CDR Tickets

Issue Number 4293
Summary Display PerDoctypeErrorThreshold Job Options
Created 2017-07-28 12:47:03
Issue Type Improvement
Submitted By Englisch, Volker (NIH/NCI) [C]
Assigned To Kline, Bob (NIH/NCI) [C]
Status Closed
Resolved 2018-04-10 17:45:00
Resolution Fixed
Path /home/bkline/backups/jira/ocecdr/issue.212125
Description

We have no easy way to display and/or modify the new element PerDoctypeErrorThreshold option causing publishing failures on the lower tiers.

We need to be able to display or modify these options as part of the Job Options when submitting a publishing job on the lower tiers.

Comment entered 2018-03-13 17:01:55 by Kline, Bob (NIH/NCI) [C]

We have a number of ways to tackle this. We've only got one document type left standing (Summary) which uses a per-doctype error threshold. So we could move the information from the PerDoctypeErrorThreshold block into a SubsetParameter with ParmName of MaxSummaryErrors. If we ever had another document type which needed such treatment, then we'd have a parameter like MaxKumquatErrors. Or we could create a new table along the lines of

CREATE TABLE per_doctype_error_threshold_override
   (pub_proc INTEGER NOT NULL FOREIGN KEY ...,
    doc_type INTEGER NOT NULL FOREIGN KEY ...,
   threshold INTEGER NOT NULL)
PRIMARY KEY (pub_proc, doc_type)

(or maybe a string for the doc_type column), and we'd modify the publishing.Job class to accept more information in the constructors, and modify the cdr.publish() function, as well as the CGI web admin interface which calls that function.

What's your preference?

Comment entered 2018-03-13 17:18:47 by Englisch, Volker (NIH/NCI) [C]

It seems the second option is a little overkill. I was thinking about option 1 when I created the ticket. All I wanted was a way to see and/or update the value if necessary.

Is that vague enough? :-)
Which option do you think is easier to implement?

Comment entered 2018-03-13 17:25:25 by Kline, Bob (NIH/NCI) [C]

Which option do you think is easier to implement?

The first one.

Comment entered 2018-03-13 17:34:06 by Englisch, Volker (NIH/NCI) [C]

Then we agree: Option 1 is the winner.

I thought option 1 to be easier to implement but today you finished a gazillion tickets I can't even keep up with the emails for the tickets.

Comment entered 2018-03-13 17:42:12 by Kline, Bob (NIH/NCI) [C]

No worries. This one will slow me down. :-)

Comment entered 2018-03-14 11:27:13 by Kline, Bob (NIH/NCI) [C]

: You've got the Primary publishing control document checked out.

Comment entered 2018-03-14 11:31:46 by Englisch, Volker (NIH/NCI) [C]

On DEV I'm guessing?

Comment entered 2018-03-14 11:34:10 by Englisch, Volker (NIH/NCI) [C]

It's unlocked now.

Comment entered 2018-03-14 11:45:06 by Kline, Bob (NIH/NCI) [C]
Comment entered 2018-04-10 15:08:19 by Englisch, Volker (NIH/NCI) [C]

When I'm trying to start an Export publishing job I'm getting an error message:

invalid 'MaxSummaryErrors' value '0'

This is happening for both, DEV and QA.

Comment entered 2018-04-10 17:23:09 by Kline, Bob (NIH/NCI) [C]

I believe this is fixed on DEV. Please try again, and if it works, I'll check it in and copy the fixed publishing control document to QA.

Comment entered 2018-04-10 17:38:21 by Englisch, Volker (NIH/NCI) [C]

Yes, it works now on DEV.

Comment entered 2018-04-10 17:45:00 by Kline, Bob (NIH/NCI) [C]

Correct validation specification installed on QA.

https://github.com/NCIOCPL/cdr-publishing/commit/34b33af

Comment entered 2018-04-11 14:49:43 by Englisch, Volker (NIH/NCI) [C]

... and it works on QA.

Comment entered 2018-05-15 12:12:55 by Englisch, Volker (NIH/NCI) [C]

This option is now available on PROD. Closing ticket.

Elapsed: 0:00:00.001327