FALSE) with default
"photobiology.auto.hinges.limit" (wavelength step in nm) with new default 0.5 nm. Default behaviour is changed both for a bug in
irrad() functions where the limit was 1.1 nm instead of 0.7 nm and as a result of changing to a smaller default value. Existing and new options
"photobiology.use.cached.mult" are now used throughout.
Bug fix: in version 0.5.1x the behaviour of the product of a BSWF and a source spectrum was changed to return a response spectrum. This was not consistent with the idea that a BSWF is used for quantifying radiation rather predicting a biological response. The result should remain as radiation and continue to be expressed in energy or photon based units. This has been fixed. Starting from the current version, if the
waveband is used with the
'*' operator is a BSWF (tested with function
is.effective() ) the resulting spectrum is tagged as being
irrad() methods for
source.spct have been updated to ‘recognize’ effective spectral irradiance when supplied as input.
rbindspct() modified so that if at least one of the spectra in the list is of effective irradiance, but not all spectra have been calculated with the same BSWF and normalization, a factor named BSWF is added with the BSWFs retrieved from the spectra used as levels. The optional factor selected through parameter
idfactor has now its levels always reflecting the order of the spectra in the input list, even when using a named input list of spectra. Several rather minor bugs were fixed, including improved handling of comments and attributes.
summary() methods for spectra have been updated to report the BSWF that have been used if source spectra contain effective spectral irradiances.
Long-standing unsolved problem partly fixed:
interpolate_spct() now in most cases applies smoothing to the spectrum before attempting interpolation if the vector of output wavelengths has length > 1, but is sparser than then original spectral data.
A few inconsistencies in the formal parameters among similar functions were fixed by adding the missing parameters. Also some problems in the documentation for a few functions were fixed.
User guide updated.
Mostly backwards compatible, but some default values have changed to minimize errors at the expense of speed of computation. Also function
is_effective(), which is most frequently but not always used internally was renamed
is.effective(). The changed behaviour of operator
* may also require small changes to user code.
Preliminary support for plotting effective irradiance spectra.
plot() methods now use options
"photobiology.label.wavebands" as default if available. The default for trimming has changed to
TRUE, as this seems not too dangerous as labels show when trimming has taken place.
User Guide updated.
NOTE: Sources, and Windows binaries for R 3.1.x, are now available in the repository.