12 releases

Uses new Rust 2021

0.0.13 Jul 10, 2022
0.0.12 Jul 1, 2022
0.0.11 Jun 29, 2022
0.0.7 May 25, 2022
0.0.4 Apr 29, 2022

#2243 in Magic Beans

Download history 169/week @ 2022-06-10 161/week @ 2022-06-17 278/week @ 2022-06-24 260/week @ 2022-07-01 142/week @ 2022-07-08 234/week @ 2022-07-15 171/week @ 2022-07-22 105/week @ 2022-07-29 139/week @ 2022-08-05 143/week @ 2022-08-12 88/week @ 2022-08-19 58/week @ 2022-08-26 55/week @ 2022-09-02 125/week @ 2022-09-09 45/week @ 2022-09-16 23/week @ 2022-09-23

254 downloads per month

Apache-2.0

365KB
7K SLoC

Holochain Deterministic Integrity


lib.rs:

Holochain Deterministic Integrity (HDI) is Holochain's data model and integrity toolset for writing zomes.

The logic of a Holochain DNA can be divided into two parts: integrity and coordination. Integrity is the part of the hApp that defines the data types and validates data manipulations. Coordination encompasses the domain logic and implements the functions that manipulate data.

Examples

An example of an integrity zome with data definition and data validation can be found in the wasm workspace of the Holochain repository: https://github.com/holochain/holochain/blob/develop/crates/test_utils/wasm/wasm_workspace/integrity_zome/src/lib.rs.

Data definition

The DNA's data model is defined in integrity zomes. They comprise all data type definitions as well as relationships between those types. Integrity zomes are purely definitions and do not contain functions to manipulate the data. Therefore a hApp's data model is encapsulated and completely independent of the domain logic, which is encoded in coordinator zomes.

The MVC (model, view, controller) design pattern can be used as an analogy. The application’s integrity zomes comprise its model layer — everything that defines the shape of the data. In practice, this means three things:

  • entry type definitions
  • link type definitions
  • a validation callback that constrains the kinds of data that can validly be called entries and links of those types (see also validate).

The coordination zomes comprise the application's controller layer — the code that actually writes and retrieves data, handles countersigning sessions and sends and receives messages between peers or between a cell and its UI. In other words, all the zome functions, init functions, remote signal receivers, and scheduler callbacks will all live in coordinator zomes.

Advantages of this approach are:

  • The DNA hash is constant as long as the integrity zomes remain the same. The peer network of a DNA is tied to its hash. Changes to the DNA hash result in a new peer network. Changes to the domain logic enclosed in coordinator zomes, however, do not affect the DNA hash. Hence the DNAs and therefore hApps can be modified without creating a new peer network on every deployment.
  • Integrity zomes can be shared among DNAs. Any coordinator zome can import an integrity zome's data types and implement functions for data manipulation. This composability of integrity and coordinator zomes allows for a multitude of permutations with shared integrity zomes, i. e. a shared data model.

Data validation

The second fundamental part of integrity zomes is data validation. For every operation that can be performed on the data, a validation rule can be specified. Both data types and data values can be validated. All of these validation rules are written in a central callback which is called by the Holochain engine for each operation.

Dependencies

~3–5.5MB
~108K SLoC