csv

A CSV parser and stringifier written entirely in Lumen.

V1 uses a flat string[] representation plus row/column helpers. This keeps the API usable before nested records and richer table structures are standardized.

Use

import { parse, get, rowCount, colCount, stringify } from "https://lumen-lang.org/package/std-contrib/csv/csv.ts";

let src = "name,age\nAda,36\n";

let fields = parse(src);
let rows = rowCount(src);
let cols = colCount(src, 0);
let name = get(src, 1, 0, "");

let out = stringify(["name", "note", "Ada, Lovelace", "said \"hello\""], 2);
let byHeader = getByHeader(src, 1, "name", "");

API

API	Meaning
`parse(src: string): string[]`	Parse all fields into row-major order
`rowCount(src: string): int`	Count rows
`colCount(src: string, row: int): int`	Count columns in a row
`get(src, row, col, fallback): string`	Return one field or `fallback`
`headerIndex(src, name): int`	Return the column index for a header row
`getByHeader(src, row, header, fallback): string`	Return one field using row 0 as headers
`stringify(fields: string[], columns: int): string`	Serialize row-major fields
`parseDelimited(src, delimiter): string[]`	Parse with a custom one-character delimiter
`getDelimited(src, row, col, fallback, delimiter): string`	Read with a custom delimiter
`stringifyDelimited(fields, columns, delimiter): string`	Serialize with a custom delimiter

npm inspiration

This package is inspired by the shape of common npm CSV libraries:

csv-parse and the broader csv project split parsing and stringifying into focused APIs.
Papa Parse exposes parse and unparse; this package uses parse and stringify with Lumen-friendly flat arrays.
fast-csv supports delimited-value parsing/formatting beyond only commas, so V1 includes one-character delimiter helpers.

Streaming, callback APIs, workers, object records, and transform hooks are left out until Lumen has the surrounding runtime and data-shape support.

Supported V1 syntax

comma delimiter
quoted fields
escaped quotes as ""
LF and CRLF line endings
newlines inside quoted fields
empty fields
trailing newline
stringification with required quoting
one-character custom delimiters
header lookup helpers

Out of scope for V1

multi-character delimiters
object records
streaming
comments
automatic type casting
automatic delimiter detection

Test:

lumen test packages/csv/csv.ts