Files
OmniRoute/tests/unit/stream-handler.test.ts
Diego Rodrigues de Sa e Souza 8169b97d84 Release v3.8.18 (#3482)
* chore(release): open v3.8.18 development cycle

* fix(catalog): stop Codex CLI model-catalog refresh from erroring (#3481)

Codex's model-catalog refresh (codex_models_manager) does
GET /v1/models?client_version=<v> and decodes a JSON object with a
TOP-LEVEL `models` array. OmniRoute answers in the OpenAI-standard
`{object,data}` shape, so codex fails with "missing field `models`"
and logs "failed to refresh available models" on every startup.

Detect codex clients via the `originator` / `user-agent` = `codex_*`
headers they send and add an EMPTY top-level `models: []` so the decode
succeeds. Non-codex OpenAI clients keep the byte-identical `{object,data}`
response.

The array is intentionally empty: codex replaces its built-in per-model
agent prompt (`base_instructions`, ~21k chars) with whatever a populated
entry carries for the selected model, so emitting our catalog would drop
the agent prompt to nothing and break codex's agent behaviour (verified
empirically against codex 0.137). An empty list keeps codex on its
built-in model info — same inference as before, minus the error.

Validated end-to-end with the real handler against codex 0.137:
"failed to refresh available models" → 0 occurrences, instructions
preserved (built-in Codex agent prompt, not empty).

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* chore: ignore quality reports and local prompt artifacts

Add generated quality gate reports, metrics files, and local setup prompt
artifacts to .gitignore to prevent committing environment-specific or
temporary files.

* fix(provider): detect Responses API format when body has `input` but … (#3490)

Integrated into release/v3.8.18

* fix(sse): normalize numeric provider ids to strings (#3451)

Integrated into release/v3.8.18

* feat(browserPool): resolve Playwright proxy from proxy_registry DB (#3492)

Integrated into release/v3.8.18

* fix(theoldllm): generate X-Request-Token server-side, drop Playwright (#3491)

Integrated into release/v3.8.18

* feat(plugins): add lifecycle hooks and theme-manager plugin (#3473)

Integrated into release/v3.8.18

* fix(combo): parallel pre-screen + circuit-breaker fast-exit for priority combos (#3169)

Integrated into release/v3.8.18

* feat(ui): unifi active and finished requests into single view #1422 (#3401)

Integrated into release/v3.8.18

* docs(changelog): record #3401, #3473, #3492, #3490, #3451, #3491, #3169 under v3.8.18

* feat(docs): add doc accuracy gate + refresh AGENTS.md counts (#3510)

Integrated into release/v3.8.18

* fix(sse): drop empty-choices chunks without usage instead of injecting retry text (#3513)

PR #3422 ('allow OpenAI usage-only empty choices chunks') reintroduced the
assistant-content injection '[OmniRoute] Upstream returned an empty response.
Please retry.' for empty `choices: []` chunks that carry no valid usage. Clients
(Goose/opencode) feed that text back as a turn and spin in a retry loop -- the
exact regression #3400 had fixed by dropping the chunk.

Restore the drop behavior for the no-usage case while preserving #3422's
standards-compliant forwarding of usage-only `include_usage` final chunks.
Realign the mislabeled stream-utils test (it asserted the injection) and add a
dedicated regression guard.

Reported-by: @mochizzan
Refs: #3502, #3388, #3400, #3422

* fix(authz): fall back to URL token when Authorization isn't a usable Bearer (#3504)

Integrated into release/v3.8.18

* fix(playground): authenticate via session, test key policy by id (#3503)

Integrated into release/v3.8.18

* docs(changelog): record #3510, #3504, #3503 under v3.8.18

* fix: llama base url normalization (#3519)

* docs(changelog): reconcile v3.8.18 — add #3519, #3513, #3435-repair, gitignore chore (full commit↔changelog coverage)

* fix(opencode-plugin): bound regex quantifiers in normaliseFreeLabel (polynomial-ReDoS)

CodeQL js/polynomial-redos: unbounded \s* before an anchored \s*$ allowed
O(n²) backtracking on attacker-influenced display names. Bounded to {0,8}/{1,8}
(ample for any real label spacing). Plugin builds + 254 tests green.

* fix(types): restore clean typecheck:core for v3.8.18 release gate

- getPendingRequests() typed to real shape (was widened to object) → fixes
  unknown 'count' in the unified-requests view (#3401)
- streamChunks log payload cast to its declared type (callLogs.ts)
- preScreenTargets aligned to canonical IsModelAvailable signature (#3169),
  Promise.resolve-normalized so .catch never hits a bare boolean

All 5 gates green: lint(0 err) + typecheck:core + cycles + docs-all + unit + vitest(146).

---------

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Andrey Borodulin <borodulin@gmail.com>
Co-authored-by: Dmitrii Safronov <zimniy@cyberbrain.cc>
Co-authored-by: Paijo <14921983+oyi77@users.noreply.github.com>
Co-authored-by: PizzaV <103120356+pizzav-xyz@users.noreply.github.com>
Co-authored-by: Markus Hartung <mail@hartmark.se>
Co-authored-by: Felipe Almeman <4226997+zhiru@users.noreply.github.com>
2026-06-09 15:56:24 -03:00

433 lines
12 KiB
TypeScript

import test from "node:test";
import assert from "node:assert/strict";
import {
createDisconnectAwareStream,
createStreamController,
pipeWithDisconnect,
} from "../../open-sse/utils/streamHandler.ts";
import { FORMATS } from "../../open-sse/translator/formats.ts";
import {
clearPendingRequests,
getPendingRequests,
trackPendingRequest,
} from "../../src/lib/usage/usageHistory.ts";
const encoder = new TextEncoder();
const decoder = new TextDecoder();
const PENDING_REQUEST_CLEARED_MARKER = "__omniroutePendingRequestCleared";
async function readStreamText(stream) {
const reader = stream.getReader();
const chunks = [];
while (true) {
const { done, value } = await reader.read();
if (done) break;
chunks.push(value);
}
return decoder.decode(
chunks.length === 1 ? chunks[0] : Uint8Array.from(chunks.flatMap((chunk) => Array.from(chunk)))
);
}
test("createDisconnectAwareStream converts upstream errors into SSE error chunks", async () => {
const upstreamError = Object.assign(new Error("provider exploded"), { statusCode: 429 });
const transformStream = {
readable: new ReadableStream({
start(controller) {
controller.error(upstreamError);
},
}),
writable: {
getWriter() {
return {
abort() {},
};
},
},
};
const stream = createDisconnectAwareStream(transformStream, createStreamController());
const text = await readStreamText(stream);
assert.match(text, /"finish_reason":"error"/);
assert.match(text, /"message":"provider exploded"/);
assert.match(text, /"code":"rate_limit_exceeded"/);
assert.match(text, /\[DONE\]/);
});
test("createDisconnectAwareStream: Gemini 503 high-demand error becomes SSE error chunk with message preserved", async () => {
const geminiMsg =
"[503]: This model is currently experiencing high demand. Spikes in demand are usually temporary. Please try again later.";
const upstreamError = Object.assign(new Error(geminiMsg), { statusCode: 503 });
const transformStream = {
readable: new ReadableStream({
start(controller) {
controller.error(upstreamError);
},
}),
writable: {
getWriter() {
return {
abort() {},
};
},
},
};
const stream = createDisconnectAwareStream(transformStream, createStreamController());
const text = await readStreamText(stream);
assert.match(text, /"finish_reason":"error"/);
assert.match(text, /"message":"\[503\]: This model is currently experiencing high demand/);
assert.match(text, /"type":"server_error"/);
assert.match(text, /"code":"server_error"/);
assert.match(text, /\[DONE\]/);
});
test("createDisconnectAwareStream emits Responses API failure events for Responses clients", async () => {
const upstreamError = Object.assign(new Error("responses stream\ndied"), { statusCode: 503 });
const transformStream = {
readable: new ReadableStream({
start(controller) {
controller.error(upstreamError);
},
}),
writable: {
getWriter() {
return {
abort() {},
};
},
},
};
const stream = createDisconnectAwareStream(
transformStream,
createStreamController({ clientResponseFormat: FORMATS.OPENAI_RESPONSES })
);
const text = await readStreamText(stream);
assert.match(text, /event: response\.failed/);
assert.match(text, /"type":"response\.failed"/);
assert.match(text, /"message":"responses stream\\ndied"/);
assert.match(text, /"type":"server_error"/);
assert.match(text, /"code":"server_error"/);
assert.doesNotMatch(text, /chat\.completion\.chunk/);
assert.doesNotMatch(text, /"finish_reason":"error"/);
assert.doesNotMatch(text, /\[DONE\]/);
});
test("createDisconnectAwareStream keeps newlines escaped inside SSE data fields", async () => {
const upstreamError = Object.assign(new Error("line one\nline two\rline three"), {
statusCode: 400,
});
const transformStream = {
readable: new ReadableStream({
start(controller) {
controller.error(upstreamError);
},
}),
writable: {
getWriter() {
return {
abort() {},
};
},
},
};
const stream = createDisconnectAwareStream(
transformStream,
createStreamController({ clientResponseFormat: FORMATS.OPENAI_RESPONSES })
);
const text = await readStreamText(stream);
assert.match(text, /^event: response\.failed\ndata: \{"type":"response\.failed"/);
assert.match(text, /"message":"line one\\nline two\\rline three"/);
assert.doesNotMatch(text, /^line two/m);
assert.doesNotMatch(text, /^line three/m);
});
test("createDisconnectAwareStream treats legacy OpenAI response format alias as Responses", async () => {
const upstreamError = Object.assign(new Error("legacy responses alias died"), {
statusCode: 429,
});
const transformStream = {
readable: new ReadableStream({
start(controller) {
controller.error(upstreamError);
},
}),
writable: {
getWriter() {
return {
abort() {},
};
},
},
};
const stream = createDisconnectAwareStream(
transformStream,
createStreamController({ clientResponseFormat: FORMATS.OPENAI_RESPONSE })
);
const text = await readStreamText(stream);
assert.match(text, /event: response\.failed/);
assert.match(text, /"type":"rate_limit_error"/);
assert.match(text, /"code":"rate_limit_exceeded"/);
assert.doesNotMatch(text, /chat\.completion\.chunk/);
assert.doesNotMatch(text, /\[DONE\]/);
});
test("createDisconnectAwareStream emits Claude SSE errors for Claude clients", async () => {
const upstreamError = Object.assign(new Error("claude stream died"), { statusCode: 502 });
const transformStream = {
readable: new ReadableStream({
start(controller) {
controller.error(upstreamError);
},
}),
writable: {
getWriter() {
return {
abort() {},
};
},
},
};
const stream = createDisconnectAwareStream(
transformStream,
createStreamController({ clientResponseFormat: FORMATS.CLAUDE })
);
const text = await readStreamText(stream);
assert.match(text, /event: error/);
assert.match(text, /"type":"error"/);
assert.match(text, /"type":"api_error"/);
assert.match(text, /"message":"claude stream died"/);
assert.doesNotMatch(text, /"code"/);
assert.doesNotMatch(text, /chat\.completion\.chunk/);
assert.doesNotMatch(text, /"finish_reason":"error"/);
assert.doesNotMatch(text, /\[DONE\]/);
});
test("createDisconnectAwareStream keeps newlines escaped for Claude SSE errors", async () => {
const upstreamError = Object.assign(new Error("claude line one\nclaude line two"), {
statusCode: 502,
});
const transformStream = {
readable: new ReadableStream({
start(controller) {
controller.error(upstreamError);
},
}),
writable: {
getWriter() {
return {
abort() {},
};
},
},
};
const stream = createDisconnectAwareStream(
transformStream,
createStreamController({ clientResponseFormat: FORMATS.CLAUDE })
);
const text = await readStreamText(stream);
assert.match(text, /^event: error\ndata: \{"type":"error"/);
assert.match(text, /"message":"claude line one\\nclaude line two"/);
assert.doesNotMatch(text, /^claude line two/m);
});
test("createDisconnectAwareStream cancel propagates disconnect reason and aborts the writer", async () => {
let aborted = false;
let disconnectEvent = null;
const transformStream = {
readable: new ReadableStream({
pull() {},
cancel() {},
}),
writable: {
getWriter() {
return {
abort() {
aborted = true;
},
};
},
},
};
const controller = createStreamController({
onDisconnect(event) {
disconnectEvent = event;
},
});
const stream = createDisconnectAwareStream(transformStream, controller);
await stream.cancel("client-gone");
await new Promise((resolve) => setTimeout(resolve, 2050));
assert.equal(aborted, true);
assert.equal(controller.isConnected(), false);
assert.equal(disconnectEvent.reason, "client-gone");
assert.ok(disconnectEvent.duration >= 0);
});
test("createDisconnectAwareStream uses the default cancel reason when none is provided", async () => {
let disconnectEvent = null;
const transformStream = {
readable: new ReadableStream({
cancel() {},
}),
writable: {
getWriter() {
return {
abort() {},
};
},
},
};
const controller = createStreamController({
onDisconnect(event) {
disconnectEvent = event;
},
});
const stream = createDisconnectAwareStream(transformStream, controller);
await stream.cancel();
assert.equal(disconnectEvent.reason, "cancelled");
});
test("createDisconnectAwareStream closes immediately when the controller is already disconnected", async () => {
const controller = createStreamController();
controller.handleDisconnect("preclosed");
const stream = createDisconnectAwareStream(
{
readable: new ReadableStream({
pull(inner) {
inner.enqueue(encoder.encode("ignored"));
},
}),
writable: {
getWriter() {
return {
abort() {},
};
},
},
},
controller
);
const reader = stream.getReader();
const first = await reader.read();
assert.equal(first.done, true);
});
test("createStreamController aborts after delayed disconnect and tolerates abort/unknown errors", async () => {
const controller = createStreamController();
const errorOnlyController = createStreamController();
controller.handleDisconnect();
controller.handleDisconnect("ignored-repeat");
errorOnlyController.handleError(new DOMException("aborted", "AbortError"));
errorOnlyController.handleError({ statusCode: 418 });
await new Promise((resolve) => setTimeout(resolve, 2050));
assert.equal(controller.signal.aborted, true);
assert.equal(controller.isConnected(), false);
assert.equal(errorOnlyController.signal.aborted, false);
});
test("pipeWithDisconnect pipes transformed bytes and marks the controller complete", async () => {
const source = new ReadableStream({
start(controller) {
controller.enqueue(encoder.encode("hello"));
controller.close();
},
});
const providerResponse = new Response(source);
const controller = createStreamController();
const stream = pipeWithDisconnect(providerResponse, new TransformStream(), controller);
const text = await readStreamText(stream);
assert.equal(text, "hello");
assert.equal(controller.isConnected(), false);
});
test("pipeWithDisconnect clears pending requests when the upstream stream errors", async () => {
clearPendingRequests();
const provider = "openai";
const model = "gpt-stream-error";
const connectionId = "conn-stream-error";
const modelKey = `${model} (${provider})`;
trackPendingRequest(model, provider, connectionId, true);
const source = new ReadableStream({
start(controller) {
controller.error(Object.assign(new Error("socket closed"), { statusCode: 502 }));
},
});
const stream = pipeWithDisconnect(
new Response(source),
new TransformStream(),
createStreamController({ provider, model, connectionId })
);
const text = await readStreamText(stream);
const pending = getPendingRequests();
assert.match(text, /"message":"socket closed"/);
assert.equal(pending.byModel[modelKey], 0);
assert.equal(pending.details[connectionId], undefined);
});
test("pipeWithDisconnect does not double-clear transform errors already accounted for", async () => {
clearPendingRequests();
const provider = "openai";
const model = "gpt-marked-error";
const connectionId = "conn-marked-error";
const modelKey = `${model} (${provider})`;
trackPendingRequest(model, provider, connectionId, true);
trackPendingRequest(model, provider, connectionId, true);
trackPendingRequest(model, provider, connectionId, false);
const markedError = Object.assign(new Error("already cleared"), {
[PENDING_REQUEST_CLEARED_MARKER]: true,
});
const source = new ReadableStream({
start(controller) {
controller.error(markedError);
},
});
const stream = pipeWithDisconnect(
new Response(source),
new TransformStream(),
createStreamController({ provider, model, connectionId })
);
await readStreamText(stream);
const pending = getPendingRequests();
assert.equal(pending.byModel[modelKey], 1);
assert.equal(pending.byAccount[connectionId][modelKey], 1);
});