@openpond/evals 0.10.0 → 0.10.1

package/CONTRACT.md

@@ -143,3 +143,9 @@ and excludes missing rewards. An empty scorable population remains null.
 `aggregateTasksetRunInWorker` from `@openpond/evals/metrics/node` verifies custom
 module bytes and uses the existing bounded isolated worker. A fixture metric
 does not qualify a model for learning or acceptance.
+
+`aggregateTasksetRunInProcess` and `executeTasksetMetricInProcess` use an owned
+Node subprocess for hosts without Node worker APIs. It strips authored TypeScript
+inside its deadline, evaluates the verified source in the same deterministic
+isolate, and exits before success, failure or cancellation settles. Importing
+the metrics entrypoint does not require Node's type-stripping API on the host.

package/dist/javascript-isolate-process.js

@@ -6,10 +6,10 @@ import { assertBoundedTaskJson } from "./task-schema.js";
 export async function executeJavaScriptIsolateInProcess(input) {
     input.signal?.throwIfAborted();
     validateJavaScriptIsolateInput(input);
-    const { signal, ...data } = input;
+    const { signal, stripTypeScript, ...data } = input;
     const error = (suffix) => new Error(`${input.errorPrefix}_${suffix}`);
     // The authored source remains JSON data passed to QuickJS, never host code.
-    const program = `globalThis.__openpondIsolateInput = JSON.parse(${JSON.stringify(JSON.stringify(data))});\n${javascriptIsolateProcessSource}`;
+    const program = `globalThis.__openpondIsolateInput = JSON.parse(${JSON.stringify(JSON.stringify(data))});\n${stripTypeScript ? 'globalThis.__openpondIsolateInput.source = require("node:module").stripTypeScriptTypes(globalThis.__openpondIsolateInput.source, { mode: "strip" });\n' : ""}${javascriptIsolateProcessSource}`;
     return new Promise((resolve, reject) => {
         const child = spawn("node", ["--max-old-space-size=64", "--input-type=commonjs", "-"], {
             stdio: ["pipe", "pipe", "pipe"],

package/dist/metrics-node.js

@@ -1,5 +1,5 @@
-import { stripTypeScriptTypes } from "node:module";
 import { executeJavaScriptIsolateInWorker } from "./javascript-isolate-node.js";
+import { executeJavaScriptIsolateInProcess } from "./javascript-isolate-process.js";
 import { aggregateTasksetEvaluationReceipts, aggregateTasksetRunReceipts, executeTasksetMetric } from "./metrics.js";
 /** Verifies original source before stripping TS, then waits for worker termination
  * on success, error or cancellation. Authored imports and host APIs are unavailable. */
@@ -12,8 +12,22 @@ export function aggregateTasksetEvaluationInWorker(input) {
 export function aggregateTasksetRunInWorker(input) {
     return aggregateTasksetRunReceipts(input, executeMetric);
 }
-const executeMetric = ({ source, module, exportName, scores, timeoutMs, signal }) => executeJavaScriptIsolateInWorker({
-    source: /\.[cm]?ts$/.test(module) ? stripTypeScriptTypes(source, { mode: "strip" }) : source,
-    exportName, value: scores, timeoutMs, signal,
-    maxResultBytes: 1_024, deterministic: true, errorPrefix: "metric",
-});
+/** Hosted callers use the same Node process ownership as verifier/environment
+ * execution. Type stripping happens in that process within its deadline. */
+export function executeTasksetMetricInProcess(input) {
+    return executeTasksetMetric(input, executeProcessMetric);
+}
+export function aggregateTasksetRunInProcess(input) {
+    return aggregateTasksetRunReceipts(input, executeProcessMetric);
+}
+const executeMetric = async ({ source, module, exportName, scores, timeoutMs, signal }) => {
+    const javascript = /\.[cm]?ts$/.test(module)
+        ? (await import("node:module")).stripTypeScriptTypes(source, { mode: "strip" }) : source;
+    return executeJavaScriptIsolateInWorker({
+        source: javascript,
+        exportName, value: scores, timeoutMs, signal,
+        maxResultBytes: 1_024, deterministic: true, errorPrefix: "metric",
+    });
+};
+const executeProcessMetric = ({ source, module, exportName, scores, timeoutMs, signal }) => executeJavaScriptIsolateInProcess({ source, exportName, value: scores, timeoutMs, signal, stripTypeScript: /\.[cm]?ts$/.test(module),
+    maxResultBytes: 1_024, deterministic: true, errorPrefix: "metric" });

package/dist/types/javascript-isolate-process.d.ts

@@ -1,4 +1,6 @@
 import { type JavaScriptIsolateInput } from "./javascript-isolate.js";
 /** A Node subprocess owns the interpreter; completion always waits for its exit. */
-export declare function executeJavaScriptIsolateInProcess(input: JavaScriptIsolateInput): Promise<unknown>;
+export declare function executeJavaScriptIsolateInProcess(input: JavaScriptIsolateInput & {
+    stripTypeScript?: boolean;
+}): Promise<unknown>;
 //# sourceMappingURL=javascript-isolate-process.d.ts.map

package/dist/types/javascript-isolate-process.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"javascript-isolate-process.d.ts","sourceRoot":"","sources":["../../../../src/javascript-isolate-process.ts"],"names":[],"mappings":"AAEA,OAAO,EAAmE,KAAK,sBAAsB,EAAE,MAAM,yBAAyB,CAAC;AAGvI,oFAAoF;AACpF,wBAAsB,iCAAiC,CAAC,KAAK,EAAE,sBAAsB,GAAG,OAAO,CAAC,OAAO,CAAC,CAwDvG"}
+{"version":3,"file":"javascript-isolate-process.d.ts","sourceRoot":"","sources":["../../../../src/javascript-isolate-process.ts"],"names":[],"mappings":"AAEA,OAAO,EAAmE,KAAK,sBAAsB,EAAE,MAAM,yBAAyB,CAAC;AAGvI,oFAAoF;AACpF,wBAAsB,iCAAiC,CAAC,KAAK,EAAE,sBAAsB,GAAG;IAAE,eAAe,CAAC,EAAE,OAAO,CAAA;CAAE,GAAG,OAAO,CAAC,OAAO,CAAC,CAwDvI"}

package/dist/types/metrics-node.d.ts

@@ -6,4 +6,8 @@ import type { EvaluationResult } from "./runs.js";
 export declare function executeTasksetMetricInWorker(input: TasksetMetricExecutionInput): Promise<TasksetMetricResult>;
 export declare function aggregateTasksetEvaluationInWorker(input: TasksetEvaluationInput): Promise<EvaluationResult>;
 export declare function aggregateTasksetRunInWorker(input: TasksetRunEvaluationInput): Promise<TasksetMetricResult>;
+/** Hosted callers use the same Node process ownership as verifier/environment
+ * execution. Type stripping happens in that process within its deadline. */
+export declare function executeTasksetMetricInProcess(input: TasksetMetricExecutionInput): Promise<TasksetMetricResult>;
+export declare function aggregateTasksetRunInProcess(input: TasksetRunEvaluationInput): Promise<TasksetMetricResult>;
 //# sourceMappingURL=metrics-node.d.ts.map

package/dist/types/metrics-node.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"metrics-node.d.ts","sourceRoot":"","sources":["../../../../src/metrics-node.ts"],"names":[],"mappings":"AAEA,OAAO,EAAyF,KAAK,sBAAsB,EAAE,KAAK,yBAAyB,EAAE,KAAK,2BAA2B,EAA8B,MAAM,cAAc,CAAC;AAChP,OAAO,KAAK,EAAE,mBAAmB,EAAE,MAAM,oBAAoB,CAAC;AAC9D,OAAO,KAAK,EAAE,gBAAgB,EAAE,MAAM,WAAW,CAAC;AAElD;wFACwF;AACxF,wBAAgB,4BAA4B,CAAC,KAAK,EAAE,2BAA2B,GAAG,OAAO,CAAC,mBAAmB,CAAC,CAE7G;AAED,wBAAgB,kCAAkC,CAAC,KAAK,EAAE,sBAAsB,GAAG,OAAO,CAAC,gBAAgB,CAAC,CAE3G;AAED,wBAAgB,2BAA2B,CAAC,KAAK,EAAE,yBAAyB,GAAG,OAAO,CAAC,mBAAmB,CAAC,CAE1G"}
+{"version":3,"file":"metrics-node.d.ts","sourceRoot":"","sources":["../../../../src/metrics-node.ts"],"names":[],"mappings":"AAEA,OAAO,EAAyF,KAAK,sBAAsB,EAAE,KAAK,yBAAyB,EAAE,KAAK,2BAA2B,EAA8B,MAAM,cAAc,CAAC;AAChP,OAAO,KAAK,EAAE,mBAAmB,EAAE,MAAM,oBAAoB,CAAC;AAC9D,OAAO,KAAK,EAAE,gBAAgB,EAAE,MAAM,WAAW,CAAC;AAElD;wFACwF;AACxF,wBAAgB,4BAA4B,CAAC,KAAK,EAAE,2BAA2B,GAAG,OAAO,CAAC,mBAAmB,CAAC,CAE7G;AAED,wBAAgB,kCAAkC,CAAC,KAAK,EAAE,sBAAsB,GAAG,OAAO,CAAC,gBAAgB,CAAC,CAE3G;AAED,wBAAgB,2BAA2B,CAAC,KAAK,EAAE,yBAAyB,GAAG,OAAO,CAAC,mBAAmB,CAAC,CAE1G;AAED;4EAC4E;AAC5E,wBAAgB,6BAA6B,CAAC,KAAK,EAAE,2BAA2B,GAAG,OAAO,CAAC,mBAAmB,CAAC,CAE9G;AAED,wBAAgB,4BAA4B,CAAC,KAAK,EAAE,yBAAyB,GAAG,OAAO,CAAC,mBAAmB,CAAC,CAE3G"}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@openpond/evals",
-  "version": "0.10.0",
+  "version": "0.10.1",
   "description": "Portable Tasksets, graders, evaluation runs, receipts, conformance fixtures, and Work evidence for OpenPond Harnesses",
   "license": "MIT",
   "type": "module",