Issue (probably known) with ADL Workbench

Technical (archive)
1

Hi all,

  I'm quite new in openEHR, testing last released version of ADL
workbench I received an error (ERROR line 87 [last cADL token =
V_C_DOMAIN_TYPE]: (VACSIT) cannot add C_DV_QUANTITY object with
rm_type_name=DV_QUANTITY to singly-valued attribute value because
attribute already has child with same RM type) when I tried to load the
openEHR-HER-ITEM_TREE.medication.v1.adl (draft version). I don't know if
this is known or unknown but just in case.

  If I remove from line 73 to line 98 (C_DV_QUANTITY multiple
references, just leaving one of them) the archetype is perfectly loaded.
The same happened with
openEHR-EHR-ITEM_TREE.medication-formulation.v1.adl. In this case the
C_DV_QUANTITY values are empty and you need to include values on them.

  Thanks and regards.
  Carlos.

2

Hi Carlos

Welcome to openEHR. This is a known issue, not with the Workbench but
with the archetype itself. This is pretty old and breaks a validation
rule that was not enforced properly in the Archetype Editor in the
past. I have actually fixed or worked around most of the validation
errors that AWB reports but have not had time to commit the fixes to
CKM yet. As soon as I have a proper web connect I will do so at least
for that archetype.

The problem is a bit obscure for an openEHR newbie but is related to a
bit of ADL syntax that is not currently supported in the Archetype
Editor but which is required if multiple constraints of the same
datatype are applied to a single element. You should also be aware
that our intention is to replace the current CKM medication archetype
with others based on the NEHTA medication archetypes.

Ian

Dr Ian McNicoll
Clinical modelling consultant Ocean Informatics
Mobile +44 (0) 775 209 7859
Skype imcnicoll

3

Thanks Ian, I will wait then, I will also have a look at the
NEHTA archetypes. I will continue asking about how to include medication
properly in the clinical list.

  Regards.
  Carlos.

4

Hi Carlos - for easy reference:

· The NEHTA Instruction - http://dcm.nehta.org.au/ckm/OKM.html#showarchetype_1013.1.838

· The NEHTA Action - http://dcm.nehta.org.au/ckm/OKM.html#showarchetype_1013.1.876

Regards

Heather

5

Great! Thanks Heather! I see they are slightly different than CKM ones. I will have a look at them and let’s see.

Regards.

Carlos.

6

Hi Carlos,

I have just uploaded the corrected openEHR-EHR-ITEM_TREE.medication.v1 archetype to CKM.

Ian

7

Hi all,

I’ve created an archetype with a DvDateTime node with a partial time constraint using the Archetype Editor.
Opening the archetype with the ADL WB I get this error message:

------------- compiling openEHR-EHR-COMPOSITION.orden_de_estudio_de_laboratorio.v1 -------------
ERROR - line 84: invalid date/time constraint pattern; allowed patterns: YYYY-MM-DDTHH:MM:SS, YYYY-MM-DDTHH:MM:??, YYYY-??-??T??:??:??, YYYY-MM-DDTHH:??:XX, YYYY-MM-DDTHH:MM:XX [last cADL token = V_ISO8601_DATE_TIME_CONSTRAINT_PATTERN]
(Parse failed) (ARCHETYPE_PARSER.parse_archetype)

The constraint looks like this:

ELEMENT[at0009] occurrences matches {0..1} matches { – fecha esperada del resultado
value matches {
DV_DATE_TIME matches {
value matches {yyyy-mm-ddTHH:??:??}
}
}
}

8

I’m getting another error in a constraint_binding:

------------- compiling openEHR-EHR-INSTRUCTION.estudio_laboratorio.v1 -------------
ERROR - line 154: parse error [last dADL token = ‘:’]
line 154: parse error [last dADL token = ‘:’]Error in attribute value [last dADL token = ‘:’]
(Parse failed) (ARCHETYPE_PARSER.parse_archetype)

This is the binding created by the Archetype Editor when I select LOINC from the terminology list:

constraint_bindings = <
[“LOINC”] = <
items = <
[“ac0001”] = terminology:LOINC

This is the constrainted DV_CODED_TEXT:

ELEMENT[at0006] occurrences matches {0..1} matches { – tipo de estudio
value matches {
DV_CODED_TEXT matches {
defining_code matches {[ac0001]} – Estudios de laboratorio LOINC
}
}
}

9

Hi Pablo,

This is loading properly with my version of the openEHR Archetype editor 2.2.779 and the current Workbench. I am not getting errors from the Partial Date constraint either.

Are you using the openEHR or Ocean version of the archetype editor?

Ian

Dr Ian McNicoll
office +44 (0)1536 414 994
fax +44 (0)1536 516317
mobile +44 (0)775 209 7859
skype ianmcnicoll
ian.mcnicoll@oceaninformatics.com

Clinical Modelling Consultant, Ocean Informatics, UK
Director openEHR Foundation www.openehr.org/knowledge
Honorary Senior Research Associate, CHIME, UCL
SCIMP Working Group, NHS Scotland
BCS Primary Health Care www.phcsg.org

10

Hi Pablo,

As far as I can tell this is legal ADL. It is passing the validation on the most recent version of AWB.

From the spec..

• for partial date/times, any part of the date/time up to the month may be
missing, not just seconds and minutes as in the standard;

I am sure Thomas will be along later to tell me that I am wrong :slight_smile:

Ian

11

Hi Pablo,

This is loading properly with my version of the openEHR Archetype editor 2.2.779 and the current Workbench. I am not getting errors from the Partial Date constraint either.

Are you using the openEHR or Ocean version of the archetype editor?

Ian

12

The current ADL Workbench from http://www.openehr.org/svn/ref_impl_eiffel/TRUNK/apps/adl_workbench/doc/web/index.html accepts this too.

I think you must have a very old version of ADL Workbench, Pablo.

Peter

13

Ian McNicoll wrote:

Are you using the openEHR or Ocean version of the archetype editor?

That would make no difference, Ian. The same ADL is generated by openEHR and Ocean Archetype Editors.

I'm sure that the problem is that Pablo must be using a very old version of ADL Workbench.

Peter

14

Hi Ian, Peter was right, the issue was because I’ve installed ADLWB v1.4.1.595 in my notebook.

I downloaded the ADLWB from here: http://wiki.oceaninformatics.com/confluence/display/TTL/ADL+Workbench+Releases
And this page lead my to the old download page: http://wiki.oceaninformatics.com/confluence/display/TTL/ADL+Workbench

That last link is the first result when “adl workbench” is searched using google. Maybe that page could be updated to link the new ADLWB download page: http://www.openehr.org/svn/ref_impl_eiffel/TRUNK/apps/adl_workbench/doc/web/index.html

Thanks a lot.

15

pablo pazos wrote:

Hi Ian, Peter was right, the issue was because I've installed ADLWB v1.4.1.595 in my notebook.

I downloaded the ADLWB from here: http://wiki.oceaninformatics.com/confluence/display/TTL/ADL+Workbench+Releases
And this page lead my to the old download page: http://wiki.oceaninformatics.com/confluence/display/TTL/ADL+Workbench

That last link is the first result when "adl workbench" is searched using google. Maybe that page could be updated to link the new ADLWB download page: http://www.openehr.org/svn/ref_impl_eiffel/TRUNK/apps/adl_workbench/doc/web/index.html

Maybe google is doing that because the proper download page (on openehr.org) has a heading of "AWB Home" and a title of "ADL 1.5 Workbench". I guess google gets as confused by acronyms as I do :wink:

I've edited the ADL Workbench page that you found on the Ocean wiki. The "Current Release" now links to the openehr.org download page.

Thanks for pointing this out, Pablo.

Peter

16

I’m glad to help. I thought maybe other people is downloading old WB versions.