Fork of github:chrisdickinson/varint with BigInt support
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Dominic Tarr 76ce046de9 make a quick benchmark script 9 years ago correct usage notes 9 years ago
bench.js make a quick benchmark script 9 years ago
decode.js instead of throwing, return undefined in invalid (too short) buffers 9 years ago
encode.js use bitwise check instead of >= when number is small 9 years ago
index.js export encodingLength 9 years ago
length.js export encodingLength 9 years ago
package.json 1.2.0 9 years ago
test.js test invalid buffers, set bytesRead = 0, and return undefined 9 years ago


encode whole numbers to an array of protobuf-style varint bytes and also decode them.

var varint = require('varint')

var bytes = varint.encode(300) // === [0xAC, 0x02]
varint.decode(bytes) // 300
varint.decode.bytesRead // 2 (the last decode() call required 2 bytes)


varint = require('varint')

varint.encode(num[, output=[], offset=0]) -> array

encodes num into either the array given by offset or a new array at offset and returns that array filled with integers.

varint.decode(data[, offset=0]) -> number

decodes data, which can be either a buffer or array of integers, from position offset or default 0 and returns the decoded original integer.


if you also require the length (number of bytes) that were required to decode the integer you can access it via varint.decode.bytesRead. this is an integer property that will tell you the number of bytes that the last .decode() call had to use to decode.


similar to bytesRead when encoding a number it can be useful to know how many bytes where written (especially if you pass an output array). you can access this via varint.encode.bytesWritten which holds the number of bytes written in the last encode.


returns the number of bytes this number will be encoded as, up to a maximum of 8.

usage notes

If varint is passed a buffer that does not contain a valid end byte, then decode will return undefined, and decode.bytesRead will be set to 0. If you are reading from a streaming source, it's okay to pass an incomplete buffer into decode, detect this case, and then concatenate the next buffer.