utf8_format() and
utf8_print(), no extra arguments allowed.Replace [v]sprintf() with [v]snprintf()
(#67).
Add direct include for snprintf() (@MichaelChirico,
#43).
Add ellipsis before optional args (#74).
Show NEWS on CRAN page (#42, #71).
Add pkgdown reference index.
Use roxygen2 (#68, #69) with Markdown.
Update maintainer e-mail address.
Fix compiler warnings (@Antonov548, #37).
utf8_width() now reports correct widths for narrow
emojis (#9).utf8lite.Fix build on Solaris (#7, reported by @krlmlr).
Fix rendering of emoji ZWJ sequences like
"\U1F469\U200D\U2764\UFE0F\U200D\U1F48B\U200D\U1F469".
Make output_utf8() always return TRUE
on Windows, so that characters in the user’s native locale don’t get
escaped by utf8_encode(). The downside of this change is
that on Windows, utf8_width() reports the wrong values for
characters outside the user’s locale when stdout() is
redirected by knitr or another process.
When truncating long strings strings via
utf8_format(), use an ellipsis that is printable in the
user’s native locale ("\u2026" or“…”`).
utf8_format() with non-NULL
width argument.as_utf8().output_ansi() and output_utf8()
functions to test for output capabilities.Add utf8 argument to utf8_encode(),
utf8_format(), utf8_print(), and
utf8_width() for precise control over assumed output
capabilities; defaults to the result of
output_utf8().
Add ability to style backslash escapes with the
escapes arguments to utf8_encode() and
utf8_print(). Switch from “faint” styling to no styling by
default.
Slightly reword error messages for
as_utf8().
Fix (spurious) rchk warnings.
utf8_width() determining width of non-ASCII
strings when LC_CTYPE=C.as_utf8() (the R
version is still present).Split off functions as_utf8(),
utf8_valid(), utf8_normalize(),
utf8_encode(), utf8_format(),
utf8_print(), and utf8_width() from corpus package.
Added special handling for Unicode grapheme clusters in formatting and width measurement functions.
Added ANSI styling to escape sequences.
Added ability to style row and column names in
utf8_print().