sorry, it obviously is java but I actually meant to say 'Is it possbile
to convert MultiServerStartupListener servlet to _scala_ too?'
On 02/01/2013 04:48 PM, Michal Linhard wrote:
MultiServerStartupListener is written in Java.
but maybe it would be ok to initialize ManagerInstance.instance
directly in ServerInstance.start()
actually the code could me much simpler that way :)
maybe one reason to let StartupListener to do it's work is that it's
the code that it's being tested as well.
In case of multiple servers the ManagerInstance could be initialized
directly...
On 02/01/2013 03:38 PM, Adrian Nistor wrote:
> Is it possbile to convert MultiServerStartupListener servlet to java
> too? Otherwise it looks like an odd crossbreed :) Do we really need a
> servlet to bootstrap this or just a ServletContextListener will do?
>
> On 02/01/2013 03:57 PM, Michal Linhard wrote:
>> On 01/31/2013 01:37 PM, Mircea Markus wrote:
>>> On 31 Jan 2013, at 12:28, Michal Linhard wrote:
>>>
>>>> In this special case I'm gonna try to rewrite the test case as an
>>>> excercise - it doesn't seem like that much extra work, and I'd
>>>> rewrite
>>>> it anyway I'm not satisfied with it's current form even if it was
to
>>>> stay in Java.
>>> -1. If there's a choice it should be Java.
>> So I did some Scala excercise, which I quite enjoyed, but I have to
>> admit, it's not practical to require Scala in all the test code.
>>
>> So how about this:
>>
https://github.com/mlinhard/infinispan/commit/297c275a22f582eafb1aaf1b0eb...
>>
>>
>> @Galder can you please review the Scala part ?
>>
>> MultiServerStartupListener - new class that I started in Java - this
>> will make the precedent of commits of Java code into server modules,
>> that may come out of this discussion
>> ServerInstance - continued in Scala
>> TwoServerTest - written in scala, but trivial, most of the code is
>> copypaste from other test cases, there is quite straightforward
>> line-by-line correspondence to java.
>>
>> m.
>>
>