Schema

Schemas define how each row from CSV/Sheet will be transformed into.

Let's assume this table is the input CSV data:

ID	Name
1	A
2	B

Column

Column() map value from the column in the CSV/Sheets.

If we describe schema as:

import { Column, asString } from 'sheethuahua';

const schema = Column('Name', asString());

In each row, I want a value from column "Name" and parse is as a string.

Then we expect output to be like:

['A', 'B'];

Object

Object() define an object data structure.

If we describe schema as:

import { Column, Object, asNumber, asString } from 'sheethuahua';

const schema = Object({
	id: Column('ID', asNumber()),
	name: Column('Name', asString()),
});

Then we expect output to be like:

[
	{ id: 1, name: 'A' },
	{ id: 2, name: 'B' },
];

Tuple

Tuple() define an array of the exact set of items.

If we describe schema as:

import { Tuple, Object, asNumber, asString } from 'sheethuahua';

const schema = Tuple([
	id: Column('ID', asNumber()),
	name: Column('Name', asString()),
])

Then we expect output to be like:

[
	[1, 'A'],
	[2, 'B'],
];

Nested

Both Object and Tuple can be nested.

import { Column, Object, Tuple, asNumber, asString } from 'sheethuahua';

const schema = Object({
	id: Column('ID', asNumber()),
	name: Column('Name', asString()),
	contact: Object({
		email: Column('Email Address', asString()),
		phone: Column('Phone Number', asString().optional()),
	}),
	links: Tuple([Column('Link 1', asString()), Column('Link 2', asString())]),
});

NOTE

Any column in CSV that is not referenced in the schema will be ignored.

Infer Type

Use StaticDecode to infer schema type

import { type StaticDecode } from 'sheethuahua';

// type Person: {
//     id: number;
//     name: string;
// }
type Person = StaticDecode<typeof schema>;