Path: ...!news.mixmin.net!weretis.net!feeder8.news.weretis.net!reader5.news.weretis.net!news.solani.org!.POSTED!not-for-mail
From: Mild Shock <janburse@fastmail.fm>
Newsgroups: comp.lang.prolog
Subject: Re: Is Scryer Prologs failure measurable? (Was: Holy Grail makes
 People Disappear)
Date: Sat, 10 Aug 2024 13:41:21 +0200
Message-ID: <v97jkv$118im$1@solani.org>
References: <v8gc6e$l44p$1@solani.org> <v952rb$10h3e$1@solani.org>
 <v97f6r$11l7k$1@solani.org> <v97h77$11m5e$1@solani.org>
 <v97het$11mct$1@solani.org> <v97i30$11mn6$1@solani.org>
 <v97icn$11mpq$1@solani.org> <v97j54$11n9o$1@solani.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Injection-Date: Sat, 10 Aug 2024 11:41:19 -0000 (UTC)
Injection-Info: solani.org;
	logging-data="1090134"; mail-complaints-to="abuse@news.solani.org"
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101
 Firefox/91.0 SeaMonkey/2.53.18.2
Cancel-Lock: sha1:5N8ipzmJJZ74GiLnIoQFxpBeozs=
In-Reply-To: <v97j54$11n9o$1@solani.org>
X-User-ID: eJwNwoERwEAEBMCW4h5HOU/ov4Rkdu24eFPdXG1/HaWIqUPI6TKTmGS9shL3JvAgkK27pazr9Nxpb8XkkB9J2BV6
Bytes: 3329
Lines: 75

Hi,

Currently I have first hand experience how
difficult it is to model single treaded async
on top of multi-threaded Prolog,

since I want to bring Dogelog Player async
also to formerly Jekejeke Prolog. It wasn't
so much a problem in the Dogelog Player for

Java version, since I could replicate the
JavaScript design. But it is a little
nasty in the formerly Jekejeke Prolog case,

since Thread identity from multi-threaded
Prolog and Task identity from single treaded
async are not the same, you have to dig

deeper into stackfull and stackless coroutines,
to get the Task identity notion correctly.
I wasn't able to do it yet for formerly

Jekejeke Prolog, there is still a pending ticket.

Bye

Mild Shock schrieb:
> Hi,
> 
> Generally I don't believe in things like
> SWI-Prolog Janus Python integration. This
> type of integration has the advantage
> 
> that SWI-Prolog can run in full speed,
> on the other hand it has the disadvantage
> that one cannot take advantage of Python async,
> 
> because SWI-Prolog itself has no async. I
> am talking about the single threaded async here.
> My experience single treaded async and multi-threaded
> 
> Prolog are quite different pairs of shoes. So
> when doing the C# backend for Dogelog Player,
> I will be faced again to provide a complete
> 
> async solution for the async part of Novacore.
> 
> Woa!
> 
> Bye
> 
> Mild Shock schrieb:
>> Hi,
>>
>> A few days age we improved the cross referencer,
>> to deal with the many target platforms. The
>> old cross referencer for example showed:
>>
>> xor/3        nova/Eval.java
>>               nova/eval.mjs
>>               nova/eval.py
>>
>> Meaning the evaluable function xor/3 is
>> available in Java, JavaScript and Python.
>> What we now do, we simply show:
>>
>> xor/3        nova/eval.*
>>
>> Soon the wildcard will mean C# as well,
>> and not only Java, JavaScript and Python.
>> Extending the proof of concept for Prolog
>>
>> compiler 100% written in Prolog itself.
>>
>> Bye