# csvnorm > Command line tool to normalize CSV and *SV files. Latest version **1.2.0** (published 2026-05-27) · ISC license · 0 weekly downloads ## Install ```sh npm install csvnorm pnpm add csvnorm yarn add csvnorm bun add csvnorm ``` Provides the command `csvnorm`. ## Health **Score 60/100 (C)** — status: active. Positive: has types; esm support; no vulnerabilities; high quality score. Warnings: low downloads. ## Facts | | | |---|---| | Version | 1.2.0 | | Published | 2026-05-27 | | First published | 2017-04-13 | | Weekly downloads | 0 | | License | ISC | | TypeScript types | bundled | | Module format | ESM + CommonJS | | Dependencies | 13 | | Unpacked size | 128.4 KB | | Known vulnerabilities | 0 (+1 in 1 direct dependencies) | | Install scripts | no | | GitHub stars | 6 | | Author | Adrian Sieber | | Maintainers | adius | | Keywords | csv, normalize | ## Links - npm: https://www.npmjs.com/package/csvnorm - Repository: https://github.com/ad-si/csvnorm - Homepage: https://github.com/ad-si/csvnorm#readme - Issues: https://github.com/ad-si/csvnorm/issues - npm.io page: https://npm.io/package/csvnorm ## Dependencies (13) - [execa](https://npm.io/package/execa.md) ^9.6.1 - [tempy](https://npm.io/package/tempy.md) ^3.2.0 - [yargs](https://npm.io/package/yargs.md) ^18.0.0 - [chardet](https://npm.io/package/chardet.md) ^2.1.1 - [fs-extra](https://npm.io/package/fs-extra.md) ^11.3.5 - [csv-parse](https://npm.io/package/csv-parse.md) ^6.2.1 - [iconv-lite](https://npm.io/package/iconv-lite.md) ^0.7.2 - [@types/node](https://npm.io/package/@types/node.md) ^25.9.1 - [@types/execa](https://npm.io/package/@types/execa.md) ^2.0.2 - [@types/yargs](https://npm.io/package/@types/yargs.md) ^17.0.35 - [csv-stringify](https://npm.io/package/csv-stringify.md) ^6.7.0 - [@types/fs-extra](https://npm.io/package/@types/fs-extra.md) ^11.0.4 - [@types/tempfile](https://npm.io/package/@types/tempfile.md) ^3.0.0 ## Alternatives - [csv-to-markdown-table](https://npm.io/package/csv-to-markdown-table.md) — 47.0K weekly downloads - [@sapphire/ratelimits](https://npm.io/package/@sapphire/ratelimits.md) — 4.4K weekly downloads - [js-csvparser](https://npm.io/package/js-csvparser.md) — 2.0K weekly downloads - [@adadapted/js-sdk](https://npm.io/package/@adadapted/js-sdk.md) — 251 weekly downloads - [@grapecity/spread-sheets-sparklines](https://npm.io/package/@grapecity/spread-sheets-sparklines.md) — 103 weekly downloads ## Recent versions - 1.2.0 (latest) — 2026-05-27 - 1.1.0 — 2022-03-01 - 1.0.0 — 2022-03-01 - 0.10.0 — 2020-10-26 - 0.9.1 — 2020-01-18 - 0.9.0 — 2020-01-18 - 0.8.0 — 2020-01-02 - 0.7.0 — 2019-12-08 - 0.6.0 — 2019-01-20 - 0.5.0 — 2019-01-08 - 0.4.3 — 2017-10-20 - 0.4.2 — 2017-10-10 - 0.4.1 — 2017-10-10 - 0.4.0 — 2017-07-03 - 0.3.1 — 2017-06-28 - … 2 more at https://npm.io/package/csvnorm/versions ## README # Csvnorm Command line tool to normalize CSV and \*SV files. Steps: - Convert to UTF-8 encoding - Replace separator with `,` - Reformat - Date columns to ISO8601 - Number columns to `1456.25` - Currency columns to `1539.16 $` ## Encoding Detection The input encoding is detected in this order: 1. **Byte Order Mark (BOM)** — UTF-8, UTF-16 LE/BE, and UTF-32 LE/BE BOMs are recognized and stripped from the output. 2. **UTF-16 heuristic** — Files without a BOM are identified by the alternating zero-byte pattern of ASCII text encoded as UTF-16 LE or BE. 3. **Strict UTF-8 validation** — The byte stream is checked against the UTF-8 grammar (rejecting overlong sequences and surrogate halves). 4. **[chardet](https://github.com/runk/node-chardet) fallback** — Used to discriminate between single-byte encodings such as ISO-8859-x and windows-125x. 5. **windows-1252** as the last resort if every previous step is inconclusive. The detected encoding is shown in the interpretation summary when the output is a terminal, and can be overridden via `--encoding`. ## CLI Version ### Installation ```sh npm install --global csvnorm ``` ### Usage ```txt Usage: csvnorm [Options] INFILE [> OUTFILE] csvnorm [Options] < INFILE [> OUTFILE] Options: --date-format Specify an additional prioritized input date format [string] --encoding Overwrite detected input encoding [string] --in-place Normalize CSV file in place [boolean] [default: false] --iso-datetime Output datetimes with format YYYY-MM-DD[T]HH:mm:ss.SSS[Z] [boolean] [default: false] --skip-start Skip lines at the start of the input [number] [default: 0] --skip-end Skip lines at the end of the input [number] [default: 0] --version Show version number [boolean] --help Show help [boolean] Examples: csvnorm input.csv > normalized.csv Normalize a CSV file cat input.csv | csvnorm > normalized.csv Pipe and normalize a CSV file csvnorm --date-format "dd/mm/yyyy" i.csv Normalize a CSV file with an unusual date format ``` ## Node Module ### Installation ```sh npm install --save csvnorm ``` `csvnorm` is an ES module and requires Node.js 20 or later. ### Usage With files: ```js import csvnorm from "csvnorm" csvnorm({ filePath: csvFilePath, inPlace: true, }) ``` With streams: ```js import csvnorm from "csvnorm" csvnorm({ readableStream: process.stdin, writableStream: process.stdout, }) ``` **Warning:** Numbers from `1,000` to `999,999` with 3 decimal places are parsed as floats (`1.000` to `999.999`). Only numbers larger than `1,000,000` or numbers with less or more than 3 decimal places can be unambiguously parsed as integers (`1000000`). --- _Source: https://npm.io/package/csvnorm · Machine-readable twin of the npm.io package page. Health data is recomputed on every publish._