hypertrie-index

Run an indexing function over a hypertrie.

API

const hypertrieIndex = require('hypertrie-index')

const indexer = hypertrieIndex(hypertrie, opts)

hypertrie is a hypertrie instance. opts is an object of options (with their defaults):

• map: The map function to run on each batch.
• batchSize: Max number of entries to pass into the map function.
• storeState: function (buf, cb), fetchState: function (cb): Function to store and retrieve the current state. buf is a buffer that should be stored as-is in some storage backend, and be passed back into cb(null, buf) in fetchState.
• live: true Keep watching for changes
• prefix: '': Set a prefix to only index and watch a part of the trie.
• transformNode: false: Pass all messages through hypertrie.transformNode

The map function will receive an array of change messages. The messages are in the format returned from hypertrie.diff: { left: { key, value, seq }, right: { key, value, seq } }. If passed through hypertrie.transformNode, they are in the format { key, value, seq, deleted, previousValue } where deleted is false for puts and true for deletes. In case of updates previousValue will contain the previous value.

indexer.pause()

Pause the indexing.

indexer.resume()

Resume the indexing.

indexer.on('indexed', function (batch) {}

Emitted after each round of indexing. batch is the array of change messages.

indexer.on('finished', function () {})

Emitted when indexing finishes.