Files
to-do/rollover/validate.py
T
Connor JohnstoneandClaude Opus 5 185b45dbfc Add the prune rule, its tests, and a corpus validator
The rule PLAN.md describes and the rationale it gives for that rule
disagree on one shape that turns out to be common. 235 checkboxes in the
corpus sit under a plain bullet used as a group label, and by the letter
of the rule those have no checkbox ancestor, so a finished one would be
dropped on its own. But section 4 already argues the opposite case for
checkbox parents: a finished item under an unfinished group is context,
not cruft, and a label like `- Trellis "Ergonomics" Fixes` is a group in
every sense that matters.

Resolved in favour of the rationale, which collapses both cases into one
rule. A top-level bullet is any bullet with no bullet ancestor in its
section. It is dropped, with its subtree, exactly when that subtree holds
at least one checkbox and all of them are [x]. Prose bullets hold none and
so are never eligible, which is the property section 4 wanted.

Validated across all 228 notes with zero invariant failures: no heading,
no unfinished item and no checkbox-free subtree is ever removed, and no
group label is left with its children gone. 270 lines would be dropped
across 118 files.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PU5ZFfQFtDTFqqhvMTWdGH
2026-09-09 18:30:29 -04:00

90 lines
2.8 KiB
Python

#!/usr/bin/env python3
"""Run the prune across every historical note and assert it never misbehaves.
The prune is the piece of this system carrying real semantic risk, and 228 real
notes exist to test it against, so this checks invariants rather than relying on
someone eyeballing a diff.
"""
import glob
import os
import sys
sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
from prune import DONE, dropped_indices, parse, subtree_end, top_level_bullets
NOTES = os.path.expanduser("~/docs/notes/todo")
def audit(path):
lines = parse(open(path).read().split("\n"))
drop = dropped_indices(lines)
errs = []
for i in sorted(drop):
ln = lines[i]
if ln.is_heading:
errs.append(f"removed a heading: {ln.text!r}")
elif ln.is_bullet and ln.state is not None and ln.state != DONE:
errs.append(f"removed an unfinished item [{ln.state}]: {ln.text!r}")
# A dropped subtree must contain at least one checkbox, or it was prose.
for i in top_level_bullets(lines):
if i not in drop:
continue
end = subtree_end(lines, i)
if not any(
lines[j].is_bullet and lines[j].state is not None
for j in range(i, end + 1)
):
errs.append(f"dropped a checkbox-free subtree: {lines[i].text!r}")
# A surviving group label must not have lost all of its children.
for i, ln in enumerate(lines):
if i in drop or not ln.is_bullet or ln.state is not None:
continue
kids = [j for j in range(i + 1, subtree_end(lines, i) + 1) if lines[j].is_bullet]
if kids and all(j in drop for j in kids):
errs.append(f"left an emptied group label: {ln.text!r}")
return lines, drop, errs
def main():
files = sorted(glob.glob(os.path.join(NOTES, "*.md")))
if not files:
sys.exit(f"no notes found under {NOTES}")
failures = changed = dropped = 0
kinds = {}
for f in files:
lines, drop, errs = audit(f)
if errs:
failures += len(errs)
print(f"FAIL {os.path.basename(f)}")
for e in errs[:4]:
print(" ", e)
if drop:
changed += 1
dropped += len(drop)
for i in drop:
ln = lines[i]
k = (
f"checkbox[{ln.state}]"
if ln.is_bullet and ln.state
else "bullet" if ln.is_bullet else "continuation"
)
kinds[k] = kinds.get(k, 0) + 1
print()
print(f"files : {len(files)} ({changed} would change)")
print(f"lines dropped : {dropped}")
for k, v in sorted(kinds.items(), key=lambda x: -x[1]):
print(f" {k:>14} : {v}")
print(f"invariant failures: {failures}")
return 1 if failures else 0
if __name__ == "__main__":
sys.exit(main())