Skip to content

wooorm/parse-english

Repository files navigation

parse-english

Build Coverage Downloads Size

Natural language parser, for the English language, that produces nlcst.

Contents

What is this?

This package exposes a parser that takes English natural language and produces a syntax tree.

When should I use this?

If you want to handle English natural language as syntax trees manually, use this.

Alternatively, you can use the retext plugin retext-english, which wraps this project to also parse natural language at a higher-level (easier) abstraction.

For Dutch or most Latin-script languages, you can instead use parse-dutch or parse-latin.

Install

This package is ESM only. In Node.js (version 16+), install with npm:

npm install parse-english

In Deno with esm.sh:

import {ParseEnglish} from 'https://esm.sh/parse-english@7'

In browsers with esm.sh:

<script type="module">
  import {ParseEnglish} from 'https://esm.sh/parse-english@7?bundle'
</script>

Use

import {ParseEnglish} from 'parse-english'
import {inspect} from 'unist-util-inspect'

const tree = new ParseEnglish().parse(
  'Mr. Henry Brown: A hapless but friendly City of London worker.'
)

console.log(inspect(tree))

Yields:

RootNode[1] (1:1-1:63, 0-62)
└─0 ParagraphNode[1] (1:1-1:63, 0-62)
    └─0 SentenceNode[23] (1:1-1:63, 0-62)
        ├─0  WordNode[2] (1:1-1:4, 0-3)
        │    ├─0 TextNode "Mr" (1:1-1:3, 0-2)
        │    └─1 PunctuationNode "." (1:3-1:4, 2-3)
        ├─1  WhiteSpaceNode " " (1:4-1:5, 3-4)
        ├─2  WordNode[1] (1:5-1:10, 4-9)
        │    └─0 TextNode "Henry" (1:5-1:10, 4-9)
        ├─3  WhiteSpaceNode " " (1:10-1:11, 9-10)
        ├─4  WordNode[1] (1:11-1:16, 10-15)
        │    └─0 TextNode "Brown" (1:11-1:16, 10-15)
        ├─5  PunctuationNode ":" (1:16-1:17, 15-16)
        ├─6  WhiteSpaceNode " " (1:17-1:18, 16-17)
        ├─7  WordNode[1] (1:18-1:19, 17-18)
        │    └─0 TextNode "A" (1:18-1:19, 17-18)
        ├─8  WhiteSpaceNode " " (1:19-1:20, 18-19)
        ├─9  WordNode[1] (1:20-1:27, 19-26)
        │    └─0 TextNode "hapless" (1:20-1:27, 19-26)
        ├─10 WhiteSpaceNode " " (1:27-1:28, 26-27)
        ├─11 WordNode[1] (1:28-1:31, 27-30)
        │    └─0 TextNode "but" (1:28-1:31, 27-30)
        ├─12 WhiteSpaceNode " " (1:31-1:32, 30-31)
        ├─13 WordNode[1] (1:32-1:40, 31-39)
        │    └─0 TextNode "friendly" (1:32-1:40, 31-39)
        ├─14 WhiteSpaceNode " " (1:40-1:41, 39-40)
        ├─15 WordNode[1] (1:41-1:45, 40-44)
        │    └─0 TextNode "City" (1:41-1:45, 40-44)
        ├─16 WhiteSpaceNode " " (1:45-1:46, 44-45)
        ├─17 WordNode[1] (1:46-1:48, 45-47)
        │    └─0 TextNode "of" (1:46-1:48, 45-47)
        ├─18 WhiteSpaceNode " " (1:48-1:49, 47-48)
        ├─19 WordNode[1] (1:49-1:55, 48-54)
        │    └─0 TextNode "London" (1:49-1:55, 48-54)
        ├─20 WhiteSpaceNode " " (1:55-1:56, 54-55)
        ├─21 WordNode[1] (1:56-1:62, 55-61)
        │    └─0 TextNode "worker" (1:56-1:62, 55-61)
        └─22 PunctuationNode "." (1:62-1:63, 61-62)

API

This package exports the identifier ParseEnglish. There is no default export.

ParseEnglish()

Create a new parser.

ParseEnglish extends ParseLatin. See parse-latin for API docs.

Algorithm

All of parse-latin is included, and the following support for the English natural language:

  • unit abbreviations (tsp., tbsp., oz., ft., and more)
  • time references (sec., min., tues., thu., feb., and more)
  • business Abbreviations (Inc. and Ltd.)
  • social titles (Mr., Mmes., Sr., and more)
  • rank and academic titles (Dr., Rep., Gen., Prof., Pres., and more)
  • geographical abbreviations (Ave., Blvd., Ft., Hwy., and more)
  • American state abbreviations (Ala., Minn., La., Tex., and more)
  • Canadian province abbreviations (Alta., Qué., Yuk., and more)
  • English county abbreviations (Beds., Leics., Shrops., and more)
  • common elision (omission of letters) (’n’, ’o, ’em, ’twas, ’80s, and more)

Types

This package is fully typed with TypeScript. It exports no additional types.

Compatibility

Projects maintained by me are compatible with maintained versions of Node.js.

When I cut a new major release, I drop support for unmaintained versions of Node. This means I try to keep the current release line, parse-english@^7, compatible with Node.js 16.

Security

This package is safe.

Related

Contribute

Yes please! See How to Contribute to Open Source.

License

MIT © Titus Wormer