"hms"
objects no longer have a common type with character
and numeric
. Combining such values previously threw a warning, now throws an error (#94).vec_cast()
and as_hms()
throw error instead of a warning if input can’t be parsed (#68).unique.hms()
method (#98, @joethorley).as_hms()
is a generic again (#81).LazyData
in DESCRIPTION
."hms"
objects no longer have a common type with character
and numeric
. Combining such values previously threw a warning, now throws an error (#94).vec_cast()
and as_hms()
now throw error instead of a warning if input can’t be parsed (#68).as_hms()
is a generic again (#81).round_hms()
and trunc_hms()
gain digits
argument (#78, @hglanz).as_hms()
and vec_cast()
now correctly treat objects of class "difftime"
with integer
mode (#84).vec_default_ptype2()
, remove vec_ptype2.hms.unspecified()
(#80, @romainfrancois).vec_ptype2.hms.default()
forwards to vec_default_ptype2()
for compatibility with vctrs 0.2.1.as.data.frame.hms()
, handeld by vctrs.character
vectors to hms
now also trigger a warning if the cast succeeds in the first element of the vector but fails for other elements.Now based on vctrs >= 0.2.0 (#61). This adds support for vec_cast()
and vec_ptype2()
. Some operations (such as subset assignment) are now stricter. The new_hms()
constructor permits safe construction of hms
objects from a numeric vector.
The new is_hms()
deprecates the existing is.hms()
. as.hms()
is deprecated in favor of vec_cast()
or the new as_hms()
(which is just a thin wrapper around vec_cast()
).
Always show seconds in a pillar if they are different from zero (#54).
Values with nonzero hours, seconds and split-seconds are now displayed correctly in tibbles (#56), even with a very small distance to the full second (#64).
NA
when converting to character
(#51, @jeroen).as.hms.POSIXt()
now defaults to the current time zone, the previous default was "UTC"
and can be restored by calling pkgconfig::set_config("hms::default_tz", "UTC")
.hms
columns in tibbles in color on terminals that support it (#43).round_hms()
and trunc_hms()
for rounding or truncating to a given multiple of seconds (#31).parse_hms()
and parse_hm()
to parse strings in “HH:MM:SS” and “HH:MM” formats (#30).as.hms.POSIXt()
gains tz
argument, default "UTC"
(#28).as.hms.character()
and parse_hms()
accept fractional seconds (#33).hms()
now works correctly if all four components (days, hours, minutes, seconds) are passed (#49).hms()
creates a zero-length object of class hms
that prints as "hms()"
.hms(integer())
and as.hms(integer())
both work and are identical to hms()
.c()
now returns a hms (#41, @qgeissmann).?hms
.hms()
with a character argument (#29).lubridate
test for compatibility with 1.6.0 (#23, @vspinu).NA
(#22).Minor fixes and improvements.
hms
class (#16).format.hms()
right-justifies the output by padding with spaces from the left, as.character.hms()
remains unchanged.First CRAN release.
difftime
class."secs"
issues a warning.-
.Date
, POSIXt
and hms
classes.lubridate
package (#5).hms()
(with rigorous argument checks)as.hms()
for character
, numeric
, POSIXct
and POSIXlt
as.xxx.hms()
for character
, numeric
(implicitly), POSIXct
and POSIXlt
is.hms()
as.data.frame.hms()
(forwards to as.data.frame.difftime()
)format.hms()
print.hms()
(returns unchanged input invisibly)