The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
parse_lat() and parse_lon() (and the
functions built on them) now return NA with a warning when
minutes or seconds are not within 0/60, instead of silently converting
them, e.g. “10°30,12345” (#60). Negative minutes or seconds, and a
decimal degree or minute followed by another slot (e.g. “45.5 30”), now
also give NA with a warning. A comma is a separator, not a
decimal mark; this is now documented.parse_parts_lat() and parse_parts_lon()
(and pz_degree(), pz_minute(),
pz_second()) no longer lose the sign of values between -1
and 0: the sign now sits on the first non-zero part, e.g. -0.5 gives deg
0, min -30. They also no longer return 60 seconds, e.g. 10.35 is now 10
deg 21 min 0 sec, not 10 deg 20 min 60 sec.parse_llstr() now gives one clear warning (“could not
split into latitude and longitude”) for each string it cannot split, and
keeps the other rows. Before, each such string gave two confusing “no
digits” warnings about an internal marker.parse_lon_lat() and parse_hemisphere() now
clean symbols like ° the same way parse_lat()
and parse_lon() do (before, “40°25’” gave 40).1e-5 are no longer misread
as two slots, and a number too large for a double now gives
NA with a warning instead of shifting the other slots.parse_lon_lat() and parse_hemisphere() now
accept numeric and integer inputs, as documented.parse_llstr() now returns NA for unrecognized formats
instead of two empty strings.parse_parts_lat() and
parse_parts_lon() when a value could not be parsed (NA was
cast to an integer).std::regex), numbers are read with std::stod
instead of a new stream for each token, and parse_parts_*()
no longer allocates an R list for each coordinate.
parse_lat(), parse_lon(),
parse_llstr() and parse_parts_lat() are about
9 to 12 times faster than in 0.4.4..Call() wrapper that forced the C locale (#31)
is removed, because it only worked around a std::regex
bug.parse_hemisphere() detects the sign of a value with
std::signbit() instead of converting it to a string.bench) to detect slowdowns,
and a larger test suite (about 98% coverage).pak::pkg_install().parzer core objective of parsing messy coordinates in
character strings to convert them to decimal numeric values. Suggestion
and work by @robitalecscrub(),
used in most exported functions in parzer (#30) work by @AlbanSagouis.Call() in which we use
withr::with_locale() if the user is on a Windows operating
system (#31) (#32) work by @yutannihilationparse_llstr(): on older R versions where
stringsAsFactors=TRUE by default this function was
returning strings as factors from an internal function that caused a
problem in a subsequent step in the function (#29)parse_llstr() to parse a string that
contains both latitude and longitude (#3) (#24) (#26) (#28) work by
@AlbanSagouisscrub() internal function that strips certain
characters to include more things to scrub (#25) work by @AlbanSagouisparse_parts_lat()/parse_parts_lon() functions
where an NA was causing warnings on the cpp side; on cpp side, now check
for NA and return list of NAs instead of NAs passing through other code
(#23)These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.