Loading the Registry from the JSON file

[kyxhik]

Hello, I was wondering if there is documentation about how to load the registry programmatically from the getResolverDataset json file.

The project I am working on has the following dependency
<dependency> <groupId>org.identifiers.registry</groupId> <artifactId>registry-lib</artifactId> <version>1.1.1</version> </dependency>

And it is loading the resources like this:

RegistryDatabase.loadFromFile(file);

There is a loadFromInputStream version as well. But I am getting the following error, which clearly shows these methods accept the old xml files.

[Fatal Error] :1:1: Content is not allowed in prolog. Exception in thread "main" java.lang.RuntimeException: org.xml.sax.SAXParseException; lineNumber: 1; columnNumber: 1; Content is not allowed in prolog.
Thanks

[renatocjn]

Hello @kyxhik, thank you for the interest in identifiers.org.

I'm am unaware of the library that you are currently using. Could you provide of the location where you found it?

The library that we currently have up can be found at https://mvnrepository.com/artifact/org.identifiers.cloud/libapi, and I don't think this library has a way to access the registry dataset. However, we are currently considering deprecating it as maven central stats shows it unused and it requires considerable work to manage it.

The current recommendation for Java users that wish to use the identifiers.org registry data is to parse it directly using their prefered libraries. A benefit of this is that if you already have a dependency to parse json files, you won't need another one. In our source code, we use Jackson ObjectMapper or Spring's RestTemplate.

Here are some links to related source code:

• https://github.com/identifiers-org/monorepo/tree/master/webservices/commons/src/main/java/org/identifiers/cloud/commons/messages
• https://github.com/identifiers-org/monorepo/blob/master/webservices/validator-registry/src/main/java/org/identifiers/cloud/hq/validatorregistry/registryhelpers/ResolutionDatasetFetcher.java

Here's an example with Jackson and Apache's webclient.

import org.apache.hc.client5.http.fluent.Request;
import com.fasterxml.jackson.core.type.TypeReference;
import com.fasterxml.jackson.databind.ObjectMapper;
import org.identifiers.reverseresolution.models.api.ResolverDatasetPayload;
import org.identifiers.reverseresolution.models.api.ServiceResponse;

...

var datasetAsString = Request.get("https://registry.api.identifiers.org/resolutionApi/getResolverDataset")
                             .execute().returnContent().asString();

ObjectMapper objectMapper = new ObjectMapper();
var typeRef = new TypeReference<ServiceResponse<ResolverDatasetPayload>>() {};
return objectMapper.readValue(datasetAsString, typeRef)
                   .getPayload().getNamespaces();

[kyxhik]

Dear @renatocjn

Thank you for the elaborate response. I am basically trying to parse the elements of the json similarly to how the previous identifiers xml export could be parsed.

I understand the method for obtaining namespaces, but I am just unsure as to what dependency is needed to import the org.identifiers.reverseresolution.api. classes.

I was trying objectMapper.readtree(inputstream)

[renatocjn]

Glad I could help with accessing the namespaces.

About the reverse resolution classes. This feature is still in a beta stage as we see what works and what doesn't, so the the API may change significantly as we move forward. Thus, we didn't publish documentation on the API to prevent people from using it and later having to change their pipeline significantly. Plus, the current implementation is not very useful for automation as user verification is strongly advised. For now we are hoping that people use it from our front page and provide us with feedback.

Could you elaborate a bit on your use case? Maybe it could help us drive this feature forward. If you do not wish to explose it publicly, feel free to use our email at identifiers-org@ebi.ac.uk.

[kyxhik]

Thank you, @renatocjn. I just sent an email.