kv-toolbox

A set of tools for working with Deno KV.

Batched Atomic

A set of APIs for dealing with the limitation of atomic commit sized in Deno KV, where currently only 10 operations operations can be part of a commit.

batchedAtomic()

Similar to Deno.Kv#atomic(), but will batch individual transactions across as many atomic operations as necessary.

The commit() method will return a promise which resolves with an array of results based on how many batches the operations was broken up into.

Blob

A set of APIs for storing arbitrarily sized blobs in Deno KV. Currently Deno KV has a limit of key values being 64k. The set() function breaks down a blob into chunks and manages sub-keys to store the complete value. The get() function reverses that process, and remove() will delete the key, sub-keys and values.

set()

Similar to Deno.Kv.prototype.set(), in that it stores a blob value with an associated key. In order to deal with the size limitations of values, set() will transparently chunk up the blob into parts that can be handled by Deno KV.

The blob can be a byte ReadableStream or array buffer like.

get()

Similar to Deno.Kv.prototype.get(), in that it retrieves a blob value based on the provided key. If a previous blob value has been set with set(), it will be retrieved.

By default the value is resolved as a Uint8Array but if the option stream is set to true, then a byte ReadableStream is provided to read out the blob.

remove()

Similar to Deno.Kv.prototype.delete(), in that it deletes a blob key and value form the data store. If a blob value isn't stored for the given key, it resolves in a noop.

Keys

APIs for dealing with Deno KV keys.

equals()

Compares the quality of two Deno.KvKeys, returning true if they are equal and false if they are not. This is more specialized than other forms of deeply equal comparison.

startsWith()

Determines if the key starts with the prefix provided, returning true if it does, otherwise false.

keys()

Similar to Deno.Kv.prototype.list(), in that is takes a selector, but instead of returning an async iterator of matched values, it resolves with an array of matching keys.

unique()

Resolves with an array of unique sub keys/prefixes for the provided prefix.

This is useful when storing keys and values in a hierarchical/tree view, where you are retrieving a list and you want to know all the unique descendants of a key in order to be able to enumerate them.

uniqueCount()

Resolves with an array of values which contain the unique sub keys/prefixes for the provided prefix along with a count of how many keys there are.

This is useful when storing keys and values in a hierarchical/tree view, where you are retrieving a list and you want to know all the unique descendants of a key (and the count of keys that match that prefix) in order to be able to enumerate them or provide information about them.


Copyright 2023 - 2024 Kitson P. Kelly - All rights reserved.

MIT License.