Fix char.trunc with NA (#6442)

* omit NA in indices

* revisit

* better test

* change order

* NEWS

---------

Co-authored-by: Michael Chirico <[email protected]>

Author: joshhwuu

NEWS.md

@@ -6,6 +6,10 @@
 
 1. In `DT[, variable := value]`, when value is class `POSIXlt`, we automatically coerce it to class `POSIXct` instead, [#1724](https://github.com/Rdatatable/data.table/issues/1724). Thanks to @linzhp for the report, and Benjamin Schwendinger for the fix.
 
+## BUG FIXES
+
+1. Using `print.data.table()` with character truncation using `datatable.prettyprint.char` no longer errors with `NA` entries, [#6441](https://github.com/Rdatatable/data.table/issues/6441). Thanks to @r2evans for the bug report, and @joshhwuu for the fix.
+
 ## NOTES
 
 1. Tests run again when some Suggests packages are missing, [#6411](https://github.com/Rdatatable/data.table/issues/6411). Thanks @aadler for the note and @MichaelChirico for the fix.

R/print.data.table.R

@@ -247,7 +247,7 @@ char.trunc = function(x, trunc.char = getOption("datatable.prettyprint.char")) {
   nchar_width = nchar(x, 'width') # Check whether string is full-width or half-width, #5096
   nchar_chars = nchar(x, 'char')
   is_full_width = nchar_width > nchar_chars
-  idx = pmin(nchar_width, nchar_chars) > trunc.char
+  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

@@ -18508,6 +18508,8 @@ local({
     c(paste0(ja_ko, " ", paste0(ja_n, dots), " ", paste0(accented_a, dots)),
     paste0(c(ja_ko, ja_n, accented_a), dots, collapse=" "),
     paste0(c(ja_ko, ja_n, accented_a), dots, collapse=" ")))
+  # test for data.table with NA, #6441
+  test(2253.20, options=list(datatable.prettyprint.char = 1L), data.table(a = c("abc", NA)), output="      a\n1: a...\n2: <NA>")
 })
 
 # allow 1-D matrix in j for consistency, #783