[comp.object] Persistence in c++

mwirth@gara.une.oz.au (Michael A. Wirth) (06/19/91)

Does anybody have any code for object persistence in c++? I have some examples
in Turbo Pascal, but do not know how to implement it in c++.

						Thanks 
							Michael.


-------------------------------------------------------------------------
Michael A. Wirth
Senior Technical Advisor
Duval College                  mwirth@gara.une.oz.au
University of New England      mwirth@neumann.une.oz.au
Armidale, N.S.W, 2351.
A U S T R A L I A
-------------------------------------------------------------------------

rob@cs.man.ac.uk (Robert Marshall) (06/20/91)

In article <7087@gara.une.oz.au> mwirth@gara.une.oz.au (Michael A. Wirth) writes:

>   Does anybody have any code for object persistence in c++? I have some examples
 >  in Turbo Pascal, but do not know how to implement it in c++.
Back at the end of May
In article <1991May28.164803.1@manse.cs.man.ac.uk> marshall@manse.cs.man.ac.uk (Robert Marshall) writes:
Lines: 21

>   Does anyone know of any public domain C++ generic classes which support
>   persistency?
I suppose I ought to give a summary of responses.

Many thanks to all who replied:
	jls@netcom.com (Jim Showalter)
	Stephen J Bevan <bevan@cs.man.ac.uk>
	Paul Ganschow <pag@sv001.sandiego.ncr.com>
	m.flower@trl.OZ.au (Michael Flower)
	brian@babbage.csus.edu (Brian Witt)
	ericp@bach
	Niklas Hallqvist <niklas@appli.se>
	ed@ccs.yorku.ca
	perty@nmpcad.se (Per Lundholm)
	
Most people who suggested something mentioned NIHCL (as we at present only
have access to g++ that's a problem)

Stephen J Bevan <bevan@cs.man.ac.uk> suggested:
|No, but you could use Modula III or Eiffel, both of which allow
|objects to be saved to disk and then retrieved.  You can even get a
|copy of the former for free by ftp.

Paul Ganschow <pag@sv001.sandiego.ncr.com> mentioned Code Farms Inc.:
|A company called Code Farms Inc. produces classes which provide support for persistent objects. Their address is
|Code Farms Inc.
|7214 Jock Trail
|Richmond, Ont., K0A 2Z0, Canada
|
|Tel # 613-838-4829
|Fax # 613-838-3316
|eMail: jiri@dgbt.doc.ca
|
|I talked to them at SCOOP West and found them to be knowledgeable about
| giving objects persistence.  I have not as yet used their products.

m.flower@trl.OZ.au (Michael Flower) mentioned RogueWave:
|The RogueWave libraries support persistency. They are not public domain, but are
|so cheap that they almost are. A single license is US$50 and a site license
|US$150 (and that includes source!).
|
|You can contact rwave!keffer@CS.ORST.EDU for details.

brian@babbage.csus.edu (Brian Witt) suggested the following:
|Check out the "E" language by Joel Richardson (currently of IBM)
|and Michael J. Carey (Univ of Wisconsin - Madison).  I found an
|article in "Persistent Object Systems", Workshops in Computing:
|John Rosenberg and David Koch (Eds.), Springer-Verlag, 1990.
|IBBN 3-540-19626-9.  Page 175, Abstract:
|
|	The E language is an extension of C++ providing, amoung
|	other features, database types and persistent objects.
|	The basis of peristence in E is a new storage class for
|	variables, and physical I/O is based on a load/store
|	model of long-term storage layer.  This paper describes
|	in detail the implementation o fthe first E compiler
|	and discusses our current research direction.
|
|(typos are mine)
|
|They describe a database model EXTRA and the EXCESS query language
|using the EXODUS storage manager.  A stack example the pops from
|a persistant stack.  The main pops, and then pushes ten numbers.
|The first time it runs, nothing is printed.  The second time it
|runs, the program pop's the ten numbers pushed last time you
|ran the program!  They provide hacks to cfront to recognize their
|special storage types, which change the variable "type" slightly.
|
||From the author addresses' listing:
| M. J. Carry:   	carey@cs.wisc.edu
| Joel Richarson:	jolr@ibm.com
|
|
|The book describes other systems presented at the workshop in Newcastle,
|Australia 1989.  Also mentioned is the PS-Algol system, which also
|implements persistance.  It seems PS-Algol is almost (?) a commercial
|system.  Ie, it is real, and not a theory paper.

Niklas Hallqvist <niklas@appli.se> suggested some home written classes.

Thanks again.

Robert
--
------------------------------------------------------------------
Robert A.J.Marshall,		EMAIL: rmarshall@cs.man.ac.uk
Room 3.08, IT Building, Department of Computer Science,
University of Manchester, Oxford Road, Manchester, M13 9PL, U.K.
Tel: (+44) 61-275 6269          Fax: (+44) 61-275 6280
------------------------------------------------------------------