[
https://issues.apache.org/jira/browse/DAFFODIL-1584?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16869518#comment-16869518
]
Michael Beckerle commented on DAFFODIL-1584:
--------------------------------------------
Daffodil is not conforming to this behavior. Daffodil uses the minLength in
UnparseTargetLengthInBitsEv and computes the max of the length returned by the
expression, and minLength to obtain the target length with which to
pad/truncate.
This needs to be passed the lengthKind so that if it is explicit the minLength
is not used.
Changing bug title to reflect status of this as a bug.
> Verify unparser specified length padding behavior - whether minLength is used.
> ------------------------------------------------------------------------------
>
> Key: DAFFODIL-1584
> URL: https://issues.apache.org/jira/browse/DAFFODIL-1584
> Project: Daffodil
> Issue Type: New Feature
> Components: Back End
> Reporter: Michael Beckerle
> Assignee: Michael Beckerle
> Priority: Major
> Fix For: 2.4.0
>
>
> Tests like unparseDelimitedPaddedString09 and unparseDelimitedPaddedString09
> (note: poorly named because these tests aren't about delimited anything, or
> the delimters are beside the point)
> These tests have been updated to expect the length to be the maximum of the
> length property value (literal or evaluated expression), and the minLength
> facet, for strings, and dfdl:textOutputMinLength for other simple types.
> However, this is under discussion by the DFDL workgroup, so may change.
> Erratum 5.18 is the one that is subject to clarification of this behavior.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
