[jira] [Created] (CAMEL-23657) Enhance component metadata to mark destination-identity query parameters for topology matching

["Claus Ibsen (Jira)" <ji...@apache.org>]

Claus Ibsen created CAMEL-23657:
-----------------------------------

             Summary: Enhance component metadata to mark destination-identity query parameters for topology matching
                 Key: CAMEL-23657
                 URL: https://issues.apache.org/jira/browse/CAMEL-23657
             Project: Camel
          Issue Type: Improvement
          Components: camel-core
            Reporter: Claus Ibsen


The route topology service (CAMEL-23656) matches endpoints across routes using scheme:context-path, stripping all query parameters. This works for the vast majority of components (88%) where the destination identity (queue name, topic name, bucket, etc.) is in the URI context path.

However, some components encode the destination identity in a required query parameter rather than the path. For these components, two URIs with the same scheme:context-path but different query parameter values actually point to different destinations, and topology matching would incorrectly consider them the same.

This ticket proposes enhancing the component metadata (the component JSON schema) to allow marking specific query parameters as "destination identity" parameters. The route topology service can then include those marked parameters when building canonical URIs for matching.

Components where destination is in a query parameter (not the path):

||Scheme||Path contains||Destination query param||
|couchbase|host:port|bucket (required)|
|mongodb|connection bean ref|database + collection|
|mongodb-gridfs|connection bean ref|database (required)|
|spring-rabbitmq|exchange name|routingKey + queues|
|dapr|operation type|topic|
|hwcloud-obs|operation type|bucketName|
|iggy|topic name|streamName|

Additionally, there are ~12 components (kubernetes-*, web3j) where the path identifies the server/cluster rather than the destination, with namespace or contract address in query parameters.

And ~19 components (azure-servicebus, azure-eventhubs, azure-cosmosdb, knative, cql, etc.) where path parameters are technically optional but usually provided by users.

This is a follow-up to CAMEL-23656.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)