leveldown-plus

lighter wrapper around leveldown: adds encodings and pull-streams.

I wanted to see if this would be faster that levelup, which now has several layers of api. tl;dr streams are faster (mainly because of pull-streams) but other apis don't change. This had improved some flumeview-level benchmarks, but made no difference to bench-ssb

api: LevelDownPlus(leveldown, opts) => db

return a leveldown-plus database, given a leveldown instance and options object with keyEncoding, valueEncoding.

open (cb)

open the database ready for reading

batch (ops, cb)

write a batch. (levelup style) there are no other write methods.

get (key, opts?, cb)

get a value for the given key.

post (range, listener) => remove

calls listener when there is a write within range (which is gt, lt, gte, lte range)

stream (opts)

returns pull stream, use gt, lt, gte, lte ranges, and pull-level options limit,reverse,live,old, and keyEncoding, valueEncoding

License

MIT