Cacher (Pocket Dimension)

The PocketDimension allows Modules to store data for quick indexing and searching later on. Additionally, it may be useful to cache frequently used data so that requests can be made faster. The PocketDimension makes this easy by sorting the data per module, as well as allowing querying by specific information stored inside the payload of the Data Object for a quick turnaround for search results and other data lookups. Data added to the PocketDimension will be versioned using the standard version attached to the Data Object. When data is queried we will always return the latest Version (the greatest number). You can think of this as a giant void to throw data into and query from later.

The PocketDimension itself is built on an abstract interface, so we can switch the storage method on a whim. This can include things like Redis, The Graph, FlatFile DBs, etc.

One last thing to note, data stored within a dimension is never mutated (only replaced with newer versions), so it can be useful to see how long ago a piece of data was updated by checking the timestamp on that bit of data.

Pushing Data

Data can simply be dumped into a dimension, there is not much to worry about other than making sure you're storing the data in the correct dimension. Each module should have at least one of its own dimensions, it's very rare that two or more modules will share a dimension.

Updating data is super easy, any time the PocketDimension receives an object with an ID that already exists in the dimension, it will version the Data Object and load it in.

If we were to create the following data for example:

{
  id: "c59dda79-f2db-4247-b768-b0e093c838d3",
  version: 0,
  timestamp: 1643250387466,
  size: 387192,
  type: "messaging",
  payload: ...,
}

And store it in the dimension providing the messaging dimension type, and the Data Object...

PocketDimension::add_data(DataType::Module(Module::Messaging), DataObject)

It will simply be added to the dimension. If we then update the data we can simply call PocketDimension::add_data(Module::MESSAGING, ...) again. Retrieving the data via some query will return the expected object with version: 1, and version: 2, for each iteration of the data in the dimension.

Pulling Data

Data can then be retrieved by providing a query using PocketDimension::get_data.

PocketDimension::get_data(DataType::Module(Module::Messaging), None)

Executing

You MUST provide either a query method or run the TBD method to find data. This is to prevent unnecessarily querying large amounts of data.

Limiting

You can provide a u16 to the limit method to control how many results are returned from the scan.

TODO

Filtering

You can filter down the results by scanning for specific key values comparators. You'll provide the key the filter against, the value to expect, and the modifier. to filter by.

TODO

The modifier can be one of the following: TODO

Query Chaining

You can chain several filters together using the and chain.

Example:

TODO

Example Response

[
    {
      id: "c59dda79-f2db-4247-b768-b0e093c838d3",
      version: 1,
      timestamp: 1643250387466,
      size: 387192,
      type: "messaging",
      payload: ...,
    },
    {
      id: "c59dda79-f2db-4247-b768-b0e093c838d3",
      version: 2,
      timestamp: 1643250387466,
      size: 387192,
      type: "messaging",
      payload: ...,
    }
]