This wiki has undergone a migration to Confluence found Here
<meta name="googlebot" content="noindex">

CQL 1 1 DSTU Update

From HL7Wiki
Jump to navigation Jump to search

Return to Clinical Decision Support Work Group

Instructions

The CDS Work Group has received numerous comments on the CQL DSTU specification. These comments have been discussed and resolutions approved, and the changes applied to an updated version of the specification, CQL Release 1.1.

The CDS Work Group applied HL7 Guidance to determine the appropriate review process and selected the peer review DSTU Update Process. The review period begins April 18th, 2016, and will be two weeks, closing Monday, May 5th. Any review comments received during that time will be discussed at an upcoming CDS WG call.

The ballot review package is available here (members only):

CQL R1.1 Update Review Package

For an overview of the changes, please see this presentation:

CQL DSTU Updates

Comments can be submitted either by updating this page and adding a comment to the table below, or by submitting a DSTU comment here.

Comments

Name Item Existing Wording Proposed Wording Comment
Chris Moesel 3.1.7 Operator Precedence = <> ~ !~ = != ~ !~ If we are introducing !~ as not equivalent, then I think it makes sense to have != be not equal, rather than <>. This is consistent with many programming languages.
Chris Moesel 3.2.3 Function Resolution Note that although CQL supports forward declaration for expressions, the same is not true for function definitions. Forward declaration for function definitions is not currently supported by CQL. This limitation seems arbitrary and makes CQL more difficult to author. Why don't we support forward declaration of functions?
Chris Moesel 3.4.4.2 Implicit Conversions The second page of the table is difficult to read because the table headers are lost. We should repeat the table headers when tables get split across pages, or push the whole table to the second page.
Chris Moesel 3.4.4.2 Implicit Conversions The conversion from a tuple to a structured type requires that the set of elements in the tuple type be a subset (possibly proper) of the elements in the structured type. The conversion from a tuple to a structured type requires that the set of elements in the tuple type be the same set or a subset of the elements in the structured type. I think I know where you're trying to go with the (possibly proper) parenthetical, but I find it confusing. I think the suggested wording makes it clearer.
Chris Moesel 4.11 Structured Values Indexers specified in paths must be literal integer or string values. Indexers specified in paths must be literal integer values. The specification only indicates integers as proper indexers -- I'm not sure how/where you would use a string indexer.
Chris Moesel 9.6.12 Multiply For multiplication operations involving quantities, the input quantities must be of the same dimension (not necessarily the same units), and the resulting quantity will have the appropriate unit. Should multiplication require the same dimension? Doesn't 3 'cm' * 12 'cm2' == 36 'cm3'? Or is that just too potentially complex?