Adds optional column count display to print.data.table()

NEWS.md

@@ -34,6 +34,8 @@
 
 7. Rows can now be deleted by reference using `DT[i, .ROW := NULL]`, avoiding a full copy of the table for large row-removal operations, [#635](https://github.com/Rdatatable/data.table/issues/635). This has been one of data.table's most requested features. Target rows must be selected with the `i` expression, `by`/`keyby` are not supported, and keys/indices are cleared after deletion. The new experimental helper `setallocrow()` prepares columns for by-reference row operations. Thanks @arunsrinivasan for the feature request, @ben-schwen for the implementation, and @aitap for review and assistance.
 
+8. `print.data.table()` gains a `show.ncols` argument and `datatable.show.ncols` option to print the number of columns (labeled as `ncol:`) in the header, [#6663](https://github.com/Rdatatable/data.table/issues/6663). When `trunc.cols=TRUE`, this information is merged into the truncation message at the top, and the redundant footer message is suppressed. Thanks to @KyleHaynes for the suggestion and @venom1204 for the implementation.
+
 ### BUG FIXES
 
 1. `fread()` with `skip=0` and `(header=TRUE|FALSE)` no longer skips the first row when it has fewer fields than subsequent rows, [#7463](https://github.com/Rdatatable/data.table/issues/7463). Thanks @emayerhofer for the report and @ben-schwen for the fix.

R/onLoad.R

@@ -90,6 +90,7 @@
     datatable.print.keys=TRUE,          # for print.data.table
     datatable.print.trunc.cols=FALSE,   # for print.data.table
     datatable.show.indices=FALSE,       # for print.data.table
+    datatable.show.ncols=FALSE,         # for print.data.table
     datatable.allow.cartesian=FALSE,    # datatable.<argument name>
     datatable.join.many=TRUE,           # mergelist, [.data.table #4383 #914
     datatable.dfdispatchwarn=TRUE,      # not a function argument

R/print.data.table.R

@@ -8,6 +8,7 @@ print.data.table = function(x, topn=getOption("datatable.print.topn"),
                print.keys=getOption("datatable.print.keys"),
                trunc.cols=getOption("datatable.print.trunc.cols"),
                show.indices=getOption("datatable.show.indices"),
+               show.ncols=getOption("datatable.show.ncols", FALSE),
                quote=FALSE,
                na.print=NULL,
                timezone=FALSE, ...) {
@@ -20,6 +21,7 @@ print.data.table = function(x, topn=getOption("datatable.print.topn"),
   if (length(trunc.cols) != 1L || !is.logical(trunc.cols) || is.na(trunc.cols))
     stopf("Valid options for trunc.cols are TRUE and FALSE")
   stopifnot(isTRUEorFALSE(class))
+  stopifnot(isTRUEorFALSE(show.ncols))
   if (col.names == "none" && class)
     warningf("Column classes will be suppressed when col.names is 'none'")
   if (!shouldPrint(x)) {
@@ -43,6 +45,8 @@ print.data.table = function(x, topn=getOption("datatable.print.topn"),
   if (!is.numeric(topn)) topn = 5L
   topnmiss = missing(topn)
   topn = max(as.integer(topn),1L)
+  show_trunc_message = isTRUE(trunc.cols)
+  n_col = ncol(x)
   if (print.keys) {
     if (!is.null(ky <- key(x)))
     catf("Key: <%s>\n", toString(ky))
@@ -74,7 +78,6 @@ print.data.table = function(x, topn=getOption("datatable.print.topn"),
   }
   n_x = nrow(x)
   if ((topn*2L+1L)<n_x && (n_x>nrows || !topnmiss)) {
-    toprint = rbindlist(list(head(x, topn), tail(x, topn)), use.names=FALSE)  # no need to match names because head and tail of same x, and #3306
     rn = c(seq_len(topn), seq.int(to=n_x, length.out=topn))
     printdots = TRUE
     idx = c(seq_len(topn), seq(to=nrow(x), length.out=topn))
@@ -86,6 +89,9 @@ print.data.table = function(x, topn=getOption("datatable.print.topn"),
     printdots = FALSE
     if (show.indices) toprint = cbind(toprint, index_dt)
   }
+  if (show.ncols && !isTRUE(trunc.cols)) {
+    trunc_cols_message(ncol=n_col)
+  }
   require_bit64_if_needed(x)
   classes = classes1(toprint)
   trunc.char = getOption("datatable.prettyprint.char")
@@ -121,8 +127,13 @@ print.data.table = function(x, topn=getOption("datatable.print.topn"),
     cons_width = getOption("width")
     cols_to_print = widths < cons_width
     not_printed = colnames(toprint)[!cols_to_print]
+    if (show.ncols) {
+      trunc_cols_message(not_printed, abbs, class, col.names, ncol=n_col)
+      show_trunc_message = FALSE
+    }
+
     if (!any(cols_to_print)) {
-      trunc_cols_message(not_printed, abbs, class, col.names)
+      if (show_trunc_message) trunc_cols_message(not_printed, abbs, class, col.names)
       return(invisible(x))
     }
     # When nrow(toprint) = 1, attributes get lost in the subset,
@@ -134,7 +145,7 @@ print.data.table = function(x, topn=getOption("datatable.print.topn"),
     if (col.names != "none") cut_colnames = identity
     cut_colnames(print(x, right=TRUE, quote=quote, na.print=na.print))
     # prints names of variables not shown in the print
-    if (trunc.cols) trunc_cols_message(not_printed, abbs, class, col.names)
+    if (show_trunc_message) trunc_cols_message(not_printed, abbs, class, col.names)
   }
   if (printdots) {
     if (isFALSE(row.names)) {
@@ -291,14 +302,17 @@ toprint_subset = function(x, cols_to_print) {
   }
 }
 # message for when trunc.cols=TRUE and some columns are not printed
-trunc_cols_message = function(not_printed, abbs, class, col.names){
+trunc_cols_message = function(not_printed=character(0), abbs=NULL, class=FALSE, col.names="auto", ncol=NULL){
   n = length(not_printed)
-  if (class && col.names != "none") classes = paste0(" ", tail(abbs, n)) else classes = ""
-  catf(
-    ngettext(n, "%d variable not shown: %s\n", "%d variables not shown: %s\n"),
-    n, brackify(paste0(not_printed, classes)),
-    domain=NA
-  )
+  if (n == 0L) {
+    if (!is.null(ncol)) catf("Number of columns: %d\n", ncol)
+    return()
+  }
+  classes = if (class && col.names != "none") paste0(" ", tail(abbs, n)) else ""
+  if (is.null(ncol)) { catf(ngettext(n, "%d variable not shown: %s\n", "%d variables not shown: %s\n"), n, brackify(paste0(not_printed, classes)), domain=NA)
+  } else {
+    catf(ngettext(n, "Number of columns: %d, of which %d is not shown: %s\n", "Number of columns: %d, of which %d are not shown: %s\n"), ncol, n, brackify(paste0(not_printed, classes)), domain=NA)
+  }
 }
 
 # Maybe add a method for repr::repr_text.  See https://github.com/Rdatatable/data.table/issues/933#issuecomment-220237965

inst/tests/tests.Rraw

@@ -21797,3 +21797,14 @@ test(2378.92, .Call(CresizeVector, x, 2:3), error = "must be length 1 non-NA non
 test(2378.93, .Call(CresizeVector, x, -2L), error = "must be length 1 non-NA non-negative integer value")
 test(2378.94, .Call(CresizeVector, x, NA_integer_), error = "must be length 1 non-NA non-negative integer value")
 test(2378.95, .Call(CresizeVector, NULL, 2L), error = "must be a vector")
+
+#6663 Option to print the number of columns
+test(2379.1, print(as.data.table(iris)), output="Number of columns: 5", options=list(datatable.show.ncols=TRUE))
+test(2379.2, print(setkey(as.data.table(iris), Sepal.Length)), output="Number of columns: 5", options=list(datatable.show.ncols=TRUE))
+test(2379.3, { dt = data.table(a=1:3, b=1:3, c=1:3, d=1:3, e=1:3); any(grepl("^[0-9]+ variables? not shown", capture.output(print(dt, trunc.cols=TRUE)))) }, FALSE, options=list(width=20, datatable.show.ncols=TRUE))
+test(2379.4, print(as.data.table(iris), trunc.cols=TRUE), output="Number of columns: 5, of which 5 are not shown: [Sepal.Length, Sepal.Width, Petal.Length, Petal.Width, Species]", options=list(width=10, datatable.show.ncols=TRUE))
+test(2379.5, any(grepl("Number of columns:", capture.output(print(as.data.table(iris))))), FALSE, options=list(datatable.show.ncols=FALSE))
+test(2379.6, print(as.data.table(iris), trunc.cols=TRUE, class=TRUE), output="Number of columns: 5, of which 5 are not shown: [Sepal.Length <num>, Sepal.Width <num>, Petal.Length <num>, Petal.Width <num>, Species <fctr>]", options=list(width=10, datatable.show.ncols=TRUE))
+test(2379.7, { dt = data.table(matrix(1:200, nrow=25)); out = capture.output(print(dt, trunc.cols=TRUE, class=TRUE)); grepl("<int>", out[length(out)]) }, TRUE, options=list(width=30, datatable.show.ncols=TRUE))
+test(2379.8, any(grepl("variable", capture.output(print(data.table(a=1), trunc.cols=TRUE)))), FALSE, options=list(width=80, datatable.show.ncols=FALSE))
+test(2379.9, print(setindex(data.table(a=1:2, b=3:4), a), show.ncols=TRUE, show.indices=TRUE), output="Number of columns: 2")

man/data.table-options.Rd

@@ -33,6 +33,7 @@
    \item{\code{datatable.print.keys}}{A logical, default \code{FALSE}. If \code{TRUE}, the table's
      keys are printed above the data.}
    \item{\code{datatable.show.indices}}{A logical, default \code{TRUE}. A synonym for \code{datatable.print.keys} for historical reasons.}
+   \item{\code{datatable.show.ncols}}{A logical, default \code{FALSE}. If \code{TRUE}, the number of columns is printed in the header.}
    \item{\code{datatable.print.trunc.cols}}{A logical, default \code{FALSE}. If \code{TRUE} and a
      table has more columns than fit on the screen, it truncates the middle columns.}
    \item{\code{datatable.prettyprint.char}}{An integer, default \code{100L}. The maximum number of

man/print.data.table.Rd

@@ -25,6 +25,7 @@
     print.keys=getOption("datatable.print.keys"),       # default: TRUE
     trunc.cols=getOption("datatable.print.trunc.cols"), # default: FALSE
     show.indices=getOption("datatable.show.indices"),   # default: FALSE
+    show.ncols=getOption("datatable.show.ncols", FALSE), # default: FALSE
     quote=FALSE,
     na.print=NULL,
     timezone=FALSE, \dots)
@@ -44,6 +45,7 @@
   \item{trunc.char}{The number of characters at which character columns and list-column summaries are truncated. If \code{NULL} (the default), it is dynamically calculated based on \code{getOption("width")}.}
   \item{class}{ If \code{TRUE}, the resulting output will include above each column its storage class (or a self-evident abbreviation thereof). When combined with \code{col.names="auto"} and tables >20 rows, classes will also appear at the bottom.}
   \item{row.names}{ If \code{TRUE}, row indices will be printed alongside \code{x}. }
+  \item{show.ncols}{ If \code{TRUE}, the number of columns is printed in the header. }
   \item{col.names}{ One of three flavours for controlling the display of column names in output. \code{"auto"} includes column names above the data, as well as below the table if \code{nrow(x) > 20} (when \code{class=TRUE}, column classes will also appear at the bottom). \code{"top"} excludes this lower register when applicable, and \code{"none"} suppresses column names altogether (as well as column classes if \code{class = TRUE}. }
   \item{print.keys}{ If \code{TRUE}, any \code{\link{key}} and/or \code{\link[=indices]{index}} currently assigned to \code{x} will be printed prior to the preview of the data. }
   \item{trunc.cols}{ If \code{TRUE}, only the columns that can be printed in the console without wrapping the columns to new lines will be printed (similar to \code{tibbles}). }