README.Rmd

---
title: decs922 R package
output: github_document
---

This is a package of odds and ends originally created for the class
Data Exploration, DECS-922, now DECS-461, at the [Kellogg School of
Management](https://www.kellogg.northwestern.edu/). 

```{r, echo=FALSE}
library(decs922)
```



# Document production

* Templates to perform routine setup steps to produce beamer slides
  and pdf homework assignments.  Upon package installation these
  become automatically available from RStudio's `rmarkdown` templates
  list, as templates named "`decs922 Beamer presentation`" and "`decs922
  homework submissions`". 
  
  Note that the `decs922 homework
  submission ` template contains a mini-tutorial concerning
  aspects of producing a document with markdown, including
  examples of cross-referencing of tables and figures.
* A `knitr` chunk engine, `soln_notes`, which allows you to insert
  optional text within a standalone chunk with the engine
  `soln_notes`. When evaluated, the text will be rendered in italic
  and (by default) preceded by `Solution notes:`. The goal is to make
  it easier to create homework solutions. The implementation is
  currently functional but buggy.
* **`nbsp`** which takes a string as input and returns the
  string with spaces replaced by non-breaking spaces.
  
# Utility functions

* **`translate_datestr`** parses a string for POSIX
  date codes and returns documentation for the codes (taken from `?strftime`).
  
  ```{r}
  translate_datestr('%Y-%M-%d')
  ```

* **`data_summary`** reports for a data set the number
  of observations, number of missing values, the class of each
  variable, and various summary statistics. I view this as
  pedagogical, as there are *many* similar functions available, such
  as `summary`, `psych::describe`, etc.  Using the
  built-in dataset `iris`:
  
  
  ```{r}
  data_summary(iris)
  ```
* **`wordle_assist`** finds 5-letter words that match against excluded letters and positional
  included letters (a vector named with the letter, with positive values
  specifying confirmed positions and negative values specifying
  positional exclusions). The function matches against 
  the length-5 words in the [`words` package](https://CRAN.R-project.org/package=words). For example, suppose you have played `arose`, `until` and `chump`, and you've learned that there is an `s` in position 1 but not in position 4; there is an `h` in position 2; and there is an `o` in position 3. Enter that information like this:
  ```{r}
  wordle_assist(no = 'areuntilcump', yes = list(s = c(1, -4), h = 2, o = 3))
  ```
  The function returns the remaining possible words.
* **`wordle_sub`** given 1 to 5 letters, return 5-letter words containing
  those letters, in any order.
  ```{r}
  wordle_sub(ltrs='cawk')
  ```

* **`colorkey`** given a regex, returns matching values from
  `colors()`.  For example, to find a version of "orange" which does
  not contain "red" in the color name:

    ```{r, out.width='70%'}
    colorkey(colortext = 'orange', excludetext = 'red')
    ```