CDR Tickets

Issue Number 3283
Summary [Glossary] Modification to Publication Filters
Created 2010-12-16 14:28:45
Issue Type Improvement
Submitted By Juthe, Robin (NIH/NCI) [E]
Assigned To Englisch, Volker (NIH/NCI) [C]
Status Closed
Resolved 2011-02-09 14:37:07
Resolution Fixed
Path /home/bkline/backups/jira/ocecdr/issue.107611
Description

BZISSUE::4973
BZDATETIME::2010-12-16 14:28:45
BZCREATOR::Robin Juthe
BZASSIGNEE::Volker Englisch
BZQACONTACT::Margaret Beckwith

As discussed in today's status meeting, a case came up recently whereby a pending patient definition was drafted for a term that already had an approved HP definition. Because the document was set to publish for the approved term and definition, there was no way to prevent the patient definition from being published since the publication is dependent on the term name status in the term name document and not the definition status in the GTC. The interim solution was to remove the patient definition from the GTC until it is approved.

The long-term solution discussed in our meeting was to modify the publication filter to also look at the definition status in the GTC in order to publish only those definitions with an Approved or Revision Pending definition.

Comment entered 2010-12-21 15:55:30 by Englisch, Volker (NIH/NCI) [C]

BZDATETIME::2010-12-21 15:55:30
BZCOMMENTOR::Volker Englisch
BZCOMMENT::1

There exist four definition status values for the concept status:
Approved
New pending
Rejected
Revision pending

Should the new filter only pick up a definition status that's 'Approved'?
Are we making this change for Spanish definitions only or are we going to check the definition status for the English concept as well?

Comment entered 2010-12-22 12:57:07 by Englisch, Volker (NIH/NCI) [C]

BZDATETIME::2010-12-22 12:57:07
BZCOMMENTOR::Volker Englisch
BZCOMMENT::2

(In reply to comment #1)
> Should the new filter only pick up a definition status that's 'Approved'?
> Are we making this change for Spanish definitions only

In my mind I've answered both of these questions with 'Yes'. I am only picking up Spanish terms with the TranslatedStatus of 'Approved' and I am only checking the Spanish definition.

This is ready for review on MAHLER.

Comment entered 2011-01-05 12:19:33 by Osei-Poku, William (NIH/NCI) [C]

BZDATETIME::2011-01-05 12:19:33
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::3

(In reply to comment #2)
> (In reply to comment #1)
> > Should the new filter only pick up a definition status that's 'Approved'?
> > Are we making this change for Spanish definitions only
> In my mind I've answered both of these questions with 'Yes'. I am only picking
> up Spanish terms with the TranslatedStatus of 'Approved' and I am only checking
> the Spanish definition.
> This is ready for review on MAHLER.

I have tested this on Mahler using pub preview and comparing results with pub. preview on Franck and it appears to be working fine. But I think this testing is probably not enough. You will probably run a diff report before promoting to Bach but do you want us make some changes on either Mahler or Franck and then you can do a test publishing?

Comment entered 2011-01-05 14:49:34 by Englisch, Volker (NIH/NCI) [C]

BZDATETIME::2011-01-05 14:49:34
BZCOMMENTOR::Volker Englisch
BZCOMMENT::4

(In reply to comment #3)
> You will probably run a diff report before promoting to Bach

Absolutely, publishing filter changes always trigger a diff report.

If you want to update a few specific documents on FRANCK that would trigger the problem that would not be a bad idea. Otherwise, all that a diff report would be able to determine is that nothing else broke (depending on the type of changes, of course).

Let me know after maybe two or three documents have been updated and then I'll run the diff reports.

Comment entered 2011-01-05 15:39:05 by Osei-Poku, William (NIH/NCI) [C]

BZDATETIME::2011-01-05 15:39:05
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::5

(In reply to comment #1)
> There exist four definition status values for the concept status:
> Approved
> New pending
> Rejected
> Revision pending
> Should the new filter only pick up a definition status that's 'Approved'?
> Are we making this change for Spanish definitions only or are we going to check
> the definition status for the English concept as well?

Could you clarify your second question about Spanish definitions? It appears to me that the original request is to modify the software to consider the English definition status in cases where we have both HP and Patient definitions.

Comment entered 2011-01-05 16:05:20 by Englisch, Volker (NIH/NCI) [C]

BZDATETIME::2011-01-05 16:05:20
BZCOMMENTOR::Volker Englisch
BZCOMMENT::6

(In reply to comment #5)
> It appears to
> me that the original request is to modify the software to consider the English
> definition status in cases where we have both HP and Patient definitions.

I believe that the intended filter change was to consider the TranslationStatus for the Spanish patient definition. This is what I have implemented. If a TranslatedStatus is not 'Approved' the patient definition will not be picked up.

It is, however, possible to have an approved patient definition and a pending HP definition. The HP definition would be picked up as usual but I'm not sure if such a situation could occur in practice.

Did I get this wrong?

Comment entered 2011-01-05 16:43:57 by Osei-Poku, William (NIH/NCI) [C]

BZDATETIME::2011-01-05 16:43:57
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::7

(In reply to comment #6)
> (In reply to comment #5)
> > It appears to
> > me that the original request is to modify the software to consider the English
> > definition status in cases where we have both HP and Patient definitions.
> I believe that the intended filter change was to consider the TranslationStatus
> for the Spanish patient definition. This is what I have implemented. If a
> TranslatedStatus is not 'Approved' the patient definition will not be picked
> up.

I believe the way you have applied the above logic, should also be applied to the English - Patient Definition Status as well since that was the cause of the initial problem.

Also, in terms of your first question in comment #1, you should include Revision Pending. That is, the filter should also pickup definition status of both Approved and Revision Pending.

> It is, however, possible to have an approved patient definition and a pending
> HP definition. The HP definition would be picked up as usual but I'm not sure
> if such a situation could occur in practice.

From what I understand, the above scenario is unlikely to occur since the HP definitions are usually entered with an approved status.

Comment entered 2011-01-13 11:50:30 by Beckwith, Margaret (NIH/NCI) [E]

BZDATETIME::2011-01-13 11:50:30
BZCOMMENTOR::Margaret Beckwith
BZCOMMENT::8

I am sure that Robin will weigh in here also, but I just want to say that it is quite possible that there could be cases in the future where the HP definition is pending. We are hoping to be able to publish the Genetics Dictionary directly from the CDR data, so I am assuming that definitions could be entered that would then need to be approved by the Board, but maybe not.

Comment entered 2011-01-20 09:00:58 by Juthe, Robin (NIH/NCI) [E]

BZDATETIME::2011-01-20 09:00:58
BZCOMMENTOR::Robin Juthe
BZCOMMENT::9

We discussed this in the last status meeting, but I think that although all of the genetics HP terms are entered as approved for the time being, this could certainly change when we eventually publish the dictionary terms straight from the CDR. So a concept document that has an approved patient definition and a pending HP definition (or vice versa) should not publish both to Cancer.gov. Each TermDefinition block to be published should have a status of Approved or Revision Pending.

Comment entered 2011-01-25 18:33:45 by Englisch, Volker (NIH/NCI) [C]

BZDATETIME::2011-01-25 18:33:45
BZCOMMENTOR::Volker Englisch
BZCOMMENT::10

I've modified the filter so that a TermDefinition or a TranslatedTermDefinition will only be displayed if the DefinitionStatus or TranslatedStatus are set to 'Approved' or 'Revision pending'.
Please note that it is now possible for the filter to fail processing a document in the case when all DefinitionStatus values have been set to either 'Rejected' or 'New pending'.

The following filter has been modified:
CDR616048 - Vendor Filter: GlossaryTermName
(I've tested with document CDR46288)

This is ready for review on MAHLER.

Comment entered 2011-01-26 12:34:42 by Osei-Poku, William (NIH/NCI) [C]

BZDATETIME::2011-01-26 12:34:42
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::11

>
> This is ready for review on MAHLER.

Tested this on Mahler and everything seems to be working fine. We will also like to do more tests on Franck.

Comment entered 2011-01-26 15:58:53 by Englisch, Volker (NIH/NCI) [C]

BZDATETIME::2011-01-26 15:58:53
BZCOMMENTOR::Volker Englisch
BZCOMMENT::12

The following filter has been copied to FRANCK:
CDR616048: Vendor Filter: GlossaryTermName - R10009

I ran a diff report on FRANCK and there were no difference between the publishing jobs before and after the filter migration except that the document CDR339331.xml failed to publish with the new filter. Everything else was identical.

William, please let me know if you'd like to edit a few documents and have me rerun the publishing job on FRANCK again.

Comment entered 2011-01-27 11:00:12 by Osei-Poku, William (NIH/NCI) [C]

BZDATETIME::2011-01-27 11:00:12
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::13

>
> William, please let me know if you'd like to edit a few documents and have me
> rerun the publishing job on FRANCK again.

Yes. We will get some documents ready for you shortly.

Comment entered 2011-01-27 11:24:21 by Englisch, Volker (NIH/NCI) [C]

BZDATETIME::2011-01-27 11:24:21
BZCOMMENTOR::Volker Englisch
BZCOMMENT::14

(In reply to comment #12)
> except that the document CDR339331.xml failed to publish with the new filter.

I wanted to make sure you realize that this document will need to be modified on BACH or it'll fail publication as soon as the new filter has been put in place.
The problem is that the TermName's concept doesn't have any status listed.

Comment entered 2011-01-28 10:08:43 by Osei-Poku, William (NIH/NCI) [C]

BZDATETIME::2011-01-28 10:08:43
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::15

(In reply to comment #14)
> (In reply to comment #12)
> > except that the document CDR339331.xml failed to publish with the new filter.
>
> I wanted to make sure you realize that this document will need to be modified
> on BACH or it'll fail publication as soon as the new filter has been put in
> place.
> The problem is that the TermName's concept doesn't have any status listed.
Thanks! Yes this document has been updated on Franck and Bach.

(In reply to comment #13)
> >
> > William, please let me know if you'd like to edit a few documents and have me
> > rerun the publishing job on FRANCK again.
>
> Yes. We will get some documents ready for you shortly.

Please proceed to run the publishing job on Franck. The following documents (concepts) have been updated.

CONCEPT TERM NAME
CDR0000623412 CDR0000460130
CDR0000622731 CDR0000339342
CDR0000671176 CDR0000671177

Comment entered 2011-01-28 11:05:51 by Englisch, Volker (NIH/NCI) [C]

BZDATETIME::2011-01-28 11:05:51
BZCOMMENTOR::Volker Englisch
BZCOMMENT::16

(In reply to comment #15)
> Please proceed to run the publishing job on Franck. The following documents
> (concepts) have been updated.
>
> CONCEPT TERM NAME
> CDR0000623412 CDR0000460130
> CDR0000622731 CDR0000339342
> CDR0000671176 CDR0000671177

There were three differences found (I will attach the diff report shortly). Differences were found for the documents
CDR339331
CDR339342
CDR671177

but not for the document listed about
CDR0000460130

Comment entered 2011-01-28 11:06:47 by Englisch, Volker (NIH/NCI) [C]

BZDATETIME::2011-01-28 11:06:47
BZCOMMENTOR::Volker Englisch
BZCOMMENT::17

Comment entered 2011-01-28 11:06:47 by Englisch, Volker (NIH/NCI) [C]

Attachment GlossaryDiff.txt has been added with description: Diff Report for Glossary Filter

Comment entered 2011-01-28 12:24:47 by Osei-Poku, William (NIH/NCI) [C]

BZDATETIME::2011-01-28 12:24:47
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::18

(In reply to comment #16)
> There were three differences found (I will attach the diff report shortly).
> Differences were found for the documents
> CDR339331
> CDR339342
> CDR671177
>
> but not for the document listed about
> CDR0000460130

We intentionally created different scenarios for the three documents for the test. The reason for no difference for CDR0000460130 is because a new HP definition was added with an assigned dictionary of Genetics and I believe Genetics don’t get picked up for publishing.
Can the changes from the other three be previewed on the test site?

Comment entered 2011-01-28 12:39:10 by Englisch, Volker (NIH/NCI) [C]

BZDATETIME::2011-01-28 12:39:10
BZCOMMENTOR::Volker Englisch
BZCOMMENT::19

(In reply to comment #18)
> Can the changes from the other three be previewed on the test site?

Not with the current publishing job. This job was run in order to create the diff report (which is attached to this issue for your review). If you need to have the glossaries loaded to the test site I'll have to run another publishing job.
I thought that previewing the diff output would be sufficient for three documents but I certainly can rerun and load another publishing event or I could attach the before and after XML files as another way to compare the documents.

Let me know which you would prefer.

Comment entered 2011-01-28 13:33:49 by Osei-Poku, William (NIH/NCI) [C]

BZDATETIME::2011-01-28 13:33:49
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::20

Please run the publishing job again. This time also load to the test site. We will test with the following 'updated' documents (contain duplicates).

TERM NAME
CDR0000425595
CDR0000425594
CDR0000330204
CDR0000330203
CDR0000330202
CDR0000330201
CDR0000330180
CDR0000636400
CDR0000560143
CDR0000560142
CDR0000560141
CDR0000560140
CDR0000346515
CDR0000045140
CDR0000460130
CDR0000339342
CDR0000671177
CDR0000460130

Comment entered 2011-01-28 16:10:33 by Englisch, Volker (NIH/NCI) [C]

BZDATETIME::2011-01-28 16:10:33
BZCOMMENTOR::Volker Englisch
BZCOMMENT::21

The documents have been loaded to our test preview site.
previewGK.cancer.gov

Comment entered 2011-01-31 12:09:51 by Osei-Poku, William (NIH/NCI) [C]

BZDATETIME::2011-01-31 12:09:51
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::22

(In reply to comment #21)
> The documents have been loaded to our test preview site.
> previewGK.cancer.gov

I have reviewed the documents and didn't find any problems. The program appears to be doing the right thing. Please promote to Bach.

Comment entered 2011-02-02 17:05:21 by Englisch, Volker (NIH/NCI) [C]

BZDATETIME::2011-02-02 17:05:21
BZCOMMENTOR::Volker Englisch
BZCOMMENT::23

The following filter has been copied to FRANCK and BACH:
CDR616048: Vendor Filter: GlossaryTermName - R10009

I will monitor the publishing output and close the issue if everything looks OK after Friday's publishing event.

Comment entered 2011-02-09 14:37:07 by Englisch, Volker (NIH/NCI) [C]

BZDATETIME::2011-02-09 14:37:07
BZCOMMENTOR::Volker Englisch
BZCOMMENT::24

No publishing problems have been identified since last Wednesday.

Closing issue.

Attachments
File Name Posted User
GlossaryDiff.txt 2011-01-28 11:06:47 Englisch, Volker (NIH/NCI) [C]

Elapsed: 0:00:00.002146