I38 generic var templates #40
Conversation
This commit adds the 'as.df' parameter to relevant functions, allowing users to choose whether results should be returned as data frames, including the exact date, or start and end dates of climate extremes. Additionally, utility functions like 'ymd.dates' and 'exact.date' are introduced for date handling.
There was a problem hiding this comment.
Great work, Quintin! The code is very clean and clear.
The tests look solid, a lot of obvious cases covered, but I have not dug in to think through all the non-obvious ones. Are there any that occur to you after the fact?
A kind of generic question: Is there a prettifier/formatter for R? It seems possible it might be useful to standardize the code format, as I see a few indentation mistakes/inconsistencies and the like. (For application in a separate PR, I would think.)
R/generic_stats.R
Outdated
| #' If `include.exact.dates = TRUE`, exact dates are returned for max/min statistics. | ||
| #' The function can also filter the data based on a specified degree range of directions (using `direction.range`). | ||
| #' | ||
| #' @seealso \code{\link{compute.stat.scalar}}, \code{\link{compute.gen.stat}} |
There was a problem hiding this comment.
It's possible something might be a bit wonky with the docs ... this doesn't look as if it corresponds to what's in the doc PDF you sent me, see pp 64-65 of that document.
There was a problem hiding this comment.
Would you mind providing a few more details on this? If it pertains to section reordering, that behavior is expected from roxygen.
There was a problem hiding this comment.
Everything looks right down to the Details section bottom of p. 64. Then text reading
Value
A list containing the computed statistic for magnitude and, if applicable, the computed statistic for
direction. For circular statistics (e.g., circular_mean, circular_sd), the result is returned for
directions in degrees. If include.exact.dates = TRUE, the function returns exact dates for the
max/min statistics.
Note
This function is designed for vector climate data, where the data includes both a magnitude and
direction component. For scalar data, use compute.stat.scalar instead.
appears to be inserted before the See Also section on p. 65, where everything appears to pick up normally again.
Maybe this is roxygen, but it doesn't semantically make sense to me.
There was a problem hiding this comment.
Thank you for pointing out this discrepancy. I've adjusted the roxygen comments so that their ordering matches the generated PDFs.
Roxygen automatically reorders the documentation tags to match the base R standard, as mentioned here:
"Note also that the order in which tags appear in your roxygen comments (or even in handwritten .Rd files) does not dictate the order in rendered documentation. The order of presentation is determined by tooling within base R."
I looked for an official documentation order, and the closest I found was this Writing R Extensions manual, but it doesn't encompass the full list of available tags in roxygen.
Thanks, Rod!
I will continue to think of test cases while I work on #39. They will likely be focused on input validation.
There are two complementary packages, styler and lintr. Styler will prettify and lintr ensures conformity with a specific style (the Tidyverse style) and can be added to our CI workflow. I'll open a new issue for this. |
Description
This PR introduces new classes and functionality to handle generic scalar and vector climate variables, allowing users to compute statistics for a broader range of climate data.
Generic Scalar and Vector Classes
ClimdexGenericScalarandClimdexGenericVectorclasses for scalar variables (e.g., humidity, snow depth) and vector variables (e.g., wind speed and direction).Scalar Data Handling
compute.stat.scalar()function to compute statistics (mean, max, min, sum, sd, var) for scalar data.climdexGenericScalar.csv) and construct raw scalar objects (climdexGenericScalar.raw).Vector Data Handling
compute.stat.vector()function for vector data, supporting statistics: max, min, mean, sum, circular_mean, sd, and circular_sd."polar","cartesian", and"cardinal".direction.range.climdexGenericVector.csv) and construct raw vector objects (climdexGenericVector.raw).Circular Statistics Support
compute_circular_mean()andcompute_circular_sd()for directional data (e.g., wind direction), using thecircularpackage.Data Filtering by Direction
filter_by_direction_range()to filter vector data based on direction range.Utility Functions
convert_cartesian_to_polar()convert_polar_to_cartesian()convert_degrees_to_cardinal()for vector data in different coordinate formats.
Tests Added
circular_meanandcircular_sd.