Stream: implementers
Topic: version question
David Hay (May 27 2016 at 02:57):
If I ask for history on a non-existent resource - eg /Condition/200/_history where there is no Condition with an id of 200, then is the correct response a 404, or an empty bundle with an OperationOutcome?
Jason Walonoski (May 27 2016 at 11:48):
It could also be a 410 if the Condition used to exist, and it was deleted. According to http://hl7.org/fhir/2016May/http.html#history, since you don't supply a [vid], it should be 200 with empty bundle and OperationOutcome.
Jason Walonoski (May 27 2016 at 11:49):
Not sure that makes sense though.
David Hay (May 27 2016 at 17:53):
Yes, that was the point raised by the dev - that the empty bundle was the correct response, but it did feel a bit strange if the underlying resource did not exist - hence the question...
Grahame Grieve (May 27 2016 at 21:34):
I don't think that section says that a 200 empty bundle is the correct response; it just doesn't address the point. And there's a typoe in there as well
Grahame Grieve (May 27 2016 at 21:36):
can one of you create a task to clarify this?
Grahame Grieve (May 27 2016 at 21:36):
I think that if the resource doesn't exist (and never existed) then the response should be a 404 not found, and we should clarify that
Grahame Grieve (May 27 2016 at 21:36):
some if you do a type level history for a not supported type
Grahame Grieve (May 27 2016 at 21:36):
"The principal reason a resource might be missing is that the resource was changed by some other channel rather than via the RESTful interface" - should be "The principal reason a request might be missing..."
David Hay (May 30 2016 at 08:16):
http://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=10093
Erich Schulz (May 30 2016 at 08:20):
it seems like a reasonably finite set of possibilities...
Erich Schulz (May 30 2016 at 08:22):
located "on trolley A, in ambularnce B in the hull of plane/ship C, mored at birth D" would be about as bad as it gets
Erich Schulz (May 30 2016 at 08:25):
with A, B & C all being intances of nestable vehicles anc D being a physical location
Erich Schulz (May 30 2016 at 08:25):
B and C would both have "sub locations"
Erich Schulz (May 30 2016 at 08:25):
so could A in some countries....
Brian Postlethwaite (May 31 2016 at 01:16):
Wrong thread Erich I think, but yes, this is a good (bad) example of the potential complexity of locations.
Erich Schulz (May 31 2016 at 01:30):
arg sorry! will double check my threads from now on!
Last updated: Apr 12 2022 at 19:14 UTC
