TODO.txt

stack run -- goat
stack run -- wormbreeder

::TODO::
-add some punnet square sorta functions to genotype
-new test cases
  -verify results of SkellyNode on worm test
    -verify reflected sides are indeed reflections of each other locally
    -verify that reflected sides are indeed reflections when parents are centered
-fix missing-home-module warning
-finish lookAt function, and make its interface less weird
-basic test cases
-finish goat breeding app MVP
  -just randomly breed and see what happens?
    -I think I did this already
  -try breeding to maximize volume and see what happens
  -try breeding to maximize bending joints and see what happens
-consider moving parts of Builder.hs into Genetics
DONE-switch benchmarks to relude
  DONE-do benchmarks for skellygen
DONE-do a package cleanup
DONE-switch to relude FFS
IGNORE-fix circleci [![CircleCI](https://circleci.com/gh/pdlla/animalclub.svg?style=svg)](https://circleci.com/gh/pdlla/animalclub)
DONE-change scale to V3 instead of M33
  -see comments in reduceBoneTrans in AnimalScript.hs
IGNORE-switch LocalMesh to use storable vector
  -not really worth doing right now, needs a little extra work to actually make it more performant
DONE-finish exporting genetics stuff to C++
DONE-switch to hspec
DONE/FAIL-look into adding type level size parameter to DNA
DONE-figure out why applicative do causes things to crash
DONE-switch indentation to 2 spaces
  -in vim: :%s;^\(\s\+\);\=repeat(' ', len(submatch(0))/2);g
DONE-look into switching DNA to use fixed vector (type level dimension)
  -it's not compatible with Data.Vector.Storable and not intended for large dimensions
DONE-refactor
  DONE-consider deleting Hierarchical
  DONE-move goat.hs:goatHelper into Builder.hs
DONE D:-do constraint parameterization of Float (so we can switch to Double)
DONE-combine all math libraries into single Math.hs module...
DONE-fix or change the isRoot parameter in AnimalNode
DONE-math
  IGNORE-organize Skellygen/Math ugg
  IGNORE/FAIL-finish invTRS
  DONE-it seems like the S component could have all sorts of stuff going on in it (not just shears)
    -confirm
    -update comments
    -see if this breaks our inherit and invTRS functions
IGNORE-BoneTrans refactor
  -TRS should always have a pure scale component (V3 instead of M33)
  FAIL-applyBoneTrans modifies TRS directly
DONE-refactor
  IGNORE-rename AnimalNode' to AnimalNodeInternal
  DONE-expand skellyfunc to host all possibilities
  DONE-make AnimalNode/BoneName/BoneTrans stuff have better indexing
    -`Bone T.Text [Tags]`
    -add a comment explaining how reflecting bones in AnimalNode works
      -the current implementation is weird, come up with something better
DONE-add clear documentation to how skellygen/AnimalNode skeleton hierarchy stuff works
IGNORE-go through some modules and rename things so that they don't have their modules as prefix
DONE-make separate testing library for matrix.linear
DONE-wtf are you using LHS coordinates for? Switch to RHS if possible or make it agnostic
IGNORE-add comment explaining default animal orientation assumptions
DONE-review skellygen and see if you want to redesign cuz it's weird af