Using decaf API in user code?

[MarcSchroeder <sc...@dfki.de>]

Hi,

as a Java programmer who needs to write a wrapper around ActiveMQ-CPP in
C++, I am of course very happy to see familiar concepts such as
"synchronized()" available as part of the "decaf" namespace under
ActiveMQ-CPP 2.2.

Indeed, when mixing threads between ActiveMQ and my code, I guess it is a
good idea to use the same threading mechanism as ActiveMQ, so reusing the
stuff from decaf::util::concurrent seems like the right thing to do.

So here is my question: How stable can I expect the 'decaf' API to be? Is it
safe for me to use this, i.e., can I hope that changes to the decaf API in
future versions of ActiveMQ-CPP will be backward-compatible?

I am asking because of the incompatible API change between ActiveMQ-CPP 2.1
and 2.2.

Thanks for this nice piece of software, and best regards,
Marc 
-- 
View this message in context: http://www.nabble.com/Using-decaf-API-in-user-code--tp19436487p19436487.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.

Re: Using decaf API in user code?

[MarcSchroeder <sc...@dfki.de>]

Thanks Nate and Tim,

that clarifies the situation. Given the added value and simplicity of use, I
will use the current decaf code and will be able to live with minor code
adaptations in the future.

Regards,
Marc
-- 
View this message in context: http://www.nabble.com/Using-decaf-API-in-user-code--tp19436487p19450057.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.

Re: Using decaf API in user code?

[Timothy Bish <ta...@gmail.com>]

Thanks for the kind words, now to answer your question.

In general the CMS interface in ActiveMQ-CPP is the only portion that
gets close scrutiny for ABI compatibility concerns right now.  That
being said we try not to make to many API changes underneath between
minor versions, however, since the decaf API is still quite immature and
is intended to slowly become more closely aligned with that of the Java
SDK there are some things that will most likely change in the future.
Functions and Classes in decaf are currently being added and changed on
an as needed basis since the real goal is to provide a solid ActiveMQ
client.  

In the near term I don't see many changes happening to the Threading
classes in decaf, however down the road I would like to make things a
bit more Java-sih by replacing mutex with the Locks package in the Java
1.5 API.  In general there probably won't be anything nearly as drastic
as the changes that occurred between 2.1 and 2.2.

As decaf matures it would be great if it could be pulled out and made a
sub project of ActiveMQ-CPP so that more attention would be paid to ABI
rules but since its still pretty young that probably won't happen any
time soon.  

Regards
Tim.

On Thu, 2008-09-11 at 07:55 -0700, MarcSchroeder wrote:
> 
> Hi,
> 
> as a Java programmer who needs to write a wrapper around ActiveMQ-CPP in
> C++, I am of course very happy to see familiar concepts such as
> "synchronized()" available as part of the "decaf" namespace under
> ActiveMQ-CPP 2.2.
> 
> Indeed, when mixing threads between ActiveMQ and my code, I guess it is a
> good idea to use the same threading mechanism as ActiveMQ, so reusing the
> stuff from decaf::util::concurrent seems like the right thing to do.
> 
> So here is my question: How stable can I expect the 'decaf' API to be? Is it
> safe for me to use this, i.e., can I hope that changes to the decaf API in
> future versions of ActiveMQ-CPP will be backward-compatible?
> 
> I am asking because of the incompatible API change between ActiveMQ-CPP 2.1
> and 2.2.
> 
> Thanks for this nice piece of software, and best regards,
> Marc 

RE: Using decaf API in user code?

["Mittler, Nathan" <na...@sensis.com>]

Hi Marc,
I'm glad to see you're interest in decaf - we've worked hard to make
programming in C++ painless (well, maybe less painful) to Java
developers :)

As far as the stability of the decaf API goes, I think at this point the
dust has settled a bit and we don't really have any plans for changing
APIs - adding to them perhaps, but not changing.

Regards,
Nate

> -----Original Message-----
> From: MarcSchroeder [mailto:schroed@dfki.de]
> Sent: Thursday, September 11, 2008 7:55 AM
> To: users@activemq.apache.org
> Subject: Using decaf API in user code?
> 
> 
> 
> Hi,
> 
> as a Java programmer who needs to write a wrapper around ActiveMQ-CPP
in
> C++, I am of course very happy to see familiar concepts such as
> "synchronized()" available as part of the "decaf" namespace under
> ActiveMQ-CPP 2.2.
> 
> Indeed, when mixing threads between ActiveMQ and my code, I guess it
is a
> good idea to use the same threading mechanism as ActiveMQ, so reusing
the
> stuff from decaf::util::concurrent seems like the right thing to do.
> 
> So here is my question: How stable can I expect the 'decaf' API to be?
Is
> it
> safe for me to use this, i.e., can I hope that changes to the decaf
API in
> future versions of ActiveMQ-CPP will be backward-compatible?
> 
> I am asking because of the incompatible API change between
ActiveMQ-CPP
> 2.1
> and 2.2.
> 
> Thanks for this nice piece of software, and best regards,
> Marc
> --
> View this message in context:
http://www.nabble.com/Using-decaf-API-in-
> user-code--tp19436487p19436487.html
> Sent from the ActiveMQ - User mailing list archive at Nabble.com.


This message is intended only for the addressee and may contain information that is confidential or privileged. Unauthorized use is strictly prohibited and may be unlawful. If you are not the intended recipient, or the person responsible for delivering to the intended recipient, you should not read, copy, disclose or otherwise use this message. If you have received this email in error, please delete and advise immediately.