[
https://issues.apache.org/jira/browse/DAFFODIL-1989?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Josh Adams reassigned DAFFODIL-1989:
------------------------------------
Assignee: Josh Adams
> lengthKind 'explicit' with expression should not use minLength nor
> textOutputMinLength
> --------------------------------------------------------------------------------------
>
> Key: DAFFODIL-1989
> URL: https://issues.apache.org/jira/browse/DAFFODIL-1989
> Project: Daffodil
> Issue Type: Bug
> Components: Back End, Middle "End"
> Affects Versions: 2.1.0
> Reporter: Michael Beckerle
> Assignee: Josh Adams
> Priority: Major
> Fix For: 2.3.0
>
>
> Per DFDL Erratum 5.18, if the dfdl:lengthKind is explicit, the facet
> minLength is NOT used to determine the target length. The target length is
> the value of the length expression (or constant) directly. The minLength
> facet is only used for validation (if enabled).
> Code in Daffodil e.g., UnparseTargetLengthInCharactersEv, MinLengthInBitsEv,
> and UnparseTargetLengthInBitsEv, all take minLength into account even for
> lengthKind explicit, and do not consider whether the expression is a constant
> or not. (That check occurs elsewhere, so need not be taken into account, but
> minLength needs to be disregarded as far as the target length is concerned,
> but still kept around for validation purposes.)
> Similarly, for types other than string/hexBinary that are text,
> textOutputMinLength should be ignored if dfdl:lengthKind is explicit, though
> it is an error if the length is a constant and is out of range for
> textOutputMinLength. (Similarly for minLength/maxLength, except that is an
> SDW)
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
