b", "d"))[["a"]], droplevels(DT[1:5,a], c("b", "d")))
-}
+test(2214.03, fdroplevels(x[1:5], c("b", "d")), droplevels(x[1:5], c("b", "d")))
+test(2214.04, fdroplevels(x[1:5], letters[1:5]), droplevels(x[1:5], letters[1:5]))
+test(2214.05, droplevels(DT, exclude=c("b", "d"))[["a"]], droplevels(DT[1:5,a], c("b", "d")))
 test(2214.06, droplevels(DT)[["a"]], droplevels(DT[1:5,a]))
 test(2214.07, droplevels(DT, 1)[["a"]], x[1:5])
 # 2214.08 was of defunct in.place argument