jonocarroll

## Goal: reproduce e.g. https://www.reddit.com/r/oddlysatisfying/comments/uc054a/lissajous_polygons

using Plots
import GeometryBasics: Point

## https://jcarroll.xyz/2022/04/07/interpolation-animation-in.html 
interpolate(a, b) = t -> ((1.0 - t) * a + t * b)

## define the vertices of an N-gon

jonocarroll

starts_with_ordered <- function(match, decreasing = FALSE) {
  all_vars <- tidyselect::peek_vars()
  matches <- tidyselect::starts_with(match)
  named_matches <- setNames(matches, all_vars[matches])
  named_matches[order(names(named_matches), decreasing = decreasing)]
}

mtcars |>
  dplyr::select(starts_with_ordered("c", decreasing = FALSE)) |> 
  head()

jonocarroll

library(SummarizedExperiment)

## create a SummarizedExperiment
se <- SummarizedExperiment(assays = 
                             list(
                               counts = matrix(0:99, 10, 10), 
                               cpm = matrix(200:299, 10, 10)
                             )
                           )
## add tabular metadata with assay-specific properties

jonocarroll

## Inspired by 
##  https://twitter.com/ted_dunning/status/1435027697386721280?s=20&t=cDVb0XOQRJeOjXoTrOz54w

using Plots 

## These are required to be imported to write methods for them

import Base:*
import Base:+

jonocarroll

## How should we write non-reactive (business logic) functions so that they 
## are maximally compatible with shiny/reactive processing?

## Here's a function which performs business logic. It may be used at the 
## console or it may be used in a shiny app.
## A shiny app will probably provide the right type of input so the 
## input validations are perhaps not as critical, but a typical design 
## is to stop() if a critical condition is not met during calculations.
var_top_cars <- function(d, minval) {
  stopifnot("mpg not found" = utils::hasName(d, "mpg"))

jonocarroll

library(ggplot2)
library(dplyr)

## fake data
d <- mtcars %>%
  group_by(cyl) %>%
  summarise(lower = min(disp),
            med = median(disp),
            upper = max(disp))
p <- ggplot(d, aes(x = factor(cyl), y = med)) +

jonocarroll

library(COVID19)
library(dplyr)
library(zoo) # rollmean
library(ggplot2)
library(ggeasy) # easy_add_legend_title

#### deaths ####

## gather USA data at state level and identify weekends
us <- covid19("USA", level = 2) %>% 

jonocarroll

# Fork of https://gist.github.com/johnburnmurdoch/1d23978fc8213ff656d9f629608dd1f5/revisions
# modified to work with https://github.com/nacnudus/google-location-coronavirus

# Install and load required packages
# install.packages("needs")
# library(needs)
# needs(tidyverse, magrittr, animation, pdftools, png, scales)
library(tidyverse)
library(magrittr)
library(animation)

jonocarroll

library(htmltools)
library(d3r)
library(eulerr)

ui <- list(
  attachDependencies(
    tagList(),
    d3_dep_v5()
  ),
  titlePanel("Interactive Set Selector"),

jonocarroll

fastsort <- function(x, partial) {
  y <- if (length(partial) <= 10L) {
    partial <- .Internal(qsort(partial, FALSE))
    .Internal(psort(x, partial))
  }
  else {
    .Internal(qsort(x, FALSE))
  }
  y
}