Issue 3136: typedefs for TimeStamp and TimeSpan that are never used                                                                                                                                                 (coas-ftf)
Source: Cognition Group, Inc. (Dr. David Forslund, forslund@cognitiongroup.biz)
Nature: Uncategorized Issue
Severity: 
Summary: DsObservationValues.idl has typedefs for TimeStamp and TimeSpan that are 
never used.  Why not delete them since they create problems in the IDL2Java 
mapping by creating *Helper classes than cause name collisions?
DsObservationQualifers.idl has a typedef for TimeStamp that also is not used.

Can we remove these references?   They can have no impact on any 
application that uses these interfaces.

Resolution: rejected
Revised Text: 
Actions taken:
December 17, 1999: received issue
February 27, 2001: closed issue
Discussion: 
This is not an issue.
They have been typedef correctly to refer back to the one required idl file called DsObservationAccess.idl

End of Annotations:=====
X-Sender: u076663@cic-mail.lanl.gov
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.2.2 
Date: Fri, 17 Dec 1999 10:09:32 -0700
To: tcculpepper@mmm.com, coas-ftf@omg.org
From: David Forslund <dwf@lanl.gov>
Subject: Re: issues document
Cc: issues@omg.org
In-Reply-To: <86256832.0075CE08.00@em-stpmta-01.mmm.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
X-UIDL: ]nLe9Kcf!!(^%e9R>M!!

Here is a clean up issue.

DsObservationValues.idl has typedefs for TimeStamp and TimeSpan that
are 
never used.  Why not delete them since they create problems in the
IDL2Java 
mapping by creating *Helper classes than cause name collisions?
DsObservationQualifers.idl has a typedef for TimeStamp that also is
not used.

Can we remove these references?   They can have no impact on any 
application that uses these interfaces.

Dave
At 02:21 PM 11/23/99 -0700, tcculpepper@mmm.com wrote:


>I have enclosed the issues document. Please review.
>I have also enclosed some changes I think we need in the IDL.
>
>Chuck,
>Can you please look at issue 3042 and help get the right places for
thses
>insertions. Also, we will need to look at
>all the places ObservationData is referenced in the text. Where we
talk
>about ObservationData we will need to add
>some text, maybe some from what Tim wrote would work.
>Thanks
>(See attached file: COASFTFIssuesResolutions.doc)(See attached file:
IDL
>Changes.doc)
>Tom
>

X-Sender: u076663@cic-mail.lanl.gov
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.2.2 
Date: Fri, 17 Dec 1999 13:02:25 -0700
To: charles.carman@philips.com
From: David Forslund <dwf@lanl.gov>
Subject: Re: issues document
Cc: <coas-ftf@omg.org>
In-Reply-To: <0056910002965451000002L112*@MHS>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
X-UIDL: KUed94=Pd9iP7e9Cid!!

At 01:14 PM 12/17/99 -0600, charles.carman@philips.com wrote:
>Dave,
>
>What sort of name collisions are you having?
>These typedefs were used instead of defining new structures to hold
time 
>values.
>For completeness, we felt that ObservationValue should have some time

>related definitions, and we chose the existing definitions from 
>ObservationAccess.

But if they aren't used, they are unnecessary, since they are simply 
pointing to the ObservationAccess.   The name collisions are in the
Helper 
generated classes.  The classes themselves aren't there because of the

typedef, but Java needs Helper and Holder classes and at least the
Helper 
class is put in each package which has the corresponding typedef.
This 
requires one to use the fully qualified name for every one of those 
typedefs, to disambiguate it from the others.  If a different local
name 
was used in the typedef, the problem wouldn't occur.  The idl2java
creates 
a org.omg.DsObservationAccess.TimesSpanHelper and a 
org.omg.DsObservationValues.TimeSpanHelper class, for example. Thus
one 
can't simply refer to TimeSpanHelper if you have imported both of
those 
packages.   Since the TimeSpan is not used in ObservationValues and
simply 
points back to the TimeSpan in ObservationAccess, it isn't needed.
Is it clearer now?



>Are the name collisions in the IDL or in the resulting Java classes.
If 
>we defined new structures with the same names, would you still have
the 
>problem?
>
>I had problems with names when implementing a server using IDL in
both the 
>OMG name space and a Philips name space.  Even though the IDL was
fully 
>qualified, sometimes the Java class names were not correctly
qualified.  I 
>am not sure what the answer to
>this problem is.

That is a different problem which different orb vendors handle  in 
different ways, but is really either a bug or a configuration problem
with 
any particular orb product.

thanks,

Dave

>Charles S. Carman
>   Philips Medical Systems
>     2171 Landings Drive
>       Mountain View, CA 94043
>         +1.650.426.2509