In KoderKow/kowr: Utility Functions

embed_png <- function(path, dpi = NULL) {
  meta <- attr(png::readPNG(path, native = TRUE, info = TRUE), "info")
  if (!is.null(dpi)) meta$dpi <- rep(dpi, 2)
  knitr::asis_output(paste0(
    "<img src='", path, "'",
    " width=", 400,
    " height=", 300,
    " />"
  ))
}
knitr::opts_chunk$set(comment = "#>", collapse = TRUE)
library(tibble)

Function Examples

Load the Package

library(kowr)

---

back_to_forward()

Primarily for Windows users. When copying file paths in Windows it will use a backslash to separate files, this will throw an error in R. The process of manually changing the backslashes to forward slashes can get highly repetitive. The back_to_forward() function will take what is in the user's clipboard and paste the path with the wanted forward slashes. There are two approaches to this.

Example file path saved in clipboard: C:\Documents\Newsletters\Summer2018.csv

Option 1) Run back_to_forward() in the Console

Run back_to_forward() in the console. This will output a string with forward slashes.

Option 2) Attatch back_to_forward() to a Hotkey

back_to_forward() is also an add-in called Back to Forward Slash. In RStudio near the top click Addins and you will see the kowr package with the add-in. If you click this add-in it will paste the file path wherever your cursor is located within RStudio. This is not the most ideal way to use this. Since backward_to_forward() is an add-in we can assign the add-in Back to Forward Slash to a hotkey.

1. Go to Tools > Modify Keyboard Shortcuts... in RStudio
2. Sort by Name and find Back to Forward Slash
3. Assign a hotkey to it, on my Mac I used CMD + /
4. Test out the hotkey with the path in your clipboard!

Arguments

• render: A Logical. Defaults to TRUE. Exists primarily for the sake of internal testing.
• text: A String. Default uses the text in your clipboard. This should not be altered from the default. Exists primarily for the sake of internal testing.

---

clear_plotly_options()

Default plot_ly() output:

library(plotly)

p <- plot_ly(data = mtcars, x = ~mpg, y = ~hp, type = "scatter", mode = "markers")

p

embed_png("plotly_with_options.png")

Remove all buttons on top of the plot:

p %>% 
  clear_plotly_options()

embed_png("plotly_no_options.png")

Keep "Zoom in" and "Zoom out" buttons:

p %>%
  clear_plotly_options(buttons_to_keep = c("zoomIn2d", "zoomOut2d"))

embed_png("plotly_zoom_options.png")

Arguments:

• plotly_object A plotly object.
• buttons_to_keep A string. Default is NULL. Use a vector of strings for multiple buttons to keep. Below is a list of items with the plotly code that corresponds to the text displayed on the plotly plot when hovering over the buttons.
• toImage: Download plot as a png
• zoom2d: Zoom
• pan2d: Pan
• select2d: Box Select
• lasso2d: Lasso Select
• zoomIn2d: Zoom in
• zoomOut2d: Zoom out
• autoScale2d: Autoscale
• resetScale2d: Reset axes
• toggleSpikelines: Toggle Spike Lines
• hoverClosestCartesian: Show closest data on hover
• hoverCompareCartesian: Compare data on hover
• keep_logo A logical. Default is FALSE. Set to TRUE to keep the "Produced with Plotly" button located in the top right of the plot.

---

round_up()

Round whole numbers up. This function will always round up to the number place second to the left:

round_up(x = 123456)

Numbers with only 2 numeric places (tens), then it will round on the tens position value:

round_up(x = 19)

If needed, we can change the value to a dollar amount. Using as_dollar = TRUE will change the numeric value to a character:

round_up(x = 1991, format = "dollar")

Arguments

• x A numeric value.
• format A boolean. Takes three different options to round up and stylize:
• normal: The default, return the number with no formatting.
• dollar: Return the number with a leading dollar sign along with commas.
• comma: Return the number with commas.

---

snake_to()

This is useful for when the user wants to use the columns in a presentation such as reports, plots, or dashboards.

I find myself using stringr::str_to_title() for a clean way of presenting columns in documents and dashboards. This is the reason the default argument for snake_to()'s output is "title". This will only work with snake case (my preferred naming syntax). If you use snake_to() on camelCase then it will capitalize it as if it was one word.

dat <- tibble(
  this_will_get_cleaned    = c(1, 2, 3),
  thisWillNot    = c("a", "b", "c"),
  cleans_snake_only = c("q", "w", "e"),
  notCamcel = c(10, 12, 14)
)

dat %>%
  snake_to()

Sentence case:

dat %>%
  snake_to(format = "sentence")

Title case and keep column names only:

dat %>%
  snake_to(names_only = TRUE)

Title case and capitalize the acronyms

acronyms = c("GET")
dat %>% 
  snake_to(acronyms = acronyms)

ggplot titles:

snake_to() will also clean up ggplot x and y axes. Pipe the saved ggplot object into snake_to() and it will clean up snake case titles. If there is a legend the title will be capitalized using str_to_title().

library(ggplot2)

p <- dat %>% 
  ggplot(aes(x = thisWillNot, y = this_will_get_cleaned)) +
  geom_col()

p

p %>% snake_to()

Arguments

• object A data.frame, ggplot or character vector. A data.frame will have transformed column names. A ggplot object will have transformed X and Y axes.
• format A string. The desired target (default is "title") case with options including:
• "title" produces title case
• "lower" produces lower case
• "normal" do not transform the string
• "sentence" produces sentence case
• "upper" produces upper case
• acronyms A Character. Default NULL. For when acronyms exist in the column names that need to be capitalized. Pass a character vector for when there is more than one acronym. Upper and/or lower case acronyms in this parameter will be accepted. This will only capitalize the wanted words, words that may contain the acronyms letter will NOT be capitalized.
• names_only A Logical. Default FALSE. If TRUE, snake_to() will return a vector of transformed column names.

KoderKow/kowr documentation built on July 19, 2021, 4:18 p.m.