3 Fixes for 3 issues discovered by M.Taylor

[Bonnarel]

Not familiar with this. Hope this change proposal is correct. Could be the base of an erratum

[mbtaylor]

I'm happy with the fix for issue #1 .

The fix for #2 needs changing - the word "fixed" is OK but the "{\em [?...?]}" is not supposed to be there (I just added that markup to emphasise that there was a problem).

Issue #3 - really? The edit in the PR removes the ref attribute from the ID PARAM, so I can't see where its value would ever come from, or what purpose it serves. I thought the change would be to add ID="primaryID" to one of the columns in the table. But maybe I don't understand self-describing services. Can somebody else review?

[Bonnarel]

HI Mark, #2 <#2> I let the bracket stuff because I wanted Pat to validate the "fixed" term. He wrote this original section. #3 <#3> : well, in a self describing service the service descriptor is there to tell you which input parameters the service understands without specifying value ranges. The case described in 1.0 is for an empty request. So the input ID parameter should come from elsewhere (ivoa:ids or a priori knowledge of the ID "range" or "set" . The output table coming after is also empty. It gives you an apriori hint of the structure of the possible query responses. We could also put a ref to the "ID" column there (because the iD column in the response is supposed to repeat the input ID value) but I think it's more confusing. Anyway SIAP service was a bad choice to explain self described services. So this fix is probably for an erratum for 1.0. In 1.1 there is a full rewriting of this section proposed Cheers François Le 17/10/2019 à 11:46, Mark Taylor a écrit :

…

I'm happy with the fix for issue #1 <#1> . The fix for #2 <#2> needs changing - the word "fixed" is OK but the "{\em [?...?]}" is not supposed to be there (I just added that markup to emphasise that there was a problem). Issue #3 <#3> - really? The edit in the PR removes the ref attribute from the ID PARAM, so I can't see where its value would ever come from, or what purpose it serves. I thought the change would be to add ID="primaryID" to one of the columns in the table. But maybe I don't understand self-describing services. Can somebody else review? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#4>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AMP5LTGXFZIW425JBYHWAZLQPAX6VANCNFSM4IFN7X7Q>.

[pdowler]

One small change.

[pdowler]

For the vospace service descriptor example, I would just remove "and inputs are fixed" because it doesn't add anything.

[Bonnarel]

I modified the change for REStful according to Pat's demand. This text is for an erratum, I guess

[Bonnarel]

This change is for RESTful fix. All is valid for an erratumon 1.0

[Bonnarel]

There were 3 changes in this PR. 2 were typos or unachieved sentences. they are still to be changed in the master. The third was an error in the "Self-described" service. But this subsection has been totally changed (the example is another service). So the change is now irrelevant to the PR. this has been solved. But it still remain the need to create errata for the 3 "bugs" in 1.0.