@genroc/eval-node 0.0.0-edge.f16dfcef → 0.0.0-edge.f33f7a37
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +114 -306
- package/dist/eval.js +9 -17
- package/dist/import.js +60 -91
- package/dist/realm.js +12 -17
- package/dist/worker.js +61 -76
- package/eval.ts +13 -23
- package/import.ts +76 -105
- package/package.json +1 -1
- package/realm.ts +12 -17
- package/worker.ts +66 -82
package/dist/import.js
CHANGED
|
@@ -1,17 +1,11 @@
|
|
|
1
1
|
#!/usr/bin/env node
|
|
2
|
-
// The
|
|
3
|
-
// with
|
|
4
|
-
//
|
|
5
|
-
//
|
|
6
|
-
// Two modes, one binary: "types" writes the declarations an editor needs and returns no
|
|
7
|
-
// code; "build" typechecks and bundles. A separate types hook would mean a second `tsc`
|
|
8
|
-
// over the same project.
|
|
9
|
-
import { spawn } from "node:child_process";
|
|
2
|
+
// The typed-phase resolver: manifest on stdin, `{"values": [...]}` on stdout, diagnostic on stderr
|
|
3
|
+
// with a non-zero exit (specs/source-resolution.md). Two modes share one `tsc` pass: "generate"
|
|
4
|
+
// writes declarations only, "resolve" typechecks and bundles.
|
|
10
5
|
import { existsSync } from "node:fs";
|
|
11
|
-
import { access,
|
|
6
|
+
import { access, writeFile } from "node:fs/promises";
|
|
12
7
|
import { builtinModules } from "node:module";
|
|
13
|
-
import { dirname, join,
|
|
14
|
-
import { fileURLToPath } from "node:url";
|
|
8
|
+
import { dirname, join, resolve } from "node:path";
|
|
15
9
|
import commonjs from "@rollup/plugin-commonjs";
|
|
16
10
|
import json from "@rollup/plugin-json";
|
|
17
11
|
import { nodeResolve } from "@rollup/plugin-node-resolve";
|
|
@@ -30,11 +24,6 @@ async function exists(path) {
|
|
|
30
24
|
return false;
|
|
31
25
|
}
|
|
32
26
|
}
|
|
33
|
-
/** Creates the parent directory, which `.genroc-cache/` relies on: nothing else makes it. */
|
|
34
|
-
async function write(path, content) {
|
|
35
|
-
await mkdir(dirname(path), { recursive: true });
|
|
36
|
-
await writeFile(path, content);
|
|
37
|
-
}
|
|
38
27
|
// ── JSON Schema → TypeScript ───────────────────────────────────────────────────
|
|
39
28
|
/** Only genroc's keyword set is handled; anything else its strict decoder would have
|
|
40
29
|
* refused before this ran (internal/schema, allowedKeywords). */
|
|
@@ -127,7 +116,7 @@ function declarations(at) {
|
|
|
127
116
|
emitted.push(`export type ${identifier(next)} = ${body};`);
|
|
128
117
|
}
|
|
129
118
|
return [
|
|
130
|
-
"// Generated by genroc. Do not edit - regenerate with `genctl
|
|
119
|
+
"// Generated by genroc. Do not edit - regenerate with `genctl generate`.",
|
|
131
120
|
`// ${where.name} (${address(site.pointer)})`,
|
|
132
121
|
"",
|
|
133
122
|
...emitted,
|
|
@@ -144,9 +133,8 @@ function typesPathFor(scriptPath) {
|
|
|
144
133
|
return scriptPath.replace(/\.[^.\/]+$/, "") + ".genroc.d.ts";
|
|
145
134
|
}
|
|
146
135
|
// ── typecheck ──────────────────────────────────────────────────────────────────
|
|
147
|
-
/** The
|
|
148
|
-
*
|
|
149
|
-
* the project root: above it is not this project. */
|
|
136
|
+
/** The tsconfig the author's editor reads, so editor and apply agree. Stops at the project
|
|
137
|
+
* root. */
|
|
150
138
|
async function nearestTsconfig(from, root) {
|
|
151
139
|
for (let dir = from;; dir = dirname(dir)) {
|
|
152
140
|
const candidate = join(dir, "tsconfig.json");
|
|
@@ -157,11 +145,7 @@ async function nearestTsconfig(from, root) {
|
|
|
157
145
|
}
|
|
158
146
|
}
|
|
159
147
|
async function typecheck(sites) {
|
|
160
|
-
//
|
|
161
|
-
// coexist with it. The suffix is what keeps the scratch area out of its way.
|
|
162
|
-
const dir = join(root, ".genroc-cache");
|
|
163
|
-
await write(join(dir, ".gitignore"), "*\n");
|
|
164
|
-
// One tsc per distinct base config: `extends` takes a single base, so merging two would
|
|
148
|
+
// One program per distinct base config: `extends` takes a single base, so merging two would
|
|
165
149
|
// check each script under the other author's options.
|
|
166
150
|
const groups = new Map();
|
|
167
151
|
for (const at of sites) {
|
|
@@ -172,10 +156,10 @@ async function typecheck(sites) {
|
|
|
172
156
|
else
|
|
173
157
|
groups.set(base, [at]);
|
|
174
158
|
}
|
|
175
|
-
|
|
159
|
+
const diagnostics = [];
|
|
176
160
|
for (const [base, group] of groups) {
|
|
177
161
|
const config = {
|
|
178
|
-
...(base ? { extends:
|
|
162
|
+
...(base ? { extends: base } : {}),
|
|
179
163
|
compilerOptions: {
|
|
180
164
|
noEmit: true,
|
|
181
165
|
strict: true,
|
|
@@ -186,42 +170,36 @@ async function typecheck(sites) {
|
|
|
186
170
|
// `lib` DESCRIBES the realm and is written after `extends` so a base cannot widen it:
|
|
187
171
|
// a worker thread has no document, whatever an author's config claims.
|
|
188
172
|
lib: ["esnext", "webworker"],
|
|
189
|
-
// `types` is the author's
|
|
190
|
-
//
|
|
191
|
-
// base config there is nothing to opt in with, so the default stays none.
|
|
173
|
+
// `types` is the author's opt-in to node globals, which the realm has. With no base
|
|
174
|
+
// config there is nothing to opt in with, so it stays none.
|
|
192
175
|
...(base ? {} : { types: [] }),
|
|
193
176
|
},
|
|
194
|
-
files: group.flatMap((s) => [
|
|
195
|
-
relative(dir, s.file),
|
|
196
|
-
relative(dir, typesPathFor(s.file)),
|
|
197
|
-
]),
|
|
177
|
+
files: group.flatMap((s) => [s.file, typesPathFor(s.file)]),
|
|
198
178
|
// `files` overrides the base's, but a base `include` survives beside it and would
|
|
199
179
|
// drag the author's whole tree in, to be checked under the worker lib.
|
|
200
180
|
include: [],
|
|
201
181
|
};
|
|
202
|
-
|
|
203
|
-
|
|
204
|
-
|
|
182
|
+
// Never written; anchors `@types` lookup at the root. Not `tsconfig.json`: that is often the
|
|
183
|
+
// base it extends, and tsc refuses a circular extends.
|
|
184
|
+
const parsed = ts.parseJsonConfigFileContent(config, ts.sys, root, undefined, join(root, "tsconfig.genroc.json"));
|
|
185
|
+
const program = ts.createProgram({
|
|
186
|
+
rootNames: parsed.fileNames,
|
|
187
|
+
options: parsed.options,
|
|
188
|
+
projectReferences: parsed.projectReferences,
|
|
189
|
+
configFileParsingDiagnostics: parsed.errors,
|
|
190
|
+
});
|
|
191
|
+
diagnostics.push(...parsed.errors, ...ts.getPreEmitDiagnostics(program));
|
|
205
192
|
}
|
|
206
|
-
|
|
207
|
-
|
|
208
|
-
|
|
209
|
-
|
|
210
|
-
|
|
211
|
-
|
|
212
|
-
|
|
213
|
-
|
|
214
|
-
|
|
215
|
-
|
|
216
|
-
proc.stderr.on("data", (c) => (err += c));
|
|
217
|
-
const code = await new Promise((resolve, reject) => {
|
|
218
|
-
proc.on("error", reject);
|
|
219
|
-
proc.on("close", (c) => resolve(c ?? 1));
|
|
220
|
-
});
|
|
221
|
-
if (code !== 0) {
|
|
222
|
-
// tsc reports on stdout; the exit code IS the type check, so this is the diagnostic
|
|
223
|
-
// genctl surfaces and the reason a failed import never produces a string.
|
|
224
|
-
die([out, err].filter(Boolean).join("\n").trimEnd());
|
|
193
|
+
if (diagnostics.length > 0) {
|
|
194
|
+
// The diagnostics are the whole type check: this is what genctl surfaces, and the reason a
|
|
195
|
+
// failed import never produces a string.
|
|
196
|
+
die(ts
|
|
197
|
+
.formatDiagnostics(diagnostics, {
|
|
198
|
+
getCanonicalFileName: (f) => f,
|
|
199
|
+
getCurrentDirectory: () => root,
|
|
200
|
+
getNewLine: () => "\n",
|
|
201
|
+
})
|
|
202
|
+
.trimEnd());
|
|
225
203
|
}
|
|
226
204
|
}
|
|
227
205
|
// ── bundle ─────────────────────────────────────────────────────────────────────
|
|
@@ -248,11 +226,9 @@ const BUILTIN = new Set([
|
|
|
248
226
|
...builtinModules,
|
|
249
227
|
...builtinModules.map((m) => `node:${m}`),
|
|
250
228
|
]);
|
|
251
|
-
/** Resolves
|
|
252
|
-
*
|
|
253
|
-
*
|
|
254
|
-
* A package resolving to a `.d.ts` is declined — that is a type, not the implementation —
|
|
255
|
-
* which leaves node_modules to nodeResolve. */
|
|
229
|
+
/** Resolves through TYPESCRIPT under the typecheck's config, so a `paths` alias that compiles
|
|
230
|
+
* also bundles. A `.d.ts` resolution is declined (a type, not code), leaving node_modules to
|
|
231
|
+
* nodeResolve. */
|
|
256
232
|
function tsResolve(configPath) {
|
|
257
233
|
let options = {};
|
|
258
234
|
if (configPath) {
|
|
@@ -273,14 +249,12 @@ function tsResolve(configPath) {
|
|
|
273
249
|
},
|
|
274
250
|
};
|
|
275
251
|
}
|
|
276
|
-
/**
|
|
277
|
-
*
|
|
278
|
-
* Bundling is entirely the importer's job, so a definition version pins its code forever. */
|
|
252
|
+
/** A self-contained ES module whose default export is the author's own, unwrapped. Bundling here
|
|
253
|
+
* means a definition version pins its code forever. */
|
|
279
254
|
async function bundle(at) {
|
|
280
255
|
const site = at.site;
|
|
281
|
-
// Builtins are
|
|
282
|
-
//
|
|
283
|
-
// of a module that will not be there, which bundles clean and fails at runtime.
|
|
256
|
+
// Builtins are externalised; anything else unresolved is a REFUSAL, since rollup's default
|
|
257
|
+
// leaves an import that bundles clean and fails at runtime.
|
|
284
258
|
const built = await rollup({
|
|
285
259
|
input: at.file,
|
|
286
260
|
external: (id) => BUILTIN.has(id),
|
|
@@ -288,8 +262,7 @@ async function bundle(at) {
|
|
|
288
262
|
tsResolve(await nearestTsconfig(dirname(at.file), root)),
|
|
289
263
|
nodeResolve({ extensions: [".ts", ".tsx", ".mjs", ".js", ".json"] }),
|
|
290
264
|
commonjs(),
|
|
291
|
-
//
|
|
292
|
-
// did natively; without it rollup hands the JSON to the JS parser.
|
|
265
|
+
// Inlines `.json` imports; without it rollup hands JSON to the JS parser.
|
|
293
266
|
json(),
|
|
294
267
|
transpile,
|
|
295
268
|
],
|
|
@@ -334,20 +307,18 @@ function address(pointer) {
|
|
|
334
307
|
.join("")
|
|
335
308
|
.replace(/^\./, "");
|
|
336
309
|
}
|
|
337
|
-
const located = manifest.processes.flatMap((where) => where.sites.map((site) =>
|
|
338
|
-
site
|
|
339
|
-
|
|
340
|
-
|
|
341
|
-
})
|
|
342
|
-
|
|
343
|
-
|
|
344
|
-
|
|
345
|
-
//
|
|
310
|
+
const located = manifest.processes.flatMap((where) => where.sites.map((site) => {
|
|
311
|
+
const [script, ...rest] = site.args;
|
|
312
|
+
if (script === undefined || rest.length > 0) {
|
|
313
|
+
die(`${join(where.dir, where.file)}: ${address(site.pointer)}: $import takes one script, ` +
|
|
314
|
+
`and is given ${site.args.length} arguments`);
|
|
315
|
+
}
|
|
316
|
+
return { site, where, file: resolve(where.dir, script) };
|
|
317
|
+
}));
|
|
318
|
+
// The `code`-field contract is this resolver's, not genctl's, so it checks the directive landed
|
|
319
|
+
// there: in a child call (the scaffold's shape) or an external task's input.
|
|
346
320
|
for (const at of located) {
|
|
347
|
-
//
|
|
348
|
-
// it the directive sits is the pointer. A child call is the shape the scaffold generates — to a
|
|
349
|
-
// process that forwards to the evaluator — and an external task is the same request made
|
|
350
|
-
// directly.
|
|
321
|
+
// The action's kind is a field; where in it the directive sits is the pointer.
|
|
351
322
|
const kind = at.site.level === "action" ? (at.site.action ?? "") : "";
|
|
352
323
|
const slot = at.site.pointer.slice(-2).join(".");
|
|
353
324
|
if ((kind !== "child" && kind !== "external") || slot !== "input.code") {
|
|
@@ -356,12 +327,11 @@ for (const at of located) {
|
|
|
356
327
|
`${kind ? `a ${kind} task's ` : ""}\`${slot}\`.`);
|
|
357
328
|
}
|
|
358
329
|
}
|
|
359
|
-
// genctl
|
|
360
|
-
// does is the one that must say when it is not there.
|
|
330
|
+
// genctl does not know the argument is a file, so this resolver reports a missing one.
|
|
361
331
|
for (const at of located) {
|
|
362
332
|
if (!existsSync(at.file)) {
|
|
363
333
|
die(`${join(at.where.dir, at.where.file)}: ${address(at.site.pointer)}: ` +
|
|
364
|
-
`"$import: ${at.site.
|
|
334
|
+
`"$import: ${at.site.args[0]}" names no file (looked at ${at.file})`);
|
|
365
335
|
}
|
|
366
336
|
}
|
|
367
337
|
// One script at two sites with different input types is a refusal, not a union: the union
|
|
@@ -378,15 +348,14 @@ for (const at of located) {
|
|
|
378
348
|
byPath.set(at.file, at);
|
|
379
349
|
}
|
|
380
350
|
for (const at of byPath.values()) {
|
|
381
|
-
await
|
|
351
|
+
await writeFile(typesPathFor(at.file), declarations(at));
|
|
382
352
|
}
|
|
383
|
-
if (manifest.mode === "
|
|
384
|
-
console.log(JSON.stringify(manifest, null, 2));
|
|
353
|
+
if (manifest.mode === "generate") {
|
|
385
354
|
process.exit(0);
|
|
386
355
|
}
|
|
387
356
|
await typecheck([...byPath.values()]);
|
|
388
|
-
const
|
|
357
|
+
const values = [];
|
|
389
358
|
for (const at of located) {
|
|
390
|
-
|
|
359
|
+
values.push(await bundle(at));
|
|
391
360
|
}
|
|
392
|
-
process.stdout.write(JSON.stringify({
|
|
361
|
+
process.stdout.write(JSON.stringify({ values }));
|
package/dist/realm.js
CHANGED
|
@@ -1,9 +1,5 @@
|
|
|
1
|
-
// The evaluation realm
|
|
2
|
-
//
|
|
3
|
-
// eval.ts owns the budget and does the killing; nothing here knows about time.
|
|
4
|
-
//
|
|
5
|
-
// Everything that touches the script's VALUE lives on this side of the boundary — loading,
|
|
6
|
-
// classifying, serialising — because this is the only realm the value exists in.
|
|
1
|
+
// The evaluation realm: one Worker per execution, killable mid-loop by eval.ts, which owns the
|
|
2
|
+
// budget. Everything touching the script's VALUE lives here, the only realm it exists in.
|
|
7
3
|
var __rewriteRelativeImportExtension = (this && this.__rewriteRelativeImportExtension) || function (path, preserveJsx) {
|
|
8
4
|
if (typeof path === "string" && /^\.\.?\//.test(path)) {
|
|
9
5
|
return path.replace(/\.(tsx)$|((?:\.d)?)((?:\.[^./]+?)?)\.([cm]?)ts$/i, function (m, tsx, d, ext, cm) {
|
|
@@ -14,9 +10,8 @@ var __rewriteRelativeImportExtension = (this && this.__rewriteRelativeImportExte
|
|
|
14
10
|
};
|
|
15
11
|
import { registerHooks } from "node:module";
|
|
16
12
|
import { parentPort } from "node:worker_threads";
|
|
17
|
-
//
|
|
18
|
-
//
|
|
19
|
-
// the way it does everywhere else.
|
|
13
|
+
// IMPORTED under our own URL rather than compiled from a string, so frames carry the author's
|
|
14
|
+
// line numbers and builtin imports resolve normally.
|
|
20
15
|
const SCRIPT_URL = "script:main";
|
|
21
16
|
const STACK_BYTES = 2_048;
|
|
22
17
|
// The only channel a load hook has to the source. Written per request and read once — a realm
|
|
@@ -41,7 +36,10 @@ function scriptStack(err) {
|
|
|
41
36
|
}
|
|
42
37
|
function describe(err, kind) {
|
|
43
38
|
if (err instanceof Error) {
|
|
44
|
-
|
|
39
|
+
// A subclass inherits `name` from Error.prototype, so `class LimitExceeded extends Error {}`
|
|
40
|
+
// would report "Error"; a name the script set still wins.
|
|
41
|
+
const name = err.name === "Error" && err.constructor.name ? err.constructor.name : err.name;
|
|
42
|
+
return { kind, name, message: err.message, stack: scriptStack(err) };
|
|
45
43
|
}
|
|
46
44
|
// A script may throw a non-Error (`throw {code: "x"}`), so name/message must not assume one.
|
|
47
45
|
return { kind, name: "Thrown", message: safeText(err) };
|
|
@@ -54,9 +52,8 @@ function safeText(v) {
|
|
|
54
52
|
return String(v);
|
|
55
53
|
}
|
|
56
54
|
}
|
|
57
|
-
/**
|
|
58
|
-
*
|
|
59
|
-
* which is the script throwing. */
|
|
55
|
+
/** Unparseable code or an unresolvable import is broken code; anything else the import throws is
|
|
56
|
+
* the module's top level running, i.e. the script throwing. */
|
|
60
57
|
function unloadable(err) {
|
|
61
58
|
const code = err?.code;
|
|
62
59
|
return err instanceof SyntaxError || (typeof code === "string" && code.startsWith("ERR_MODULE"));
|
|
@@ -97,10 +94,8 @@ async function run(req) {
|
|
|
97
94
|
return { ok: false, failure: describe(err, "nonserializable") };
|
|
98
95
|
}
|
|
99
96
|
}
|
|
100
|
-
//
|
|
101
|
-
//
|
|
102
|
-
// write's callback fires once the queue ahead of it has drained, which makes the reply a barrier
|
|
103
|
-
// for `console` and a direct process.stdout.write alike: both are this stream.
|
|
97
|
+
// eval.ts terminates this thread the moment the reply lands, while stdio may still be in the
|
|
98
|
+
// pipe. An empty write's callback fires once the queue drains, making the reply a barrier.
|
|
104
99
|
function flush(stream) {
|
|
105
100
|
return new Promise((resolve) => stream.write("", () => resolve()));
|
|
106
101
|
}
|
package/dist/worker.js
CHANGED
|
@@ -1,50 +1,44 @@
|
|
|
1
1
|
#!/usr/bin/env node
|
|
2
|
-
// The queue worker: claims parked `external` script tasks
|
|
3
|
-
//
|
|
4
|
-
// nothing about the queue, which is what keeps the containment strategy swappable.
|
|
5
|
-
//
|
|
6
|
-
// See README.md for the contract, and specs/external-task-queue.md for the queue itself.
|
|
2
|
+
// The queue worker: claims parked `external` script tasks, evaluates each in its own realm, and
|
|
3
|
+
// answers. The only genroc-facing half (README.md; specs/external-task-queue.md).
|
|
7
4
|
import { readFileSync } from "node:fs";
|
|
8
5
|
import { Cancelled, evaluate } from "./eval.js";
|
|
9
6
|
const SERVER = (process.env.GENROC_SERVER ?? "http://localhost:8448").replace(/\/$/, "");
|
|
10
7
|
const WORKER_ID = process.env.WORKER_ID ?? `evaluator-${process.pid}`;
|
|
11
|
-
//
|
|
12
|
-
//
|
|
13
|
-
//
|
|
14
|
-
// trust least. specs/api-auth.md §5.
|
|
15
|
-
//
|
|
16
|
-
// Sent as a header rather than in the URL because Node's fetch REFUSES a URL carrying
|
|
17
|
-
// credentials ("Request cannot be constructed from a URL that includes credentials"), so the
|
|
18
|
-
// basic-auth-in-the-URL trick that works for genctl is not available here.
|
|
19
|
-
// GENROC_TOKEN_FILE is the mounted-secret shape: a credential in a file rather than an
|
|
20
|
-
// environment variable, so it stays out of `docker inspect` and out of the process environment
|
|
21
|
-
// any child inherits. The inline variable wins when both are set.
|
|
8
|
+
// Needs only the `worker` permission (specs/api-auth.md §5): this credential sits on the machine
|
|
9
|
+
// you trust least. A header, because Node's fetch refuses credentials in a URL. The file form
|
|
10
|
+
// keeps it out of `docker inspect` and child environments; the inline variable wins.
|
|
22
11
|
const TOKEN = process.env.GENROC_TOKEN ??
|
|
23
12
|
(process.env.GENROC_TOKEN_FILE ? readFileSync(process.env.GENROC_TOKEN_FILE, "utf8").trim() : "");
|
|
24
13
|
const authHeaders = TOKEN ? { authorization: `Bearer ${TOKEN}` } : {};
|
|
25
|
-
//
|
|
26
|
-
//
|
|
27
|
-
// evaluator accepted every one of them. Here it claims what it can run and no more, so a
|
|
28
|
-
// backlog is a queue rather than 200 threads fighting over a core.
|
|
14
|
+
// The worker claims only what it can run, so a backlog is a queue rather than threads fighting
|
|
15
|
+
// over a core.
|
|
29
16
|
const CONCURRENCY = Number(process.env.CONCURRENCY ?? 4);
|
|
30
17
|
const POLL_MS = Number(process.env.POLL_MS ?? 250);
|
|
31
18
|
// The visibility timeout. Short, and renewed while work is in flight: a worker that dies
|
|
32
19
|
// should return its task quickly rather than holding it for the whole budget.
|
|
33
20
|
const LEASE_MS = Number(process.env.LEASE_MS ?? 30_000);
|
|
34
21
|
const RENEW_MS = Math.max(1_000, Math.floor(LEASE_MS / 3));
|
|
35
|
-
|
|
36
|
-
|
|
37
|
-
|
|
38
|
-
//
|
|
39
|
-
//
|
|
22
|
+
// The answer has to land before the task's deadline fires, or a script that overran surfaces as
|
|
23
|
+
// external.timeout ("nobody answered") rather than as its own `timeout`.
|
|
24
|
+
const ANSWER_MARGIN_MS = 250;
|
|
25
|
+
// script-node's names; a renamed copy sets PROCESS. The claim requires both: this worker answers
|
|
26
|
+
// one task's contract, so it must never be handed another's.
|
|
27
|
+
const PROCESS_FILTER = process.env.PROCESS ?? "script-node";
|
|
28
|
+
const TASK_FILTER = process.env.TASK ?? "eval_node";
|
|
29
|
+
if (!PROCESS_FILTER || !TASK_FILTER) {
|
|
30
|
+
console.error("PROCESS and TASK must both name the queue this worker serves; neither may be empty.");
|
|
31
|
+
process.exit(1);
|
|
32
|
+
}
|
|
33
|
+
// Large values arrive as refs; a bundle is one object per definition version, fetched once.
|
|
34
|
+
// Refs are content hashes, so the cache never invalidates.
|
|
40
35
|
const objectCache = new Map();
|
|
41
36
|
async function fetchObject(ref) {
|
|
42
37
|
const cached = objectCache.get(ref);
|
|
43
38
|
if (cached !== undefined)
|
|
44
39
|
return cached;
|
|
45
|
-
// Left to throw
|
|
46
|
-
//
|
|
47
|
-
// releases the claim, so the task returns to the queue.
|
|
40
|
+
// Left to throw: a task whose input cannot be fetched must not run against a missing value.
|
|
41
|
+
// The caller releases the claim.
|
|
48
42
|
const res = await fetch(`${SERVER}/api/objects/${encodeURIComponent(ref)}`, { headers: authHeaders });
|
|
49
43
|
if (!res.ok)
|
|
50
44
|
throw new Error(`fetch object ${ref}: HTTP ${res.status}`);
|
|
@@ -62,15 +56,15 @@ async function fetchObject(ref) {
|
|
|
62
56
|
/** Put each listed value back where its path says it belongs. Paths are arrays of keys, so this
|
|
63
57
|
* needs no parser: the whole reason they are not JSON Pointer strings. */
|
|
64
58
|
async function resolveObjects(job) {
|
|
65
|
-
let input = job.
|
|
59
|
+
let input = job.external_input;
|
|
66
60
|
for (const e of job.objects ?? []) {
|
|
67
61
|
const value = await fetchObject(e.ref);
|
|
68
62
|
if (e.path.length === 0) {
|
|
69
63
|
input = value;
|
|
70
64
|
continue;
|
|
71
65
|
}
|
|
72
|
-
// The path is rooted at the entry and starts with "
|
|
73
|
-
const rest = e.path[0] === "
|
|
66
|
+
// The path is rooted at the entry and starts with "external_input", the value being rebuilt.
|
|
67
|
+
const rest = e.path[0] === "external_input" ? e.path.slice(1) : e.path;
|
|
74
68
|
if (rest.length === 0) {
|
|
75
69
|
input = value;
|
|
76
70
|
continue;
|
|
@@ -84,10 +78,8 @@ async function resolveObjects(job) {
|
|
|
84
78
|
return input;
|
|
85
79
|
}
|
|
86
80
|
async function call(path, body) {
|
|
87
|
-
// A network error is a REPLY
|
|
88
|
-
//
|
|
89
|
-
// unhandled rejection here kills it for a condition the next poll would clear. Status 0
|
|
90
|
-
// says "never reached the server", which is distinct from anything genroc answers.
|
|
81
|
+
// A network error is a REPLY (status 0, "never reached the server"), not a throw: a worker
|
|
82
|
+
// outlives restarts and deploys, and an unhandled rejection would kill it.
|
|
91
83
|
let res;
|
|
92
84
|
try {
|
|
93
85
|
res = await fetch(SERVER + path, {
|
|
@@ -109,49 +101,39 @@ async function call(path, body) {
|
|
|
109
101
|
}
|
|
110
102
|
return { ok: res.ok, status: res.status, data };
|
|
111
103
|
}
|
|
112
|
-
/**
|
|
113
|
-
*
|
|
114
|
-
|
|
115
|
-
function asEvalRequest(input) {
|
|
104
|
+
/** A task input that is not an EvalRequest is the definition's fault, reported as compile_error:
|
|
105
|
+
* the nearest permanent kind, since no retry fixes the definition. */
|
|
106
|
+
function asEvalRequest(input, timeoutMs) {
|
|
116
107
|
if (typeof input !== "object" || input === null)
|
|
117
108
|
return "the task input is not an object";
|
|
118
109
|
const r = input;
|
|
119
110
|
if (typeof r.code !== "string")
|
|
120
111
|
return "the task input has no `code` string";
|
|
121
|
-
return {
|
|
122
|
-
code: r.code,
|
|
123
|
-
input: r.input,
|
|
124
|
-
timeout_ms: typeof r.timeout_ms === "number" ? r.timeout_ms : undefined,
|
|
125
|
-
};
|
|
112
|
+
return { code: r.code, input: r.input, timeout_ms: timeoutMs };
|
|
126
113
|
}
|
|
127
114
|
const inFlight = new Map();
|
|
128
115
|
let running = true;
|
|
129
|
-
//
|
|
130
|
-
//
|
|
131
|
-
// buries the one line that mattered. Announce the TRANSITIONS instead: going away, and coming
|
|
132
|
-
// back. Silence in between is the report that nothing changed.
|
|
116
|
+
// Announce reachability TRANSITIONS only: polling several times a second would otherwise emit
|
|
117
|
+
// thousands of lines during a restart.
|
|
133
118
|
let serverReachable = true;
|
|
134
119
|
async function claim(n) {
|
|
135
120
|
const { ok, status, data } = await call("/api/external-tasks/claim", {
|
|
136
121
|
worker_id: WORKER_ID,
|
|
137
122
|
limit: n,
|
|
138
123
|
lease_ms: LEASE_MS,
|
|
139
|
-
|
|
140
|
-
|
|
124
|
+
process: PROCESS_FILTER,
|
|
125
|
+
task: TASK_FILTER,
|
|
141
126
|
});
|
|
142
127
|
if (!ok) {
|
|
143
|
-
// A credential problem is not transient, and polling through it looks like a healthy
|
|
144
|
-
// worker
|
|
145
|
-
// instead, so a supervisor restarts it and the failure is visible where it happened.
|
|
128
|
+
// A credential problem is not transient, and polling through it looks like a healthy idle
|
|
129
|
+
// worker. Exit, so a supervisor restarts it visibly.
|
|
146
130
|
if (status === 401 || status === 403) {
|
|
147
131
|
console.error(`claim rejected (${status}): ${JSON.stringify(data)}\n` +
|
|
148
132
|
`The server requires authentication. Set GENROC_TOKEN to a token with the 'worker' ` +
|
|
149
133
|
`permission — mint one with: genctl token create --perms worker --label evaluator -q`);
|
|
150
134
|
process.exit(1);
|
|
151
135
|
}
|
|
152
|
-
// status 0
|
|
153
|
-
// network blip. Not an error to act on — the next poll clears it — so it is reported once
|
|
154
|
-
// and then waited out.
|
|
136
|
+
// status 0 (see call): a restart or blip the next poll clears, so it is reported once.
|
|
155
137
|
if (status === 0) {
|
|
156
138
|
if (serverReachable) {
|
|
157
139
|
serverReachable = false;
|
|
@@ -174,10 +156,8 @@ async function release(token) {
|
|
|
174
156
|
if (!ok)
|
|
175
157
|
console.error(`release failed: ${JSON.stringify(data)}`);
|
|
176
158
|
}
|
|
177
|
-
/**
|
|
178
|
-
*
|
|
179
|
-
* not fit `raises`), and guessing again would only pick a second wrong answer. Release it, so
|
|
180
|
-
* the task returns to the queue and an operator sees it waiting rather than silently gone. */
|
|
159
|
+
/** A refused outcome is not retried with another: the worker does not satisfy the declared
|
|
160
|
+
* contract. Released, so an operator sees the task waiting rather than gone. */
|
|
181
161
|
async function answer(token, outcome) {
|
|
182
162
|
const { ok, data } = await call("/api/external-tasks/resolve", { token, ...outcome });
|
|
183
163
|
if (ok)
|
|
@@ -186,6 +166,9 @@ async function answer(token, outcome) {
|
|
|
186
166
|
await release(token);
|
|
187
167
|
}
|
|
188
168
|
async function run(job, signal) {
|
|
169
|
+
// The task's own timeout is the script's budget, moved onto the monotonic clock so a wall-clock
|
|
170
|
+
// step mid-run cannot stretch it. Fetching the objects below spends the same deadline.
|
|
171
|
+
const due = job.deadline === undefined ? undefined : performance.now() + (Date.parse(job.deadline) - Date.now());
|
|
189
172
|
let resolved;
|
|
190
173
|
try {
|
|
191
174
|
resolved = await resolveObjects(job);
|
|
@@ -197,7 +180,16 @@ async function run(job, signal) {
|
|
|
197
180
|
await release(job.token);
|
|
198
181
|
return;
|
|
199
182
|
}
|
|
200
|
-
|
|
183
|
+
let timeoutMs;
|
|
184
|
+
if (due !== undefined) {
|
|
185
|
+
timeoutMs = Math.floor(due - performance.now() - ANSWER_MARGIN_MS);
|
|
186
|
+
if (timeoutMs <= 0) {
|
|
187
|
+
// Too late to run anything: the deadline fires as external.timeout, which is the truth.
|
|
188
|
+
await release(job.token);
|
|
189
|
+
return;
|
|
190
|
+
}
|
|
191
|
+
}
|
|
192
|
+
const req = asEvalRequest(resolved, timeoutMs);
|
|
201
193
|
if (typeof req === "string") {
|
|
202
194
|
await answer(job.token, {
|
|
203
195
|
error: { code: "compile_error", message: req, data: { name: "BadTaskInput" } },
|
|
@@ -209,18 +201,15 @@ async function run(job, signal) {
|
|
|
209
201
|
result = await evaluate(req, signal);
|
|
210
202
|
}
|
|
211
203
|
catch (err) {
|
|
212
|
-
// Cancelled is not a fault: the
|
|
213
|
-
//
|
|
214
|
-
// waiting out a lease nobody is serving.
|
|
204
|
+
// Cancelled is not a fault. Still released: the row is terminal so nothing re-claims it,
|
|
205
|
+
// but the release stops it waiting out a lease nobody serves.
|
|
215
206
|
if (err instanceof Cancelled) {
|
|
216
207
|
console.log(`cancelled: ${job.token}`);
|
|
217
208
|
await release(job.token);
|
|
218
209
|
return;
|
|
219
210
|
}
|
|
220
|
-
// The RUNNER faulted,
|
|
221
|
-
//
|
|
222
|
-
// it puts the task in front of a different worker instead of burning the definition's
|
|
223
|
-
// on_error budget on this one's bad day.
|
|
211
|
+
// The RUNNER faulted, the one retryable class. Releasing is how a queue spells "retryable",
|
|
212
|
+
// and it reaches another worker without spending the definition's on_error budget.
|
|
224
213
|
console.error(`evaluator fault on ${job.token}: ${err instanceof Error ? err.message : String(err)}`);
|
|
225
214
|
await release(job.token);
|
|
226
215
|
return;
|
|
@@ -255,17 +244,13 @@ async function renewLoop() {
|
|
|
255
244
|
});
|
|
256
245
|
if (!ok)
|
|
257
246
|
continue;
|
|
258
|
-
//
|
|
259
|
-
//
|
|
260
|
-
// keep burning a core on work an operator has already stopped. The release rides run()'s
|
|
261
|
-
// own catch, so nothing is released twice.
|
|
247
|
+
// Abort first rather than let a stopped script burn a core; the release rides run()'s
|
|
248
|
+
// catch, so nothing is released twice.
|
|
262
249
|
for (const token of (data?.cancelled ?? [])) {
|
|
263
250
|
inFlight.get(token)?.abort.abort();
|
|
264
251
|
}
|
|
265
|
-
// Lost:
|
|
266
|
-
//
|
|
267
|
-
// is too short for what these scripts actually take. Deliberately NOT released: the claim
|
|
268
|
-
// is the new holder's, and releasing would bump the epoch out from under it.
|
|
252
|
+
// Lost: another worker holds it now and our answer will be refused; logged as the sign that
|
|
253
|
+
// LEASE_MS is too short. NOT released: that would bump the epoch out from under the holder.
|
|
269
254
|
const lost = (data?.lost ?? []);
|
|
270
255
|
if (lost.length) {
|
|
271
256
|
console.error(`lost ${lost.length}/${tokens.length} claims; a lease lapsed under load`);
|