---
title: "14.1 Vectors and Lists"
---

```{r}
#| label: setup
#| include: false
knitr::opts_chunk$set(
  echo = TRUE,
  message = FALSE,
  warning = FALSE,
  fig.width = 6,
  fig.height = 3.6,
  fig.align = "center",
  dpi = 120,
  error = TRUE,
  fig.path = "_figs/vectors-lists/"
)
library(tidyverse)
```

[ keeps structure. [[ extracts one. A data frame is a list of columns.

## 1. Two kinds of vectors

R has two kinds of vectors: **atomic** vectors and **lists**.

![](files/graphics/data-structures-overview.png)

An atomic vector holds one type: logical, integer, double, character (plus complex and raw, which we will not use). A list can hold mixed types, including other lists.

```{r}
#| label: chunk-1
x <- c(1L, 10L, 2L)
typeof(x)
length(x)
```

`c(1, 10, 2)` is double. The `L` makes an integer. `typeof()` is the storage type. `length()` is how many elements.

A factor is an integer with labels. A `Date` is a double. Both still behave like atomic vectors when you subset them.

---

## 2. `[` keeps structure; `[[` extracts one

`[` returns the same kind of object, names included. `[[` pulls out **one** element and drops the wrapper. `$` is `[[` for a name.

```{r}
#| label: chunk-2
x <- c(horse = 7, man = 1, dog = 8)
x[2:3]
x[c("man", "horse")]
x[-3]
x[[3]]
```

`x[3]` is still a named vector of length 1. `x[[3]]` is the number `8`.

A logical vector of the same length picks the `TRUE` positions:

```{r}
#| label: chunk-3
x[c(TRUE, FALSE, TRUE)]
```

---

## 3. Recycling and names

R recycles a shorter vector to the length of the longer one:

```{r}
#| label: chunk-4
x <- c(1, 2, 3, 4)
x + 10
# same idea as x + c(10, 10, 10, 10)
```

Recycling a non-scalar (`x + c(10, 20)`) is legal and easy to misread. Prefer a scalar, or write a vector of the same length.

`names(x)` is a character vector. You can set names when you create the object:

```{r}
#| label: chunk-5
c(Yoshi = 10L, Mario = 31L)
```

---

## 4. Lists hold mixed types

```{r}
#| label: chunk-6
my_list <- list(
  x = "a",
  y = 1,
  z = c(1L, 2L, 3L)
)
```

`[` on a list returns a **list**. `[[` or `$` returns the element inside.

```{r}
#| label: chunk-7
my_list[1]
my_list[[1]]
my_list$y
```

Remove an element by assigning `NULL`:

```{r}
#| label: chunk-8
my_list$x <- NULL
```

---

## 5. A data frame is a list of columns

A tibble is a list. Each element is a column. That is why `df[[2]]` is the second column and `df$species` is the `species` column.

A list can also hold a tibble as one of its elements. Subset the list to get the tibble; then use dplyr on that tibble.

---

## 6. Practice

1. Create

```{r}
#| label: chunk-9
x <- c(
  Yoshi = 10L,
  Mario = 31L,
  Luigi = 72L,
  Peach = 11L,
  Toad  = 38L
)
```

Extract Yoshi and Peach four ways: positive integers, negative integers, a logical vector, and names.

1. Build a wedding-style list: a `venue`, a `guest` tibble (name, meal, age), a `bride`, a `groom`, and a `date`. Remove one guest row from the embedded tibble. Add a named vector that decodes the meal codes. Extract the venue two ways.