Has anybody successfully built and run CPRS using Delphi 10 Lite

[Raja Anumele]

I have managed to enter Patient name in Mandarin and Japanese by modifying the
M routine XLFMAME7.m. This has the FORMAT routine. I did not try Russian because
I do not know Russian Language yet.

To select the patient whose name is in Chinese Simplified I need to display the
name in Chinese. I got the source for CPRS from VA ftp site but am not successful
in building the CPRS Using Delph 10 Lite.

CPRS as implemented in Delphi is good usable product.
I did build the OpenVistaCIS from medsphere but I am having trouble connecting
to the WorldVistA server. Initial screen is full of text and my laptop display is only
1920x1200 pixels. I cannot see the bottom of the screen. I can probably remove
the text that is displayed but CPRS works very well and would like to build CPRS
using Delphi 10 Lite.

Regards
Raja

[Ben Mehling]

On Sun, Jun 21, 2009 at 9:33 PM, Raja Anumele <adr...@gmail.com> wrote:

I did build the OpenVistaCIS from medsphere but I am having trouble connecting
to the WorldVistA server. Initial screen is full of text and my laptop display is only
1920x1200 pixels. I cannot see the bottom of the screen. 

I do not think anyone has ported the extensions necessary for CIS from OpenVista Server to WorldVistA EHR. 

At the VistA Community Meeting we showed several screenshots with CIS running in French, Russian, Arabic and Chinese.  The presentation that includes some screen captures is not yet posted, but I'm sure will be soon.

- Ben

[Jonathan Tai]

On Sun, 2009-06-21 at 21:33 -0700, Raja Anumele wrote:
> I did build the OpenVistaCIS from medsphere but I am having trouble
> connecting to the WorldVistA server. Initial screen is full of text
> and my laptop display is only 1920x1200 pixels. I cannot see the
> bottom of the screen.

This issue has been fixed in CIS mainline revision 1316. You can grab
the code from Launchpad using bzr.

https://code.launchpad.net/~fireball-medsphere/openvista-cis/mainline

As Ben mentioned, there are extensions that CIS requires that have not
been ported from OpenVista to WorldVistA, so I would expect CIS to throw
a missing RPC/routine exception as soon as you log in.

- Jon

[K.S. Bhaskar]

I don't see an XLFMAME7.m in WorldVistA EHR, OpenVista or FOIA VistA.
Which flavor & release of VistA are you using and what changes did you
need to make?

Regards
-- Bhaskar

[David Whitten]

I expect that Raja was thinking of the routine XLFNAME7

It is part of the Name Component subroutines, and was last changed in the VA by patch XU*8*343. By the way, this effort adds entries into the NAME COMPONENT File (#20) and allows names to be recognized as being made up of components rather than just being a simple string of characters:

Select OPTION: ENTER OR EDIT FILE ENTRIES 



INPUT TO WHAT FILE: NAME COMPONENTS//
EDIT WHICH FIELD: ALL// ?
 Answer with FIELD NUMBER, or LABEL
 Do you want the entire FIELD List? Y  (Yes)
   Choose from:
   .01          FILE
   .02          FIELD
   .03          IENS
   1            FAMILY (LAST) NAME
   2            GIVEN (FIRST) NAME
   3            MIDDLE NAME
   4            PREFIX
   5            SUFFIX
   6            DEGREE
   7            SOURCE NAME FORMAT FLAGS
   11           NOTES ABOUT NAME

Raja,
since you are thinking about this, I assume you loaded a version of MUMPS that can do international characters, probably using Unicode.  Could you tell us how you are able to recognize the various parts of a person's name in Japanese and Chinese?
Did you have to make any modifications to globals, what if any, storage concerns have you addressed. I would expect that you had to change things like collating sequences for the PATIENT File (#2) and the NEW PERSON File (#200) at minimum, so the names would sort properly for reports. Did you have to change lookup in FileMan to allow for partial name lookups, or is the way that George Timson currently does it in FileMan sufficient? How do the pattern matches for Chinese and Japanese characters work? Does everything match a "U" for upper case? or is there such a thing as lowercase characters?   How do  you handle that there are different widths of characters compared to ASCII characters? Does your MUMPS implementation already do that, or do you have to modify code to make sure that columns line up properly?

I think it interesting that you have taken on this task, and would appreciate a message sharing your design criteria and how you handle some of these issues. (Of course, more details are fine too)

Best Regards,
David

[Raja Anumele]

Dear David,
  You are correct. I did mean XLFNAME7.m. This is what checks the name is made
  of standard last name, first name separated by commas and converts to upper
  case English letters. To make VistA take Chinese names I just added the
  statement Q XUNAME as the first executable statement for the FORMAT entry
  point. This allows names to be preserved. The names are all encoded as
  UTF-8 which is a variable length string. I  am using GT.M that works wonderfully
  well wtih UTF-8. Even the length function works well with GT.M.
  The UTF-8 characters are preserved well in the globals. If you use D ^%G
  to print the DPAT we can see the name correctly in chinese.
  The name lookup appears to work correctly too. I had added only two names.
  On chinese name zhang, hao tian and one japanese yama, fuji.
  张,好天 and 山,富士.

  I hope you can read these in Simplified Chinese and Japanese.
 
  The motto of WorldVistA is to propogate VistA to the entire world and that means
   in all the world languages. GT.M is wonderful for this. Thanks to Bhaskar
   பாஷ்கர். I found out how to setup GT.M to handle UTF-8.

  I am trying to get the GUI part selected before listing things to do.

  Now VistA is all console commands. What people need now is GUI.
  This means Console Commands have to move to GUI, preferably web based.
  When Prophet Mohammed did not go to the mountain the moutain went to
  Prophet Mohammed, as the saying goes.
  If VistA Console Commands cannot go to GUI then the GUI must go to the
  Console Commands. I am working on making this possible using scripting.
 
  Regards
  Raja

[Gregory Woodhouse]

Could it be XLFNAME?

Sent from my iPhone