Python rewrite of the Urb/Homemaker stack
Find a file
Claude f5286dde3d
Make both failing tests assert their intent, not stale artefacts
The suite is green for the first time this session: 376 passed, 0 failed.

test_collapse_insearch_reproduces_94g_finish_time_result hard-coded both
endpoints of the 17 result -- 15 fails before collapse, 12 after. Those
were measured before 39.4, when harbor's effective programme was silently
32 instances because codes like cr1 were read as generic circulation; the
same layout now scores 82. But the guarantee the test exists to protect,
per its own docstring, is that in-search collapse reaches the SAME layout
as finish-time collapse on fixed geometry -- and two independent constants
never checked that. They can both drift and stay equal, or both hold and
mask an inequality.

Rewritten to compute both sides live and assert they agree, plus that
collapse does not make the layout worse. Measured: 82 -> 58 in-search, and
finish-time collapse independently reaches 58 at iters=3 and iters=6. The
invariant holds; only the constants were stale. Restating the reference
figure itself remains homemaker-py-ut5.

test_classify_fail_tier_covers_full_corpus globbed examples/**/*.fails and
asserted checked > 0. Git tracks ZERO .fails -- they are artefacts the
scorer writes beside a .dom -- so its docstring described files that by
design never exist in the repo, and it passed only on a machine that had
already run the scorer. Split into: a test that GENERATES fails by scoring
three corpus layouts picked for breadth (requiring >= 8 distinct kinds so
it cannot silently narrow); a test that an unclassifiable string actually
raises; and an opportunistic .fails sweep that never requires them.
Verified by moving every .fails out of the tree and re-running.

Closes homemaker-py-1ue.

Lint at parity (46).

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MJ84Feep79Hhm3E4zZJmnB
2026-08-29 07:11:25 +00:00
.beads Make both failing tests assert their intent, not stale artefacts 2026-08-29 07:11:25 +00:00
.claude Scaffold homemaker-py with validated geometry port 2026-06-10 20:50:20 +01:00
examples Checkpoint long searches; the cold-start runs were lost to a reclaimed box 2026-08-29 05:43:58 +00:00
experiments Checkpoint long searches; the cold-start runs were lost to a reclaimed box 2026-08-29 05:43:58 +00:00
src/homemaker_layout Checkpoint long searches; the cold-start runs were lost to a reclaimed box 2026-08-29 05:43:58 +00:00
tests Make both failing tests assert their intent, not stale artefacts 2026-08-29 07:11:25 +00:00
.gitignore bd init: initialize beads issue tracking 2026-06-11 23:27:11 +01:00
AGENTS.md §39.7: access requirements become a declared usage: attribute (homemaker-py-sel) 2026-08-26 13:39:41 +00:00
CLAUDE.md §39.7: access requirements become a declared usage: attribute (homemaker-py-sel) 2026-08-26 13:39:41 +00:00
DESIGN.md Records Room ruled a store; stays exempt 2026-08-28 23:22:02 +00:00
pyproject.toml homemaker-py-2g7.5: CP-SAT exact room-code assignment (seeder + reassign op) 2026-08-04 09:19:36 +01:00
README.md §39.7: access requirements become a declared usage: attribute (homemaker-py-sel) 2026-08-26 13:39:41 +00:00

homemaker-layout

Programme-driven building-layout search over slicing trees. A clean-room Python successor to the Perl Urb project, intended to eventually be 100% Python.

Why a rewrite

Urb represents a building as a binary slicing tree where room sizes are derived top-down from division ratios. That makes room area an emergent property of every cut above it, which:

  • gives the genome low locality (a cut near the root rescales every descendant),
  • makes target room sizes nearly impossible to hit, so the gaussian size penalty dominates fitness, and
  • defeats crossover (transplanted subtrees lose their proportions).

homemaker inverts this: leaves carry target dimensions from the programme and division ratios are solved bottom-up for a fixed topology. The evolutionary search then only explores topology + types + adjacency.

Phase plan

  1. Solver experiment: port Urb's geometry, re-solve ratios from programme targets, score the result against the original via the Perl oracle.
  2. Native Python fitness (retire the Perl oracle).
  3. Memetic search: canonical slicing genome + high-locality operators + Nelder-Mead inner loop.
  4. Penalty reshaping: lexicographic (-n_fails, fitness) outer-search comparison.
  5. Representation upgrade: canonical slicing encoding + bottom-up shape feasibility, scaled to larger programmes.
  6. Search-quality experiments (current): a long running series of opt-in levers tried against the harbor-house, health-centre, and programme-house example corpora — leaf-sharing, finish-time cell→room collapse, ruin-and-recreate LNS, 2-opt polish, multi-use/co-located leaves, adjacency-graph and bubble-diagram fitness signals, and more. Most of these are negative/null results kept as opt-in flags or reference code rather than defaults. See DESIGN.md §11 onward for the full, numbered experiment log with methodology and results for each.

Layout

  • src/homemaker_layout/dom.py — read/write Urb .dom YAML into a Node tree.
  • src/homemaker_layout/geometry.py — faithful port of Urb's top-down geometry.
  • src/homemaker_layout/programme.py — parse patterns.config space requirements.
  • src/homemaker_layout/solver.py — bottom-up ratio solve (scipy).
  • src/homemaker_layout/fitness.py — native Python fitness evaluator.
  • src/homemaker_layout/fitness_cmd.pyhomemaker-fitness CLI (drop-in for urb-fitness.pl).
  • src/homemaker_layout/collapse_cmd.pyhomemaker-collapse CLI: finish-time global cell→room relabel of a .dom.
  • src/homemaker_layout/graph.py — leaf-adjacency graph for programme-driven checks.
  • src/homemaker_layout/genome.py — topology genome: base-floor tree + per-storey deltas.
  • src/homemaker_layout/operators.py — high-locality mutation and subtree crossover.
  • src/homemaker_layout/innerloop.py — ratio optimisation inner loop (Nelder-Mead / CMA-ES).
  • src/homemaker_layout/driver.py — memetic search outer loop.
  • src/homemaker_layout/evolve.pyhomemaker-evolve CLI entry point.
  • src/homemaker_layout/oracle.py — legacy Perl shim, kept for cross-validation only.
  • src/homemaker_layout/bubble.py — 3D bubble-diagram adjacency fitness-signal prototype (DESIGN.md §27); validated null, not wired into fitness.py — reference only.

Room codes and reserved names

Leaf types live in three namespaces that share a first character. Only the first is enforced; the other two are conventions the fitness function reads, so a room's spelling can change how it is scored.

1. Generic structural types — C, O, S (reserved). The leaves the search itself creates: C circulation, O outside, S sahn (an outside court that also serves as circulation). Always uppercase. A programme code spelled exactly C, O or S is rejected at load.

2. Programme room codes — anything else, lowercase. k1, b1, cr1, of, and single-character codes like r or t. These may start with any letter: since DESIGN.md §39.4 the generic tests match C/O/S exactly, so naming a room cr1 no longer makes it circulation. (Before that fix it did — and silently dropped it from the required-space check entirely.)

3. Access requirements — the usage: attribute. Every space declares one of living, kitchen, bedroom, toilet, utility, none. Mandatory, no fallback, and a missing or unknown value is a load error. It replaced a first-character convention (b/t/l/k) under which a room silently inherited another room's connectivity rules from its spelling — la1 "Laundry Room" was trimmed as a living room (DESIGN.md §39.7).

spaces:
  la1:
    usage: utility          # controlled, drives engine behaviour
    name: Laundry Room      # free text, building-specific

A usage value exists only where the engine treats it differently, so the vocabulary is closed: a new access class means new code, not new config. Check a programme with:

python experiments/audit_programme_config.py

which reports reserved-name collisions, the usage class each code picks up, and whether each room's size/width/proportion/crinkliness targets are mutually satisfiable at all.