Subtrees First in Dogelog Player (Re: ANN: Dogelog Player 1.3.4 (Interactive Debugger))

For best possible experience, we investigated
rational trees printing. The result is a top-level
which does not fall back to print <cyclic> for cyclic
terms anymore, but display a set of cycle defining
equations. The required native support is minimal,
a single new built-in same_term/2.

Alain Colmerauers “Prolog and Infinite Trees“ (1982),
shows two solutions for cycle breaking. Starting
with a prototype that generates some non-monster
equations, we end up with producing monster equations
for the benefit of the interactive debugger and
the answer display.

Subtrees First in Dogelog Player
https://medium.com/@janburse_2989/2c72018758d6

Mild Shock schrieb:
> Dear All,
> 
> We are happy to announce a new edition of
> the Dogelog player:
> 
> - Enhanced library(fancy):
>    As a convenience we provide the additional
> predicate print/[1,2] which checks the flags of
> the output stream. If the output stream has the
> mask MASK_DST_CANS set, it will automatically
> colorize the output. This is for example used in
> the new interactive debugger to print goals.
> 
> - Enhanced library(tools):
> Besides clause listing, the file also provides an
> interactive debugger. The interactive debugger can
> be invoked for a goal by the meta predicate trace/1.
> The interactive debugger is implemented by means of
> meta programming and therefore might executed slower
> than call/1. On the other hand it allows inspecting
> the Byrd Box ports, already supporting useful
> command such as skip.
> 
> - Enhanced library(session):
> The interactive debugger is still minimal, for example
> it does not allow spy points yet. So the leap command
> will leap into the next answer substitution. The end-user
> might not want to continue with the next answer substitution
> in leap mode. Therefore the top-level is now trace/1 aware,
> in that it provides debugger commands as well, such
> as switching to creep mode.
> 
> Have Fun!
> 
> Jan Burse, http://www.herbrand.ai/ , 07.07.2025