simon-git: tilings (main): Simon Tatham

[Commits to Tartarus hosted VCS]

TL;DR:
  d73afde --findcoords: exit if the output NFA is empty.
  c53e9a4 Typo in a comment.
  4f6d559 Add arrows to the p3-whole tiling.
  a9b198b Remove an outdated comment.
  1a177c8 A second version of the Hat H7/H8 system.
  fa3be13 Convert spectre9 and hat10 into coloured H7/H8.

Repository:     https://git.tartarus.org/simon/tilings.git
On the web:     https://git.tartarus.org/?p=simon/tilings.git
Branch updated: main
Committer:      Simon Tatham <anakin at pobox.com>
Date:           2024-07-04 08:00:18

commit d73afded5a3d5cd66fa5125bb4ed90c510b3ccfe
web diff https://git.tartarus.org/?p=simon/tilings.git;a=commitdiff;h=d73afded5a3d5cd66fa5125bb4ed90c510b3ccfe;hp=2f30e561c27b05a690c5e1fb8bb068f1067b59b1
Author: Simon Tatham <anakin at pobox.com>
Date:   Mon Jul 1 19:31:52 2024 +0100

    --findcoords: exit if the output NFA is empty.
    
    Otherwise, the attempt to find a shortest repeating cycle will try
    forever. Probably that could be fixed too, but this is a simple change
    that makes the command-line mode work.

 toplevel.sage | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

commit c53e9a4d98531558004b97a20f6a72ce7af55c03
web diff https://git.tartarus.org/?p=simon/tilings.git;a=commitdiff;h=c53e9a4d98531558004b97a20f6a72ce7af55c03;hp=d73afded5a3d5cd66fa5125bb4ed90c510b3ccfe
Author: Simon Tatham <anakin at pobox.com>
Date:   Tue Jul 2 20:36:08 2024 +0100

    Typo in a comment.

 infinitary.sage | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

commit 4f6d559eb71bc82a228496105d299dedeaa699bd
web diff https://git.tartarus.org/?p=simon/tilings.git;a=commitdiff;h=4f6d559eb71bc82a228496105d299dedeaa699bd;hp=c53e9a4d98531558004b97a20f6a72ce7af55c03
Author: Simon Tatham <anakin at pobox.com>
Date:   Tue Jul 2 20:36:35 2024 +0100

    Add arrows to the p3-whole tiling.
    
    Unfortunately, one of the arrowheads is absurdly enormous. So also in
    this commit I scale each arrowhead to match its arrow.

 p3-whole.tl | 10 ++++++++--
 svg.py      |  2 +-
 2 files changed, 9 insertions(+), 3 deletions(-)

commit a9b198bb05c3fe92f5571839d2f3f51e7e076ebf
web diff https://git.tartarus.org/?p=simon/tilings.git;a=commitdiff;h=a9b198bb05c3fe92f5571839d2f3f51e7e076ebf;hp=4f6d559eb71bc82a228496105d299dedeaa699bd
Author: Simon Tatham <anakin at pobox.com>
Date:   Wed Jul 3 13:08:32 2024 +0100

    Remove an outdated comment.
    
    When I wrote the description of the Spectre H7/H8 system, the spurs
    were so awkward that even the _recursive_ algorithm couldn't deal with
    them. This comment was documenting some thoughts on possible hints I
    could add to the .tl file to help it out.
    
    But commit d452a4e0aec61d9 turned out to be all I needed to make the
    recursion work automatically, so I never had to uncomment this
    lot. (And then I developed the improved adjmatcher which could use
    that in turn.)

 spectre-h7h8.tl | 23 -----------------------
 1 file changed, 23 deletions(-)

commit 1a177c8960c9384c04d82dcf9e5c205920e8f06e
web diff https://git.tartarus.org/?p=simon/tilings.git;a=commitdiff;h=1a177c8960c9384c04d82dcf9e5c205920e8f06e;hp=a9b198bb05c3fe92f5571839d2f3f51e7e076ebf
Author: Simon Tatham <anakin at pobox.com>
Date:   Wed Jul 3 13:07:52 2024 +0100

    A second version of the Hat H7/H8 system.
    
    In this version, the reflected hat is fused with its _other_
    neighbour. It wasn't obvious from the actual hat paper which was
    intended, and either one works.
    
    But this layout is convenient because it matches the layout of the
    hexagons in hat10.tl.

 hats-h7h8b.tl | 124 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 toplevel.sage |   1 +
 2 files changed, 125 insertions(+)

commit fa3be1395f5cbac2ff129df6433efa1c917a1ad5
web diff https://git.tartarus.org/?p=simon/tilings.git;a=commitdiff;h=fa3be1395f5cbac2ff129df6433efa1c917a1ad5;hp=1a177c8960c9384c04d82dcf9e5c205920e8f06e
Author: Simon Tatham <anakin at pobox.com>
Date:   Wed Jul 3 13:13:05 2024 +0100

    Convert spectre9 and hat10 into coloured H7/H8.
    
    These two new tiling files are combinatorially the same as the
    hexagon-based systems I've been using all along. But geometrically,
    they're the same as the H7/H8 systems (the new h7h8b, in the hats
    case); I've just cloned the non-special expansion a few times and
    named it after the various hexagon types. So you have the same set of
    metatiles, expanding in the same way; it's just that now the metatiles
    are hat- or spectre-shaped (plus one double-hat or mystic), instead of
    hexagonal.
    
    Mainly the idea is to demonstrate that this works, and relate the two
    expansions more explicitly for analysis purposes. But a side benefit,
    is that these expansions are a little bit simpler, in one sense - you
    don't need the complicated set of all-different rules for how each hex
    turns into an output tile (and in particular how its edges match up),
    because it's completely obvious.
    
    Of course, one could consider that that simplicity is cancelled out by
    the extra-weird spurs in the H7/H8 world!

 hats-h7h8-10.tl   | 147 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 spectre-h7h8-9.tl | 135 +++++++++++++++++++++++++++++++++++++++++++++++++
 toplevel.sage     |   2 +
 3 files changed, 284 insertions(+)