From: gavino
Subject: MUMPS in lisp?
Date: 
Message-ID: <f4ba6585-313d-4780-ac55-b4f7271023da@d4g2000prg.googlegroups.com>
http://en.wikipedia.org/wiki/MUMPS

I think MUMPS in lisp would rock and give lisp some db punch!

I know prolog was implemented in lisp and APL in scheme so how about a
database language?

From: Leslie P. Polzer
Subject: Re: MUMPS in lisp?
Date: 
Message-ID: <99788bbe-af04-4d73-a009-c8929ce91fd7@u10g2000prn.googlegroups.com>
On Jan 19, 1:52 am, gavino <·········@gmail.com> wrote:
> http://en.wikipedia.org/wiki/MUMPS
>
> I think MUMPS in lisp would rock and give lisp some db punch!

That's an interesting link, thanks.


> I know prolog was implemented in lisp and APL in scheme so how about a
> database language?

Great! When will you release it? :)

  Leslie
From: George Neuner
Subject: Re: MUMPS in lisp?
Date: 
Message-ID: <32m5p39s6dt2sbbsic4a0tthnlrjdcvrm4@4ax.com>
On Fri, 18 Jan 2008 16:52:05 -0800 (PST), gavino <·········@gmail.com>
wrote:

>http://en.wikipedia.org/wiki/MUMPS
>
>I think MUMPS in lisp would rock and give lisp some db punch!
>
>I know prolog was implemented in lisp and APL in scheme so how about a
>database language?

Having actually used MUMPS I would say something much different than
"it rocks".  I remember it as a write-only language ... simple to use
but very difficult to debug.  Lisp is a much better language.  What
value do you think MUMPS would bring?

Lisp already has implementations of and interfaces to virtually every
type of database that exists.  Though I can't find a hierarchical
database (as in MUMPS) searching just now, I'm sure that one or more
exist somewhere.

George
--
for email reply remove "/" from address
From: vanekl
Subject: Re: MUMPS in lisp?
Date: 
Message-ID: <fmun1j$1go$1@aioe.org>
George Neuner wrote:
> On Fri, 18 Jan 2008 16:52:05 -0800 (PST), gavino <·········@gmail.com>
> wrote:
...
> Lisp already has implementations of and interfaces to virtually every
> type of database that exists.  Though I can't find a hierarchical
> database (as in MUMPS) searching just now, I'm sure that one or more
> exist somewhere.
> 
> George
> --
> for email reply remove "/" from address

postgresql allows for hierarchical table structures, and lisp has an interface
(i think it's called pg, IIRC).
From: George Neuner
Subject: Re: MUMPS in lisp?
Date: 
Message-ID: <c1r5p39rmo2720n2aavuca0hr8v3t2mpud@4ax.com>
On Sun, 20 Jan 2008 05:45:03 +0000, vanekl <·····@acd.net> wrote:

>George Neuner wrote:
>> On Fri, 18 Jan 2008 16:52:05 -0800 (PST), gavino <·········@gmail.com>
>> wrote:
>...
>> Lisp already has implementations of and interfaces to virtually every
>> type of database that exists.  Though I can't find a hierarchical
>> database (as in MUMPS) searching just now, I'm sure that one or more
>> exist somewhere.
>> 
>> George
>> --
>> for email reply remove "/" from address
>
>postgresql allows for hierarchical table structures, and lisp has an interface
>(i think it's called pg, IIRC).

Postgres's hierarchical tables aren't quite the same.  In MUMPS the
table itself is a 2 dimensional linked list structure:

tables
 |- table1
      |-------  col1       - col2       - ... - coln
      |         |            |                  |
      |- rec1 - <data 1,1> - <data 1,2> - ... - <data 1,n>
      |         |            |                  |
      |- rec2 - <data 2,1> - <data 2,2> - ... - <data 2,n>
      |         |            |                  | 
      |- :
      |         |            |                  |
      |- recm - <data m,1> - <data m,2> - ... - <data m,n>

A record is a linked list of disk pages containing field data.  Each
data item is also linked into a separate list headed by attribute.
Several fields may share a page so the links between them are of the
form page#:offset rather than a more typical file offset to the item.

MUMPS is really a network database (in the DB topology meaning of
"network" ... not the LAN/WAN meaning) which happens to impose a
hierarchical organization rather than allow a more general graph
structure.

George
--
for email reply remove "/" from address
From: Alex Mizrahi
Subject: Re: MUMPS in lisp?
Date: 
Message-ID: <47932bdd$0$90270$14726298@news.sunsite.dk>
 ??>> http://en.wikipedia.org/wiki/MUMPS
 ??>>
 ??>> I think MUMPS in lisp would rock and give lisp some db punch!
 ??>>
 ??>> I know prolog was implemented in lisp and APL in scheme so how about a
 ??>> database language?

 GN> Having actually used MUMPS I would say something much different than
 GN> "it rocks".  I remember it as a write-only language ... simple to use
 GN> but very difficult to debug.  Lisp is a much better language.  What
 GN> value do you think MUMPS would bring?

 GN> Lisp already has implementations of and interfaces to virtually every
 GN> type of database that exists.  Though I can't find a hierarchical
 GN> database (as in MUMPS) searching just now, I'm sure that one or more
 GN> exist somewhere.

i don't think MUMPS as language is valuable, but database stuff they have is 
quite interesting IMHO.
sutff like SQL and ORM/object persistent are popular nowadays, but they are 
overkill for many purposes. MUMPS style persistent multidimensional arrays 
might be a better solution.

i suspect M-style data storage is not that popular because there are no 
simple, "no bullshit" implementations.
at least i was not able to find solution for C/C++ when i was starting a new 
project -- only thing i've found was complete M implemented in C++, that was 
quite fat, and it used BDB iirc.
it's overkill to include thing like this into a small project (just like 
it's overkill to use SQL..). i've hacked my own little storage thingie which 
used file system for storage -- i.e. when you write
db["invoices"][315]["date"] it was reading file invoices/315/date. however 
later i had to replace it with embedded SQL..
From: Pascal J. Bourguignon
Subject: Re: MUMPS in lisp?
Date: 
Message-ID: <7cbq7evwqg.fsf@pbourguignon.anevia.com>
gavino <·········@gmail.com> writes:

> http://en.wikipedia.org/wiki/MUMPS
>
> I think MUMPS in lisp would rock and give lisp some db punch!

http://thedailywtf.com/Articles/Classics-Week-A-Case-of-the-MUMPS.aspx


-- 
__Pascal Bourguignon__
·························@anevia.com
http://www.anevia.com