Merge branch 'master' into forder_segfault

Author: ben-schwen

.ci/.lintr.R

@@ -9,10 +9,10 @@ linters = c(dt_linters, all_linters(
   packages = "lintr", # TODO(lintr->3.2.0): Remove this.
   # eq_assignment_linter(),
   brace_linter(allow_single_line = TRUE),
+  implicit_integer_linter(allow_colon = TRUE),
   # TODO(michaelchirico): Activate these incrementally. These are the
   #   parameterizations that match our style guide.
   # implicit_assignment_linter(allow_lazy = TRUE, allow_scoped = TRUE),
-  # implicit_integer_linter(allow_colon = TRUE),
   # system_time_linter = undesirable_function_linter(c(
   #   system.time = "Only run timings in benchmark.Rraw"
   # )),
@@ -26,7 +26,8 @@ linters = c(dt_linters, all_linters(
   #   setwd = NULL
   # )),
   undesirable_operator_linter(),
-  # TODO(lintr#2441): Use upstream implementation.
+  # TODO(lintr#2765): Use upstream implementation.
+  # assignment_linter(operator = "="),
   assignment_linter = NULL,
   absolute_path_linter = NULL, # too many false positives
   # TODO(lintr#2442): Use this once x[ , j, by] is supported.
@@ -70,43 +71,39 @@ linters = c(dt_linters, all_linters(
 ))
 rm(dt_linters)
 
-# TODO(lintr#2172): Glob with lintr itself.
-exclusions = c(local({
-  exclusion_for_dir <- function(dir, exclusions) {
-    files = file.path("..", list.files(dir, pattern = "\\.(R|Rmd|Rraw)$", full.names=TRUE))
-    stats::setNames(rep(list(exclusions), length(files)), files)
-  }
-  c(
-    exclusion_for_dir("tests", list(
-      quotes_linter = Inf,
-      # TODO(michaelchirico): Enforce these and re-activate them one-by-one.
-      implicit_integer_linter = Inf,
-      infix_spaces_linter = Inf,
-      undesirable_function_linter = Inf
-    )),
-    exclusion_for_dir(c("vignettes", "vignettes/fr"), list(
-      quotes_linter = Inf,
-      sample_int_linter = Inf
-      # strings_as_factors_linter = Inf
-      # system_time_linter = Inf
-    )),
-    exclusion_for_dir("inst/tests", list(
-      library_call_linter = Inf,
-      numeric_leading_zero_linter = Inf,
-      undesirable_operator_linter = Inf, # For ':::', possibly we could be more careful to only exclude ':::'.
-      # TODO(michaelchirico): Enforce these and re-activate them one-by-one.
-      comparison_negation_linter = Inf,
-      condition_call_linter = Inf,
-      duplicate_argument_linter = Inf,
-      equals_na_linter = Inf,
-      missing_argument_linter = Inf,
-      paste_linter = Inf,
-      rep_len_linter = Inf,
-      sample_int_linter = Inf,
-      seq_linter = Inf,
-      unnecessary_lambda_linter = Inf
-    ))
+exclusions = list(
+  `../tests` = list(
+    quotes_linter = Inf,
+    # TODO(michaelchirico): Enforce these and re-activate them one-by-one.
+    implicit_integer_linter = Inf,
+    infix_spaces_linter = Inf,
+    undesirable_function_linter = Inf
+  ),
+  `../vignettes*` = list(
+    # assignment_linter = Inf,
+    implicit_integer_linter = Inf,
+    quotes_linter = Inf,
+    sample_int_linter = Inf
+    # strings_as_factors_linter = Inf
+    # system_time_linter = Inf
+  ),
+  `../inst/tests` = list(
+    library_call_linter = Inf,
+    numeric_leading_zero_linter = Inf,
+    undesirable_operator_linter = Inf, # For ':::', possibly we could be more careful to only exclude ':::'.
+    # TODO(michaelchirico): Enforce these and re-activate them one-by-one.
+    comparison_negation_linter = Inf,
+    condition_call_linter = Inf,
+    duplicate_argument_linter = Inf,
+    equals_na_linter = Inf,
+    missing_argument_linter = Inf,
+    paste_linter = Inf,
+    rep_len_linter = Inf,
+    sample_int_linter = Inf,
+    seq_linter = Inf,
+    unnecessary_lambda_linter = Inf
+  ),
+  `../inst/tests/froll.Rraw` = list(
+    dt_test_literal_linter = Inf # TODO(michaelchirico): Fix these once #5898, #5692, #5682, #5576, #5575, #5441 are merged.
   )
-}),
-  list(`../inst/tests/froll.Rraw` = list(dt_test_literal_linter = Inf)) # TODO(michaelchirico): Fix these once #5898, #5692, #5682, #5576, #5575, #5441 are merged.
 )

.ci/README.md

@@ -12,11 +12,15 @@ 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-310-cran` - R 3.1.0 on Linux, stated R dependency version.
+- `test-lin-ancient-cran` - Stated R dependency version (currently 3.3.0) 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.
 - `test-win-old` - `r-oldrel` on Windows.
-- `test-mac-rel` - macOS build not yet available, see [#3326](https://github.com/Rdatatable/data.table/issues/3326) for status
+- `test-mac-rel` - `r-release` on macOS.
+- `test-mac-old` - `r-oldrel` on macOS.
+
+The CI steps for the tests are [required](https://github.com/Rdatatable/data.table/blob/55eb0f160b169398d51f138131c14a66c86e5dc9/.ci/publish.R#L162-L168) to be named according to the pattern `test-(lin|win|mac)-<R version>[-<suffix>]*`, where `<R version>` is `rel`, `dev`, `old`, `ancient`, or three digits comprising an R version (e.g. `362` corresponding to R-3.6.2).
 
 Tests jobs are allowed to fail, summary and logs of test jobs are later published at _CRAN-like checks_ page, see artifacts below.
 

.ci/atime/tests.R

@@ -231,7 +231,20 @@ test.list <- atime::atime_test_list(
     },
     expr = data.table:::melt(DT, measure.vars = measure.vars),
     Slow = "fd24a3105953f7785ea7414678ed8e04524e6955", # Parent of the merge commit (https://github.com/Rdatatable/data.table/commit/ed72e398df76a0fcfd134a4ad92356690e4210ea) of the PR (https://github.com/Rdatatable/data.table/pull/5054) that fixes the issue
-    Fast = "ed72e398df76a0fcfd134a4ad92356690e4210ea"), # Merge commit of the PR (https://github.com/Rdatatable/data.table/pull/5054) that fixes the issue
+    Fast = "ed72e398df76a0fcfd134a4ad92356690e4210ea"), # Merge commit of the PR (https://github.com/Rdatatable/data.table/pull/5054) that fixes the issue  # Test case created directly using the atime code below (not adapted from any other benchmark), based on the issue/fix PR https://github.com/Rdatatable/data.table/pull/5054#issue-930603663 "melt should be more efficient when there are missing input columns."
+
+  # Test case created from @tdhock's comment https://github.com/Rdatatable/data.table/pull/6393#issuecomment-2327396833, in turn adapted from @philippechataignon's comment https://github.com/Rdatatable/data.table/pull/6393#issuecomment-2326714012
+  "fwrite refactored in #6393" = atime::atime_test(
+    setup = {
+      set.seed(1)
+      NC = 10L
+      L <- data.table(i=1:N)
+      L[, paste0("V", 1:NC) := replicate(NC, rnorm(N), simplify=FALSE)]
+      out.csv <- tempfile()
+    },
+    expr = data.table::fwrite(L, out.csv, compress="gzip"),
+    Before = "f339aa64c426a9cd7cf2fcb13d91fc4ed353cd31", # Parent of the first commit https://github.com/Rdatatable/data.table/commit/fcc10d73a20837d0f1ad3278ee9168473afa5ff1 in the PR https://github.com/Rdatatable/data.table/pull/6393/commits with major change to fwrite with gzip.
+    PR = "3630413ae493a5a61b06c50e80d166924d2ef89a"), # Close-to-last merge commit in the PR.
 
   tests=extra.test.list)
 # nolint end: undesirable_operator_linter.

.ci/linters/c/alloc_linter.R

@@ -4,8 +4,9 @@
 #   2. Check the next line for a check like 'if (!x || !y)'
 alloc_linter = function(c_obj) {
   lines = c_obj$lines
-  # Be a bit more precise to avoid mentions in comments
-  alloc_lines = grep(R"{=\s*([(]\w+\s*[*][)])?[mc]alloc[(]}", lines)
+  # Be a bit more precise to avoid mentions in comments, and allow
+  #   malloc(0) to be used for convenience (e.g. #6757)
+  alloc_lines = grep(R"{=\s*([(]\w+\s*[*][)])?[mc]alloc[(][^0]}", lines)
   if (!length(alloc_lines)) return()
   # int *tmp=(int*)malloc(...); or just int tmp=malloc(...);
   alloc_keys = lines[alloc_lines] |>

.ci/linters/c/cocci_linter.R

@@ -0,0 +1,20 @@
+cocci_linter = if (!nzchar(Sys.which("spatch"))) function(...) {} else function(c_obj) {
+  bad <- FALSE
+  for (spfile in list.files(".ci/linters/cocci", full.names = TRUE)) {
+    # Coccinelle parser gets confused sometimes, so ignore stderr and the exit code
+    out = suppressWarnings(system2(
+      "spatch",
+      shQuote(c(
+        "--sp-file", spfile, c_obj$path, "--recursive-includes",
+        "-I", R.home("include"), "-I", "src"
+      )),
+      stdout = TRUE, stderr = FALSE
+    ))
+    if (length(out) > 0) {
+      cat(sprintf("In file '%s', Coccinelle patch '%s' recommends the following changes:\n", c_obj$path, spfile))
+      writeLines(out)
+      bad <- TRUE
+    }
+  }
+  if (bad) stop("Please apply the changes above or fix the linter")
+}

.ci/linters/cocci/redundant_cast_to_same_pointer_type.cocci

@@ -0,0 +1,6 @@
+@@
+type T;
+T* E;
+@@
+- (T*)
+  E

.ci/linters/md/heading_id_linter.R

@@ -0,0 +1,26 @@
+any_mismatch = FALSE
+
+# ensure that ids are limited to alphanumerics and dashes
+# (in particular, dots and underscores break the links)
+check_header_ids = function(md) {
+  # A bit surprisingly, some headings don't start with a letter.
+  # We're interested in those that set an id to link to, i.e., end with {#id}.
+  heading_captures = regmatches(md, regexec("^#+ \\S.*[{]#([^}]*)[}]$", md))
+  lines_with_id = which(lengths(heading_captures) > 0)
+  ids = vapply(heading_captures[lines_with_id], `[`, '', 2)
+  # ids must start with a letter and consist of alphanumerics or dashes.
+  good_ids = grepl('^[A-Za-z][A-Za-z0-9-]*$', ids)
+  for (line in lines_with_id[!good_ids]) cat(sprintf(
+    "On line %d, bad heading id '%s':\n%s\n",
+    line, heading_captures[[line]][2], heading_captures[[line]][1]
+  ))
+  !all(good_ids)
+}
+
+any_error = FALSE
+for (vignette in list.files('vignettes', pattern = "[.]Rmd$", recursive = TRUE, full.name = TRUE)) {
+  cat(sprintf("Checking vignette file %s...\n", vignette))
+  rmd_lines = readLines(vignette)
+  any_error = check_header_ids(rmd_lines) || any_error
+}
+if (any_error) stop("Please fix the vignette issues above.")

.dev/CRAN_Release.cmd

@@ -3,22 +3,22 @@
 ###############################################
 
 # 1) Update messages for new release
-## (a) Update C template file: src/data.table.pot
-##     ideally, we are including _() wrapping in
-##     new PRs throughout dev cycle, and this step
-##     becomes about tying up loose ends
-## Check the output here for translatable messages
-xgettext -o /dev/stdout ./*.c \
-  --keyword=Rprintf --keyword=error --keyword=warning --keyword=STOP --keyword=DTWARN --keyword=Error --keyword=DTPRINT --keyword=snprintf:3
-
-## (b) Update R template file: src/R-data.table.pot
-##  NB: this relies on R >= 4.0 to remove a bug in update_pkg_po
-Rscript -e "tools::update_pkg_po('.')"
+dt_custom_translators = list(
+  R = 'catf:fmt|1',
+  # TODO(MichaelChirico/potools#318): restore snprintf:3 here too
+  src = c('STOP:1', 'DTWARN:1', 'DTPRINT:1')
+)
+message_db =
+  potools::get_message_data(custom_translation_functions = dt_custom_translators)
+potools::check_cracked_messages(message_db)
+potools::check_untranslated_cat(message_db)
+potools::check_untranslated_src(message_db)
+
+## (b) Update R template files (po/*.pot)
+potools::po_extract(custom_translation_functions = dt_custom_translators)
 
 # 2) Open a PR with the new templates & contact the translators
-#   * zh_CN: @hongyuanjia
-#   * pt_BR: @rffontenelle
-#   * es: @rikivillalba
+#    using @Rdatatable/<lang>, e.g. @Rdatatable/chinese
 ## Translators to submit commits with translations to this PR
 ##   [or perhaps, if we get several languages, each to open
 ##    its own PR and merge to main translation PR]
@@ -114,12 +114,6 @@ grep -P "\t" ./src/*.c
 grep -n "[^A-Za-z0-9]T[^A-Za-z0-9]" ./inst/tests/tests.Rraw
 grep -n "[^A-Za-z0-9]F[^A-Za-z0-9]" ./inst/tests/tests.Rraw
 
-# All integers internally should have L suffix to avoid lots of one-item coercions
-# Where 0 numeric is intended we should perhaps use 0.0 for clarity and make the grep easier
-# 1) tolerance=0 usages in setops.R are valid numeric 0, as are anything in strings
-# 2) leave the rollends default using roll>=0 though; comments in PR #3803
-grep -Enr "^[^#]*(?:\[|==|>|<|>=|<=|,|\(|\+)\s*[-]?[0-9]+[^0-9L:.e]" R | grep -Ev "stop|warning|tolerance"
-
 # Never use ifelse. fifelse for vectors when necessary (nothing yet)
 grep -Enr "\bifelse" R
 
@@ -135,10 +129,6 @@ grep -Fn "tryCatch" ./inst/tests/*.Rraw
 # All % in *.Rd should be escaped otherwise text gets silently chopped
 grep -n "[^\]%" ./man/*.Rd
 
-# if (a & b) is either invalid or inefficient (ditto for replace & with |);
-#   if(any(a [&|] b)) is appropriate b/c of collapsing the logical vector to scalar
-grep -nr "^[^#]*if[^&#]*[^&#\"][&][^&]" R | grep -Ev "if\s*[(](?:any|all)"
-
 # seal leak potential where two unprotected API calls are passed to the same
 # function call, usually involving install() or mkChar()
 # Greppable thanks to single lines and wide screens

.dev/lsan.supp

@@ -0,0 +1 @@
+leak:libfontconfig.so

.devcontainer/r-devel-alpine/Dockerfile

@@ -0,0 +1,22 @@
+FROM docker.io/rhub/r-minimal:devel
+
+RUN apk update \
+  && apk add --no-cache \
+     gcc git musl-dev openmp pkgconf tzdata zlib-dev \
+  && echo 'options("repos"="https://cloud.r-project.org")' >> /usr/local/lib/R/etc/Rprofile.site
+
+ENV TZDIR=/usr/share/zoneinfo
+
+COPY DESCRIPTION .
+
+RUN Rscript -e '                                          \ 
+read.dcf("DESCRIPTION", c("Imports", "Suggests")) |>      \
+  tools:::.split_dependencies() |>                        \
+  names() |>                                              \
+  setdiff(tools:::.get_standard_package_names()$base) |>  \
+  install.packages(repos="https://cloud.r-project.org")   \
+'
+
+# setup cc()
+WORKDIR /root
+COPY .devcontainer/.Rprofile .