Pairs must be keysorted and may also contain variables as keys.
Examples:
?- group_pairs_by_key([1-a,1-b,2-c], Ps).
Ps = [1-[a,b],2-[c]].
?- group_pairs_by_key([X-a,X-b,2-c], Ps).
Ps = [X-[a,b],2-[c]].
In run_top_level: rename variables, make them partial strings
The rename is mostly cosmetic: all of argv is passed, and if argv
contains flags, goals, etc, it's more than filenames.
Using partial strings instead of atom came from discussions about how to
pass goals in the CLI, and can thus be considered preliminary work for
that.
In toplevel.pl: convert partial string to atom before passing it to
`use_module`.
Signed-off-by: Stephan Renatus <srenatus@chef.io>
At the moment, library(format) seems to be a fitting place.
In the eventual library organization, portray_clause/1 and
related predicates may be moved to their own dedicated library.
library(clpz) implements declarative integer arithmetic.
The most important predicates for reasoning about integers are:
(#=)/2 equality
(#\=)/2 disequality
(#<)/2 less than
(#>)/2 greater than
(#=<)/2 less than or equal to
(#>=)/2 greater than or equal to
In addition, the library provides several global constraints, such as
all_distinct/1 and global_cardinality/2, and reification predicates
that reflect the truth values of constraints into integer variables.
Enumeration predicates such label/1 and labeling/2 can be used to
search for solutions over finite domains.
Almost all Prolog programs also reason about integers. Therefore, I
recommend to add this library to your .scryerrc configuration file so
that declarative integer arithmetic is available in all your programs.
More information about CLP(ℤ):
https://www.metalevel.at/prolog/clpz
Enjoy!
This is a bit pedantic, it's unlikely that the executable would be
called `-v` or `--version`. However, it seems slightly more correct to
disregard argv[0] when checking flags.
Signed-off-by: Stephan Renatus <srenatus@chef.io>
