Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.prolog > #14803 > unrolled thread

ANN: Dogelog Player 1.3.6 (Canonical Compare)

Back to article view | Back to comp.lang.prolog

Contents

  ANN: Dogelog Player 1.3.6 (Canonical Compare) Mild Shock <janburse@fastmail.fm> - 2025-08-21 15:44 +0200
    Jaffar's Unification in Dogelog Player (Was: ANN: Dogelog Player 1.3.6 (Canonical Compare)) Mild Shock <janburse@fastmail.fm> - 2025-09-23 16:19 +0200
      Gap Buffers for Dogelog Player (Was: Jaffar's Unification in Dogelog Player) Mild Shock <janburse@fastmail.fm> - 2025-10-02 14:05 +0200

#14803 — ANN: Dogelog Player 1.3.6 (Canonical Compare)

Dear All,

We are happy to announce a new edition
of the Dogelog Player:

- Frozen Terms:
We extended our garbage collector marking bits
to Prolog compounds. Unlike Prolog variables, where
setting all bits is used to indicate change set
membership, we use it to indicate frozen objects.
This makes the garbage collector and certain
built-ins such as copy_term/2 etc.. aware of
program sharing (PS).

- Marking Algorithms:
We now provide native implementations of (==)/2,
copy_term/2, etc.. with sharing and cycle detection.
We didn't deploy additional marking bits and/or
pointers inside the Prolog terms, instead used
additional datastructures, leaving the Prolog terms
untouched. The compare/3 implementation is
not a total order.

- Canonical Compare:
The library(lists) has experimental predicates
term_decompose/3 and term_canonical/2, still written
in 100% Prolog. We showcase their usage in
library(sequence) and library(aggregate). The end-
user gets by default a structure compare, but can
also have a canonical compare, both being
total and natural orders.

Have Fun!

Jan Burse, http://www.herbrand.ai/ , 21.08.2025

[toc] | [next] | [standalone]

#14837 — Jaffar's Unification in Dogelog Player (Was: ANN: Dogelog Player 1.3.6 (Canonical Compare))

We are currently in our third iteration of
unification for rational trees. What first
begun as a separate routine inside library(math),
has now recently become a novel take on the
host language routine unify() inside the
Dogelog Player Prolog system.

We report how we replaced the union find map
data structure inside unify() by some Prolog
compound pointer swizzling as suggested by Jaxon
Jaffar in 1984, arriving at a more competetive
rational tree unification realization.

Jaxon Jaffar (1984) already proposed Hydra test
cases. They take very less memory but are very
large when written. We conducted according tests
with Dogelog Player and Scryer Prolog. For
N = 16'384, 65'536, 261'144 the Rust Prolog

has a minor advantage over the JavaScript Prolog,
but interestingly losing it for N = 1'048'576.

See also:

Jaffar's Unification in Dogelog Player
https://qiita.com/j4n_bur53/items/b0b754f02d88680d3249

Mild Shock schrieb:
> Dear All,
> 
> We are happy to announce a new edition
> of the Dogelog Player:
> 
> - Frozen Terms:
> We extended our garbage collector marking bits
> to Prolog compounds. Unlike Prolog variables, where
> setting all bits is used to indicate change set
> membership, we use it to indicate frozen objects.
> This makes the garbage collector and certain
> built-ins such as copy_term/2 etc.. aware of
> program sharing (PS).
> 
> - Marking Algorithms:
> We now provide native implementations of (==)/2,
> copy_term/2, etc.. with sharing and cycle detection.
> We didn't deploy additional marking bits and/or
> pointers inside the Prolog terms, instead used
> additional datastructures, leaving the Prolog terms
> untouched. The compare/3 implementation is
> not a total order.
> 
> - Canonical Compare:
> The library(lists) has experimental predicates
> term_decompose/3 and term_canonical/2, still written
> in 100% Prolog. We showcase their usage in
> library(sequence) and library(aggregate). The end-
> user gets by default a structure compare, but can
> also have a canonical compare, both being
> total and natural orders.
> 
> Have Fun!
> 
> Jan Burse, http://www.herbrand.ai/ , 21.08.2025

[toc] | [prev] | [next] | [standalone]

#14874 — Gap Buffers for Dogelog Player (Was: Jaffar's Unification in Dogelog Player)

Dogelog Player is a Prolog system for the
JavaScript, Python and Java platform. We
recently ventured into cyclic terms, re-
discovering Jaffar’s Unification for
binary operations such as (==)/2, (=)/2,
etc.. . Here we talk about unary operations
such as term_variables/2, ground/2, etc..

While we could already adopt a Peter Deutsch
algorithm for the Dogelog Player garbage
collection, freeing us from native stack
constraints. For performance reasons of unary
operations such as term_variables/2, ground/2,
etc.. gap buffers known from Emacs showed up.

See also:

Gap Buffers for Dogelog Player
https://medium.com/2989/bf3bc861ae04

Mild Shock schrieb:
> 
> We are currently in our third iteration of
> unification for rational trees. What first
> begun as a separate routine inside library(math),
> has now recently become a novel take on the
> host language routine unify() inside the
> Dogelog Player Prolog system.
> 
> We report how we replaced the union find map
> data structure inside unify() by some Prolog
> compound pointer swizzling as suggested by Jaxon
> Jaffar in 1984, arriving at a more competetive
> rational tree unification realization.
> 
> Jaxon Jaffar (1984) already proposed Hydra test
> cases. They take very less memory but are very
> large when written. We conducted according tests
> with Dogelog Player and Scryer Prolog. For
> N = 16'384, 65'536, 261'144 the Rust Prolog
> 
> has a minor advantage over the JavaScript Prolog,
> but interestingly losing it for N = 1'048'576.
> 
> See also:
> 
> Jaffar's Unification in Dogelog Player
> https://qiita.com/j4n_bur53/items/b0b754f02d88680d3249
> 
> Mild Shock schrieb:
>> Dear All,
>>
>> We are happy to announce a new edition
>> of the Dogelog Player:
>>
>> - Frozen Terms:
>> We extended our garbage collector marking bits
>> to Prolog compounds. Unlike Prolog variables, where
>> setting all bits is used to indicate change set
>> membership, we use it to indicate frozen objects.
>> This makes the garbage collector and certain
>> built-ins such as copy_term/2 etc.. aware of
>> program sharing (PS).
>>
>> - Marking Algorithms:
>> We now provide native implementations of (==)/2,
>> copy_term/2, etc.. with sharing and cycle detection.
>> We didn't deploy additional marking bits and/or
>> pointers inside the Prolog terms, instead used
>> additional datastructures, leaving the Prolog terms
>> untouched. The compare/3 implementation is
>> not a total order.
>>
>> - Canonical Compare:
>> The library(lists) has experimental predicates
>> term_decompose/3 and term_canonical/2, still written
>> in 100% Prolog. We showcase their usage in
>> library(sequence) and library(aggregate). The end-
>> user gets by default a structure compare, but can
>> also have a canonical compare, both being
>> total and natural orders.
>>
>> Have Fun!
>>
>> Jan Burse, http://www.herbrand.ai/ , 21.08.2025
> 

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.prolog

csiph-web