|Requirements for Accessing VistA Dev Environment for M[UMPS]
|Page 1 of 1|
|Author:||Robp [ Tue Aug 16, 2011 2:37 pm ]|
|Post subject:||Requirements for Accessing VistA Dev Environment for M[UMPS]|
I work for the VA (NW HSR&D within Region 20) for the last year and a half - and have mostly used their operational data store (VISN20) for accessing data for various projects. I'm at a point where I'll need to retrieve information that currently is not in VISN20, or it is in a form that is unsuitable.
On my spare time I have done some research on the VistA file system - especially M[UMPS]. I have even downloaded a DOS MUMPS interpreter to get a feel of the software. Though I'm mostly a SQL programmer (along with many scripting languages), I'm finding the MUMPS language and the hierarchial database structure to be quite simple.
Though my initial request for accessing VistA data directly has been met with a response of "no", or "you're probably be limited to FileMan", I'm still convinced that using MUMPS directly on VistA Files - for accessing data only - is by far the best approach.
My question at this stage is to understand where MUMPS is run on? What software and configuration is needed, as well as what platform would this work be accomplished on? For example, my prior job, I was able to access many database systems on Unix via Java and Perl. Scripting extract jobs was always faster (and more straight forward) than the software those systems would provide.
I'm trying to identify the questions to ask in order to get this type of scenario rolling.
Any help is very much appreciated.
|Author:||jollis [ Sun Aug 21, 2011 9:07 pm ]|
|Post subject:||Re: Requirements for Accessing VistA Dev Environment for M[U|
As far as software, you'd need at least one of the supported MUMPS implementations. I'd recommend FIS GT.M for Linux, or InterSystems Cache for Windows.
It is indeed possible to access the VistA globals "directly" (the FileMan documentation covers the structure of the globals), although the DBS API is probably a far cleaner (and far less risky) approach. The DBS API is also well-documented.
As far as scripting "from the outside", MUMPS implementations tend to be rather insular, in that they provide little in the way of connecting to the outside world. There are third-party tools from FIS and M/Gateway that could possibly help you with this, but as you would end up banging the FM globals directly using these tools, you'd be unlikely to have much of a productivity boost over writing the translation tools directly in MUMPS using the DBS API. YMMV.
There are also FileMan tools which can export FM files to other external formats... this might be worth looking into as well.
|Page 1 of 1||All times are UTC - 8 hours [ DST ]|
|Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group