[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: [MoM] Packaging fis-get





On 01/30/2012 08:20 AM, Karsten Hilbert wrote:
On Sun, Jan 29, 2012 at 06:14:29PM -0500, Bhaskar, K.S wrote:

So, when I install fis-gtm 5.3-017g today and next week
Debian delivers a package containing the code for 5.5-000
will there be a way for me to upgrade my database/M code ?
If so, then I fail to see the problem with Debian providing
the new GT.M version to me via a package ?  And later
providing me with a 5.6-002z package allowing me to run,
say,

	/sbin/fis-gtm-upgrade
[KSB]  Yes, that's the general idea.  But it's a little more
complicated than that because, for example, you may want to build new
shared libraries for the code, and occasionally other configuration
files may need to be tweaked or extended.
Absolutely. Just like with any other Debian package.

Let me try to derive the most basic question again so we can
perhaps find an answer:

- VistA runs on Mumps.

- GT.M provides a Mumps environment

[KSB2] The last two statements are correct.

- GT.M is written in Mumps and thusly needs to be compiled
   by a Mumps compiler

- this presents a chicken-egg problem in that Debian does
   not allow packages to be included without compiling them
   from source

[KSB2] GT.M is almost entirely written in C with a few small bits in assembler. Some utility programs are written in MUMPS. When building GT.M, some C source programs are generated from text files using a MUMPS program - this is where GT.M is used to build GT.M, and it is a text to text transformation.


- the easy way out would be if

1) there existed a Mumps compiler written in C/C++/Pascal/...
2) which is compatible with the DFSG
3) which can - however deficiently - compile GT.M

So, is there ?

[KSB2] There are certainly other MUMPS implementations. I don't know how well those map DFSG guidelines and I don't know how well they might work to bootstrap GT.M. I can't use any of them because we have a policy of not touching any other MUMPS implementation so that all GT.M development is done in a clean room. But someone else who is interested certainly can look at them.

However, I don't really understand the problem with bootstrapping. It only needs to be done once and that bootstrap has already been accomplished. So, we can now work on creating the real GT.M package.

In my opinion, if you really want to eliminate the bootstrap, the obvious solution is to use awk or perl for the small number of files involved. Or, for the initial bootstrap, just take the generated C files from an existing GT.M for the bootstrap.

Regards
-- Bhaskar


Karsten

--
GT.M - Rock solid. Lightning fast. Secure. No compromises.

_____________
The information contained in this message is proprietary and/or confidential. If you are not the intended recipient, please: (i) delete the message and all copies; (ii) do not disclose, distribute or use the message in any manner; and (iii) notify the sender immediately. In addition, please be aware that any message addressed to our domain is subject to archiving and review by persons other than the intended recipient. Thank you.


Reply to: