search

cts2 / cts2-specification

The CTS2 OMG Specification
http://informatics.mayo.edu/cts2
11 stars 2 forks source link

OMG Issue 18482: "readContext" missing on ResolvedValueSetResolution functions #126

Open hsolbrig opened 11 years ago

hsolbrig commented 11 years ago

The 'resolve' functions in ResolvedValueSetResolution lack a readContext - this is an oversight and needs to be corrected.

http://www.omg.org/issues/cts2-rtf#Issue18482

kevinpeterson commented 11 years ago

Need to add this to the PIM/WSDL/WADL as described above.

hsolbrig commented 11 years ago

The only attribute of readContext that could apply to ResolvedValueSet is the language. The description field of a resolved value set is language specific, so we can now have multiple resolved sets that differ by language. Proposed solution - add an optional resolution language to the ResolvedValueSetHeader and add it as an explicit parameter to the read and query operations.

hsolbrig commented 11 years ago

Rethinking the above comment. The identity of a value set should be the same whether its output consists of just a URI, a URI and a namespace/name and/or designation. The real extension is that the loader needs to be supplied with an optional language parameter and verbiage / error messages need to be supplied about what to do if the set of URI's is different in one set than in the other. Verbiage also needs to be supplied with respect to namespace/name - when it applies, how it is transmitted, etc. (Does a loader need a namespace map?)

cendle commented 11 years ago

Reason For Change

The 'resolve' functions in ResolvedValueSetResolution lack a readContext - this is an oversight and needs to be corrected.

The identity of a value set should be the same whether its output consists of just a URI, a URI and a namespace/name and/or designation. The real extension is that the loader needs to be supplied with an optional language parameter and verbiage / error messages need to be supplied about what to do if the set of URI's is different in one set than in the other. Verbiage also needs to be supplied with respect to namespace/name - when it applies, how it is transmitted, etc. (Does a loader need a namespace map?) ### Severity: Minor ### Proposed Disposition: Deferred It is proposed that this issue be deferred to the next RTF due to complexity and timing of the current RTF.