The summary() for collections of spectra is just that for list objects. Some of the metadata or info in summary() for spectra could be added.

The improved approach implemented in v0.10.15 in e2q() and q2e() needs to be implements also in Afr2T, A2T, etc., any2T, etc.

T2A() seems to not work correctly when Tfr == 0 and Rfr != 0. The problem could be caused by a wrongly set TfrType attribute in the data I was using, or could be an error in the computations. In the first case a validity test would be missing and in the second case, a plain bug.

The missing Afr2A() would have been useful for testing and diagnosing the problem, and adding it should be the first step.

Minimal tests will be added before release of version 0.10.17, throrough ones with 0.11.0

Once again package 'photobiology' was affected by code-breakiing changes in 'dplyr' and 'tidyr'. In this case indirectly as the change was made in 'tidyeval' and possibly not caught by tests of reverse dependencies.

I think the time has come for removing these dependencies from 'photobiology' and minimizing other dependencies on the 'tidyverse', as most of the unplanned and urgent maintenance work has been caused in recent years by their inestability, while they do not provide much of an advantage compared to base R.

This is clearly seen in the results of some operations with functions from package 'ooacquire', it remains to be checked if this bug is due to code in 'photobiology', 'ooacquire' or elsewhere.

Adding these tests makes it neccessary to add example data for multiple spectra in long form and collections of spectra for all classes derived from "generic.spct" and from "generic.mspct". Current code was tested manually during development of these new functions.

This will help ensure that tests cover most of the code. The current approx. 5000 unit tests do not necessarily ensure full coverage.

There is a need to interpolate wavelengths of spectra within a collection to allow some parallel or rowwise operations. We need a function to achieve this efficiently now that we are using thinning of spectra to reduce their stored size. See also #11 as this two issues need to be resolved so that the new function can work together. I should have thought of this before starting to use wl-thinned spectra in the data-only packages!

Extracting from a collection of spectra using negative indexing sets the dimension to the length of the index vector, rather than to the number of remaining members.

Add the equation used to calculate normalized difference indexes to the details sections of the help page.
for NDVI = (NIR - R) / (NIR + R) or $NDxV = \frac{f(wb_1) - f(wb_1)}{f(wb_1) + f(wb_1)}$

Collections of spectra containing spct objects with multiple spectra in long form trigger errors in summary functions like q_irrad(), q_ratio(), etc. The error message from plyr::ldply() is obscure, and unhelpful. At the very least an informative message is needed. Best solution would be to write a function to unnest these spectra, by replacing the spct object by multiple spct objects within the collection. This could be an exported function. By default, applied with a warnig or message. See also #12 as they both affect similar use cases and the solution should solve also mixed cases.

e_ratio() and q_ratio() compute the ratio based on irradiances, i.e., based on the integral of irradiance for two different wavebands. This is in most cases what we are interested in. Occasionally, we may want to compute ratios based on mean spectral irradiance of the two wavebands. Only when the wavelength range covered by both wavebands is the same, the two approaches yield the same value.

Adding a parameter to switch between the two approaches could be sometimes useful, and simple to implement.

As we have a NDxI() function and q_ratio(), e_ratio() and ratio() functions we should add support for computing fractions based on pairs of wavelengths: we can just add an option to q_ratio(), e_ratio() and ratio(). Maybe named qty.out with values "ratio", "spct.ratio", and "fraction" and "spct.fraction", thus addressing also #13.

The first two conversion functions are in package 'ggspectra' but should be moved to 'photobiology'. Wavelength is used frequently in chemistry and physics, "wavelength" expressed in eV (the energy of a photon) and this needs to be also implemented in 'ggspectra'.

Using filter_spct objects to store spectral molar cross sections is clumsy unless we add a way to store this information in an attribute. This is likely to be enough. Creating a class derived from filter_spct could also work as it would inherit most of the functionality. In either case this seems a rather major undertaking as it may break old code. NEEDS TO BE DONE BEFORE FIRST COMPLETE DRAFT OF THE BOOK.

It seems sunrise_time is not correct for non-UTC times with recent versions of lubridate or R, as shown in the User Guide.

library(lubridate)
library(photobiology)

my.geocode <- data.frame(lat = 60.16, lon = 24.93, address = "Helsinki")
my.datetime <- as_datetime("2022-07-10 12:00:00", tz="EET")

sunrise_time(my.datetime, geocode = my.geocode, tz = "UTC")
[1] "2022-07-10 01:13:46 UTC"

sunrise_time(my.datetime, geocode = my.geocode, tz = "EET")
[1] "2022-07-10 01:13:46 EEST"

> sessionInfo()
R version 4.2.0 (2022-04-22)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 20.04.4 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3
LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/liblapack.so.3

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C             LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] photobiology_0.10.2 tibble_3.1.7        lubridate_1.8.0    

loaded via a namespace (and not attached):
 [1] compiler_4.2.0    pillar_1.7.0      prettyunits_1.1.1 remotes_2.4.2     tools_4.2.0       testthat_3.1.4    pkgbuild_1.3.1   
 [8] pkgload_1.2.4     memoise_2.0.1     lifecycle_1.0.1   pkgconfig_2.0.3   rlang_1.0.2       DBI_1.1.3         cli_3.3.0        
[15] fastmap_1.1.0     withr_2.5.0       dplyr_1.0.9       generics_0.1.2    desc_1.4.1        fs_1.5.2          vctrs_0.4.1      
[22] devtools_2.4.3    rprojroot_2.0.3   tidyselect_1.1.2  glue_1.6.2        R6_2.5.1          processx_3.6.1    fansi_1.0.3      
[29] sessioninfo_1.2.2 callr_3.7.0       purrr_0.3.4       magrittr_2.0.3    ps_1.7.1          ellipsis_0.3.2    usethis_2.1.6    
[36] assertthat_0.2.1  utf8_1.2.2        cachem_1.0.6      crayon_1.5.1      brio_1.1.3   

However, regressing to earlier versions of lubridate and R but using the same photobiology version results in the correct time.

library(lubridate)
library(photobiology)

my.geocode <- data.frame(lat = 60.16, lon = 24.93, address = "Helsinki")
my.datetime <- as_datetime("2022-07-10 12:00:00", tz="EET")

sunrise_time(my.datetime, geocode = my.geocode, tz = "UTC")
[1] "2022-07-10 01:13:46 UTC"

sunrise_time(my.datetime, geocode = my.geocode, tz = "EET")
[1] "2022-07-10 04:13:46 EEST"

sessionInfo()
R version 3.6.3 (2020-02-29)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Debian GNU/Linux 10 (buster)

Matrix products: default
BLAS/LAPACK: /usr/lib/x86_64-linux-gnu/libopenblasp-r0.3.5.so

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               LC_TIME=en_US.UTF-8       
 [4] LC_COLLATE=en_US.UTF-8     LC_MONETARY=en_US.UTF-8    LC_MESSAGES=C             
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  LC_ADDRESS=C              
[10] LC_TELEPHONE=C             LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] photobiology_0.10.2 tibble_3.0.1        lubridate_1.7.8    

loaded via a namespace (and not attached):
 [1] Rcpp_1.0.4.6      compiler_3.6.3    pillar_1.4.3      prettyunits_1.1.1 remotes_2.1.1    
 [6] tools_3.6.3       testthat_2.3.2    digest_0.6.25     pkgbuild_1.0.6    pkgload_1.0.2    
[11] memoise_1.1.0     lifecycle_0.2.0   pkgconfig_2.0.3   rlang_0.4.5       cli_2.0.2        
[16] rstudioapi_0.11   withr_2.2.0       dplyr_0.8.5       generics_0.0.2    desc_1.2.0       
[21] fs_1.4.1          vctrs_0.2.4       devtools_2.3.0    rprojroot_1.3-2   tidyselect_1.0.0 
[26] glue_1.4.0        R6_2.4.1          processx_3.4.2    fansi_0.4.1       sessioninfo_1.1.1
[31] purrr_0.3.4       callr_3.4.3       magrittr_1.5      backports_1.1.6   ps_1.3.2         
[36] ellipsis_0.3.0    usethis_1.6.0     assertthat_0.2.1  crayon_1.3.4  

Tetens and Magnus formulations are simpler but Goff-Gratch may be preferred under some conditions.

Implemented in version 0.9.25 for saturation water VP, inverse not yet implemented. Need to find the inverse equation formulation or derive it.

Added forward and inverse implementations of Wexler. See http://www.decatur.de/javascript/dew/resources/its90formulas.pdf. ITS90 of forward equation needs the replacement of constants. Currently forward follows ITS68 as the original formulation, while the only inverse available follows ITS90.

See https://www.eol.ucar.edu/system/files/VaporPressure.pdf for the equation of Murphy and Koop.

This looks fascinating, and rather easy to implement. It will also teach me about color vision.

https://se.mathworks.com/matlabcentral/fileexchange/69853-convert-spectral-radiance-to-rgb-for-animal-eyes

https://dx.doi.org/10.1093%2Fcz%2Fzow076

Could be the start of ‘photobiologyAnimals’.

We're about to release dplyr 1.0.8 and it seems that the released version of photobiology has issues:

# photobiology

<details>

* Version: 0.10.8
* GitHub: https://github.com/aphalo/photobiology
* Source code: https://github.com/cran/photobiology
* Date/Publication: 2021-12-08 11:50:02 UTC
* Number of recursive dependencies: 60

Run `cloud_details(, "photobiology")` for more info

</details>

## Newly broken

*   checking examples ... ERROR
    ```
    Running examples in ‘photobiology-Ex.R’ failed
    The error most likely occurred in:
    
    > ### Name: A.illuminant.spct
    > ### Title: CIE A illuminant data
    > ### Aliases: A.illuminant.spct
    > ### Keywords: datasets
    > 
    > ### ** Examples
    > 
    > A.illuminant.spct
    Object: source_spct [97 x 2]
    Wavelength range 300 to 780 nm, step 5 nm 
    Label: CIE A standard illuminant, normalized to one at 560 nm 
    Time unit 1s
    Spectral data normalized to 1 at 560 nm 
    
    Error: 'trunc_mat' is not an exported object from 'namespace:dplyr'
    Execution halted
    ```

*   checking dependencies in R code ... WARNING
    ```
    Missing or unexported object: ‘dplyr::trunc_mat’
    ```

*   checking re-building of vignette outputs ... WARNING
    ```
    Error(s) in re-building vignettes:
    --- re-building ‘userguide-1-radiation.Rmd’ using rmarkdown
    News at https://www.r4photobiology.info/
    
    Attaching package: 'lubridate'
    
    The following objects are masked from 'package:base':
    
        date, intersect, setdiff, union
    
    ...
    
        date, intersect, setdiff, union
    
    --- finished re-building ‘userguide-2-astronomy.Rmd’
    
    SUMMARY: processing the following file failed:
      ‘userguide-1-radiation.Rmd’
    
    Error: Vignette re-building failed.
    Execution halted
    ```

it seems however fixed in the dev version.

[Moved from Bitbucket: original date 2018-06-27]
It would be good to add equations describing the various summaries to help pages and/or vignettes. This requires to make sure that portability is not affected.