Skip to content

Latest commit

 

History

History
850 lines (669 loc) · 39.7 KB

handbook.md

File metadata and controls

850 lines (669 loc) · 39.7 KB

Developer's Handbook

Introduction

This document is intended for developers, who aim at using the Eclipse Dataspace Components in their software project at the library level. The Eclipse Dataspace Components projects is not a ready-to-use application, but rather a comprehensive collection of libraries and modules, that are published as Maven artifacts, and that developers can use and extend.

Therefore, if you are a solution architect looking for a high-level description on how to integrate EDC, or a developer wanting to contribute to the project itself, this guide is not for you. More suitable resources can be found here and here respectively.

However, if you are a developer who is familiar with Java (17+) and the Gradle build system in general, and concepts like extensibility, dependency injection and modularity, then this guide is for you.

Furthermore, this guide assumes a basic understanding of the following topics:

  • JSON-LD
  • HTTP/REST
  • relational databases (PostgreSQL) and transaction management
  • git and git workflows

At a minimum, the following tools are required:

  • Java 17+
  • Gradle 8+
  • a POSIX compliant shell (bash, zsh,...)
  • a text editor
  • CLI tools like curl and git

This guide will use CLI tools as common denominator, but in many cases graphical alternatives exist (e.g. Postman, Insomnia), and most developers will likely use IDEs like IntelliJ or VSCode. Please understand that while we are aware of them and recommend their use, we simply cannot cover and explain every possible combination of OS, tool and tool version.

Terminology

  • runtime: a Java process executing code written in the EDC programming model (e.g. a control plane)
  • distribution: a specific assortment of modules, compiled into a runnable form, e.g. a JAR file, a Docker image etc.
  • launcher: a runnable Java module, that pulls in other modules to form a distribution, sometimes used synonymously with distribution.
  • connector: a control plane runtime and 1...N data plane runtimes. Sometimes used interchangeably with distribution.

Building a distribution

In the EDC terminology, a "distribution" is an executable fat jar file, that consists of a compilation of specific EDC modules. It is sometimes sloppily referred to as "runtime", which is not quite correct, but it does loosely compare to the notion of a Linux distribution.

Typically, distributions consist of a control plane and one or more data planes. The former takes care of data manipulation, contract negotiation and transfer setup, and is geared toward reliability, whereas the job of the latter is to actually shovel bits from A to B.

The EDC project provides a comprehensive collection of samples, including one that demonstrates the assembly of a very simplistic distribution.

This simplest of distributions doesn't do much yet, it cannot even be configured properly, but it serves as starting point for more complex distributions and launchers.

Subsequent chapters of this document will assume a full understanding of the concepts and terms presented in the basic samples, so please make sure you go through them carefully.

Perform a simple data transfer

Now that you've built and run your first distribution, it is time to actually do something with it. So far, our connector wasn't able to do much, so we'll have to add a few capabilities to it. In EDC, we call these capabilities extensions, and they are dynamically loaded at startup to extend the functionality of a runtime. For example, adding you favorite authentication backend can be done through extensions.

At this point, we also need to introduce the notion of a "consumer" and a "provider" connector. As the saying goes, it takes two to tango, so in this section we will run two connectors - a consumer and a provider. To that end, we need to extend our distribution and add the following components:

  • an API so we can communicate with it
  • an authentication subsystem to secure the API
  • protocol plugins so it can communicate with other connectors

Please note that although both runtimes technically could be launched off of the same distribution the same way you could install the same Linux distro on multiple computers, for demonstration purposes this sample defines two identical distributions, but located in different directories and pre-loaded with different config values.

Please work through the complete sample, look at the enclosed source code and try to modify some values and see how they affect the data transfer.

Be advised that some of the concepts presented there are only suitable in a sample/demo environment and are not suitable for a production environment, such as seeding data through code, using hardcoded API tokens or storing files on the local file system.

Transfer some more data

Great, you've just gone through a pre-canned data transfer of assets, that were already created for you. That's not very interesting, is it? Also, copying a local file from the provider to the consumer is not a very realistic scenario.

Therefor in this chapter, we will use a slightly extended setup, where the data that the provider offers, is coming out of some "private backend" service, that is purportedly inaccessible to the consumer. The consumer then is given access to it through a proxy, and can obtain the data. We call this a "consumer-pull" transfer.

Please be advised that the sample will use some concepts and terms we haven't discussed yet, such as policies and data plane registration. Bear with us, we will explain them later. For now, please just make sure you follow the http-pull sample guide.

We encourage you to play around with this sample a bit, try to add a new asset and see if you can work out what else you need in order to be able to transfer it to the consumer.

Core concepts

When EDC was originally created, there were a few fundamental architectural principles around which we designed and implemented all dataspace components. These include:

  • asynchrony: all mutations of internal data structures happen in an asynchronous fashion. While the REST requests to trigger the mutations may still be synchronous, the actual state changes happen in an asynchronous and persistent way. For example starting a contract negotiation through the API will only return the negotiation's ID, and the control plane will cyclically advance the negotiation's state.
  • single-thread processing: the control plane is designed around a set of sequential state machines, that employ pessimistic locking to guard against race conditions and other problems.
  • idempotency: requests, that do not trigger a mutation, are idempotent. The same is true when provisioning external resources.
  • error-tolerance: the design goal of the control plane was to favor correctness and reliability over (low) latency. That means, even if a communication partner may not be reachable due to a transient error, it is designed to cope with that error and attempt to overcome it.

Developers who aim to use the Eclipse Dataspace Components in their projects are well-advised to follow these principles and build their applications around them. For example, when developing an app that monitors the progress of a contract negotiation or a transfer process, it is better to listen to events rather than polling a service or an API.

There are other, less technical principles of EDC such as simplicity and self-contained-ness. We are extremely careful when adding third-party libraries or technologies to maintain a simple, fast and un-opinionated platform.

The control plane

Simply put, the control plane is the brains of a connector, whereas the data plane would be the muscle. Its tasks include handling protocol and API requests, managing various internal asynchronous processes, validating policies, performing participant authentication and delegating the data transfer to a data plane. Its job is to handle (almost) all business logic. For that, it is designed to favor reliability over low latency. It does not directly transfer data from source to destination.

API objects in detail

We've seen in previous sections that the primary way to interact with a running connector instance is the Management API. It follows conventional REST semantics and allows manipulation of certain objects within the control plane. For example, it allows to specify, which data offerings are presented to the dataspace, and under what conditions.

For this chapter we assume basic knowledge of JSON-LD, and we also recommend reading the specifications for ODRL and DCAT.

The complete OpenAPI specification for the management API can be found here.

Assets

Assets are containers for metadata, they do not contain the actual bits and bytes. Say you want to offer a file to the dataspace, that is physically located in an S3 bucket, then the corresponding Asset would contain metadata about it, such as the content type, file size, etc. In addition, it could contain private properties, for when you want to store properties on the asset, which you do not want to expose to the dataspace.

The Asset also contains a DataAddress, which can be understood as a "pointer into the physical world". In the S3 example, that DataAddress might contain the bucket name and the region. Notice that the schema of the DataAddress will depend on where the data is physically located, for instance a HttpDataAddress has different properties.

A very simplistic Asset could look like this:

{
  "@context": {
    "edc": "https://w3id.org/edc/v0.0.1/ns/"
  },
  "@id": "79d9c360-476b-47e8-8925-0ffbeba5aec2",
  "properties": {
    "somePublicProp": "a very interesting value"
  },
  "privateProperties": {
    "secretKey": "this is very secret, never tell it to the dataspace!"
  },
  "dataAddress": {
    "type": "HttpData",
    "baseUrl": "http://localhost:8080/test"
  }
}

There are a couple of noteworthy things here. First, while there isn't a strict requirement for the @id to be a UUID, we highly recommend it.

By design, the Asset is extensible, so users can store any metadata they want in it. For example, the properties object could contain a simple string value, or it could be a complex object, following some custom schema. Be aware, that unless specified otherwise, all properties are put under the edc namespace by default. There are some "well-known" properties in the edc: namespace: id, description, version, name, contenttype.

Here is an example of how an Asset with a custom property following a custom namespace would look like:

{
  "@context": {
    "edc": "https://w3id.org/edc/v0.0.1/ns/",
    "ex": "http://w3id.org/example/v0.0.1/ns/"
  },
  "@id": "79d9c360-476b-47e8-8925-0ffbeba5aec2",
  "properties": {
    "somePublicProp": "a very interesting value",
    "ex:foo": {
      "name": "Darth Vader",
      "webpage": "https://death.star"
    }
  }
  //...
}

assuming the context object found at http://w3id.org/example/v0.0.1/ns/ contains a type definition of the foo object. As a reminder: the JSON-LD context could look like this:

{
  "@context": {
    "name": "http://schema.org/name",
    "webpage": {
      "@id": "http://schema.org/url",
      "@type": "@id"
    }
  }
}

Note that upon ingress through the management API, all JSON-LD objects get expanded, because the control plane only operates on expanded JSON-LD objects. Since the Asset is an extensible type, this is necessary to avoid potential name clashes of properties.

This is important to keep in mind, because when querying for assets using the POST /assets/request endpoint, the query that is submitted in the request body must target the expanded property. For example, a query targeting the somePublicProp field from the example above would look like this:

{
  "https://w3id.org/edc/v0.0.1/ns/filterExpression": [
    {
      "https://w3id.org/edc/v0.0.1/ns/operandLeft": [
        {
          "@value": "https://w3id.org/edc/v0.0.1/ns/somePublicProp"
        }
      ],
      "https://w3id.org/edc/v0.0.1/ns/operator": [
        {
          "@value": "="
        }
      ],
      "https://w3id.org/edc/v0.0.1/ns/operandRight": [
        {
          "@value": "a very interesting value"
        }
      ]
    }
  ]
}

More information about JSON-LD Contexts and type definitions can be found here.

Policies

Policies are the EDC way of expressing that certain conditions may, must or must not be satisfied in certain situations. In that sense, policies are used to express what requirements a subject (e.g. a communication partner) must satisfy in order to be able to perform an action. For example, one such requirement - or more specifically: a duty could be that a communication partner who wants to negotiate a contract, must have headquarters in the European Union.

Policies are ODRL serialized as JSON-LD. Thus, our previous example would look like this:

{
  "@context": {
    "edc": "https://w3id.org/edc/v0.0.1/ns/"
  },
  "@type": "PolicyDefinition",
  "policy": {
    "@context": "http://www.w3.org/ns/odrl.jsonld",
    "@type": "Set",
    "duty": [
      {
        "target": "http://example.com/asset:12345",
        "action": "use",
        "constraint": {
          "leftOperand": "headquarter_location",
          "operator": "eq",
          "rightOperand": "EU"
        }
      }
    ]
  }
}

The duty object expresses the semantics of the constraint. It is a specialization of rule, which expresses either a MUST (duty), MAY (permission) or MUST NOT (prohibition) relation. The action expresses the type of action for which the rule is intended. Acceptable values for action are defined here, but in EDC you'll exclusively encounter "use".

The constraint object expresses logical relationship of a key (leftOperand), the value (righOperand) and the operator. Multiple constraints can be linked logically, see advanced policy concepts. The leftOperand and rightOperand are completely arbitrary, only the operator is limited to the following possible values: eq, neq, gt, geq, lt, leq, in, hasPart, isA, isAllOf, isAnyOf, isNoneOf.

Please note that not all operators are always allowed, for example headquarter_location lt EU is nonsensical and should result in an evaluation error, whereas headquarter_location isAnOf [EU, US] would be valid. Whether an operator is valid is solely defined by the policy evaluation function, supplying an invalid operator should raise an exception.

NB: internally we always use the expanded JSON-LD form for all processing.

Policy scopes

In casual terms, a policy scope is the "context", in which a policy is to be evaluated. There are certain pre-defied points in the code, injecting/adding additional scopes is not possible. Currently, the following scopes exist:

  • contract.negotiation: evaluated upon initial contract offer. Ensures that the consumer fulfills the contract policy.
  • transfer.process: evaluated before starting a transfer process to ensure that the policy of the contract agreement is fulfilled. One example would be contract expiry.
  • catalog: evaluated when the catalog for a particular participant agent is generated. Decides whether the participant has the asset in their catalog.
  • request.contract.negotiation: evaluated on every request during contract negotiation between two control plane runtimes. Not relevant for end users.
  • request.transfer.process: evaluated on every request during transfer establishment between two control plane runtimes. Not relevant for end users.
  • request.catalog: evaluated upon an incoming catalog request. Not relevant for end users.
  • provision.manifest.verify: evaluated during the precondition check for resource provisioning. Only relevant in advanced use cases.

A policy scope is a string that is used for two purposes:

  1. binding a scope to a rule type: implement filtering based on the action or the leftOperand of a policy. This determines for every rule inside a policy whether it should be evaluated in the given scope.
  2. binding a policy evaluation function to a scope: if a policy is determined to be "in scope" by the previous step, the policy engine invokes the evaluation function that was bound to the scope to evaluate if the policy is fulfilled.

Another way to understand policy scopes is for them to be the link between a rule and an evaluation function, and the points in the code where it needs to be evaluated.

Policy evaluation functions

If policies are a formalized declaration of requirements, policy evaluation functions are the means to evaluate those requirements. They are pieces of Java code executed at runtime. A policy on its own only expresses the requirement, but in order to enforce it, we need to run policy evaluation functions.

Upon evaluation, they receive the operator, the rightOperand (or rightValue), the rule, and the PolicyContext. A simple evaluation function that asserts the headquarters policy mentioned in the example above could look similar to this:

public class HeadquarterFunction implements AtomicConstraintFunction<Duty> {
    public boolean evaluate(Operator operator, Object rightValue, Permission rule, PolicyContext context) {
        if (!(rightValue instanceof String)) {
            context.reportProblem("Right-value expected to be String but was " + rightValue.getClass());
            return false;
        }
        if (operator != Operator.EQ) {
            context.reportProblem("Invalid operator, only EQ is allowed!");
            return false;
        }

        var participant = context.getContextData(ParticipantAgent.class);
        var participantLocation = extractLocationClaim(participant); // EU, US, etc.
        return participantLocation != null && rightValue.equalsIgnoreCase(participantLocation);
    }
}

This particular evaluation function only accepts eq as operator, and only accepts scalars as rightValue.

The ParticipantAgent is a representation of the communication counterparty that contains a set of verified claims. In the example, extractLocationClaim() would look for a claim that contains the location of the agent and return it as string.

Other policies may require other context data than the participant's location, for example an exact timestamp, or may even need a lookup in some third party system such as a customer database.

The same policy can be evaluated by different evaluation functions, if they are meaningful in different contexts (scopes).

Example: binding an evaluation function

As we've learned, for a policy to be evaluated at certain points, we need to create a policy (duh!), bind the policy to a scope, create a policy evaluation function, and we need to bind the function to the same scope. The standard way of registering and binding policies is done in an extension. For example, here we configure our HeadquarterFunction so that it evaluates our headquarter_location function whenever someone tries to negotiate a contract:

public class HeadquarterPolicyExtension implements ServiceExtension {

    @Inject
    private RuleBindingRegistry ruleBindingRegistry;

    @Inject
    private PolicyEngine policyEngine;

    private static final String HEADQUARTER_POLICY_KEY = "headquarter_location";

    @Override
    public void initialize() {
        // bind the policy to the scope
        ruleBindingRegistry.bind(HEADQUARTER_POLICY_KEY, NEGOTIATION_SCOPE);
        // create the function object
        var function = new HeadquarterFunction();
        // bind the function to the scope
        policyEngine.registerFunction(NEGOTIATION_SCOPE, Duty.class, HEADQUARTER_POLICY_KEY, function);
    }
}

Let's accept for now, that @Inject is how EDC achieves dependency injection. This example assumes, a policy object exists in the system, that has a leftOperand = headquarter_function. For details on how to create policies, please check out the OpenAPI documentation.

Advanced policy concepts

Policies are in essence containers for rules (duties, permissions and prohibitions), and rules in turn are compose of one or multiple constraints each. The following example extends the previous one by adding another duty rule: the claimant must either have headquarters in the EU, or be in business for >= 10 years. Please also note that the following example uses the compacted form because all EDC Management APIs return the compacted form.

{
  "@id": "test-policy",
  "@type": "edc:PolicyDefinition",
  "edc:createdAt": 1692858228518,
  "edc:policy": {
    "@id": "8c2ff88a-74bf-41dd-9b35-9587a3b95adf",
    "@type": "odrl:Set",
    "odrl:permission": [],
    "odrl:prohibition": [],
    "odrl:obligation": {
      "odrl:action": {
        "odrl:type": "USE"
      },
      "odrl:constraint": {
        "odrl:or": [
          {
            "odrl:leftOperand": "headquarter_location",
            "odrl:operator": {
              "@id": "odrl:eq"
            },
            "odrl:rightOperand": "EU"
          },
          {
            "odrl:leftOperand": "years_in_business",
            "odrl:operator": {
              "@id": "odrl:gteq"
            },
            "odrl:rightOperand": "10"
          }
        ]
      }
    }
  },
  "@context": {
    "dct": "https://purl.org/dc/terms/",
    "edc": "https://w3id.org/edc/v0.0.1/ns/",
    "dcat": "https://www.w3.org/ns/dcat/",
    "odrl": "http://www.w3.org/ns/odrl/2/",
    "dspace": "https://w3id.org/dspace/v0.8/"
  }
}

Similarly, the odrl:or could have been replaced with an odrl:and or odrl:xone (exclusive-or).

Contract definitions

Contract definitions are how assets and policies are linked together. It is our way of expressing which policies are in effect for an asset. So when we want to offer an asset (or several assets) in the dataspace, we use a contract definition to express under what conditions we do that. Those conditions are comprised of a contract policy and an access policy, see below for details. Those policies are referenced by ID, that means they must already exist in the policy store before creating the contract definition.

It is important to note that contract definitions are a internal objects, i.e. they never leave the realm of the provider, and they are never sent to the consumer.

  • access policy: determines whether a particular consumer is offered an asset or not. For example, we may want to restrict certain assets such that only consumers within a particular geography can see them. Consumers outside that geography wouldn't even have them in their catalog.
  • contract policy: determines the conditions for initiating a contract negotiation for a particular asset. Note that does not automatically guarantee the successful creation of a contract, it merely expresses the eligibility to start the negotiation.

Contract definitions also contain an assetsSelector, which - in broad terms - is a query expression that defines all the assets that are included in the definition, not unlike an SQL SELECT statement. With that it is possible to configure the same set of conditions (= access policy and contract policy) for a multitude of assets.

Please note that creating an assetSelector may require knowledge about the shape of an Asset and can get complex fairly quickly, so be sure to read the chapter about querying.

Here is an example of a contract definition, that defines an access policy and a contract policy for all assets, that have the foo=bar property.

{
  "@context": {
    "edc": "https://w3id.org/edc/v0.0.1/ns/"
  },
  "@type": "https://w3id.org/edc/v0.0.1/ns/ContractDefinition",
  "@id": "test-id",
  "edc:accessPolicyId": "access-policy-1234",
  "edc:contractPolicyId": "contract-policy-5678",
  "edc:assetsSelector": [
    {
      "@type": "https://w3id.org/edc/v0.0.1/ns/Criterion",
      "edc:operandLeft": "foo",
      "edc:operator": "=",
      "edc:operandRight": "bar"
    }
  ]
}

The sample expresses that every asset, that contains a property "foo" : "bar" be made available under the access policy access-policy-1234 and contract policy contract-policy-5678.

Contract negotiations

If a connector fulfills the contract policy, it may initiate the negotiation of a contract for a particular asset. During that negotiation, both parties can send offers and counter-offers that can contain altered terms (= policy) as any human would in a negotiation, and the counter-party may accept or reject them.

Contract negotiations have a few key aspects:

  • they target one asset
  • they take place between a provider and a consumer connector
  • they cannot be changed by the user directly
  • users can only be decline, terminate or cancel them

As a side note it is also important to note that contract offers are ephemeral objects as they are generated on-the-fly for a particular participant, and they are never persisted in a database and thus cannot be queried through any API.

Contract negotiations are asynchronous in nature. That means after initiating them, they become (potentially long-running) stateful processes that are advanced by an internal state machine. The current state of the negotiation can be queried and altered through the management API.

In the current iteration all contract offers are automatically accepted by both parties if the contract policy is satisfied. This may change in the future, when manual interaction hooks are implemented.

Please check out the OpenAPI documentation for an exemplary request to initiate a negotiation. The callbackAddresses object is optional and can be used to get notified about state changes of the negotiation. Read more on callbacks in the section about events and callbacks.

Contract agreements

Once a contract negotiation is successfully concluded (i.e. reaches the FINALIZED state), it "turns into" a contract agreement. Note that it is always the provider connector that gives the final approval. Contract agreements are immutable objects that contain the final, agreed-on policy, the ID of the asset that the contract was negotiated for, and the exact signing date.

Note that in future iterations contracts will be cryptographically signed to further support the need for immutability and non-repudiation.

Catalog

The catalog contains the "data offerings" of a connector and one or multiple service endpoints to initiate a negotiation for those offerings.

Every data offering is represented by a Dataset object which contains a policy and one or multiple Distribution objects. A Distribution should be understood as a variant or representation of the Dataset. For instance, if a file is accessible through multiple transmission channels from a provider (HTTP and FTP), then each of those channels would be represented as a Distribution. Another example would be image assets that are available in different file formats (PNG, TIFF, JPEG).

A DataService object specifies the endpoint where contract negotiations and transfers are accepted by the provider. In practice, this will be the DSP endpoint of the connector.

Transfer processes

A TransferProcess is a record of the data sharing procedure between a consumer and a provider. As they traverse through the system, they transition through several states (TransferProcessStates, source).

Once a contract is negotiated) and an agreement is reached, the consumer connector may send a transfer initiate request (more on that here) to start the transfer. In the course of doing that, both parties may provision additional resources, for example deploying a temporary object store, where the provider should put the data. Similarly, the provider may need to take some preparatory steps, e.g. anonymizing the data before sending it out.

This is sometimes referred to as the provisioning phase. If no additional provisioning is needed, the transfer process simply transitions through the state with a NOOP.

Once that is done, the transfer begins in earnest. Data is transmitted according to the dataDestination, that was passed in the initiate-request.

Once the transmission has completed, the transfer process will transition to the COMPLETED state, or - if an error occurred - to the TERMINATED state.

The Management API provides several endpoints to manipulate data transfers.

About Data Destinations

A data destination is a description of where the consumer expects to find the data after the transfer completes. In a " provider-push" scenario this could be an object storage container, a directory on a file system, etc. In a "consumer-pull" scenario this would be a placeholder, that does not contain any information about the destination, as the provider "decides" on which endpoint he makes the data available.

A data address is a schemaless object, and the provider and the consumer need to have a common understanding of the required fields. For example, if the provider is supposed to put the data into a file share, the DataAddress object representing the data destination will likely contain the host URL, a path and possibly a file name. So both connectors need to be "aware" of that.

The actual data transfer is handled by a data plane through extensions (called "sources" and " sinks"). Thus, the way to establish that "understanding" is to make sure that both parties have matching sources and sinks. That means, if a consumer asks to put the data in a file share, the provider must have the appropriate data plane extensions to be able to perform that transfer.

If the provider connector does not have the appropriate extensions loaded at runtime, the transfer process will fail.

Using event callbacks

In order to get timely updates about status changes of a transfer process, we could simply poll the management API by firing a GET /{tp-id}/state request every X amount of time. That will not only put unnecessary load on the connector, you may also run into rate-limiting situations, if the connector is behind a load balancer of some sort. Thus, we recommend using event callbacks.

As shown in the OpenAPI documentation they must be specified when requesting to initiate the transfer:

{
  // ...
  "callbackAddresses": [
    {
      "transactional": false,
      "uri": "http://callback/url",
      "events": [
        "contract.negotiation",
        "transfer.process"
      ],
      "authKey": "auth-key",
      "authCodeId": "auth-code-id"
    }
  ]
  //...
}

Currently, we support the following events:

  • transfer.process.deprovisioned
  • transfer.process.completed
  • transfer.process.deprovisioningRequested
  • transfer.process.initiated
  • transfer.process.provisioned
  • transfer.process.provisioning
  • transfer.process.requested
  • transfer.process.started
  • transfer.process.terminated

The connector's event dispatcher will send invoke the webhook specified in the uri field passing the event payload as JSON object.

Expressing queries with a Criterion

A Criterion is a normalized way of expressing a logical requirement between two operands and an operator. For example, we can use this to formulate a query, which selects all objects, where a particular property must have a particular value. This is similar to an SQL SELECT statement, or a policy's constraint property.

Like the constraint, a Criterion has an operandLeft, an operator and an operandRight, which determine the semantics of the query.

However, there are several very important aspects to understand:

Canonical form

The operandLeft targets the canonical form or the object. For non-extensible properties, the canonical form is equal to the object representation in Java code. For example, the canonical form of a transfer process's state field is state, because that is what the property on the Java class TransferProcess.java is called.

Thus, the resulting Criterion would look like this:

{
  "@type": "https://w3id.org/edc/v0.0.1/ns/Criterion",
  "operandLeft": "state",
  "operator": "=",
  "operandRight": 800
}

Keen readers will notice, that the TransferProcess#state property is an enum named TransferProcessStates, but the operandRight of the sample is an integer - what gives?

In Java, enums are just "named integers", so we store them as integers in the database. Consequently, the transformation layer that converts from Criterion to an actual SQL query expects the state field to be an INT. For reference, all TransferProcess state values are listed here.

More detailed information about the canonical format and SQL queries can be found here.

Supported operators

The semantic cardinality of operators is virtually limitless, and it would be impossible to maintain or test. Out-of-the box, EDC supports a limited set of operators. Please find the related documentation here.

It should be noted, that - like everything else in EDC - this can be extended through custom SQL dialects, or even custom data backends. In other words, the set of supported operators depend on the backend system interpreting the query. In the EDC PostgreSQL implementation, these operators are =, like and in.

Namespaced properties

Some objects, such as Asset don't have a fixed a schema, rather they are "extensible objects". This is comparable to a java.util.Map. That means, you can store on them whatever you like, even complex objects. That not only brings a lot of flexibility, it also introduces the potential danger of name clashes: for instance, a property content-type could potentially have different meanings in different contexts. To counter this, we always persist the keys of such dynamic objects with their fully qualified namespace. So if you are using custom properties with a custom JSON-LD namespace, you will have to use that namespace in the Criterion!

A word on JSON-LD contexts

Control plane state machines

--> gives an overview of the contract negotiation and transfer process state machines

Provisioning

The extension model

--> how to write your own first extension, explains plug-points points, metamodel annotations, SPIs, etc

  • Authentication/Authorization of the Management API
  • policies and evaluation functions
  • persistence implementations
  • provisioners
  • transaction management (?)

EDC dependency injection

--> explains how a custom extension can inject and provide objects

Policy scopes and evaluation

see the chapter about policy scopes

The data plane

Data plane selectors

--> explains how dataplane selection works and how devs can register a dataplane with the selector

Writing a DataSink and DataSource extension

--> link to samples?

The Control API

--> explains the usage of the control api, in case people want to write their own data plane

Advanced concepts

Events and callbacks

--> shows how to hook into the event observers, and how to use http callbacks

The EDC JUnit framework

--> explains how devs can leverage the In-Mem runtime feature, the dependency-injection extension, all the different tags, the postgres extension, etc.

Automatic documentation

--> explains what automatic documentation is, how to leverage it, how to download the EDC autodoc manifests and how to merge them with the downstream docs

Customize the build

--> shows how to configure and influence the build

Further references and specifications

  • DSP specification
  • OpenAPI documentation