Data wrangling

class: center, middle, inverse, title-slide

# Data wrangling
## The pipe
### Ben Baumer
### SDS 192</br>Sep 28, 2020</br>(<a href="http://beanumber.github.io/sds192/lectures/mdsr_wrangling_02-pipe.html" class="uri">http://beanumber.github.io/sds192/lectures/mdsr_wrangling_02-pipe.html</a>)

---

class: center, middle, inverse
# The pipe

---
## The pipe operator

.pull-left[

.center[![](http://hexb.in/hexagons/magrittr.png)]

- Inspired by pipe (`|`) in UNIX

- Provided by `magrittr` package

]

.pull-right[

![](https://upload.wikimedia.org/wikipedia/en/b/b9/MagrittePipe.jpg)

- [The Treachery of Images](https://en.wikipedia.org/wiki/The_Treachery_of_Images)

- Rene Magritte, 1929

]

---
## How does the pipe work?

![](../gfx/tidy-pipe.png)

---
## Using the pipe

The expression

```r
mydata %>%
  verb(arguments)
```

is the same as:

```r
verb(mydata, arguments)
```

Thus,

```r
function(x, args)
```

has the same effect as

```r
x %>% 
  function(args)
```

---
## Why the pipe?

Instead of having to read/write:

```r
select(filter(mutate(data, args1), args2), args3)
```

You can write:

```r
data %>%
  mutate(args1) %>%
  filter(args2) %>%
  select(args3)
```

---
## Little Bunny Foo Foo

---
## Coding Little Bunny Foo Foo

- Nested form:

```r
bop(scoop(hop(foo_foo, through = forest), up = field_mice), on = head)
```

- With pipes:

```r
foo_foo %>%
  hop(through = forest) %>%
  scoop(up = field_mouse) %>%
  bop(on = head)
```

.footnote[https://github.com/hadley/r4ds/blob/master/pipes.Rmd]