Match(header("name").isAMatchTo(language("simple",
"Claus")));
// TODO: this already works, but I do not know how to register a "test
language" inside a test
//assertMatches(header("size").isAMatchTo(language("math", "123 > x <
345")));
//assertDoesNotMatch(header("size").isAMatchTo(language("math", "x^2 +
x*3 > 300")));
}
Let me know what do you think, if it could be useful to open a bug to
better evaluate this or not.
Bye
Lorenzo
I ended up writing a custom AbstractJdbcMessageIdRepository which essentially
uses the table I was using previously with my "by-hand" method.
Do you think that can fly ?
(at first sight it does )
--
View this message in context:
http://camel.465427.n5.nabble.com/Transaction-design-for-a-c
Hi Charles,
I did not know about this idempotentConsumer, thanks for that.
I wonder if it pays off in my case though, as I have to query the repository
table so I can select things to send out based on what has already been
sent.
I have done that for now, which I think suffice for my needs:
Hello there,
I am writing a custom camel component which essentially has to record what
it has sent successfully so that it does not resend the same piece of info
twice.
Surely this involves a transactional behaviour. The data is read from a DB
(JDBC) and is sent to a JMS queue. All good in terms
Versions: 2.8.1
Reporter: Lorenzo Luconi Trombacchi
Priority: Minor
I found an issue with seda and stream endpoints.
Using more then 1 concurrent consumers for seda and using "stream:out" as
destination endopoint , I sometimes receive some NullPointerExceptio