[ome-devel] C/C++ client library for OME::Remote

[Zachary Pincus]

Harry,

Thanks for the feedback. To answer your questions:
I've been mainly looking at the code under org.openmicroscopy.remote, 
as per the newbie's guide on the docs site. I wasn't actually aware 
that this was deprecated.

Anyhow, my intention all along had been to use the OME XML-RPC API from 
some client program I'm developing (a mixture of Python, C++, and 
chewing gum). Looking over what documentation I could find, and at the 
code under org.openmicroscopy.remote, it appeared that there was some 
non-trivial amount of logic present in the Java OME XML-RPC client 
library, and that to profitably use the XML-RPC API from a different 
language/environment, I would need to replicate some of that logic.

This lead me to thinking that maybe it would be worthwhile to implement 
that logic in a C or C++ library, such that anyone, from any language 
that has C/C++ bindings (pretty much everything...), could make use of 
that library. (As long as I was going to be doing some of that work 
anyway, it would be nice to make something that would be useful to 
others.) I mentioned this on the OME list a week or so ago, and the 
idea met with some enthusiasm.

Now, there are a lot of potential reasons I should stay away from such 
a project. Perhaps (a) there really doesn't need to be that much 
client-side logic to use OME's XML-RPC API, or (b) that API is still 
changing too fast, or (c) such an endeavor would be too much for one 
grad student at partial effort, or (d) a lot more logic will be built 
in server-side, a la Chris's desires, or (e) any number of things.

Unfortunately, I really don't know much about where OME is going, etc, 
given that my main exposure to it has been talking with Ilya and Jason 
at a conference, some emails from them, and the docs and wiki web site 
(plus the codebase itself). That's why I've been trying to solicit some 
advice from those who know more than I about the advisability of this 
project, and on the best ways to approach it, if it would be a useful 
addition.

Any thoughts you have on the matter would be most appreciated,

Zach Pincus

Department of Biochemistry and Program in Biomedical Informatics
Stanford University School of Medicine


On Jul 19, 2004, at 7:04 AM, Harry Hochheiser wrote:

> Zach:
>
> I personally would argue for either (1) or (3) - whichever is going to 
> get you there as quickly and cleanly as possible.
>
> One question, though. when you discuss the caching in the java code, 
> are your referring to the code under org.openmicroscopy.remote?  I 
> don't know is this has been discussed here, but that code is in the 
> process of being deprecated, in favor of the org.openmicroscopy.ds, 
> org.openmicroscopy.is, and org.openmicroscopy.shoola code.
>
> These new frameworks don't handle any caching at all. This has led to 
> lots of discussion as to how we might extend them to support caching 
> and other desired features, but there's lots of work left to be done 
> in this regard.
>
> Question: why is C/C++ so important? Can you give us more of an 
> understanding about your goals and your need for C/C++?
>
> -harry
>
>
>
> On Jul 18, 2004, at 7:35 PM, Zachary Pincus wrote:
>
>> Hello,
>>
>> From the feedback I've gotten, it looks like there are three options 
>> for using OME XML-RPC outside of a Java environment:
>> ....
>