Skip to content

Latest commit

 

History

History
87 lines (66 loc) · 2.49 KB

TODO.md

File metadata and controls

87 lines (66 loc) · 2.49 KB

TODO notes for polyply package

This package is hosted on github, so any issues and bug reports should be posted there. Here we document those large-scale features that are to be added to the package.

TODO notes by feature

TODO notes by type

Tests

  • [+] Construction of a poly_frame
  • [+] merge_fn attribute of a poly_frame
  • [+] Merge a poly_frame
  • [+] Convert a poly_frame back to a list of data-frames
  • [+] Slice/extract on a poly_frame
  • [+] 'active' attribute of a poly_frame
  • [+] activate method
  • [+] filtering pass-through after activate
  • [+] lintr-checking test
    • Dev must use devtools::test() or devtools::check(check_dir = ".") to ensure that source-code is available for this test to run
    • No source code is present upon which to lint when using devtools::check()
  • [+] select columns from a given data-frame

Classes

poly_frame

  • [+] construction using as_poly_frame
  • [+] underlying structure is a list of data-frames
  • [+] store a merging function
  • [+] merge.poly_frame method
  • [+] as.list.poly_frame method
  • [+] activate.poly_frame method for numeric indices
  • [] activate.poly_frame method for by-name indexing
  • [+] activate.poly_frame method for as-variable indexing
  • [+] filter/filter_ methods
  • [+] set_merge_fn method
  • [+] mutate.poly_frame method
  • [+] select method

Functions

  • [+] as_poly_frame with bare df-list
  • [+] as_poly_frame with merging function
  • [+] poly_frame function with '...'-expanded data-frame set and a merging-function
  • [] as_poly_frame.ExpressionSet and as_poly_frame.DGEList

Documentation

  • [+] Style-guide
  • [] Vignette
    • [+] Plot brain ~ body split by Order for 'Animals' dataset
    • Explain why I think polyply provides a neater way of storing datasets that will subsequently be combined
  • [] Memory usage while working with polyply:
    • How big can a vector be in R and does this impose constraints on the size of tables we can join in-memory
  • [] Add a CONTRIBUTING.md

Example data

  • [+] Animals from MASS (suitably reformatted);
    • add a second dataframe mapping common-name to species
    • add a third data-frame mapping species to Family
    • and then Family to Order

Git structure

  • [] Add develop branch
  • [] Add initial tag to master

Continuous Integration

  • [+] Add Travis file
  • [+] Ensure lintr tests in test_package_style.R fail in Travis-CI when an inappropriately formatted file is present