bug fix: truncate wide columns in print.data.table

NEWS.md

@@ -56,6 +56,8 @@
 
 11. `between()` now supports `Date` and `IDate` bounds with default `NAbounds=TRUE`, avoiding errors like "Not yet implemented NAbounds=TRUE for this non-numeric and non-character type" when date bounds contain `NA`, [#7281](https://github.com/Rdatatable/data.table/issues/7281). Thanks @grcatlin for the report and fix, and @ben-schwen and @aitap for assistance.
 
+12. `print.data.table()` now truncates long character columns and list-column summaries by default to avoid horizontal console overflow, [#7718](https://github.com/Rdatatable/data.table/issues/7718). When `datatable.prettyprint.char` is `NULL` (the default), the truncation limit is now dynamically calculated based on the available console width. Use `options(datatable.prettyprint.char=Inf)` for the old default behavior (never truncate). Thanks @tdhock for the report and @venom1204 for the fix.
+
 ### Notes
 
 1. {data.table} now depends on R 3.5.0 (2018).

R/print.data.table.R

@@ -88,7 +88,13 @@ print.data.table = function(x, topn=getOption("datatable.print.topn"),
   }
   require_bit64_if_needed(x)
   classes = classes1(toprint)
-  toprint=format.data.table(toprint, na.encode=FALSE, timezone = timezone, ...)  # na.encode=FALSE so that NA in character cols print as <NA>
+
+  trunc.char = getOption("datatable.prettyprint.char")
+  if (is.null(trunc.char)) {
+    rn_w = if (isTRUE(row.names)) nchar(as.character(max(rn))) + 2L else 0L
+    trunc.char = max(0L, getOption("width") - rn_w - 3L)
+  }
+  toprint=format.data.table(toprint, na.encode=FALSE, timezone = timezone, trunc.char = trunc.char, ...)  # na.encode=FALSE so that NA in character cols print as <NA>
 
   # FR #353 - add row.names = logical argument to print.data.table
   if (isTRUE(row.names)) rownames(toprint)=paste0(format(rn,right=TRUE,scientific=FALSE),":") else rownames(toprint)=rep.int("", nrow(toprint))
@@ -155,12 +161,12 @@ print.data.table = function(x, topn=getOption("datatable.print.topn"),
   invisible(x)
 }
 
-format.data.table = function(x, ..., justify="none") {
+format.data.table = function(x, ..., trunc.char = getOption("datatable.prettyprint.char"), justify="none") {
   if (is.atomic(x) && !is.null(x)) { ## future R can use  if (is.atomic(x))
 
     stopf("Internal structure doesn't seem to be a list. Possibly corrupt data.table.")
   }
-  do.call(cbind, lapply(x, format_col, ..., justify=justify))
+  do.call(cbind, lapply(x, format_col, ..., trunc.char = trunc.char, justify=justify))
 }
 
 shouldPrint = function(x) {
@@ -198,13 +204,13 @@ has_format_method = function(x) {
   any(vapply_1b(class(x), f))
 }
 
-format_col.default = function(x, ...) {
+format_col.default = function(x, ..., trunc.char = getOption("datatable.prettyprint.char")) {
   if (!is.null(dim(x)))
     "<multi-column>"
   else if (is.list(x))
-    vapply_1c(x, format_list_item, ...)
+    vapply_1c(x, format_list_item, ..., trunc.char = trunc.char)
   else
-    format(char.trunc(x), ...) # relevant to #37
+    format(char.trunc(x, trunc.char = trunc.char), ...) # relevant to #37
 }
 
 # #2842 -- different columns can have different tzone, so force usage in output
@@ -221,20 +227,21 @@ format_col.POSIXct = function(x, ..., timezone=FALSE) {
 }
 
 # #3011 -- expression columns can wrap to newlines which breaks printing
-format_col.expression = function(x, ...) format(char.trunc(as.character(x)), ...)
+format_col.expression = function(x, ..., trunc.char = getOption("datatable.prettyprint.char")) {
+  format(char.trunc(as.character(x), trunc.char = trunc.char), ...)
+}
 
-format_list_item.default = function(x, ...) {
-  if (is.null(x))  # NULL item in a list column
-    "[NULL]" # not '' or 'NULL' to distinguish from those "common" string values in data
-  else if (is.atomic(x) || inherits(x, "formula")) # FR #2591 - format.data.table issue with columns of class "formula"
-    paste(c(format(head(x, 6L), ...), if (length(x) > 6L) sprintf("...[%d]", length(x))), collapse=",") # fix for #5435, #37, and #605 - format has to be added here...
+format_list_item.default = function(x, ..., trunc.char = getOption("datatable.prettyprint.char")) {
+  res = if (is.null(x))  # NULL item in a list column
+    "[NULL]"
+  else if (is.atomic(x) || inherits(x, "formula"))
+    paste(c(format(head(x, 6L), ...), if (length(x) > 6L) sprintf("...[%d]", length(x))), collapse=",")
   else if (has_format_method(x) && length(formatted<-format(x, ...))==1L) {
-    # the column's class does not have a format method (otherwise it would have been used by format_col and this
-    # format_list_item would not be reached) but this particular list item does have a format method so use it
     formatted
   } else {
     paste0("<", class1(x), paste_dims(x), ">")
   }
+  char.trunc(res, trunc.char = trunc.char)
 }
 
 # #6592 -- nested 1-column frames breaks printing
@@ -247,12 +254,14 @@ format_list_item.data.frame = function(x, ...) {
 # Current implementation may have issues when dealing with strings that have combinations of full-width and half-width characters,
 # if this becomes a problem in the future, we could consider string traversal instead.
 char.trunc = function(x, trunc.char = getOption("datatable.prettyprint.char")) {
+  if (is.null(trunc.char)) return(x)
   trunc.char = max(0L, suppressWarnings(as.integer(trunc.char[1L])), na.rm=TRUE)
   if (!is.character(x) || trunc.char <= 0L) return(x)
-  nchar_width = nchar(x, 'width') # Check whether string is full-width or half-width, #5096
-  nchar_chars = nchar(x, 'char')
+  nchar_width = nchar(x, 'width', allowNA = TRUE)
+  nchar_chars = nchar(x, 'char', allowNA = TRUE)
   is_full_width = nchar_width > nchar_chars
-  idx = !is.na(x) & pmin(nchar_width, nchar_chars) > trunc.char
+  is_full_width[is.na(is_full_width)] = FALSE
+  idx = !is.na(x) & !is.na(nchar_width) & pmin(nchar_width, nchar_chars) > trunc.char
   x[idx] = paste0(strtrim(x[idx], trunc.char * fifelse(is_full_width[idx], 2L, 1L)), "...")
   x
 }

inst/tests/tests.Rraw

@@ -21660,3 +21660,12 @@ test(2374.08, key(DT[, .(a, a)]), NULL)
 test(2374.09, key(subset(DT, select=c(a, a))), NULL)
 DT = data.table(a=1:2, a.1=3:4, val=10:11)
 test(2374.10, key(DT[, .(a.1, sum(val)), keyby=.(a, a)]), NULL)
+
+# print.data.table truncates long character columns based on width
+test(2375.1, capture.output(print(data.table(x="1234567890"))), output="1234...", options=list(width=10, datatable.prettyprint.char=NULL))
+test(2375.11, print(data.table(x="1234567890")), output="1234567890", options=list(width=10, datatable.prettyprint.char=Inf))
+test(2375.2, capture.output(print(data.table(L=list(1:20)))), output="1,2,3,4,...", options=list(width=15, datatable.prettyprint.char=NULL))
+test(2375.3, capture.output(print(data.table(x=c("short", "abcdefghijklmnopqrstuvwxyz")))), output="abcdefghijklmn...", options=list(width=20, datatable.prettyprint.char=NULL))
+test(2375.4, capture.output(print(data.table(x="abcdefghijklmnopqrstuvwxyz"))), output="abcdefghijklmnopqrstuvwxyz", options=list(width=200, datatable.prettyprint.char=NULL))
+test(2375.5, capture.output(print(data.table(id=1L, score=99.1, txt="abcdefghijklmnopqrstuvwxyz"))), output="abcdefghijklmn...", options=list(width=20, datatable.prettyprint.char=NULL))
+test(2375.6, tail(capture.output(print(data.table(x=rep("ABCDEFGHIJKLMNOPQRSTUVWXYZ", 1000000)), topn=1)), 1), output="1000000: ABCDEFGHIJKLM...", options=list(width=25, datatable.prettyprint.char=NULL))

man/print.data.table.Rd

@@ -30,17 +30,18 @@
     timezone=FALSE, \dots)
 
   format_col(x, \dots)
-  \method{format_col}{default}(x, \dots)
+  \method{format_col}{default}(x, \dots, trunc.char = getOption("datatable.prettyprint.char"))
   \method{format_col}{POSIXct}(x, \dots, timezone=FALSE)
-  \method{format_col}{expression}(x, \dots)
+  \method{format_col}{expression}(x, \dots, trunc.char = getOption("datatable.prettyprint.char"))
 
   format_list_item(x, \dots)
-  \method{format_list_item}{default}(x, \dots)
+  \method{format_list_item}{default}(x, \dots, trunc.char = getOption("datatable.prettyprint.char"))
 }
 \arguments{
   \item{x}{ A \code{data.table}. }
   \item{topn}{ The number of rows to be printed from the beginning and end of tables with more than \code{nrows} rows. }
   \item{nrows}{ The number of rows which will be printed before truncation is enforced. }
+  \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{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}. }
@@ -134,5 +135,13 @@
   iris_agg = iris[ , .(reg = list(lm(Sepal.Length ~ Petal.Length))), by = Species]
   format_list_item.lm = function(x, ...) sprintf('<lm:\%s>', format(x$call$formula))
   print(iris_agg)
+
+  # Truncation based on console width
+  old = options(width = 25, datatable.prettyprint.char = NULL)
+  data.table(x = "abcdefghijklmnopqrstuvwxyz", L = list(1:25))
+
+  # Dynamic truncation: Content shrinks as row labels grow
+  print(data.table(x = rep("ABCDEFGHIJKLMNOPQRSTUVWXYZ", 1e6)), topn = 1)
+  options(old)
 }