Some more rewording/writing in 2021-07-27-procedural-meshes.org

content/posts/2021-07-27-procedural-meshes.org

@@ -31,7 +31,7 @@ that.  Naturally, this led me to wonder how I might extend Context
 Free's model to work more generally with 3D geometry, and let me use
 it to produce procedural geometry.
 
-[[http://structuresynth.sourceforge.net/index.php][Structure Synth]] of course already exists as a straightforward
+[[http://structuresynth.sourceforge.net/index.php][Structure Synth]] of course already exists, and is a straightforward
 generalization of Context Free's model to 3D (thank you to Mikael
 Hvidtfeldt Christensen's blog [[http://blog.hvidtfeldts.net/][Syntopia]], another of my favorite things
 ever, for introducing me to it awhile ago).  See also [[https://kronpano.github.io/BrowserSynth/][BrowserSynth]].
@@ -47,9 +47,8 @@ Tools like [[https://openscad.org/][OpenSCAD]], based on [[https://www.cgal.org/
 suspect that [[https://www.opencascade.com/][Open CASCADE]] (thus [[https://www.freecadweb.org/][FreeCAD]]) also does.  In CAD work, it's
 crucial.  I experimented with similar recursive systems with some of
 these, but I quickly ran into a problem: they were made for actual
-practical applications in CAD, not for my nonsensical generative art,
+practical applications in CAD, not so much for my generative art, and
-and they scaled quite poorly with the sort of recursion I was asking
+they scaled quite poorly with the sort of recursion I was asking for.
-for.
 
 Implicit surfaces (or one of the many
 equivalent-except-for-when-it's-not names for this, e.g. F-Reps or
@@ -59,15 +58,15 @@ rendered directly on the GPU via shaders, operations like blending
 shapes or twisting them are easy... for more on this, see [[http://blog.hvidtfeldts.net/][Syntopia]]
 again, or nearly anything by [[https://iquilezles.org/][Inigo Quilez]], or look up raymarching and
 sphere tracing, or see [[https://ntopology.com/][nTopology]], or Matt Keeter's work with [[https://www.libfive.com/][libfive]]
-and [[https://www.mattkeeter.com/research/mpr/][MPR]].  They're pure magic and they're wonderfully elegant and I'll
+and [[https://www.mattkeeter.com/research/mpr/][MPR]].  They're pure magic, they're wonderfully elegant, and I'll
 probably have many other posts on them.
 
 However, there is one big issue: turning implicit surfaces to good
 meshes for rendering /is a huge pain/, and while many renderers can
-handle implicit surfaces directly, Blender's renderers cannot. I have
+handle implicit surfaces directly, Blender's renderers cannot. I will
-other posts on this as well, but for now, take it on faith.  This is
+have other posts going into more detail on this subject, but for now,
-why I did not try to use implicit surfaces for this project.  (TODO:
+take it on faith.  This is why I did not try to use implicit surfaces
-Make those posts.)
+for this project.  (TODO: Make those posts.)
 
 With these limitations in mind, around 2018 June I had started jotting
 some ideas down.  The gist is that I wanted to create
@@ -94,7 +93,8 @@ My meandering path to implementing it went something like this:
 - Write some very ad-hoc Python to generate a mesh of a parametric
   conversion of my annoying spiral isosurface from 2005 by breaking it
   into planar "slices" or "frames", which move along the geometry and
-  then are connected together at corresponding vertices.
+  then are connected together at corresponding vertices.  (TODO: Add
+  link to the automata_scratch repo, whatever it's renamed to)
 - Explore [[https://github.com/thi-ng/geom][thi.ng/geom]] and pretty quickly give up - but in the process,
   discover [[https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.42.8103][Parallel Transport Approach to Curve Framing]].
 - Implement that paper in Python, reusing the basic model from my
@@ -109,39 +109,58 @@ My meandering path to implementing it went something like this:
 - Realize that Rust is the wrong tool for the job, and rewrite *again*
   in Python but with a rather different design and mindset.
 
-(this is, of course, ignoring projects on various other tangents)
+(this is, of course, ignoring many other tangents with things like
+shaders)
 
 (TODO: Maybe split these off into sections for each one?)
 
-Somewhere in here, I concluded that my fundamental idea was
+I put some serious effort into [[https://github.com/Hodapp87/prosha][Prosha]] and was conflicted on shelving
-half-broken.  It half-worked: I was able to produce closed, manifold
+the project, but the issues didn't look easily solvable.  Part of
-meshes this way, and it could be tedious, but not *that* difficult.
+those issues were implementation issues with Rust - not that Rust
-However, all of my attempts to also produce "good" meshes this way
+could have really done anything "better" here, but that it just wasn't
-failed miserably.
+the right tool for what I was doing.  In short, I had spent a lot of
+time and effort trying to badly and unintentionally implement a Lisp
+inside of Rust instead of just picking a Lispier language, or perhaps
+using an embeddable Rust-based scripting language like [[https://github.com/koto-lang/koto][Koto]] or [[https://github.com/rhaiscript/rhai][Rhai]]. I
+had ignored that many things that functional programming left me very
+accustomed to - like first-class functions and closures - were
+dependent on garbage collection.  When I realized this and did a big
+refactor to remove this entire layer of complexity, I was left with
+very little "core" code - just a handful of library functions, and the
+actual recursive rules for the geometry I was trying to generate.
+That's good and bad: things were much simpler and vastly faster, but
+also, it felt like I had wasted quite a lot of time and effort.  I
+have some more detailed notes on this in the Prosha repository.
+
+Part of the issues also weren't Rust implementation issues - they were
+deeper issues with my original "correct-by-construction" mesh idea
+being half-broken.  It half-worked: I was able to produce closed,
+manifold meshes this way, and it could be tedious, but not *that*
+difficult.  However, all of my attempts to also produce "good" meshes
+this way failed miserably.
 
 (TODO: Can I find examples of this?)
 
-A few of the same fundamental issues kept cropping up.  One is that
+The crux is that the recursive rules I used for generating geometry
-the recursive rules I used for generating geometry (inspired heavily
+(inspired heavily by those in Context Free) were inherently based
-by those in Context Free) were inherently based around discrete steps,
+around discrete steps, generating discrete entities, like vertices,
-generating discrete entities, like vertices, edges, and face; it made
+edges, and face, and it made no sense to "partially" apply a rule,
-no sense to "partially" apply a rule, especially if that rule involved
+especially if that rule involved some kind of branching - but I kept
-some kind of branching.  I kept trying to treat it as something
+trying to treat it as something continuous for the sake of being able
-continuous for the sake of being able to "refine" the mesh to as fine
+to "refine" the mesh to as fine of detail as I wanted.  Further, I was
-of detail as I wanted.  Further, I was almost never consistent with
+almost never consistent with the nature of this continuity: sometimes
-the nature of this continuity: sometimes I wanted to treat it like a
+I wanted to treat it like a parametric curve (one continuous
-parametric curve (one continuous parameter), sometimes I wanted to
+parameter), sometimes I wanted to treat it like a parametric surface
-treat it like a parametric surface (two continuous parameters),
+(two continuous parameters), sometimes I wanted to treat it like an
-sometimes I wanted to treat it like an implicit surface
+implicit surface (with... theoretically two continuous parameters,
-(with... theoretically two continuous parameters, just not explicit
+just not explicit ones?).  It was a mess, and it's part of why my
-ones?).  It was a mess, and it's part of why my Prosha repository is a
+Prosha repository is a graveyard of branches.
-graveyard of branches.
 
 The recursive rules were still excellent at expressing arbitrarily
 complex, branching geometry - and I really wanted to keep this basic
 model around somehow.  After some reflection, I believed that the only
-way to do this was to completely separate the process of
+way to do this was to completely separate the process of meshing
-meshing/refinement/subdivision from the recursive rules.
+(refinement, subdivision, facetization...) from the recursive rules.
 
 This would have been obvious if I read the guides from [[https://graphics.pixar.com/opensubdiv/overview.html][OpenSubdiv]]
 instead of reimplementing it badly.  Their [[https://graphics.pixar.com/opensubdiv/docs/subdivision_surfaces.html][subdivision surface]]
@@ -151,7 +170,7 @@ made a lot of sense: I shouldn't be trying to generate the "final"
 mesh, I should be generating a mesh as the /control cage/, which
 guides the final mesh.  Further, I didn't even need to bother with
 OpenSubdiv's C++ API, I just needed to get the geometry into Blender,
-and Blender would handle the subdivision via OpenSubdiv.
+and Blender would handle the subdivision on-demand via OpenSubdiv.
 
 One minor issue is that this control cage isn't just a triangle mesh,
 but a triangle mesh plus edge creases.  I needed a way to get this
@@ -159,12 +178,26 @@ data into Blender.  However, the only format Blender can read edge
 creases from is [[http://www.alembic.io/][Alembic]].  Annoyingly, its [[http://docs.alembic.io/reference/index.html#alembic-intro][documentation]] is almost
 completely nonexistent, the [[https://alembic.github.io/cask/][Cask]] bindings still have spotty Python 3.x
 support, and when I tried to run their example code to produce some
-files, and Blender was crashing when importing them.  Until I shave
+files, and Blender was crashing when importing them.... and this is
-that yak, I am instead generating the mesh data directly in Blender
+all a yak to shave another day.  I instead generated the mesh data
-(via its Python interpreter), adding it to the scene, and then setting
+directly in Blender (via its Python interpreter), added it to the
-its creases via its Python API.
+scene, and then set its creases via its Python API.
 
-TODO while I'm not so tired:
+After the aforementioned refactor in Prosha, I was able to quickly
+translate the Rust code for most of my examples into Python code with
+the help of some library code I'd accumulated from the past projects.
+Debugging this mostly inside Blender also made the process vastly
+faster.  Further, because I was letting Blender handle all of the
+heavy lifting with mesh processing (and it in turn was using things
+like OpenSubdiv), the extra overhead of Python compared to Rust didn't
+matter - I was handling so much less data when only producing the
+control cage, not the full mesh.
 
-What is the aim of this post?  To explain Prosha?  To explain current
+I'm still a little stuck at how to build higher 'geometric'
-work, including Prosha?
+abstractions here and compose them.  I have felt like most of the
+model couples me tightly to low-level mesh constructs - while Context
+Free and Structure Synth definitely don't have this problem.  This is
+particularly annoying because a lot of the power of these recursive
+grammars comes from their ability to be abstracted away and composed.
+
+(TODO: Show some examples)