You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@daffodil.apache.org by "Michael Beckerle (JIRA)" <ji...@apache.org> on 2018/02/05 16:57:00 UTC

[jira] [Commented] (DAFFODIL-1884) Regression in bitOrder changing

    [ https://issues.apache.org/jira/browse/DAFFODIL-1884?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16352640#comment-16352640 ] 

Michael Beckerle commented on DAFFODIL-1884:
--------------------------------------------

It appears that priorBitOrder, which was put into parsing, to address bitorder changing issues, was never implemented in unparsing.

There's also a fragmentBitOrder var and setter in data output streams which seems like duplicate mechanism to priorBitOrder which is in data stream common state so part of both parsing and unparsing. 

fragment bit order should go away, and priorBitOrder used where appropriate in unparser code instead of finfo.bitOrder being used everywhere.



> Regression in bitOrder changing
> -------------------------------
>
>                 Key: DAFFODIL-1884
>                 URL: https://issues.apache.org/jira/browse/DAFFODIL-1884
>             Project: Daffodil
>          Issue Type: Bug
>            Reporter: Steve Lawrence
>            Assignee: Michael Beckerle
>            Priority: Major
>             Fix For: 2.1.0
>
>
> test_bitOrderOVC1 in ./daffodil-test/src/test/scala-new/edu/illinois/ncsa/daffodil/section00/general/TestUnparserGeneral2.scala currently fails. It appears to be a regression with recent changes to how bitOrder works, but was not discovered because the test is in scala-new. This test should be fixed and moved into the scala directory since the scala-new directory is no longer used and ignored.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)