[READ-ONLY] Mirror of https://github.com/glacials/splits-io. a speedrunning data store and analysis engine splits.io

buffer-writer #

Build Status

Fast & efficient buffer writer used to keep memory usage low by internally recycling a single large buffer.

Used as the binary protocol writer in node-postgres

Since postgres requires big endian encoding, this only writes big endian numbers for now, but can & probably will easily be extended to write little endian as well.

I'll admit this has a few postgres specific things I might need to take out in the future, such as addHeader

api #

var writer = new (require('buffer-writer')());

writer.addInt32(num) #

Writes a 4-byte big endian binary encoded number to the end of the buffer.

writer.addInt16(num) #

Writes a 2-byte big endian binary encoded number to the end of the buffer.

writer.addCString(string) #

Writes a string to the buffer utf8 encoded and adds a null character (\0) at the end.

var buffer = writer.addHeader(char) #

Writes the 5 byte PostgreSQL required header to the beginning of the buffer. (1 byte for character, 1 BE Int32 for length of the buffer)

var buffer = writer.join() #

Collects all data in the writer and joins it into a single, new buffer.

var buffer = writer.flush(char) #

Writes the 5 byte postgres required message header, collects all data in the writer and joins it into a single, new buffer, and then resets the writer.

thoughts #

This is kind of node-postgres specific. If you're interested in using this for a more general purpose thing, lemme know. I would love to work with you on getting this more reusable for your needs.

license #

MIT