Thanks to Sam Habiel, and Bhaskar (YottaDB), and some VA employees, Vivian has had some major fixes!

35 views
Skip to first unread message

Nancy Anthracite

unread,
Dec 31, 2021, 8:52:35 PM12/31/21
to hard...@googlegroups.com

Vivian is found here: https://vivian.worldvista.org

 

As most of you know, OSEHRA closed in mid Feb 2020 and transferred most of its web content and training assets to WorldVistA. Vivian is an important part of what was transferred. Vivian is generated by a truly remarkable but very complicated script that takes about 8 hours to run. With time, there have been issues with generating Vivian as the FOIA changed and other issues came up. Skip Ormsby helped apply a bandaid to one of the issues while we were waiting for a permanent fix. Thankfully, Sam Habiel has managed to fix most of the issues. His work was very generously funded by Bhaskar of YottaDB. In addition, the ability to generate the ICRs or Integration Control Agreements has been fixed thanks to VA employees working hard to generate the text document about the agreements to allow it to work with the existing parser used in the Vivian generation script.

 

The two major issues that were fixed by Sam were as follows:

 

- Somewhere early this year (which is just ending), the script that

installs our version of XINDEX that produces a lot more dependency tracking

information broke. It broke because FOIA (correctly) disabled the option to

restart taskman, which our script used. This is now fixed by starting

taskman using the public entry point ^ZTMB, and now we are back to using

our version of XINDEX. More information can be found here:

https://github.com/WorldVistA/XINDEX.

 

- DI*22.2*19 changed the output of the Data Dictionary (DD) Lister in ways

that Vivian couldn't cope with. The first error that we fixed is that this

patch prints subfile DD even if the parent does not point to it anymore

(issue with the domain file); Vivian correctly complained that this subfile

was never declared. I suppressed the output if the subfile is not declared

by the parent file. The second error was that a long field name will cause

the output to bunch up together (in this case "0;13VARIABLE POINTER). The

best I could do with this is to emit as much information as was salvageable.

 

A lot more detail can be found here:

-

https://github.com/WorldVistA/VistA/commit/55829255e25978bf9110695a0bdc575c7172c47a

-

https://github.com/WorldVistA/VistA/commit/b1799f41132bd2d31e61a94ff8c0170664747c67

 

Thanks very much to everyone who made these changes possible and kept things going in the meantime!

 

--

Nancy Anthracite

 

Reply all
Reply to author
Forward
0 new messages