data-wrangling.Rmd

# Data Wrangling {#wrangling}

## Read csvs

### I want to... 

Read a list of csv and turn them all into a data frame. 

### Here's how to:   

```{r message = FALSE}
files <- list.files(pattern = "csv", full.names = TRUE) %>% 
  set_names(basename(.))
full <- map_df(files, readr::read_csv, .id = "file")
```

### Ok, but why? 

`map_df` maps the read_csv on all the files, and returns a data.frame. The `.id` is used to keep track of the original file. 

### See also 


---

## Remove the NA

### I want to... 

Remove the NA from my list. 

### Here's how to:   

```{r}
l <- list(1, NA, "a", "b", NA)
discard(l, is.na)
```

### Ok, but why? 

`discard` is the opposite of `keep`, and removes all the elements that validate a condition. 

### See also 



---

## List append

### I want to... 

Add an element at the end of all my sublists. 

### Here's how to: 

```{r}
l <- list(list(1:10, 1:20), list(20:30, 40:50))
modify_depth(l, 2, ~ prepend(.x, sum(.x)))
```

### Ok, but why? 

`modify_depth` allows to map function in sublists at a given depth. 

`prepend` adds a element to the list at a givent index. If no index is given, the element is put at the beggining of the list. 

### See also

---

## Position matching

### I want to... 

Find the first argument that matches a condition. 

### Here's how to: 

```{r}
l <- 22:143
detect_index(l, ~ sqrt(.x) == 9)
```

### Ok, but why? 

`detect_index` returns the position of the first argument matching the condition. `detect` alone returns the value. 

You can pass the `.right` argument to these function, so that the search starts from the end of the list. 

### See also

---

## Cut a list 

### I want to... 

Takes all the arguments from a list until the condition is met. 

### Here's how to: 

```{r}
l <- 22:143
head_while(l, ~ sqrt(.x) < 9)
```

### Ok, but why? 

`head_while`, and its counterpart `tail_while`, take a list and perform a `head` or `tail` until a predicate is validated. 

### See also