[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [sca-j] JAVA-125: proposed resolution
Graham, See replies inline. Simon Graham Charters wrote: > > Hi Simon, thanks for taking the time to look at the proposal. I've > added some comments below. > > > Regards, > > Graham. > > > > Graham Charters PhD CEng MBCS CITP > STSM, AIM Technical Lead OSGi Expert Groups, Master Inventor, UKISA > Technical Staff Member > IBM United Kingdom Limited, MP 146, Hursley Park, Winchester, SO21 2JN, UK > Tel: (Ext) +44-1962-816527 (Int) 7-246527 (Fax) +44-1962-818999 > Internet: charters@uk.ibm.com > > > Simon Nash <oasis@cjnash.com> wrote on 09/02/2009 14:49:39: > > > [image removed] > > > > Re: [sca-j] JAVA-125: proposed resolution > > > > Simon Nash > > > > to: > > > > sca-j > > > > 09/02/2009 14:50 > > > > Graham, > > Thanks for doing this. I have two comments. > > > > 1. I don't think it should be illegal to say @Remotable on the > > Java interface as well as on <interface.java>. We allow a > > callback interface to be specified on <interface.java> when > > it is also specified using a @Callback annotation on the > > forward Java interface. This case seems extremely similar. > > > > This design was based on the belief that @Remotable was similar to > intents and if the developer has made a statement with the annotation we > shouldn't be changing that statement. If we override an @Remotable and > say "remotable=false", then local invocations would start using > by-reference semantics where an implementation may have assumed it would > receive a copy. This isn't clear cut, because the absence of @Remotable > may or may not have been a conscious choice, and therefore we can't say > whether specifying "remotable=true" is overriding a conscious decision > to leave off @Remotable, or whether it's just the first time remotable > has been considered for the interface. > I agree that if the interface says @Remotable, then it wouldn't be valid for <interface.java> to say remotable="false". I think the other combinations are all OK: interface says nothing, <interface.java> says remotable="false" interface says nothing, <interface.java> says remotable="true" interface says @Remotable, <interface.java> says remotable="true" I understood the wording in your proposal your proposal to be disallowing the third of these. My comment was intended to apply to this case, saying that it should be allowed. > > 2. Given that remotable interfaces are an assembly-level concept, > > and the base definition of <interface> is part of the assembly > > specification, it seems strange for SCA to only define the > > @remotable attribute for <interface.java>. Would it make sense > > for this attribute to be part of the base <interface> element? > > > > I believed that this had been raised in the assembly TC prior to the > @remotable attribute being added to the C/C++ specs, but Bryan has > clarified to me that this is in fact not the case. I will therefore > raise this as an issue in the Assembly TC. If that is accepted, we will > still need some statement for how the attribute and annotation interact. > Thanks for doing this. I agree that the Assembly issue does not replace the need for this sca-j issue. Simon > > > Simon > > > > Graham Charters wrote: > > > > > > Hi All, > > > > > > Please find attached a document containing a proposed resolution to > > > JAVA-125. The edits are as follows: > > > > > > 1. Section 3.1 - add and define a new @remotable attribute. Also > add a > > > new conformance statement for the relationship between the @Remotable > > > annotation and the @remotable attribute. > > > 2. Appendix A - add the attribute definition to the XML schema for > > > interface.java. > > > > > > Changes considered, but excluded: > > > 1. Considered changing section 2.1.2 but this seemed to be > specifically > > > covering the metadata in the Java code, rather than the general > concept > > > of 'remotable'. > > > 2. Considered having an @remotableCallback attribute, but given it is > > > an error to mix remote and local interfaces (section 6.7), I concluded > > > the @remotable attribute should apply to both. > > > > > > > > > > > > > > > Regards, > > > > > > Graham. > > > > > > > > > > > > Graham Charters PhD CEng MBCS CITP > > > STSM, AIM Technical Lead OSGi Expert Groups, Master Inventor, UKISA > > > Technical Staff Member > > > IBM United Kingdom Limited, MP 146, Hursley Park, Winchester, SO21 > 2JN, UK > > > Tel: (Ext) +44-1962-816527 (Int) 7-246527 (Fax) +44-1962-818999 > > > Internet: charters@uk.ibm.com > > > > > > > > > > > > > > > > ------------------------------------------------------------------------ > > > > > > / > > > / > > > > > > /Unless stated otherwise above: > > > IBM United Kingdom Limited - Registered in England and Wales with > number > > > 741598. > > > Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire > PO6 3AU/ > > > > > > > > > > > > > > > > > > > > > > > > > ------------------------------------------------------------------------ > > > > > > --------------------------------------------------------------------- > > > To unsubscribe from this mail list, you must leave the OASIS TC that > > > generates this mail. Follow this link to all your TCs in OASIS at: > > > https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php > > > > > > > > --------------------------------------------------------------------- > > To unsubscribe from this mail list, you must leave the OASIS TC that > > generates this mail. Follow this link to all your TCs in OASIS at: > > https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php > > > > > > ------------------------------------------------------------------------ > > / > / > > /Unless stated otherwise above: > IBM United Kingdom Limited - Registered in England and Wales with number > 741598. > Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU/ > > > > > >
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]