[Biojava-dev] Hi.. Calling Java from Perl

[Andy Yates]

Depends on what you're going to use but when I last did it I bought into
the Spring way of things and found that the spring manual was very good.
The WS bit is:

http://static.springsource.org/spring/docs/3.0.x/spring-framework-reference/html/ch21s05.html

It goes through doing it for JAX-WS & XFire.

There's also a JAX-WS tutorial from:

http://java.sun.com/javaee/5/docs/tutorial/doc/?wp405739&JAXWS.html#wp72279

To be honest though Google is your best friend here.

Good luck,

Andy

Abhishek Pratap wrote:
> Thanks all. I think Java WS is a way out for me then. As you said it
> would be code agnostic and will help me in updating the core code
> later.
> 
> Just a quick question . Do you happen to know of any good tutorial to
> implement a WS for a java process.
> 
> Thanks,
> -Abhi
> 
> On Wed, Jul 29, 2009 at 4:48 AM, Andy Yates<ayates at ebi.ac.uk> wrote:
>> Indeed I would agree with Mark here and go for web services as the
>> desired solution. JAX-WS & CXF are both popular frameworks for doing Web
>> Services and as far as I remember Spring had some very nice helper
>> classes for quickly exposing any Java class as a web service. Then there
>> are other remoting protocols such as Hessian, Burlap, Protocol Buffers
>> or Thrift all of which are good in their own ways.
>>
>> However Web Services should be the quickest (re implementation) way to
>> communicate with a persistent Java process.
>>
>> Personally I would stay away from Java::Inline.
>>
>> Andy
>>
>> Mark Schreiber wrote:
>>> Hi -
>>>
>>> You could try and use something like CORBA but that would be quite ugly.
>>>
>>> A nicer alternative would be to put the BioJava functionality in a web
>>> service and send sequences as FASTA or some custom format??
>>>
>>> I think WS is considered the best way for Java and .NET to talk so probably
>>> it is for Perl too.
>>>
>>> - Mark
>>>
>>> On 29 Jul 2009, 6:24 AM, "Abhishek Pratap" <abhishek.vit at gmail.com> wrote:
>>>
>>> Hi Andy
>>>
>>> Thanks for a quick reply.  I think SHELLING out will be too process
>>> intensive as we expect thousands of call to same Java method. I also
>>> read about the Perl modules Java::Inline. Is that any good ?
>>>
>>> And to answer your second question I am basically using a inhouse
>>> method which in turns used a lot of BioJava classes for DNA
>>> manipulation.
>>>
>>> Thanks,
>>> -Abhi
>>>
>>> On Tue, Jul 28, 2009 at 5:54 PM, Andy Yates<ayates at ebi.ac.uk> wrote: > Hi
>>> Abhi, > > Well to answer ...
>>>