[ 
https://issues.apache.org/jira/browse/CXF-8583?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Andriy Redko closed CXF-8583.
-----------------------------
    Resolution: Not A Problem

> OpenApiFeature Always Appends Numbered Suffix (i.e. "_1") to operationId
> ------------------------------------------------------------------------
>
>                 Key: CXF-8583
>                 URL: https://issues.apache.org/jira/browse/CXF-8583
>             Project: CXF
>          Issue Type: Bug
>          Components: JAX-RS
>    Affects Versions: 3.4.4
>            Reporter: James W. Carman
>            Priority: Minor
>
> It appears that the duplicate calls to OpenApiContext.read() are causing it 
> to think the operationIds are duplicated, thus causing it to append a "_1" 
> (or whatever the counter gets to) to the operationIds in order to make them 
> unique.
> Here's my understanding of what's causing the collisions:
> In OpenApiCustomizedResource, it calls the OpenApiContext.read() operation 
> here:
> [https://github.com/apache/cxf/blob/75fb6bb56d82f72771a9ee6ecab5d36168303f51/rt/rs/description-openapi-v3/src/main/java/org/apache/cxf/jaxrs/openapi/OpenApiCustomizedResource.java#L77]
> At the end of the method, though, it makes a call to the superclass' 
> implementation of the getOpenApi() method here:
> [https://github.com/apache/cxf/blob/75fb6bb56d82f72771a9ee6ecab5d36168303f51/rt/rs/description-openapi-v3/src/main/java/org/apache/cxf/jaxrs/openapi/OpenApiCustomizedResource.java#L106]
> The overridden BaseOpenApiResource class' implementation of getOpenApi() 
> subsequently calls OpenApiContext.read() again here:
> [https://github.com/swagger-api/swagger-core/blob/v2.1.6/modules/swagger-jaxrs2/src/main/java/io/swagger/v3/jaxrs2/integration/resources/BaseOpenApiResource.java#L51]
> It is during this second invocation of the OpenApiContext.read() method in 
> the superclass where the operatorId collisions occur.
> I am happy to submit a patch to address it, but I do ask for a little 
> guidance here on how we'd like to fix this, please. It would seem that this 
> would break existing behavior and I don't want to mess anyone up.
> Thanks,
>  
> James
>  



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

Reply via email to