add(named): the paper measure with machine checked counterexamples to its invariants (M5 metatheory)

This commit is contained in:
milner committed 2026-09-23 01:30:00 +02:00
1 parent 6235dee2ee
commit 70270ceb22
34 files changed
+4067 -2568

No files matched your search

+1 -1
View File
@@ -35,7 +35,7 @@ done < <(find theory -name '*.v' -print0 2>/dev/null)
echo "== Print Assumptions =="
WORK="$(mktemp -d ./.audit-work.XXXXXX)"
cp theory/*.v "$WORK"/
for f in ExecReducer Binding Reduction Metatheory Closure Subsystem Substitution Parallel Metaterm NamedMeta NamedEs Tests Random Enumerate; do
for f in ExecReducer Binding Reduction Metatheory Closure Subsystem Substitution Parallel Metaterm NamedMeta NamedEs NamedMeasure Tests Random Enumerate; do
echo "-- $f"
out="$(rocq compile -Q "$WORK" LambdaSub "$WORK/$f.v" 2>&1 || true)"
echo "$out"
+137 -102
View File
@@ -9,28 +9,16 @@ ROOT = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
GRAPH_DIR = os.path.join(ROOT, "graphs")
META = os.path.join(GRAPH_DIR, "theorems.json")
STATUS_FILL = {
"proved": "#d9ead3",
"stated": "#fff2cc",
"planned": "#f3f3f3",
"blocked": "#f4cccc",
}
STATUS_PEN = {
"proved": "#38761d",
"stated": "#b45f06",
"planned": "#777777",
"blocked": "#cc0000",
}
STATUS_STYLE = {
"proved": "filled",
"stated": "filled,dashed",
"planned": "dashed",
"blocked": "filled,bold",
}
INK = "#333333"
RULE_B = "#990000"
RULE_R = "#674ea7"
RULE_G = "#38761d"
INK = "#000000"
MUTED = "#000000"
FONT = "#000000"
FILL = "#ffffff"
EXT_FILL = "#ffffff"
NF_FILL = "#ffffff"
RULE_B = "#000000"
RULE_R = "#000000"
RULE_G = "#000000"
def q(s):
@@ -46,110 +34,146 @@ def esc_html(s):
return str(s).replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;")
def dependency_dot(data):
nodes = data["nodes"]
by_id = {n["id"]: n for n in nodes}
title = data["paper"]["title"] + " - theorem dependency (" + data["current"] + ")"
legend = (
"<B>" + esc_html(title) + "</B>"
"<BR/><FONT POINT-SIZE=\"9\">"
"<FONT COLOR=\"#38761d\">proved</FONT> "
"<FONT COLOR=\"#b45f06\">stated</FONT> "
"<FONT COLOR=\"#777777\">planned</FONT> "
"<FONT COLOR=\"#cc0000\">blocked</FONT>"
"</FONT>"
def node_line(n, external=False):
if external:
label = n["id"] + "\\n(" + n["milestone"] + ")"
shape = "box"
style = "rounded,dashed,filled"
fill = EXT_FILL
color = MUTED
fontcolor = FONT
pen = 0.9
else:
label = n["id"]
shape = "box" if n["kind"] == "paper" else "ellipse"
style = "rounded,filled" if n["kind"] == "paper" else "filled"
fill = FILL
color = INK
fontcolor = FONT
pen = 1.0
tip = n["id"]
if n.get("section"):
tip += " (section " + n["section"] + ")"
if n.get("file"):
tip += " " + n["file"]
tip += " " + n["status"]
return (
" %s [label=%s, shape=%s, style=%s, fillcolor=%s, color=%s, "
"penwidth=%.1f, fontcolor=%s, tooltip=%s];"
% (n["id"], q(label), shape, q(style), q(fill), q(color), pen, q(fontcolor), q(tip))
)
lines = [
"digraph theorem_deps {",
def dep_header(title):
return [
"digraph deps {",
" rankdir=LR;",
' bgcolor="white";',
" splines=spline;",
" nodesep=0.22;",
" ranksep=0.6;",
' node [fontname=Inter, fontsize=9, fontcolor="#222222"];',
' edge [fontname=Inter, fontsize=8, color="#9a9a9a", arrowsize=0.6, penwidth=0.8];',
" graph [fontname=Inter, fontsize=12, labelloc=t, label=<%s>];" % legend,
' bgcolor="#ffffff";',
" splines=polyline;",
" concentrate=true;",
" nodesep=0.2;",
" ranksep=0.9;",
' node [fontname=Inter, fontsize=9];',
' edge [fontname=Inter, fontsize=8, color="%s", fontcolor="%s", arrowsize=0.6, penwidth=0.8];' % (MUTED, FONT),
' graph [fontname=Inter, fontsize=12, labelloc=t, fontcolor="%s", label=<<B>%s</B>>];' % (FONT, esc_html(title)),
]
for n in nodes:
fill = STATUS_FILL.get(n["status"], "#f3f3f3")
pen = STATUS_PEN.get(n["status"], "#777777")
style = STATUS_STYLE.get(n["status"], "dashed")
tip = n["id"]
if n.get("section"):
tip += " (section " + n["section"] + ")"
if n.get("file"):
tip += " " + n["file"]
tip += " " + data["paper"]["html"]
if n["kind"] == "paper":
shape = "box"
style = "rounded," + style
else:
shape = "ellipse"
attrs = [
"label=%s" % q(n["label"]),
"color=%s" % q(pen),
"fillcolor=%s" % q(fill),
"style=%s" % q(style),
"fontcolor=%s" % q("#222222"),
"tooltip=%s" % q(tip),
"shape=%s" % shape,
]
lines.append(" %s [%s];" % (n["id"], ", ".join(attrs)))
for n in nodes:
for d in n.get("depends", []):
def dependency_overview_dot(data):
by_id = {n["id"]: n for n in data["nodes"]}
counts = {}
for n in data["nodes"]:
counts[n["milestone"]] = counts.get(n["milestone"], 0) + 1
milestones = [m for m in data["milestones"] if counts.get(m)]
edges = set()
for n in data["nodes"]:
for d in n["depends"]:
if d in by_id:
a = by_id[d]["milestone"]
b = n["milestone"]
if a != b and a in counts and b in counts:
edges.add((a, b))
lines = dep_header("Theorem dependency by milestone")
for m in milestones:
label = "%s\\n%d results" % (m, counts[m])
lines.append(
' %s [label=%s, shape=box, style="rounded,filled", fillcolor=%s, '
"color=%s, fontcolor=%s];" % (m, q(label), q(FILL), q(INK), q(FONT))
)
for a, b in sorted(edges):
lines.append(" %s -> %s;" % (a, b))
lines.append("}")
return "\n".join(lines) + "\n"
def dependency_subset_dot(data, title, keep_ids):
by_id = {n["id"]: n for n in data["nodes"]}
inside = [n for n in data["nodes"] if n["id"] in keep_ids]
ids = {n["id"] for n in inside}
ext_ids = sorted(
{d for n in inside for d in n["depends"] if d not in ids and d in by_id}
)
lines = dep_header(title)
for d in ext_ids:
lines.append(node_line(by_id[d], external=True))
for n in sorted(inside, key=lambda x: (x["file"], x["id"])):
lines.append(node_line(n))
keep = ids | set(ext_ids)
for n in inside:
for d in n["depends"]:
if d in keep:
lines.append(" %s -> %s;" % (d, n["id"]))
lines.append("}")
return "\n".join(lines) + "\n"
def rules_dot():
box = 'shape=box, style="rounded,filled", fillcolor="#f5f5f5", color="%s", fontcolor="%s"' % (INK, INK)
box = 'shape=box, style="rounded,filled", fillcolor="%s", color="%s", fontcolor="%s"' % (FILL, INK, FONT)
lines = [
"digraph lambda_sub_cfg {",
" rankdir=TB;",
" bgcolor=\"white\";",
" node [fontname=Inter, fontsize=10];",
" edge [fontname=Inter, fontsize=9, color=\"%s\"];" % INK,
" start [label=\"term\", shape=oval, style=filled, fillcolor=\"#f3f3f3\", color=\"%s\"];" % INK,
' bgcolor="#ffffff";',
' node [fontname=Inter, fontsize=10, fontcolor="%s"];' % FONT,
' edge [fontname=Inter, fontsize=9, color="%s", fontcolor="%s"];' % (MUTED, FONT),
' start [label="term", shape=oval, style=filled, fillcolor="%s", color="%s"];' % (FILL, INK),
" beta [label=\"(lambda x. t) u\", %s];" % box,
" es [label=\"t[x/u]\", %s];" % box,
" pure [label=\"pure term\", shape=oval, style=filled, fillcolor=\"#f3f3f3\", color=\"%s\"];" % INK,
" nf [label=\"normal form\", shape=doublecircle, style=filled, fillcolor=\"#d9ead3\", color=\"%s\"];" % RULE_G,
" start -> beta [label=\"App(Lam,_)\"];",
" start -> es [label=\"ESub\"];",
" start -> pure [label=\"BVar/FVar/Lam\"];",
" beta -> es [label=\"B\", color=\"%s\"];" % RULE_B,
" es -> es [label=\"R (one occurrence)\", color=\"%s\"];" % RULE_R,
" es -> pure [label=\"Gc (x not in fv)\", color=\"%s\"];" % RULE_G,
" pure -> pure [label=\"B under context\", color=\"%s\"];" % RULE_B,
" es -> nf [label=\"R/Gc normalisation\", color=\"%s\"];" % RULE_R,
' pure [label="pure term", shape=oval, style=filled, fillcolor="%s", color="%s"];' % (FILL, INK),
' nf [label="normal form", shape=doublecircle, style=filled, fillcolor="%s", color="%s"];' % (NF_FILL, RULE_G),
' start -> beta [label="App(Lam,_)"];',
' start -> es [label="ESub"];',
' start -> pure [label="BVar/FVar/Lam"];',
' beta -> es [label="B", color="%s"];' % RULE_B,
' es -> es [label="R (one occurrence)", color="%s"];' % RULE_R,
' es -> pure [label="Gc (x not in fv)", color="%s"];' % RULE_G,
' pure -> pure [label="B under context", color="%s"];' % RULE_B,
' es -> nf [label="R/Gc normalisation", color="%s"];' % RULE_R,
"}",
]
return "\n".join(lines) + "\n"
def reduction_dot():
box = 'shape=box, style="filled", fillcolor="#f5f5f5", color="%s", fontcolor="%s"' % (INK, INK)
box = 'shape=box, style="filled", fillcolor="%s", color="%s", fontcolor="%s"' % (FILL, INK, FONT)
lines = [
"digraph reduction_dup {",
" rankdir=LR;",
" bgcolor=\"white\";",
" node [fontname=Inter, fontsize=10];",
" edge [fontname=Inter, fontsize=9, color=\"%s\"];" % INK,
" graph [labelloc=t, fontcolor=\"%s\", label=\"(lambda x. x x) y reduction graph\"];" % INK,
' bgcolor="#ffffff";',
' node [fontname=Inter, fontsize=10, fontcolor="%s"];' % FONT,
' edge [fontname=Inter, fontsize=9, color="%s", fontcolor="%s"];' % (MUTED, FONT),
' graph [labelloc=t, fontcolor="%s", label="(lambda x. x x) y reduction graph"];' % FONT,
" n0 [label=\"(lambda x. x x) y\", %s];" % box,
" n1 [label=\"(x x)[x/y]\", %s];" % box,
" n2 [label=\"(y x)[x/y]\", %s];" % box,
" n2b [label=\"(x y)[x/y]\", %s];" % box,
" n3 [label=\"(y y)[x/y]\", %s];" % box,
" nf [label=\"y y (normal form)\", shape=doublecircle, style=filled, fillcolor=\"#d9ead3\", color=\"%s\", fontcolor=\"%s\"];" % (RULE_G, INK),
" n0 -> n1 [label=\"B\", color=\"%s\"];" % RULE_B,
" n1 -> n2 [label=\"R\", color=\"%s\"];" % RULE_R,
" n1 -> n2b [label=\"R\", color=\"%s\"];" % RULE_R,
" n2 -> n3 [label=\"R\", color=\"%s\"];" % RULE_R,
" n2b -> n3 [label=\"R\", color=\"%s\"];" % RULE_R,
" n3 -> nf [label=\"Gc\", color=\"%s\"];" % RULE_G,
' nf [label="y y (normal form)", shape=doublecircle, style=filled, fillcolor="%s", color="%s", fontcolor="%s"];' % (NF_FILL, RULE_G, FONT),
' n0 -> n1 [label="B", color="%s"];' % RULE_B,
' n1 -> n2 [label="R", color="%s"];' % RULE_R,
' n1 -> n2b [label="R", color="%s"];' % RULE_R,
' n2 -> n3 [label="R", color="%s"];' % RULE_R,
' n2b -> n3 [label="R", color="%s"];' % RULE_R,
' n3 -> nf [label="Gc", color="%s"];' % RULE_G,
" {rank=same; n2; n2b;}",
"}",
]
@@ -192,11 +216,22 @@ def main():
print("wrote", os.path.relpath(args.meta, ROOT))
else:
data = load(args.meta)
outputs = {
"dependency.dot": dependency_dot(data),
"rules.dot": rules_dot(),
"reduction.dot": reduction_dot(),
}
counts = {}
for n in data["nodes"]:
counts[n["milestone"]] = counts.get(n["milestone"], 0) + 1
outputs = {"dependency-overview.dot": dependency_overview_dot(data)}
for m in data["milestones"]:
if not counts.get(m):
continue
keep = {n["id"] for n in data["nodes"] if n["milestone"] == m}
title = "Theorem dependencies %s, %d results" % (m, counts[m])
outputs["dependency-%s.dot" % m] = dependency_subset_dot(data, title, keep)
outputs["rules.dot"] = rules_dot()
outputs["reduction.dot"] = reduction_dot()
ok = True
for name, text in outputs.items():
path = os.path.join(args.outdir, name)