CDR Tickets

BZDATETIME::2011-05-06 11:45:17
BZCOMMENTOR::Bob Kline
BZCOMMENT::1

Assigned to Alan, as #4755 was his issue.

BZDATETIME::2011-05-10 11:50:17
BZCOMMENTOR::Alan Meyer
BZCOMMENT::2

The schema for InScopeProtocols says that the cdr:ref is optional
within OrgSiteID.

Checking back through the version control repository, it looks
like it's always been that way. It appears that the OrgSiteID
element type was created in 2002, and it was optional then.

If we want it to be required I think the only thing we need to do
is make it required in the schema. The element will then fail
schema validation if it has no cdr:ref and fail link validation
if it links to the wrong kind of object. However, there are
places where it should not be required. See:

http://verdi.nci.nih.gov/tracker/show_bug.cgi?id=4755#c5

If we can come up with a clear algorithm for when it is illegal
to not have a cdr:ref, we can probably add some sort of
validation for it.

Does anyone know what the algorithm should be? Or is it the case
that this needs human judgment?

BZDATETIME::2011-05-10 16:03:11
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::3

I thought one area that could pose a problem is the external sites updates from RSS and COG but when I checked both services use a different element <ExternalSiteID> to link to the organization document. Another area I checked is the new trials that are imported from the submission portal (which are created automatically upon import). But even in that case, it should be fine to have the document fail validation. Unlike the CTGovProtocol document type, I don't know of any cases where it will be illegal to not have the attribute required in the InScopeProtocol schema. Kim, do you know of any?

BZDATETIME::2011-05-11 08:26:12
BZCOMMENTOR::Kim Eckley
BZCOMMENT::4

I can't think of a reason why the OrgSite element shouldn't require the link. As William noted, the other places we link orgs have different element names.
Is OrgSite a common element that appears in any other schema? If not, I would vote for making the document fail validation if there is no link.

BZDATETIME::2011-05-12 16:37:26
BZCOMMENTOR::Alan Meyer
BZCOMMENT::5

As discussed above and decided at our CDR status meeting, I have
modified the schema.

I was requested to do this on Franck but, for a start, I've done
it on Mahler and am running a schema only re-validation of
InScopeProtocols. I'm not updating the database, just getting
a report.

If I don't see any big surprises, I'll do it on Franck today.
If I do see surprises, I'll post the information here and hold
off doing it on Franck.

BZDATETIME::2011-05-13 13:25:34
BZCOMMENTOR::Alan Meyer
BZCOMMENT::6

(In reply to comment #5)
...
> If I don't see any big surprises, I'll do it on Franck today.
> If I do see surprises, I'll post the information here and hold
> off doing it on Franck.

I installed the modified schema on Mahler and ran a revalidation
of all InScopeProtocols. There were some errors discovered but
not an inordinate number, so I installed the schema and ran a
revalidation on Franck. The revalidation only checked schema
errors (that made it faster and it was only the schema that
changed) and I excluded blocked documents.

There were only seven errors. They were:

InScopeProtocol: 78634
InScopeProtocol: 269542
InScopeProtocol: 271306
InScopeProtocol: 271307
InScopeProtocol: 447141
InScopeProtocol: 463938
InScopeProtocol: 517475

In every case, there were many other schema errors in the documents.
I presume they are mostly or all older, no longer maintained protocols.

BZDATETIME::2011-05-13 16:44:06
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::7

I verified this on Mahler and rightly got the error message. Let's do another test on Franck.

BZDATETIME::2011-05-13 18:03:48
BZCOMMENTOR::Alan Meyer
BZCOMMENT::8

(In reply to comment #7)
> I verified this on Mahler and rightly got the error message. Let's do another
> test on Franck.

I guess my last posting was ambiguous. The results posted in
comment #6 were from Franck.

BZDATETIME::2011-05-16 09:40:49
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::9

(In reply to comment #8)
> (In reply to comment #7)
> > I verified this on Mahler and rightly got the error message. Let's do another
> > test on Franck.
>
> I guess my last posting was ambiguous. The results posted in
> comment #6 were from Franck.

I just checked the errors on Franck. They are Duplicate, Withdrawn, Out of Scope or trials that were not completely processed.

Please install the changes on Mahler and Franck.

BZDATETIME::2011-05-16 10:48:01
BZCOMMENTOR::Alan Meyer
BZCOMMENT::10

(In reply to comment #9)
...
> Please install the changes on Mahler and Franck.

The changes are already installed on Mahler and Franck. In order to run the revalidation it was necessary to have the modified schema in the database as the current schema since that's what the server programs use to do their validation.

The only thing left to do is to install it on Bach.

BZDATETIME::2011-05-16 10:54:14
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::11

(In reply to comment #10)
> (In reply to comment #9)
> The only thing left to do is to install it on Bach.

OK. Got it now :-). Please proceed to install on Bach.

BZDATETIME::2011-05-16 12:49:27
BZCOMMENTOR::Alan Meyer
BZCOMMENT::12

(In reply to comment #11)
> (In reply to comment #10)
> > (In reply to comment #9)
> > The only thing left to do is to install it on Bach.
>
> OK. Got it now :-). Please proceed to install on Bach.

After I install the schema, a new DTD will be produced to be downloaded to XMetal. I don't think this would cause anyone the slightest problem since the changes are not likely to affect actual editing of actual documents. However, just to be doubly safe, I plan to wait until after 5pm tonight to install the modified schema.

BZDATETIME::2011-05-16 18:26:11
BZCOMMENTOR::Alan Meyer
BZCOMMENT::13

The revised InScopeProtocol schema is now installed on Bach.

I'm marking the bug resolved-fixed.

BZDATETIME::2011-05-17 16:07:56
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::14

It looks like the OrgSite element has a required attribute called "CoopMember". Since the attribute is required, it is making some, if not all, of the sites come up with validation errors = For example 682791. We may have to rollback this change because I am not sure if this attribute is used elsewhere in other programs.

BZDATETIME::2011-05-17 16:30:02
BZCOMMENTOR::Alan Meyer
BZCOMMENT::15

It appears that I somehow edited the wrong line and never
noticed it.

I will fix the problem and install on Mahler right away.

BZDATETIME::2011-05-17 16:36:47
BZCOMMENTOR::Alan Meyer
BZCOMMENT::16

I think I've fixed this on Mahler.

I'm testing now.

BZDATETIME::2011-05-17 16:47:01
BZCOMMENTOR::Alan Meyer
BZCOMMENT::17

I think it's working on Mahler. The OrgSiteID is required,
CoopMember is optional.

Please check. I can fix this on Bach quickly if it's now right.

Sorry for the brain spasm. Sometimes I mess up the simple parts
when I'm focusing on more complex issues.

BZDATETIME::2011-05-17 17:03:48
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::18

(In reply to comment #17)
> I think it's working on Mahler. The OrgSiteID is required,
> CoopMember is optional.
>
> Please check. I can fix this on Bach quickly if it's now right.
>
Yes. It is OK on Mahler. Please promote to Bach.

BZDATETIME::2011-05-17 17:33:14
BZCOMMENTOR::Alan Meyer
BZCOMMENT::19

(In reply to comment #18)

> Yes. It is OK on Mahler. Please promote to Bach.

Done. I'll put it on Franck also.

BZDATETIME::2011-05-17 17:57:34
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::20

(In reply to comment #19)
> (In reply to comment #18)
>
> > Yes. It is OK on Mahler. Please promote to Bach.
>
> Done. I'll put it on Franck also.

Verified on Bach. There are no longer validation errors. Thank you!

BZDATETIME::2011-05-17 18:05:07
BZCOMMENTOR::Alan Meyer
BZCOMMENT::21

Franck is now also updated and the revised schema is under version control.

I'm setting the status back to resolved-fixed. Hopefully it really is so.

BZDATETIME::2011-05-18 13:10:40
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::22

(In reply to comment #21)
> Franck is now also updated and the revised schema is under version control.
>
> I'm setting the status back to resolved-fixed. Hopefully it really is so.

Everything seems to be working fine. I am closing this issue. Thank you!