This was a bit of an oversight initially - we should be storing the
positions that the character has seen *on the level*, rather than on the
entire game state, for obvious reasons. This introduces a GameLevel
record, which has this field, the entities, and also the up staircase
position, which we can *also* use to position the character after going
down to a level we've already visited.
New levels need to go at the *end* of the list of levels, not the
beginning - otherwise we jump to the proper position on the new level
but the current level stays the same (oops).
Cheddar now needs to be passed the --about-filter flag to toggle the
behaviour for rendering Markdown into HTML.
By default Markdown will be highlighted like normal source code (i.e.
cgit source-filtering is the default behaviour).
Per https://github.com/noinia/hgeometry/issues/28, occasionally
DelaunayTriangulation.DivideAndConquer loops infinitely - in this case,
I was able to consistently use the seed 127624940715530481, to generate
a dungeon which had the following room centroids:
[ Point2 [38.5,3.5] :+ 0
, Point2 [67.0,33.0] :+ 1
, Point2 [46.0,45.5] :+ 2
, Point2 [55.5,42.0] :+ 3
, Point2 [36.0,25.0] :+ 4
, Point2 [76.5,12.0] :+ 5
, Point2 [29.0,26.5] :+ 6
, Point2 [55.0,10.5] :+ 7
]
and cause delaunay triangulation to loop indefinitely (or at least
longer than I cared to wait for). Given the size of our graphs switching
to naive generation should be fine performance-wise, and avoids the
infinite loop.
This advice is potentially defined before the autoloads for telega
have run, which means that the macro-expansion fails and
`telega-ins-fmt` is looked up as a function.
With this setup the initialisation works as expected.
Some more pains of weening off of Dropbox is that my Emacs initialization is
sensitive to dependencies and missing require statements. I'm still debugging
everything.
Some modules called `exwm-input-set-key` before the `window-manager` module
loaded, which itself requires EXWM. This broke initialization. To get around
this I could've called `(require 'exwm)` in each of those modules. I chose to
define a `keybindings.el` module to whitelist some of my EXWM keybindings. I'm
not sure if this is the best way forward, but it is *some* way forward.
Since the tokenizing isn't working as expected, my keyboard.el function
keyboard/swap-caps-lock-and-escape was silenting failing.
I'm adding a prelude/refute in that function to make the failures noisy until
the tokenizing is properly supported.
Cameron sent over some property tests for his File.split function, which is a
part of a larger effort to port f.el, a nice library for working with file
paths, over to Haskell.
While I've done these algorithms before, I'm preparing for an on-site with
DeepMind, so I created a subdirectory called deepmind where I'm storing my
second attempts at these problems. The idea of storing them in a second
directory is to remove the urge to check my existing solutions that also exist
in this repository.
Calls the sourcehut API at builds.sr.ht to trigger a build if the
master branch changes.
The build manifest is going to be stored in the depot too, coming up
next ...
