Proposed change of XML-Syntax and DTD for ECC transitions in 2nd Edition
Ingo
A major change in the ECC-transition is expected in the 2nd editon of
the IEC 61499 standard and already implemented in FBDK.
This brings up some compliance-problems between tools supporting
different editions, which can only be fixed by parsing the condition
represented as a string in the ECTransition-Element, following
different semantics in the two versions.
A possible solution could be to split the "condition"-attribute into 2
separate attributes for the "ec_transition_event" and the
"guard_condition"-part of the transition
("ec_transition_event[guard_condition]" in the new edition), e.g.,
"ec_transition_event"-attribute and "guard_condition"attribute.
Thus a parallel support for FBs of the 1st edition and the 2nd edition
would be possible - references to different DTDs - and ease the
transition. Furthermore a more explicit separation in the DTD could
ease/replace semantic checks for the transition condition.
What is your opinion on that proposal?
Best regards
Ingo
William Dai
Recently I've done a tool which can transform between nxtStudio and
FBDK solution based on DTD based ontology model. This approach is
flexible can also apply to the old and new FBDK. One solution is that
I can provide this small utility and help transform from old FBDK to
new FBDK files. If you need this software, video demo or relevant
publication, please let me know.
Best Regards,
William Dai
> --
> You received this message because you are subscribed to the Google Groups "FBDK" group.
> To post to this group, send email to fb...@googlegroups.com.
> To unsubscribe from this group, send email to fbdk+uns...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/fbdk?hl=en.
>
>
--
Best Regards,
William Dai
Ingo
The main focus of my proposal is to make the semantics provided in the
standard as BNF more explicit in the XML-representation. The parsing
would/could be simplified in tools relying on the DTD.
The existence of two different DTDs, which are referenced within the
FBs dependent on the edition of the standard, could help to identify
the edition. Thus different treatment (in both directions) would be
possible, comparable to document-converters for office.
This would only be a side-effect for the transition-phase, which might
take a while.
best regards
Ingo
James Christensen
the syntax of the ECTransition.Condition attribute to be as defined in
the ec_transition_condition production of 61499-1 Annex B, which in
61499-1 2nd Edition will be
'1' | ec_transition_event | '[' guard_condition ']' |
ec_transition_event '[' guard_condition ']'
(References to other Parts of a Standard are understood to refer to
the most recent Edition.)
The solution for software tool migration is to recognize both "old"
and "new" syntaxes, but save only in the "new" syntax.
--
Best regards,
Jim Christensen
PS. There will probably be some changes in the LibraryElement DTD to
allow representation of ST, "Other" and LD Rung text as CDATA
sections, but this will be compatible with documents in the old
(Attribute) format, at least until 3rd(?!) Edition. A trial version of
the new DTD is at http://www.holobloc.com/xml/LibraryElement.dtd.