cc() can once again send generative code for its
output to the system clipboard via the .clip argument. This
functionality was previously removed in the 1.5.0 release due to
unreliability in it’s use on some environments (e.g. RStudio on Fedora
KDE Wayland session). Time seems to have remedied the issue.
breaking change: If an assertion succeeds it now returns it’s input invisibly. This was to enable more concise casting, e.g
x <- 1.0
x <- as.integer(assert_integerish(x))breaking change: new_package() and
it’s alias np() have been completely removed from the
package.
New function new_name() which generates unique names
for additional data frame variables (ensuring they are not already
present).
New assertion functions:
assert_character_not_na(),
assert_chr_not_na()assert_double_not_na(),
assert_dbl_not_na()assert_integer_not_na(),
assert_int_not_na()assert_integerish(), assert_whole()assert_scalar_integerish(),assert_logical_not_na(),
assert_lgl_not_na()assert_numeric_not_na(),
assert_num_not_na()assert_scalar_logical_not_na(),
assert_scalar_lgl_not_na()assert_string_not_na()assert_between()Assertions errors are now given the class “ympes-error”.
breaking change: Assertions no longer use
... to add attributes to the resulting error
conditions.
cc() now allows you to split length one character
vectors by whitespace.
breaking change: cc() no longer
allows the user to send generative code for the output to the system
clipboard with the .clip argument. This functionality was
unreliable on some systems so has been removed until a reliable
implementation can be provided.
Assertions and the new_package() functionality have
been marked experimental. Breaking changes to these functions should be
considered likely.
Added an overview vignette (see
vignette("ympes", package = "ympes")).
New function grepvrows() which is identical to
greprows() except with the default
value = TRUE.
New function greplrows() returns a logical vector
(match or not for each row of the supplied data frame).
breaking change: For consistency with
grep(), the value argument of
greprows() now defaults to FALSE. Please set
the value explicitly or use the new grepvrows()
function.
New assertion function
assert_scalar_whole().
Users can now set the condition subclass of returned errors via a
.subclass argument. In a similar vein it is now possible to
specify additional fields in the returned error condition via the
... arguments.
breaking change: Exported assertion functions are no longer prefixed with a dot.
breaking change: To allow for the additional
... arguments in the aforementioned assertions the
parameters arg/call have been renamed to
.arg/.call respectively and have also been
shifted (position-wise) within the function signatures. For example
assert_int(x, arg = deparse(substitute(x)), call = sys.call(-1L))becomes
assert_int(x, ..., .arg = deparse(substitute(x)), .call = sys.call(-1L), .subclass = NULL)fstrcapture():
perl and useBytes arguments have been
removed. These are now set internally to TRUE and
FALSE respectively.x and pattern arguments
has been switched to make the function a little more pipe-friendly.proto
argument.proto argument is missing, then all captures
will be treated as character type.Assertion functions are now (re)exported once again. Note they are all prefixed with a dot to avoid common naming collisions and, somewhat secondary, to emphasise their intended use within other functions.
breaking change. new_package() now
utilises an explicit pkg folder for the R package.
Fixes (hopefully) an issue with the internal usage of clipr
within cc() when on a Wayland system.
New function fstrcapture(). This is an (almost)
drop-in replacement for strcapture() that performs better
when using Perl-compatible regexps.
breaking change. Assertion functions have been removed. They can still be found internally (and are checked by tests) but the intention is to use them as a standalone file that is copy and pasted between packages as needed.
Licence change. Licence bumped from GPL-2 to GPL-3.
new_package() will now add
.Rbuildignore and .Rproj files with sensible
defaults.
np() is now given as an alias for
new_package().
New assertion functions:
assert_negative(),
assert_negative_or_na()
assert_positive(),
assert_positive_or_na()
assert_non_negative(),
assert_non_negative_or_na()
assert_non_positive(),
assert_non_positive_or_na()
assert_scalar_character_not_na(),
assert_scalar_chr_not_na()
assert_scalar_integer_not_na(),
assert_scalar_int_not_na()
assert_scalar_double_not_na(),
assert_scalar_dbl_not_na()
assert_scalar_numeric_not_na(),
assert_scalar_num_not_na()
Functions defunct in earlier versions of {ympes} have been removed from the package namespace.
New function new_package() for setting up a package
skeleton based on my preferred folder structure.
The old format for assertions (with the imp_ prefix;
e.g. imp_assert_int()) are now defunct and calling them
will result in an error.
Age utility functions breaks_to_interval(),
cut_ages(), split_interval_counts(),
aggregate_age_counts(),
reaggregate_interval_counts() and
ages_to_interval() are now defunct. These functions have
been migrated to the more focussed ageutils
package. As of version 0.0.1 of ageutils these have an
identical naming and usage.
New assertion functions that do not require the imp_
prefix and return NULL on success. Note that returning NULL means the
new assertions are not pipe friendly but the implementation is slightly
more efficient.
The old format for assertions (with the imp_ prefix;
e.g. imp_assert_int()) are now deprecated and will be
removed in a future release.
ages_to_interval() has been deprecated and replaced
with the function cut_ages(). This has a slightly different
API (see help(cut_ages) for more information).
New function breaks_to_interval() for transforming
breaks in to corresponding intervals (see
help(breaks_to_interval)).
breaking change. The limits
parameter has been removed from both aggregate_age_counts()
and reaggregate_age_counts(). It has been replaced by a
breaks which represents the left-hand side of your desired
interval bounds (i.e. the x value in [x, y)).
Intervals no longer span the natural numbers by default but will begin
from the minimum break input. Any ages below this value will be treated
as NA.
New function greprows() for searching rows of a data
frame.
Fixes a memory bug in ages_to_interval() that was highlighted via CRAN’s Valgrind checks.
New functions for working with age intervals:
ages_to_interval() provides categorisation of ages
based on specified right-hand interval limits.
split_interval_counts() splits counts within a age
interval in to counts for individuals years based on a given
weighting.
aggregate_age_counts() provides aggregation of
counts across ages (in years).
reaggregate_interval_counts() which is equivalent to
calling split_interval_counts() followed by
aggregate_age_counts().
The testing framework used by ympes has migrated from
testhat to tinytest.
Initial release of ympes which provides a collection of
lightweight helper functions (imps) for both interactive use and for
inclusion within other packages.
assertions for integer, double, numeric, logical, bool, character (string), list and data frame values
imp_assert_integer(),
imp_assert_int()
imp_assert_double() ,
imp_assert_dbl()
imp_assert_numeric(),
imp_assert_num()
imp_assert_logical(),
imp_assert_lgl()
imp_assert_character(),
imp_assert_chr()
imp_assert_list()
imp_assert_data_frame()
imp_assert_scalar_integer(),
imp_assert_scalar_int()
imp_assert_scalar_double(),
imp_assert_scalar_dbl()
imp_assert_scalar_numeric(),
imp_assert_scalar_num()
imp_assert_scalar_logical(),
imp_assert_scalar_lgl()
imp_assert_boolean(),
imp_assert_bool()
imp_assert_scalar_character(),
imp_assert_scalar_chr(),
imp_assert_string()
cc() for interactive quoting of comma separated
names.
plot_palette() for visualising colour
palettes.