Schema Builder

Build Schema

Finalizing a schema with .build() and reusing it across multiple workbooks.

Finish your schema with .build().

import { createExcelSchema
 } from "@chronicstone/typed-xlsx";

const schema
 = createExcelSchema
<{ firstName
: string; lastName
: string }>()
  .column
("firstName", { accessor
: "firstName" })
  .column
("lastName", { accessor
: "lastName" })
  .build
();

Calling .build() freezes the builder output into a plain schema definition that can be reused across:

several buffered workbooks
several stream exports
different column selections
different sheet layouts

In practice, this is the schema object you pass into .table({ schema, ... }).

The schema type carries all declarations

The returned SchemaDefinition type is generic over your row type, all column IDs, all group IDs, and all group context shapes. Nothing is erased by .build():

import { createExcelSchema
 } from "@chronicstone/typed-xlsx";
import type { SchemaColumnId
, SchemaGroupId
, SchemaGroupContext
 } from "@chronicstone/typed-xlsx";

type User
 = { name
: string; email
: string; orgs
: number[] };

const schema
 = createExcelSchema
<User
>()
  .column
("name", { accessor
: "name" })
  .column
("email", { accessor
: "email" })
  .group
("orgIds", (b
, ids
: number[]) => {
    for (const id
 of ids
) b
.column
(`org-${id
}`, { accessor
: (r
) => r
.orgs
.includes
(id
) });
  })
  .build
();

type Columns
 = SchemaColumnId
<typeof schema
>;

type Groups
 = SchemaGroupId
<typeof schema
>;

type Context
 = SchemaGroupContext
<typeof schema
>;

Passing this schema to .table() constrains:

rows to User[]
select.include and select.exclude to "name" | "email" | "orgIds"
context to { orgIds: number[] } when the orgIds group is selected

Summaries

Reducer-based column summaries that work in both buffered and stream mode.

Cell Styling

Static and dynamic CellStyle, header styles, border variants, number formats, and column widths.