apicurio / apicurio-registry Goto Github PK
View Code? Open in Web Editor NEWAn API/Schema registry - stores APIs and Schemas.
Home Page: https://www.apicur.io/registry/
License: Apache License 2.0
An API/Schema registry - stores APIs and Schemas.
Home Page: https://www.apicur.io/registry/
License: Apache License 2.0
Use whatever standard metrics feature Quarkus has - presumably MP metrics.
Implement the Validation Rule for AsyncAPI content.
Has anyone managed to get their own setup working against corporate security with SAML? We're trying to have SAML for all applications and have an in house version of apicurio for our development teams to use.
I attempted to wire the code directly to SAML and bypass keycloak but was unable to figure out how JBOSS security is wired up in this area as its inside the server and no matter how many google searches I took no luck finding anyone else doing this either.
Seeing the code is setup for Keycloak now wondering if keycloak can use saml as an identify provider but maybe someone else has done this before.
Use Quarkus's configuration system to create a set of profiles for various kind of deployment scenarios, including selection of the persistence layer implementation and logging.
Profiles could be e.g.: dev (h2 DB), single (PostgreSQL) cluster (Infinispan), kafka, ...
Resources:
We're going to be storing artifacts of the following formats in the registry:
Most of these are in JSON format (or YAML which can easily be converted). But at least Protobuff is not. So there is an open question about how to handle the different content types when pushing content into the registry via the API and then returning it later.
The first step is to identify the actual content types for each format. And then additionally how to handle that in the API itself. In order to make it as simple as possible, we could try to figure out the type based on the following (in order):
X-Registry-Type
Content-Type
might be application/json+openapi
When downloading or publishing an project api specification, we're seeing previously quoted values being outputted without quotes.
This is in turn leading to a bug with the fields being incorrectly rendered in tooling such as redoc — which typically converts the YAML into a JSON object
A repeatable example can be seen over at https://onlineyamltools.com/convert-yaml-to-json using the following example payloads...
Quoted Date String
Priority Order request:
value:
order:
courier_type: priority
delivery_address: 1 Union Street
delivery_postcode: "2009"
delivery_state: NSW
delivery_suburb: Pyrmont
authority_to_leave: "Yes"
delivery_date: "2016-07-26"
delivery_window: 16:00-19:00
parcel_attributes:
- qty: 1
weight: 2.1
— delivery_date results in displaying as "2016-07-26"
Unquoted Date String
Priority Order request:
value:
order:
courier_type: priority
delivery_address: 1 Union Street
delivery_postcode: "2009"
delivery_state: NSW
delivery_suburb: Pyrmont
authority_to_leave: "Yes"
delivery_date: 2016-07-26
delivery_window: 16:00-19:00
parcel_attributes:
- qty: 1
weight: 2.1
— delivery_date results in displaying as "2016-07-26T00:00:00.000Z"
If the user does not provide the artifact type via a request header, then the registry needs to probe the content to try and determine the type. This amounts to attempting to parse the content as JSON, then as protobuf. If it's JSON, then further analysis is needed to determine if it's an OpenAPI, AsyncAPI, JSON Schema, or Avro artifact.
An in-memory implementation of the storage interface.
Consult with Gunnar Morling on requirements and implementation details.
Implement the Validation Rule for Protobuff content. Rules will need to be defined for this.
A relational database (sql) implementation of the storage interface. Can use hibernate or straight jdbc (IMO).
Some resources:
Implement the Validation Rule for Avro content. Rules will need to be defined for this.
Go over the POM files, profiles, tests etc. in case there is some maintenance work needed. At the very least, we need to make sure that mvn clean install
will build, test, and package all of the supported storage variants. Right now the variants are all separated out into profiles, because the tests for "streams", "kafka", and "infinispan" will all fail because they all depend on external systems to function. We need to work on standing up those systems automatically so that tests can be run successfully.
Create an ansible-based operator for installing the registry on OpenShift.
Implement the Validation Rule for JSON Schema content. Rules will need to be defined for this.
Create a REST API for the registry that is more generic than the confluent compatible API - it should handle multiple content types and support rules instead of just compatibility.
Should help with migration from one system to the other.
Use the recommended Quarkus approach to enabling app healthchecking. I presume this is MP health.
Quarkus 0.26.1 has already changed enough that this project fails to compile against it. :(
Following bleeding edge projects is hard.
Implement the compatibility rule for OpenAPI schemas. We'll need to really analyze the avro version to implement similar logic for OpenAPI.
A kafka implementation of the storage interface.
If Avro converter is used in Kafka COnnect then Jersey REST client must/should be used. The problem is that when an atifact like /artifacts/dbserver1-key
is created the server anwers with
{"createdOn":1579862056239,"modifiedOn":1579864479545,"id":"dbserver1-key","version":3,"type":"AVRO","globalId":3}
Date fields are serialized as epoch milliseconds.
When the client tries to parse them it gets
connect_1 | SEVERE: Unable to deserialize property 'createdOn' because of: Error parsing class java.util.Date from value: 1579862056239. Check your @JsonbDateFormat has all time units for class java.util.Date type, or consider using org.eclipse.yasson.YassonProperties#ZERO_TIME_PARSE_DEFAULTING.
connect_1 | 2020-01-24 11:14:39,550 INFO || WorkerSourceTask{id=inventory-connector-0} Committing offsets [org.apache.kafka.connect.runtime.WorkerSourceTask]
connect_1 | 2020-01-24 11:14:39,550 INFO || WorkerSourceTask{id=inventory-connector-0} flushing 0 outstanding messages for offset commit [org.apache.kafka.connect.runtime.WorkerSourceTask]
connect_1 | 2020-01-24 11:14:39,550 ERROR || WorkerSourceTask{id=inventory-connector-0} Task threw an uncaught and unrecoverable exception [org.apache.kafka.connect.runtime.WorkerTask]
connect_1 | org.apache.kafka.connect.errors.ConnectException: Tolerance exceeded in error handler
connect_1 | at org.apache.kafka.connect.runtime.errors.RetryWithToleranceOperator.execAndHandleError(RetryWithToleranceOperator.java:178)
connect_1 | at org.apache.kafka.connect.runtime.errors.RetryWithToleranceOperator.execute(RetryWithToleranceOperator.java:104)
connect_1 | at org.apache.kafka.connect.runtime.WorkerSourceTask.convertTransformedRecord(WorkerSourceTask.java:287)
connect_1 | at org.apache.kafka.connect.runtime.WorkerSourceTask.sendRecords(WorkerSourceTask.java:316)
connect_1 | at org.apache.kafka.connect.runtime.WorkerSourceTask.execute(WorkerSourceTask.java:240)
connect_1 | at org.apache.kafka.connect.runtime.WorkerTask.doRun(WorkerTask.java:177)
connect_1 | at org.apache.kafka.connect.runtime.WorkerTask.run(WorkerTask.java:227)
connect_1 | at java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:515)
connect_1 | at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:264)
connect_1 | at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128)
connect_1 | at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628)
connect_1 | at java.base/java.lang.Thread.run(Thread.java:834)
connect_1 | Caused by: javax.ws.rs.ProcessingException: Error deserializing object from entity stream.
connect_1 | at org.glassfish.jersey.jsonb.internal.JsonBindingProvider.readFrom(JsonBindingProvider.java:77)
connect_1 | at org.glassfish.jersey.message.internal.ReaderInterceptorExecutor$TerminalReaderInterceptor.invokeReadFrom(ReaderInterceptorExecutor.java:233)
connect_1 | at org.glassfish.jersey.message.internal.ReaderInterceptorExecutor$TerminalReaderInterceptor.aroundReadFrom(ReaderInterceptorExecutor.java:212)
connect_1 | at org.glassfish.jersey.message.internal.ReaderInterceptorExecutor.proceed(ReaderInterceptorExecutor.java:132)
connect_1 | at org.glassfish.jersey.message.internal.MessageBodyFactory.readFrom(MessageBodyFactory.java:1071)
connect_1 | at org.glassfish.jersey.message.internal.InboundMessageContext.readEntity(InboundMessageContext.java:850)
connect_1 | at org.glassfish.jersey.message.internal.InboundMessageContext.readEntity(InboundMessageContext.java:810)
connect_1 | at org.glassfish.jersey.client.ClientResponse.readEntity(ClientResponse.java:339)
connect_1 | at org.glassfish.jersey.client.InboundJaxrsResponse$2.call(InboundJaxrsResponse.java:102)
connect_1 | at org.glassfish.jersey.internal.Errors.process(Errors.java:292)
connect_1 | at org.glassfish.jersey.internal.Errors.process(Errors.java:274)
connect_1 | at org.glassfish.jersey.internal.Errors.process(Errors.java:205)
connect_1 | at org.glassfish.jersey.process.internal.RequestScope.runInScope(RequestScope.java:365)
connect_1 | at org.glassfish.jersey.client.InboundJaxrsResponse.runInScopeIfPossible(InboundJaxrsResponse.java:240)
connect_1 | at org.glassfish.jersey.client.InboundJaxrsResponse.readEntity(InboundJaxrsResponse.java:99)
connect_1 | at org.glassfish.jersey.microprofile.restclient.MethodModel.lambda$asynchronousCall$5(MethodModel.java:266)
connect_1 | at java.base/java.util.concurrent.CompletableFuture$UniAccept.tryFire(CompletableFuture.java:714)
connect_1 | at java.base/java.util.concurrent.CompletableFuture.postComplete(CompletableFuture.java:506)
connect_1 | at java.base/java.util.concurrent.CompletableFuture.complete(CompletableFuture.java:2073)
connect_1 | at org.glassfish.jersey.client.JerseyInvocation$1.completed(JerseyInvocation.java:789)
connect_1 | at org.glassfish.jersey.client.ClientRuntime.processResponse(ClientRuntime.java:203)
connect_1 | at org.glassfish.jersey.client.ClientRuntime.access$200(ClientRuntime.java:61)
connect_1 | at org.glassfish.jersey.client.ClientRuntime$2.lambda$response$0(ClientRuntime.java:154)
connect_1 | at org.glassfish.jersey.internal.Errors$1.call(Errors.java:248)
connect_1 | at org.glassfish.jersey.internal.Errors$1.call(Errors.java:244)
connect_1 | at org.glassfish.jersey.internal.Errors.process(Errors.java:292)
connect_1 | at org.glassfish.jersey.internal.Errors.process(Errors.java:274)
connect_1 | at org.glassfish.jersey.internal.Errors.process(Errors.java:244)
connect_1 | at org.glassfish.jersey.process.internal.RequestScope.runInScope(RequestScope.java:288)
connect_1 | at org.glassfish.jersey.client.ClientRuntime$2.response(ClientRuntime.java:154)
connect_1 | at org.glassfish.jersey.client.internal.HttpUrlConnector.apply(HttpUrlConnector.java:268)
connect_1 | at org.glassfish.jersey.client.ClientRuntime.lambda$null$3(ClientRuntime.java:163)
connect_1 | at org.glassfish.jersey.internal.Errors$1.call(Errors.java:248)
connect_1 | at org.glassfish.jersey.internal.Errors$1.call(Errors.java:244)
connect_1 | at org.glassfish.jersey.internal.Errors.process(Errors.java:292)
connect_1 | at org.glassfish.jersey.internal.Errors.process(Errors.java:274)
connect_1 | at org.glassfish.jersey.internal.Errors.process(Errors.java:244)
connect_1 | at org.glassfish.jersey.process.internal.RequestScope.runInScope(RequestScope.java:288)
connect_1 | at org.glassfish.jersey.client.ClientRuntime.lambda$createRunnableForAsyncProcessing$4(ClientRuntime.java:139)
connect_1 | at org.glassfish.jersey.microprofile.restclient.ExecutorServiceWrapper.lambda$wrap$1(ExecutorServiceWrapper.java:124)
connect_1 | ... 5 more
connect_1 | Caused by: javax.json.bind.JsonbException: Unable to deserialize property 'createdOn' because of: Error parsing class java.util.Date from value: 1579862056239. Check your @JsonbDateFormat has all time units for class java.util.Date type, or consider using org.eclipse.yasson.YassonProperties#ZERO_TIME_PARSE_DEFAULTING.
connect_1 | at org.eclipse.yasson.internal.serializer.AbstractContainerDeserializer.deserializeInternal(AbstractContainerDeserializer.java:90)
connect_1 | at org.eclipse.yasson.internal.serializer.AbstractContainerDeserializer.deserialize(AbstractContainerDeserializer.java:60)
connect_1 | at org.eclipse.yasson.internal.Unmarshaller.deserializeItem(Unmarshaller.java:68)
connect_1 | at org.eclipse.yasson.internal.Unmarshaller.deserialize(Unmarshaller.java:54)
connect_1 | at org.eclipse.yasson.internal.JsonBinding.deserialize(JsonBinding.java:53)
connect_1 | at org.eclipse.yasson.internal.JsonBinding.fromJson(JsonBinding.java:93)
connect_1 | at org.glassfish.jersey.jsonb.internal.JsonBindingProvider.readFrom(JsonBindingProvider.java:75)
connect_1 | ... 44 more
connect_1 | Caused by: javax.json.bind.JsonbException: Error parsing class java.util.Date from value: 1579862056239. Check your @JsonbDateFormat has all time units for class java.util.Date type, or consider using org.eclipse.yasson.YassonProperties#ZERO_TIME_PARSE_DEFAULTING.
connect_1 | at org.eclipse.yasson.internal.serializer.AbstractDateTimeDeserializer.deserialize(AbstractDateTimeDeserializer.java:71)
connect_1 | at org.eclipse.yasson.internal.serializer.AbstractValueTypeDeserializer.deserialize(AbstractValueTypeDeserializer.java:64)
connect_1 | at org.eclipse.yasson.internal.serializer.ObjectDeserializer.deserializeNext(ObjectDeserializer.java:174)
connect_1 | at org.eclipse.yasson.internal.serializer.AbstractContainerDeserializer.deserializeInternal(AbstractContainerDeserializer.java:84)
connect_1 | ... 50 more
connect_1 | Caused by: java.time.format.DateTimeParseException: Text '1579862056239' could not be parsed at index 0
connect_1 | at java.base/java.time.format.DateTimeFormatter.parseResolved0(DateTimeFormatter.java:2046)
connect_1 | at java.base/java.time.format.DateTimeFormatter.parse(DateTimeFormatter.java:1948)
connect_1 | at java.base/java.time.ZonedDateTime.parse(ZonedDateTime.java:598)
connect_1 | at org.eclipse.yasson.internal.serializer.DateTypeDeserializer.parseWithOrWithoutZone(DateTypeDeserializer.java:83)
connect_1 | at org.eclipse.yasson.internal.serializer.DateTypeDeserializer.parseDefault(DateTypeDeserializer.java:54)
connect_1 | at org.eclipse.yasson.internal.serializer.DateTypeDeserializer.parseDefault(DateTypeDeserializer.java:34)
connect_1 | at org.eclipse.yasson.internal.serializer.AbstractDateTimeDeserializer.deserialize(AbstractDateTimeDeserializer.java:69)
connect_1 | ... 53 more
The clients expects it to be a string formatted value.
We don't really have any useful logging in registry or in the serdes right now. We should go through and add appropriate logging (info/debug/tracing).
Implement the compatibility rule for protobuff schemas. We'll need to really analyze the avro version to implement similar logic for protobuff.
Create an implementation of the "Artifacts" interface.
Implement the compatibility rule for AsyncAPI schemas. We'll need to really analyze the avro version to implement similar logic for AsyncAPI.
Kafka Connect uses a schema format that is not JSON Schema but is more similar to Avro (but is not Avro). We should add artifact type support for this type.
We need a simple, common implementation that all validation rules will need to implement. This might be as simple as passing in the content and returning either a pass/fail or perhaps a list of errors (with empty list indicating PASS).
Examples of rules that we will implement:
Implement the compatibility rule for avro schemas. This can be lifted from Perspicuus most likely.
If the user does not send the right thing to the topic when sending a message, the serdes Serializer doesn't throw a sensible error. It just throws a "404 not found" error. I think this is because it's trying to find the Schema in the registry even when the message is e.g. a simple type. It would better if the serializer could detect that the data being sent isn't appropriate for serialization (e.g. it's just a simple type) and throw an appropriate error.
Implement the compatibility rule for JSON Schema schemas. We'll need to really analyze the avro version to implement similar logic for JSON Schema.
Implement the Validation Rule for OpenAPI content.
Create a java interface for the storage layer that can be implemented by:
RH OpenShift template points to internal registry with 1.5-6 tag
https://github.com/Apicurio/apicurio-registry/blob/1.0.x-redhat/distro/openshift-template/service-registry-template.yml
Should be registry.redhat.io/fuse7-tech-preview/fuse-service-registry-rhel7 (latest tag is 1.5-9)
Seems you could also cache the install of GraalVM to speed up builds a bit.
Add the following header to all .java files in the project:
/*
* Copyright 2019 Red Hat
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
@jsenko why do we do duplicate persist here?
First with nextVersion, then with null as version ... ?!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.