Files
learnharness/scripts/replay-layout.ts
T
amosblomqvist c86430e8bf Wave 7 chunk B: remove the embeddings machinery — the ladder collapses
The owner's wave-7 ruling: embeddings die entirely (git history is the
stash); clustering is birth-session derived (chunk A, 950ddfd), labels
are session names, and the prober's graph tools replace injection.

- settings: GraphSettings collapses to { enabled } — embedding,
  embeddingModel, clustering, labeling all die. Migration story: stale
  record keys are UNREAD and dropped by the sanitizer's reconstruction;
  the next graph-section PUT (replace-whole) persists the collapsed
  shape. The settings-panel GraphSection renders one master row.
- deleted: src/server/embeddings.ts (adapters, resolveEmbeddingPortFor,
  embedNewNodes, embeddingPort, embeddingModelCatalog), the
  GET /api/models/embeddings route, scripts/embed-graph.ts, tests for
  all three, and the /api/models/embeddings picker surfaces in the UI.
- prices: EMBEDDING_PRICES overlay + callCost fallback die. cost-summary:
  the embedding stage fold dies — back to three stages (extraction/
  audit/naming) per the owner ruling; historical stages.embedding spend
  in old run records is simply not displayed.
- ports: EmbeddingPort/EmbeddingResult/EmbeddingTask die. The embeddings
  LEDGER surface (readEmbeddings/writeEmbeddings + EmbeddingsLedger +
  EMBEDDINGS_FILE + graph-deletion's vector pruning) STAYS — inert data,
  kept only so graph-context.ts compiles until Teacher's pass removes
  it; the seam dies in the follow-up interlude.
- chat-handler (approved cross-lane unwiring, wave-6 precedent): the
  extraction pipeline's embedder becomes a documented no-op (nothing
  embedded, nothing spent, ledger untouched) until Teacher's pass
  removes the phase; the graph-context injection dispatch gate, the
  graphContext deps wiring, the setContextRespawn seam, and
  turn-service's triggerGraphContext are gone — injection is dead
  code pending Teacher's file deletions.
- scripts: backfill-graph/import-notes pass the same no-op embedder;
  replay-layout derives clusters from birthSessionPartition (the same
  partition the real settle uses) instead of the embeddings ledger.
- settings-body: parseGraphSettingsBody reads the master switch only —
  old-shape keys in a body are ignored, not rejected; bad shapes 400.
- next.config: the @huggingface tracing-exclusion guards die with the
  last reference.

Cross-lane flags: extraction.ts/graph-context.ts keep stale embed
comments + the embed phase until Teacher's full pass (their parallel
interlude owns those files — zero overlap kept by design);
semantic-clusters.ts + its test stay one more round (extraction.test.ts,
Teacher's file, still imports it — they delete it with the namer cases);
ChatSettings.graphContextTokens + the graph-context-control UI stay
until Teacher+UI.

Gates: tsc clean, eslint clean, 885/885 tests, build compiles.
Live-verified on the port-3004 dev tree: PUT /api/settings/graph with
dead keys collapses on write, bad shapes 400, GET reads {enabled: true},
/api/models/embeddings 404s.
2026-09-23 15:02:55 +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);
// 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();