- Fixed clobbering pkg functions with indexing objects
- doubled up on things like
is_dblwhich is bad form - just renamed the indexing vectors not to double use the functions with the same name
- not necessary, R is smart, but was ugly
- doubled up on things like
-
Added a namespace test in 'inst'
- simple script to ensure that namespaces are kept low (or to a minimum)
- usage:
Rscript --vanilla inst/test-ns.R
-
Improvements to
enrich_test()- general simplifications
- separate out into S3 methods:
matrix,list, anddata.frame - gets a new print method and new unit tests
- the alternative param is now simply
passed direction to
fisher.test()via the new...which allows further extensibility - simplifies internal logic and params
- user is asked to consult
fisher.test()for further information
-
New unit testing helpers
-
New
jagged_tbl()function- new function from old code base which is useful here
- adds tibble package so not sure if I want to keep the dependency
-
New
file_find()wrapper- documented in filesystem docs
- thin wrapper around
ls_dir()
-
New function
write_latex_tbl()- from old code but good one to have in the back pocket
-
New
piter()function- analogue for
purrr::pmap()allowing for extensible list iteration - allows for
~formulasyntax - also added
~formulasyntax forliter()also
- analogue for
- Removed
calc_brier()from package- now lives in
libmlpackage
- now lives in
- Removed
skip_on_jenkins()
-
cleaned up and improvement
- minor param changes to
snake_case, so downstream effects are possible
- minor param changes to
-
Fixed broken http-links in p-value FDR
- multi-line
\urldoes not work withpkgdownparser
- multi-line
- Initial release! 🥳