julienetie / db64 Goto Github PK

Generate TypeScript definitions and a TS config.

Create an examples directory with an example of db64

Due to the way versioning works it is not possible to modify the schema of a database without creating a new version.

But this can be worked around by:

• Copying existing stores from the DB to be modified
• Deleting the database
• Creating a new DB with new and existing store names
• Populating the existing data back to the stores they came from

This API will be in the form of:

• db.addStore()
• db.deleteStore()

Taking one or many store names. And will do nothing if the store exists or does not exist respectively.

This feature will simplify the lifecycle management of Indexeddb.

Create a headless test setup, take influence from the index.html test file.

Common JS not yet supported

Needs a UMD build

Sourcemaps are currently commented out in create-distribution.js because they're not yet referenced in the minified files. This needs fixing.

Delete is necessary for when existing databases in a user's browser needs to be destroyed and recreated.

The only way to add a new database is to increase the version or delete and re-create all the effected databases.

The current workaround to this is:

const req = indexedDB.deleteDatabase('databaseName')
req.onsuccess = () => {
    console.log("Deleted database successfully")
}
req.onerror =  () => {
    console.log("Couldn't delete database")
}
req.onblocked = () => {
    console.log("Couldn't delete database due to the operation being blocked")
}

Proposed solutions:

db64.delete('databaseName')

Create a tag and release from the current version.

There are several opportunities to improve refactoring.
Comments can help to encourage users and contributors to understand how it works since it's a fairly simple codebase.

First example, should be?

await db64.create('Games', ['Super Nintendo', 'Gameboy'])

Add minifed ES and CJS distribution files

Show installation examples for es, cjs and script modules in readme.

Could this work with Zustand https://github.com/pmndrs/zustand/blob/main/docs/integrations/persisting-store-data.md#how-can-i-use-a-custom-storage-engine

I see that this library needs initialization so it will probably need some middleware function in order to create the store and then expose the get, set functions in order for Zustand to use them right?

Although has is demonstrated in the readme it needs to be part of the detailed API.