Croway opened a new pull request, #30075:
URL: https://github.com/apache/camel/pull/30075

   Docs and option descriptions that contradict the code, in the thread pool, 
Enrich, Simple, JAXB and SEDA pages.
   
   - **Thread pool `maxQueueSize`**: `-1` was documented as an unbounded queue, 
but `DefaultThreadPoolFactory` uses a `SynchronousQueue` for `maxQueueSize <= 
0`: no work queue, tasks go to up to `maxPoolSize` threads, then the rejected 
policy applies. `Integer.MAX_VALUE` is the unbounded queue. Fixed in 
`threading-model.adoc`, `threads-eip.adoc`, `ThreadPoolProfile`, 
`ThreadsDefinition`, `ThreadPoolProfileDefinition`, 
`AbstractCamelThreadPoolFactoryBean` and `camel.threadpool.maxQueueSize`. The 
threading model page also says that a pool grows past `poolSize` only when the 
queue is full (the default profile runs 10 tasks at a time until 1000 wait).
   - **Custom thread pool provider**: `threading-model.adoc` pointed to 
`org.apache.camel.spi.ExecutorServiceStrategy` and a "WorkManager", which do 
not exist. It now describes `ThreadPoolFactory`, used as a single Registry bean 
by Camel Main, Spring Boot and Spring XML (`DefaultConfigurationConfigurer`, 
`AbstractCamelContextFactoryBean`). The stale name is also gone from the 
Recipient List page and the Spring bean table.
   - **Threads EIP**: the `rejectedPolicy` row listed `Discard`/`DiscardOldest` 
(removed in 4.0) and not `Block`; it now lists `Abort`, `CallerRuns`, `Block` 
(`ThreadPoolRejectedPolicy`).
   - **Enrich EIP**: the example strategy named its parameters `(newExchange, 
oldExchange)`, but `Enricher` calls `aggregate(original, resource)`. It now 
uses `(original, resource)`, like the Poll Enrich page. 
`AggregationStrategies.string(":")` does not build the same strategy: it sets 
the body in `onCompletion`, which `Enricher` never calls. The page now shows 
`AggregationStrategies.bean(...)` with a POJO method (as 
`EnricherLambdaPojoAggregateTest` does) and links the latest javadoc instead of 
3.12.0.
   - **Simple `skip`**: `${skip(1)}` on a CSV text was said to start at the 2nd 
line, but a `String` body is split on commas (`ObjectHelper.createIterator`). 
The page now points to Tokenize with `skipFirst` or the CSV data format.
   - **JAXB data format** (`JaxbDataFormat` model): `prettyPrint` said "by 
default false" (it is `true`); `mustBeJAXBElement` said other bodies fail (by 
default they are written as is, `doMarshal`); `fragment` only sets 
`JAXB_FRAGMENT` (no XML declaration), partial trees use 
`partClass`/`partNamespace`; `schemaLocation` only writes `xsi:schemaLocation` 
and does not validate; `contextPath` needs an `ObjectFactory` or `jaxb.index`.
   - **SEDA**: the description said "from any Camel Context in the same JVM", 
but the queues are held per `SedaComponent`, so per CamelContext. It now says 
"from the same Camel Context".
   
   Regenerated: model and component json, catalog, XML and YAML schemas, YAML 
deserializers, camel-main metadata, component and endpoint DSL. The Spring Boot 
starter copies follow in camel-spring-boot.
   
   _Claude Code on behalf of Federico Mariani_
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to