wordwrap-stream

Streaming interface to node-wordwrap

Example

const fs = require('fs')
const wordwrap = require('wordwrap-stream')

const file = fs.createReadStream('somefile.txt')
const wrap = wordwrap({stop: 20})

file.pipe(wrap)

CLI

Install with npm install -g wordwrap-stream; all of the following examples are equivalent:

$ wordwrap --stop 20 file.txt > wrapped.txt
$ wordwrap 20 file.txt > wrapped.txt
$ cat file.txt | wordwrap --stop 20 > wrapped.txt

Use wordwrap --help for a full lits of options.

API

wordwrap(wrapOpts, [streamOpts]) -> transformStream

• wrapOpts (object) an options object, with the following keys; these options are passed through to the wordwrap instance, so check those docs for details on each:
  • stop (integer) the column at which text should be wrapped
  • start (optional, integer, default: 0) the offset at which to pad out lines
  • mode (optional, string, default: soft) if words that are longer than stop - start should be forcibly split
  • lengthFn (optional, function, default: String.length) a function that should be used to determine the length of the current chunk; by default it just uses the string's .length property, but should you choose you can pass a function that understands other lengths (like double-width characters, for instance)
• streamOpts (optional, object) options passed to the underlying transformStream; this is where you could specify the streams highWaterMark or put it into objectMode. Read the stream docs for details.

Notes

• Because wordwrap-stream only deals with chunks of data as it gets them, it needs to remember the last line to ensure that there isn't a "break" where a line is too short. It's important to end the stream (or its data source), which will flush the stream.

License

MIT. See LICENSE for details.