02_regex_and_casewhen.Rmd

---
title: "Regular expressions, if_else, case_when"
author: "John Little"
date: "`r Sys.Date()`"
output: html_notebook
---

`regex` is used to find patterns in your data.  This is helpful for data cleaning and for normalizing data, e.g. when a sub-field of data needs to be liberated from a more expansive variable.

`if_else` and `case_when` can be used to conditionally change data values.

```{r}
library(tidyverse)
```


## Regex

There's so much to say about regex.  We'll say the absolute minimum and leave you with a [link to more resources](https://cs.lmu.edu/~ray/notes/regex/).  One approach is to watch a short youtube video on the general concept of _regular expressions_.  Then use the [`stringr` library](https://stringr.tidyverse.org) to apply the regex concept in R and the Tidyverse.  For my own approach, I use the [stringr cheat sheet](https://raw.githubusercontent.com/rstudio/cheatsheets/main/strings.pdf) as a help in understanding how to apply the regex concepts in the R context.  Nonetheless, even the cheatsheet can sometimes be confusing.  Regex is a deceptively deep topic.  Start small and build.  

regex symbol | regex function
------------:|:--------------  
        `\d` | a digit  
        `$`  | the end of the pattern  
        `^`  | the beginning of the pattern  
        `\w` | a word character  
        `+`  | a multiplier:  1 or more 
        `*`  | a multiplier:  0 or more 
        `\s` | a space  
        `.`  | a wildcard  (matches all symbols)
        `()` | a capture group  

> NOTE: in R you have to escape the back-slash.  i.e `\\`.  So `\d` is written as `\\d`

```{r}
starwars |> 
  select(height, name) |>
  mutate(last_height_number = str_extract(height, "\\d$")) |> 
  mutate(first_height_number = str_extract(height, "^\\d")) |> 
  relocate(height, .after = first_height_number) |>
  mutate(first_name = str_extract(name, "^\\w+")) |> 
  mutate(numer_of_names = str_count(name, "\\s") + 1) |>
  mutate(number_of_vowels = str_count(name, "[aeiouAEIOU]")) |>
  mutate(switch_names = str_replace(name, "(\\w+)\\s(.*)", "\\2, \\1")) |>
  relocate(name, .after = first_name) |>
  mutate(double_consonants = str_detect(name, "tt|ll|ff|ss|bb|cc|rr|gg")) |>
  mutate(replace_dashes = if_else(str_detect(name, "-"), str_replace_all(name, "-", " FOO "), "")) |>
  mutate(repd = str_replace_all(name, "-", " FOO "))  
  # filter(str_detect(name, regex("skywalker", ignore_case = TRUE)))
```


## `if_else()`

```{r}
starwars |>
  filter(homeworld == "Tatooine") |>
  mutate(skin_color = if_else(sex == "none", "powder coated aluminum", skin_color ))
```

## `case_when()`

```{r}
medium_worlds <- c("Alderaan", "Coruscant", "Kamino")

starwars |>
  filter(homeworld %in% medium_worlds) |>
  select(!is.list) |>
  mutate(notes = case_when(
    homeworld == "Alderaan"  ~ "Extreme destruction",
    homeworld == "Kamino"    ~ "Super-earth",
    homeworld == "Coruscant" ~ "Exoplanet"
  )) |>
  # relocate(notes, .after = name) |>
  select(species, starts_with("n"), homeworld)
```