sorokin.oleg.edu/learnharness
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 6 Wiki Activity

fix: align compaction cuts and native usage accounting

Browse Source
This commit is contained in:
amosblomqvist
2026-10-24 19:05:02 +03:00
parent 50a2db775d
commit 624c91a437
45 changed files with 775 additions and 2329 deletions
Download Patch File Download Diff File Expand all files Collapse all files
scripts/sync-pricing.ts
-121
View File
@@ -1,121 +0,0 @@
/**
* Regenerate src/core/prices.ts tables from pi-ai's shipped per-provider
* catalogs (providers/data/*.json).
*
* Usage: npx tsx scripts/sync-pricing.ts
*
* The catalogs are pi-ai's curated, versioned snapshots — the SAME source
* piEngineFor resolves engines from and /api/models serves the picker
* from — so the money channel's lookup table can never disagree with the
* engine. Fully offline (the JSON ships in node_modules); refreshes happen
* by bumping @earendil-works/pi-ai and re-running this script. The tables
* are generated; hand edits are overwritten. A model missing from the
* catalogs gets no entry — cost renders as absent ("—"), never zero.
*
* Key format (model slots wave, 2026-09-20): uniformly provider-qualified
* ("provider:id", incl. "openrouter:<slug>"). Owner decision: no legacy
* compat — records stamped with pre-wave bare slugs simply render unpriced.
*/
import * as fs from "node:fs";
import * as path from "node:path";
const DATA_DIR = path.join(
import.meta.dirname,
"..",
"node_modules",
"@earendil-works",
"pi-ai",
"dist",
"providers",
"data",
);
const OUT = path.join(import.meta.dirname, "..", "src", "core", "prices.ts");
/** One raw catalog entry (fields we consume; the files carry more). */
interface CatalogModel {
id: string;
name: string;
/** pi-ai 1.0 (catalog schema v6): every entry is type-tagged. */
type?: "chat" | "image" | "classifier";
cost: { input: number; output: number; cacheRead: number; cacheWrite: number };
}
/** The pure walk: every catalog entry that carries pricing, keyed
* "provider:id". Exported for the drift-guard test (tests/prices-sync)
* so the guard asserts EXACTLY what this script would generate — the
* test and the generator can never disagree (owner, 2026-09-29). */
export function buildPrices(): Map<string, CatalogModel["cost"]> {
const priced = new Map<string, CatalogModel["cost"]>();
for (const file of fs.readdirSync(DATA_DIR).filter((f) => f.endsWith(".json"))) {
const provider = file.slice(0, -".json".length);
const catalog = JSON.parse(fs.readFileSync(path.join(DATA_DIR, file), "utf-8")) as Record<
string,
Record<string, CatalogModel>
>;
for (const models of Object.values(catalog)) {
for (const [id, m] of Object.entries(models)) {
// pi-ai 1.0 (catalog schema v6): keys are type-qualified
// ("chat:<id>", "image:<id>", "classifier:<id>") and entries
// carry their own `type`. We price CHAT models only — our
// usage records are chat — and keys stay "provider:id"
// (owner format, 2026-09-20): strip the chat: prefix, skip the
// rest.
if (m.type && m.type !== "chat") continue;
const plainId = id.startsWith("chat:") ? id.slice("chat:".length) : id;
// A catalog entry without pricing data gets no key at all
// (cost renders "—", never zero).
if (!m.cost) continue;
// First writer wins (an id can appear under two apis within one
// file with identical pricing; cross-file collisions cannot
// happen — keys are provider-qualified).
priced.set(`${provider}:${plainId}`, m.cost);
}
}
}
return priced;
}
function main(): void {
const priced = new Map<string, { cost: CatalogModel["cost"] }>(
[...buildPrices()].map(([k, cost]) => [k, { cost }] as const),
);
// Sort by key for a stable, diff-friendly generated table.
const rows = [...priced.entries()]
.sort(([a], [b]) => a.localeCompare(b))
.map(([key, m]) => ({
price: `\t${JSON.stringify(key)}: { input: ${m.cost.input}, output: ${m.cost.output}, cacheRead: ${m.cost.cacheRead}, cacheWrite: ${m.cost.cacheWrite} },`,
}));
const tables = `/** Token prices, USD per 1M tokens. A missing model makes cost render
* as absent ("—"), never zero. Keys are uniformly provider-qualified
* ("provider:id") — NO legacy bare-slug compat (owner decision, sync
* script 2026-09-20): pre-wave records stamped with bare slugs render
* unpriced.
* Context-window limits are NOT a table (owner amendment, 2026-09-23):
* every meter's denominator reads pi's LIVE catalog (GET /api/models /
* the activity feed's server-side join) — missing model, usage-only.
* Refresh: bump @earendil-works/pi-ai, then npx tsx scripts/sync-pricing.ts
* (reads pi-ai's shipped catalogs — the same source the engine and the
* picker use). Forgetting the re-run is exactly how models go unpriced
* (deepseek-v4.1-flash, 2026-09-29) — the drift-guard test
* (tests/prices-sync.test.ts) fails the gates until you do. */
export const PRICES: Record<string, ModelPrice> = {
${rows.map((r) => r.price).join("\n")}
};
`;
const src = fs.readFileSync(OUT, "utf-8");
const start = src.indexOf("/** Token prices");
const end = src.indexOf("// ─── Pure helpers");
if (start === -1 || end === -1) throw new Error("prices.ts markers not found — file structure changed");
fs.writeFileSync(OUT, src.slice(0, start) + tables + "\n" + src.slice(end), "utf-8");
console.log(`prices.ts regenerated from pi-ai's catalogs: ${priced.size} keys priced.`);
}
// Run only as a script, not when imported (the drift-guard test).
if (process.argv[1] && import.meta.filename && path.resolve(process.argv[1]) === import.meta.filename) {
main();
}