Errata

```
function parseCSV(input: string): {[columnName: string]: string}[] {
const lines = input.split('\n');
const [header, ...rows] = lines;
return rows.map(rowStr => {
const row: {[columnName: string]: string} = {};
rowStr.split(',').forEach((cell, i) => {
row[header[i]] = cell;
});
return row;
});
}
```

I believe this code snippet does not do what the author intended (mapping column names to values), as when `header[i]` is accessed, `header` is a string consisting of the entire header row. So given the following code:

```
console.log(parseCSV(`name,age
Will,23`))
```

The output will be:

```
[{
'n': 'Will',
'a': 23,
}]
```

Rather than what I suspect the author intended:

```
[{
'name': 'Will',
'age': 23,
}]
```

Fixed by:

```
function parseCSV(input: string): {[columnName: string]: string}[] {
const lines = input.split('\n');
const [header, ...rows] = lines;
const headerColumns = header.split(',');
return rows.map(rowStr => {
const row: {[columnName: string]: string} = {};
rowStr.split(',').forEach((cell, i) => {
row[headerColumns[i]] = cell;
});
return row;
});
}
```

Very minor!

(apologies for not listing page number: O'Reilly web viewer does not give an option to display them afaik)

Note from the Author or Editor:
Yes, this is a mistake. The header row needs to be split before being used for column names. (This is why you need tests, not just types!)

Last line of the code example is:

{ '1,2,3': 1 }

The value should be 2, not 1:

{ '1,2,3': 2 }

Line 4 is:

pt.y < polygon.bbox.y[1] || pt.y > polygon.bbox.y[1]) {

The index in the lower bound check should be 0, not 1:

pt.y < polygon.bbox.y[0] || pt.y > polygon.bbox.y[1]) {

Note from the Author or Editor:
This has been fixed in the second release (March 2020).

Line 6 is:

pt.y < x[0] || pt.y > y[1]) {

Instead of comparing with the lower x boundary, it should compare with the lower y boundary:

pt.y < y[0] || pt.y > y[1]) {

Note from the Author or Editor:
This has been fixed in the second release.

To illustrate the "chain" concept in Lodash/Underscore, two code examples are shown, intended to have an equivalent effect.

The traditional version:

_.a(_.b(_.c(v)))

The chained version:

_(v).a().b().c().value()

For these to be equivalent, the order of a/b/c must be switched in one of the examples, e.g.:

_.c(_.b(_.a(v))) and _(v).a().b().c().value()

or:

_.a(_.b(_.c(v))) and _(v).c().b().a().value()

Note from the Author or Editor:
The intent was

_.c(_.b(_.a(v))) and _(v).a().b().c().value()

I've fixed this in the book repo.

The code example is:

function pluck(records, key) {
return record.map(record => record[key]);
}

The argument is `records` in plural, while the variable that is mapped over is the undefined `record`, which also is shadowed in the arrow function. A working version would be:

function pluck(records, key) {
return records.map(record => record[key]);
}

To align with the following examples, it could be changed to:

function pluck(record, key) {
return record.map(r => r[key]);
}

Though I'd argue that the argument is an array, it should be named `records`, not `record` throughout all the examples.

Note from the Author or Editor:
They should all be `records` not `record`. I've made the fix and this will appear in the next release.

devDependencies should include @types/react

Note from the Author or Editor:
@types/lodash should be @types/react. Thanks for catching this mistake!

`npm install --sav-dev ...` should be `npm install --save-dev ...`.