I love to hear that. We should clarify the StringList problem, remove the last dependencies to the language independent spec and solve the remaining open issues. After that, the document can act as a good base for updated version of the Java- and the NET-spec. Peter. has all the concepts available, but expressed in a better way. Peter. Rajic, Hrabri schrieb:
I would like us to move on, so OO is preferable.
-Hrabri
-----Original Message----- From: owner-drmaa-wg@ggf.org [mailto:owner-drmaa-wg@ggf.org] On Behalf Of Daniel Templeton Sent: Friday, March 04, 2005 2:32 PM Cc: DRMAA Working Group Subject: Re: [drmaa-wg] RE: Latest OO binding spec v0.3
Which spec do you think the Java language binding should adhere to? The language independent spec or the OO spec? There are both technical and political reasons for both choices, and the choices are/will be mutally exclusive.
Daniel
Rajic, Hrabri wrote:
We should simplify our life a bit.
If we make the OO doc independent of/not referencing DRMAA v1.0 spec, all the nice things that were impractical to do in that doc could be rolled in.
-Hrabri
-----Original Message----- From: Daniel Templeton [mailto:Dan.Templeton@Sun.COM] Sent: Friday, March 04, 2005 12:49 PM To: Rajic, Hrabri Cc: DRMAA Working Group Subject: Re: [drmaa-wg] RE: Latest OO binding spec v0.3
Rajic, Hrabri wrote:
I have gone thru the spec. It is in a very good shape for version
0.3.
Introspection, page 3 is both a blessing and a curse. We need
standard
mechanisms, getters and setters for C++. There are few other places also mentioning just introspection, but Peter has marked few of them.
The intention is to provide guidance for both introspective and non-introspective languages. Please point out anywhere the non-introspective bit is missing.
Checked/unchecked exception list needs consensus. C++ seems to be different here too, isn't it?
As far as I know, Java is the only major language that makes that distinction. Smalltalk probably does too, since Java is borrowed many concepts from that language, but I don't forsee a Smalltalk binding anytime soon. This is a Java language binding issue, but since the OO spec is trying to accomodate everyone, it has to explicitly make allowances for all the outliers.
Comma delimited return strings needs to be converted to StringList everywhere.
Uh... That's not consistent with the DRMAA 1.0 spec. I really dislike what the 1.0 spec does, but this is one of those hot topics that will
be
difficult to change. I have brought it up before, only to be told that since it took so much bloodshed to reach an agreement in the first place, we should never revisit it. I personally see that part as the worst part of the 1.0 spec and would love to reopen the discussion.
Should we reference DRMAA 1.0 spec? How much info OO doc needs from
the
1.0 spec? If not much, it could be a good idea to make it spec 1.0 independent
This goes to my previous point. I have always forseen the OO spec as the source for the DRMAA 2.0 spec. In that regard, it should not be bound by the DRMAA 1.0 spec. However, since the OO spec is affecting changes in the Java and .Net specs, and we're holding the Java and .Net specs to the DRMAA 1.0 standard, the OO spec cannot break 1.0.
Daniel