data-import.qmd

---
title: "Data import"
---

```{r}
#| results: "asis"
#| echo: false

source("_common.R")
```

## Prerequisites {.unnumbered}

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

## 8.2.4 Exercises {.unnumbered}

1.  For reading a file delimited with `|`, use `read_delim()` with argument `delim = "|"`.

2.  All other arguments are common among the two functions.

3.  `col_positions` is an important argument since it defines the beginning and end of columns.

4.  We need to specify the `quote` argument.

    ```{r}
    read_csv("x,y\n1,'a,b'", quote = "\'")
    ```

5.  Problems with each `read_csv()` statement is shown below:\\

    -   There are only two column headers but three values in each row, so the last two get merged:

        ```{r}
        read_csv("a,b\n1,2,3\n4,5,6")
        ```

    -   There are only three column headers, first row is missing a value in the last column so gets an `NA` there, the second row has four values so the last two get merged:

        ```{r}
        read_csv("a,b,c\n1,2\n1,2,3,4")
        ```

    -   No rows are read in:

        ```{r}
        read_csv("a,b\n\"1")
        ```

    -   Each column has a numerical and a character value, so the column type is coerced to character:

        ```{r}
        read_csv("a,b\n1,2\na,b")
        ```

    -   The delimiter is `;` but it's not specified, therefore this is read in as a single-column data frame with a single observation:

        ```{r}
        read_csv("a;b\n1;3")
        ```

6.  The non-syntactic names can be read in as follows.

    ```{r}
    annoying <- tibble(
      `1` = 1:10,
      `2` = `1` * 2 + rnorm(length(`1`))
    )
    ```

    a.  Extracting the variable called `1`:

    ```{r}
    annoying |>
      select(`1`)
    ```

    b.  Plotting a scatterplot of `1` vs. `2`:

    ```{r}
    ggplot(annoying, aes(x = `2`, y = `1`)) +
      geom_point()
    ```

    c.  Creating a new column called `3`, which is `2` divided by `1`:

    ```{r}
    annoying |>
      mutate(`3` = `2` / `1`)
    ```

    d.  Renaming the columns to `one`, `two`, and `three`:

    ```{r}
    annoying |>
      mutate(`3` = `2` / `1`) |>
      rename(
        "one" = `1`,
        "two" = `2`,
        "three" = `3`
      )
    ```