Files
learnharness/scripts/replay-layout.ts
T
amosblomqvist d1b46132b9 Wave 7 close-out: the embedding ledger seam dies; notes birth via mapping lines
Part 1 — the ledger + remnants interlude (master-dispatched, owner ruling
2026-09-23: nothing configures the graph beyond chat models):

- The embedding ledger surface is gone: readEmbeddings/writeEmbeddings
  ports, the EmbeddingsLedger contract, EMBEDDINGS_FILE (config, store
  impls, graphRevision source list), and graph-deletion's vector-pruning
  step. On-disk data files stay untouched (inert history).
- The extraction embedder seam dies wholesale: ExtractionPipelineDeps
  drops the embedder dep (chat-handler + backfill/import-notes stubs go
  with it), the 'embedding' phase leaves ExtractionPhase, the usage
  capture/persist helpers die, and stages.embedding leaves
  ExtractionRunUsage (historical records keep their shape — reads are
  lenient). Extraction is two stages again, like the local-embedder era.
- Consumer-less remnants: latestQuizOutcomes + QUIZ_EVENT_KINDS,
  clusterContentId, matchClusterLabelings + Community + ClusterLabeling,
  the readClusterLabels/appendClusterLabel port seams + store impls +
  LEDGER_FILES.clusterLabels. NamingRunUsage/readNamingRuns STAY
  (standing ruling — the cost pill renders historical naming spend);
  clusterHue + MIN_CLUSTER_SIZE stay (live).
- The 'labeler' role dies: BACKGROUND_MODEL_ROLES, ROLE_SLOT, ModelRole,
  the catalog pin projections, and the UI roster strings. RolePins keys
  shrink with the union.

The notes birth fallback (owner-approved 2026-09-23, surfaced by the
interlude's live smoke: production notes imports create provenance-EMPTY
nodes (importNotes passes messages: null — the file's anchor is the
mapping line), so the provenance-only birth rule left every imported
file unclustered and invisible to the prober's session list, against the
locked file-birth ruling. A provenance-empty node now births under its
FIRST mapping line's chatId (ledger order); only notes-born nodes ever
hit the fallback (chat nodes always carry provenance; a re-encountered
node's link-append re-opens the provenance rule). Deletion-safe: mapping
lines of deleted chats are stripped alongside their provenance.

Live-verified on port 3004: a fresh notes import clusters under its
src_ id with the file path as the label, end to end with no embedder.

845/845 tests, typecheck/lint/build green.
EOF
)
2026-09-23 15:37:34 +03:00

122 lines
4.9 KiB
TypeScript

/**
* Option A simulation: replay the CURRENT graph through the incremental
* layout as if it had arrived one imported-notes file at a time — the
* organic growth path the user described (one dependency structure /
* cluster at a time, orphans collecting on the rim).
*
* Reads ONLY the ledgers (nodes / edges / mappings / sources); writes ONLY
* data/graph/positions.json — no LLM calls, no extraction, no chat records.
*
* npx tsx scripts/replay-layout.ts [--depth=vertical|radial]
*
* --depth regenerates the ledger WITH inference-depth anchoring (the
* playground mode baked into the rest state); omit it for the free layout.
*/
import { graphStore } from "../src/store/graph.ts";
// --depth=vertical|radial: bake the playground depth mode into the ledger.
const depthArg = process.argv.find((a) => a.startsWith("--depth="));
const depthMode: "off" | "vertical" | "radial" =
depthArg === "--depth=vertical" ? "vertical" : depthArg === "--depth=radial" ? "radial" : "off";
import { incrementalLayout } from "../src/core/layout.ts";
import { birthSessionPartition } from "../src/core/session-clusters.ts";
import type { GraphNode, NodeMapping, Position } from "../src/core/contracts.ts";
function main() {
const t0 = Date.now();
// Unique nodes (lines may repeat an id — first defines).
const seen = new Set<string>();
const nodes: GraphNode[] = [];
for (const line of graphStore.readNodes()) {
if (seen.has(line.id)) continue;
seen.add(line.id);
nodes.push(line);
}
const edges = graphStore.readEdges();
// The cluster source (wave 7): the birth-session partition, derived on
// read — same as the real settle (core/layout-settle.ts).
const birthClusters = birthSessionPartition(nodes, graphStore.readMappings());
// Batches: each imported file, in real import order (sources.json
// importedAt). A node belongs to the file whose mapping first claims it
// (mapping line order = arrival); nodes no mapping covers (legacy) ride
// the final batch.
const sources = [...graphStore.readSources()].sort((a, b) => a.importedAt.localeCompare(b.importedAt));
const mappings: NodeMapping[] = graphStore.readMappings();
const batchOf = new Map<string, string>(); // nodeId → srcId
for (const m of mappings) {
if (!batchOf.has(m.graphNodeId)) batchOf.set(m.graphNodeId, m.chatId);
}
const delivered = new Set<string>();
let positions = new Map<string, Position>(Object.entries(graphStore.readPositions()));
if (positions.size > 0) {
console.log("[replay] discarding existing positions ledger — full replay from empty");
positions = new Map();
}
const byId = new Map(nodes.map((n) => [n.id, n]));
const unmappedFallback = sources[sources.length - 1]?.id ?? "";
console.log(`[replay] ${nodes.length} unique nodes, ${edges.length} edge lines, ${sources.length} sources`);
let maxOldShift = 0;
let shiftReport = "";
for (const source of sources) {
const batch = nodes.filter((n) => batchOf.get(n.id) === source.id);
const isLast = source.id === sources[sources.length - 1]!.id;
if (isLast) {
// Sweep in any never-mapped nodes with the last batch.
for (const n of nodes) {
if (!batchOf.has(n.id) && !delivered.has(n.id)) {
batch.push(n);
batchOf.set(n.id, unmappedFallback);
}
}
}
for (const n of batch) delivered.add(n.id);
// Edges activate once BOTH endpoints exist (a later file can claim
// one endpoint — the edge lights up then).
const activeEdges = edges.filter((e) => delivered.has(e.from) && delivered.has(e.to));
const before = new Map(positions);
const batchNodes = nodes.filter((n) => delivered.has(n.id));
positions = incrementalLayout(batchNodes, activeEdges, positions, {
// The same partition the real settle would see for this batch:
// the birth-session clusters, filtered to delivered nodes.
clusters: birthClusters
.map((c) => ({ members: c.members.filter((id) => delivered.has(id)) }))
.filter((c) => c.members.length > 0),
...(depthMode === "off"
? {}
: {
depthAnchor: {
mode: depthMode,
strength: 0.8,
spacing: 60,
...(depthMode === "radial" ? { baseRadius: 220 } : {}),
},
}),
});
for (const [id, prev] of before) {
const now = positions.get(id)!;
const shift = Math.hypot(prev.x - now.x, prev.y - now.y);
if (shift > maxOldShift) {
maxOldShift = shift;
shiftReport = `${id} ${prev.x.toFixed(0)},${prev.y.toFixed(0)} -> ${now.x.toFixed(0)},${now.y.toFixed(0)} (batch ${batch.length} new, ${activeEdges.length} edges)`;
}
}
const label = source.filePath.split("/").pop() ?? source.id;
console.log(
`[replay] +${String(batch.length).padStart(3)} nodes ${String(activeEdges.length).padStart(4)} edges active (${label})`,
);
}
graphStore.writePositions(Object.fromEntries(positions));
console.log(
`[replay] done: ${positions.size} positions written in ${((Date.now() - t0) / 1000).toFixed(1)}s` +
` (max old-node shift within a batch: ${maxOldShift.toFixed(1)}: ${shiftReport})`,
);
}
main();