stevedlawrence opened a new pull request, #1576: URL: https://github.com/apache/daffodil/pull/1576
Commit 6650eb918e modified numeric operations so that if the args did not have any common parent type (e.g. xs:double and xs:int) then we always promoted to xs:decimal. However, this caused tiny changes in precision for some operations so could lead to minor backwards incompatibilities. This modifies the logic so we only promote numeric operations to xs:decimal if one of the args is a xs:nonNegativeInteger, xs:integer, or xs:decimal (i.e. arbitrary-precision types). If neither arg is arbitrary-precision, then it means both args have finite precision (xs:float, xs:double, xs:long, xs:unsignedLong, or subtype) and so we promote to xs:double. This should give more efficient performance, maintains fixed-precision/width operations, and maintains backwards compatibility with previous versions of Daffodil. DAFFODIL-2574 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
