Merge branch 'master' into redundantParameterCleanup

Author: badasahog

.ci/linters/rd/backtick_linter.R

@@ -0,0 +1,22 @@
+# ensure no markdown-style backticks wind up in Rd where \code is intended
+options_documentation_linter = function(rd_file) {
+  rd = tools::parse_Rd(rd_file)
+
+  error_if_backtick = function(rd_obj) {
+    if (!is.recursive(rd_obj)) {
+      if (any(grepl("`", rd_obj, fixed=TRUE))) {
+        stop(sprintf(
+          "Rd is not markdown -- backticks (`) don't render as code! Use \\code{...}.\nObserved in string '%s' in file %s",
+          trimws(rd_obj), rd_file
+        ))
+      }
+      return(invisible())
+    }
+    tags = vapply(rd_obj, \(x) attr(x, "Rd_tag") %||% "", FUN.VALUE="")
+     # backtick is valid inside R code (e.g. \examples, \code, \preformatted)
+    rd_obj = rd_obj[!tags %in% c("RCODE", "VERB")]
+    lapply(rd_obj, error_if_backtick)
+  }
+
+  invisible(error_if_backtick(rd))
+}

.ci/linters/rd/options_doc_check.R …inters/rd/options_documentation_linter.R.ci/linters/rd/options_doc_check.R renamed to .ci/linters/rd/options_documentation_linter.R .ci/linters/rd/options_doc_check.R renamed to .ci/linters/rd/options_documentation_linter.R

@@ -562,9 +562,9 @@ ls -1 *.tar.gz | grep -E 'Chicago|dada2|flowWorkspace|LymphoSeq' | TZ='UTC' para
 #  3) dllVersion() at the end of init.c
 # DO NOT push to GitHub's master branch. Prevents even a slim possibility of user getting premature version. 
 # Even release numbers must have been obtained from CRAN and only CRAN. There were too many support problems in the past before this procedure was brought in.
-du -k inst/tests                # 1.5MB before
+du -k inst/tests                # 1.5MiB before
 bzip2 inst/tests/*.Rraw         # compress *.Rraw just for release to CRAN; do not commit compressed *.Rraw to git
-du -k inst/tests                # 0.75MB after
+du -k inst/tests                # 0.75MiB after
 R CMD build .
 export GITHUB_PAT="f1c.. github personal access token ..7ad"
 Rdevel -q -e "packageVersion('xml2')"   # ensure installed

.dev/CRAN_Release.cmd

@@ -67,7 +67,7 @@ options(repos = BiocManager::repositories())
 
 options(warn=1)  # warning at the time so we can more easily see what's going on package by package when we scroll through output
 cat("options()$timeout==", options()$timeout," set by R_DEFAULT_INTERNET_TIMEOUT in .dev/.bash_aliases revdepsh\n",sep="")
-# R's default is 60. Before Dec 2020, we used 300 but that wasn't enough to download Bioc package BSgenome.Hsapiens.UCSC.hg19 (677GB) which is
+# R's default is 60. Before Dec 2020, we used 300 but that wasn't enough to download Bioc package BSgenome.Hsapiens.UCSC.hg19 (677MiB) which is
 # suggested by CRAN package CNVScope which imports data.table. From Dec 2020 we use 3600.
 
 if (is.null(utils::old.packages(.libPaths()[2]))) {

.dev/revdep.R

@@ -151,7 +151,7 @@ test-lin-rel-cran:
     _R_CHECK_CRAN_INCOMING_: "TRUE"           ## stricter --as-cran checks should run in dev pipelines continuously (not sure what they are though)
     _R_CHECK_CRAN_INCOMING_REMOTE_: "FALSE"   ## Other than no URL checking (takes many minutes) or 'Days since last update 0' NOTEs needed, #3284
     _R_CHECK_CRAN_INCOMING_TARBALL_THRESHOLD_: "7500000" ## bytes
-    _R_CHECK_PKG_SIZES_THRESHOLD_: "10"        ## MB 'checking installed package size' NOTE increased due to po
+    _R_CHECK_PKG_SIZES_THRESHOLD_: "10"        ## MiB 'checking installed package size' NOTE increased due to po
   script:
     - *install-deps
     - echo 'CFLAGS=-g -O2 -fopenmp -Wall -pedantic -fstack-protector-strong -D_FORTIFY_SOURCE=2' > ~/.R/Makevars

.gitlab-ci.yml

@@ -82,6 +82,8 @@
 
 19. Spurious warnings from internal code in `cube()`, `rollup()`, and `groupingsets()` are no longer surfaced to the caller, [#6964](https://github.com/Rdatatable/data.table/issues/6964). Thanks @ferenci-tamas for the report and @venom1204 for the fix.
  
+20. `droplevels()` works on 0-row data.tables, [#7043](https://github.com/Rdatatable/data.table/issues/7043). The result will have factor columns `factor(character())`, consistent with the data.frame method. Thanks @advieser for the report and @MichaelChirico for the fix.
+
 ### NOTES
 
 1. Continued work to remove non-API C functions, [#6180](https://github.com/Rdatatable/data.table/issues/6180). Thanks Ivan Krylov for the PRs and for writing a clear and concise guide about the R API: https://aitap.codeberg.page/R-api/.

NEWS.md

@@ -2237,7 +2237,7 @@ tail.data.table = function(x, n=6L, ...) {
 
 "[<-.data.table" = function(x, i, j, value) {
   # [<- is provided for consistency, but := is preferred as it allows by group and by reference to subsets of columns
-  # with no copy of the (very large, say 10GB) columns at all. := is like an UPDATE in SQL and we like and want two symbols to change.
+  # with no copy of the (very large, say 10GiB) columns at all. := is like an UPDATE in SQL and we like and want two symbols to change.
   if (!cedta()) {
     x = if (nargs()<4L) `[<-.data.frame`(x, i, value=value)
         else `[<-.data.frame`(x, i, j, value)

R/data.table.R

@@ -49,7 +49,7 @@ unique.data.table = function(x, incomparables=FALSE, fromLast=FALSE, by=seq_alon
 
 # Test for #2013 unique() memory efficiency improvement in v1.10.5
 # set.seed(1)
-# Create unique 7.6GB DT on 16GB laptop
+# Create unique 7.6GiB DT on 16GiB laptop
 # DT = data.table(
 #  A = sample(1e8, 2e8, TRUE),
 #  B = sample(1e8, 2e8, TRUE),

R/duplicated.R

@@ -1,6 +1,7 @@
 # 647 fast droplevels.data.table method
 fdroplevels = function(x, exclude = if (anyNA(levels(x))) NULL else NA, ...) {
   stopifnot(inherits(x, "factor"))
+  if (!length(x)) return(structure(integer(), class='factor', levels=character())) # skip factor() overhead
   lev = which(tabulate(x, nlevels(x)) & (!match(levels(x), exclude, 0L)))
   ans = match(as.integer(x), lev)
   setattr(ans, 'levels', levels(x)[lev])
@@ -15,7 +16,6 @@ droplevels.data.table = function(x, except=NULL, exclude, ...){
 }
 
 setdroplevels = function(x, except=NULL, exclude=NULL) {
-  if (!nrow(x)) return(invisible(x))
   ix = vapply_1b(x, is.factor)
   if (!is.null(except)) {
     stopifnot(is.numeric(except), except >= 1L, except <= length(x))

R/fdroplevels.R

@@ -277,13 +277,13 @@ test.data.table = function(script="tests.Rraw", verbose=FALSE, pkg=".", silent=F
     y = head(order(-diff(timings$RSS)), 10L)
     ans = timings[, diff := c(NA_real_, round(diff(RSS), 1L))][y + 1L]
     ans[, time:=NULL]  # time is distracting and influenced by gc() calls; just focus on RAM usage here
-    catf("10 largest RAM increases (MB); see plot for cumulative effect (if any)\n")
+    catf("10 largest RAM increases (MiB); see plot for cumulative effect (if any)\n")
     print(ans, class=FALSE)
     get("dev.new")(width=14.0, height=7.0)
     get("par")(mfrow=1:2)
-    get("plot")(timings$RSS, main=paste(basename(fn),"\nylim[0]=0 for context"), ylab="RSS (MB)", ylim=c(0.0, max(timings$RSS)))
+    get("plot")(timings$RSS, main=paste(basename(fn),"\nylim[0]=0 for context"), ylab="RSS (MiB)", ylim=c(0.0, max(timings$RSS)))
     get("mtext")(lastRSS<-as.integer(ceiling(last(timings$RSS))), side=4L, at=lastRSS, las=1L, font=2L)
-    get("plot")(timings$RSS, main=paste(basename(fn),"\nylim=range for inspection"), ylab="RSS (MB)")
+    get("plot")(timings$RSS, main=paste(basename(fn),"\nylim=range for inspection"), ylab="RSS (MiB)")
     get("mtext")(lastRSS, side=4L, at=lastRSS, las=1L, font=2L)
   }
 
@@ -316,7 +316,7 @@ INT = function(...) { as.integer(c(...)) }   # utility used in tests.Rraw
 
 gc_mem = function() {
   # nocov start
-  # gc reports memory in MB
+  # gc reports memory in MiB
   m = colSums(gc()[, c(2L, 4L, 6L)])
   names(m) = c("GC_used", "GC_gc_trigger", "GC_max_used")
   m