Re: [eclipse-clp-users] Controlling lifetime of variables from Java

From: Wit Jakuczun <wit.jakuczun_at_...6...>
Date: Tue, 3 Jun 2008 20:43:04 +0200
2008/6/3 Kish Shen <kisshen_at_...5...>:
> Wit Jakuczun wrote:
>> 2008/6/3 Kish Shen <kisshen_at_...5...>:
>>> In fact, even for the C/C++  interface, I strongly recommend that the
>>> user
>>> stick to the simple yield/resume model, and not to start trying to
>>> interleave C/C++ execution with ECLiPSe search -- there is probably no
>>> reason to do this as a user writing application code, unless you are
>>> extending ECLiPSe itself.
>> There was one reason. I wanted to dynamically create query.
>> The software I am writing is quite complicated and I
>> wanted to avoid writing data processing on a prolog side.
>> Now I see I should write more on prolog side. Then there
>> should be no problems.
>> Best regards

> If you mean creating dynamic predicates via assert/retreact, then in ECLiPSe
> you probably should only do this if you are writing self-modifying code
> (which is probably a bad idea in itself).
Not always. Sometimes it would be much faster if you only
alter database instead of sending megabytes each time you
post a query.

> If you just want to store data non-logically, then ECLiPSe provides better means of doing this -- there is
> a whole chapter in the User manual on this.
Ok. I will check this.

Best regards
[ Wit Jakuczun w.jakuczun_at_...116... ]
[ WLOG Solutions ]
Received on Tue Jun 03 2008 - 11:43:10 CEST

This archive was generated by hypermail 2.3.0 : Mon Aug 26 2019 - 09:14:36 CEST