RE: [sca-j] NEW ISSUE: Unannotated properties and references

Hi,

I’ve raised this new issue as JAVA-133. See http://www.osoa.org/jira/browse/JAVA-133

Thanks,

Mark

> -----Original Message-----

> From: Simon Nash [mailto:oasis@cjnash.com]

> Sent: 22 February 2009 10:44

> To: OASIS Java

> Subject: [sca-j] NEW ISSUE: Unannotated properties and references

> TARGET: Java Common Annotations and APIs

> DESCRIPTION: Unannotated properties and references

> The description of the @Property annotation includes the following

> paragraph (as amended by JAVA-117):

> Properties may also be injected via setter methods even when the

> @Property annotation is not present. However, the @Property annotation

> must be used in order to inject a property onto a non-public field.

> In the case where there is no @Property annotation, the name of the

> property is the same as the name of the field or the JavaBeans property

> name [JAVABEANS] corresponding to the setter method name.

> Similarly, the description of the @Reference annotation includes the

> following paragraph (as amended by JAVA-117):

> References may also be injected via setter methods even when the

> @Reference annotation is not present. However, the @Reference

> annotation

> must be used in order to inject a reference onto a non-public field.

> In the case where there is no @Reference annotation, the name of the

> reference is the same as the name of the field or the JavaBeans property

> name [JAVABEANS] corresponding to the setter method name.

> There are a number of problems with these paragraphs:

> 1. The requirement that the @Property or @Reference annotation must be

> used in order to inject a reference onto a non-public field

> contradicts

> the resolution of JAVA-55, which states that public or protected

> fields

> are taken into consideration for deriving properties and references

> by introspection.

> 2. It's not clear how properties and references are derived from

> unannotated setter methods. The implementation intropection rules

> laid down by the resolution to JAVA-55 define one case where this

> is possible. Is this text referring only to that case, or is it

> implying there are other cases as well?

> 3. These paragraphs shouldn't be part of the descriptions of @Property

> and @Reference, as they refer to cases where @Property and @Reference

> aren't present. The rules covering properties and references in

> these cases should be defined by C&I specs, not the JavaCAA spec.

> PROPOSAL:

> Remove the above paragraphs from the descriptions of @Property and

> @Reference.

> ---------------------------------------------------------------------

> 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

sca-j message