9router/src/sse/handlers/embeddings.js

143 lines
5 KiB
JavaScript
Raw Normal View History

feat: add /v1/embeddings endpoint (OpenAI-compatible) (#146) * feat: implement /v1/embeddings endpoint (#117) Add OpenAI-compatible POST /v1/embeddings endpoint that routes through the existing provider credential + fallback infrastructure. Changes: - open-sse/handlers/embeddingsCore.js: core handler (handleEmbeddingsCore) * Validates input (string or array), encoding_format * Builds provider-specific URL and headers for openai, openrouter, and openai-compatible providers * Handles 401/403 token refresh via executor.refreshCredentials * Returns normalized OpenAI-format response { object: 'list', data, model, usage } - cloud/src/handlers/embeddings.js: cloud Worker handler (handleEmbeddings) * Auth + machineId resolution identical to handleChat * Provider credential fallback loop with rate-limit tracking - cloud/src/index.js: wire new routes * POST /v1/embeddings (new format — machineId from API key) * POST /{machineId}/v1/embeddings (old format — machineId from URL) * test: add unit tests for /v1/embeddings endpoint - Setup vitest as test framework (tests/ directory) - embeddingsCore.test.js (36 tests): - buildEmbeddingsBody: single string, array, encoding_format, default float - buildEmbeddingsUrl: openai, openrouter, openai-compatible-*, unsupported - buildEmbeddingsHeaders: per-provider headers, accessToken fallback - handleEmbeddingsCore: input validation, success path, provider errors, network errors, invalid JSON, token refresh 401 handling - embeddings.cloud.test.js (23 tests): - CORS OPTIONS preflight - Auth: missing/invalid/old-format/wrong key → 401/400 - Body validation: bad JSON, missing model, missing input, bad model → 400 - Happy path: single string, array, delegation, CORS header, machineId override - Rate limiting: all-rate-limited → 429 + Retry-After, no credentials → 400 - Error propagation: non-fallback errors, 429 exhausts accounts Total: 59/59 tests passing Framework: vitest v4.0.18, Node v22.22.0 * feat: add Next.js API route for /v1/embeddings endpoint Wire the embeddings handler into Next.js App Router. - src/app/api/v1/embeddings/route.js: Next.js API route (POST + OPTIONS) - src/sse/handlers/embeddings.js: SSE-layer handler mirroring chat.js pattern Uses handleEmbeddingsCore from open-sse/handlers/embeddingsCore.js with the same auth, credential fallback, and token refresh logic as the chat handler. Supports REQUIRE_API_KEY env var, provider fallback loop, and consistent logging.
2026-02-18 01:24:02 -05:00
import {
getProviderCredentials,
markAccountUnavailable,
clearAccountError,
extractApiKey,
isValidApiKey,
} from "../services/auth.js";
import { getSettings } from "@/lib/localDb";
feat: add /v1/embeddings endpoint (OpenAI-compatible) (#146) * feat: implement /v1/embeddings endpoint (#117) Add OpenAI-compatible POST /v1/embeddings endpoint that routes through the existing provider credential + fallback infrastructure. Changes: - open-sse/handlers/embeddingsCore.js: core handler (handleEmbeddingsCore) * Validates input (string or array), encoding_format * Builds provider-specific URL and headers for openai, openrouter, and openai-compatible providers * Handles 401/403 token refresh via executor.refreshCredentials * Returns normalized OpenAI-format response { object: 'list', data, model, usage } - cloud/src/handlers/embeddings.js: cloud Worker handler (handleEmbeddings) * Auth + machineId resolution identical to handleChat * Provider credential fallback loop with rate-limit tracking - cloud/src/index.js: wire new routes * POST /v1/embeddings (new format — machineId from API key) * POST /{machineId}/v1/embeddings (old format — machineId from URL) * test: add unit tests for /v1/embeddings endpoint - Setup vitest as test framework (tests/ directory) - embeddingsCore.test.js (36 tests): - buildEmbeddingsBody: single string, array, encoding_format, default float - buildEmbeddingsUrl: openai, openrouter, openai-compatible-*, unsupported - buildEmbeddingsHeaders: per-provider headers, accessToken fallback - handleEmbeddingsCore: input validation, success path, provider errors, network errors, invalid JSON, token refresh 401 handling - embeddings.cloud.test.js (23 tests): - CORS OPTIONS preflight - Auth: missing/invalid/old-format/wrong key → 401/400 - Body validation: bad JSON, missing model, missing input, bad model → 400 - Happy path: single string, array, delegation, CORS header, machineId override - Rate limiting: all-rate-limited → 429 + Retry-After, no credentials → 400 - Error propagation: non-fallback errors, 429 exhausts accounts Total: 59/59 tests passing Framework: vitest v4.0.18, Node v22.22.0 * feat: add Next.js API route for /v1/embeddings endpoint Wire the embeddings handler into Next.js App Router. - src/app/api/v1/embeddings/route.js: Next.js API route (POST + OPTIONS) - src/sse/handlers/embeddings.js: SSE-layer handler mirroring chat.js pattern Uses handleEmbeddingsCore from open-sse/handlers/embeddingsCore.js with the same auth, credential fallback, and token refresh logic as the chat handler. Supports REQUIRE_API_KEY env var, provider fallback loop, and consistent logging.
2026-02-18 01:24:02 -05:00
import { getModelInfo } from "../services/model.js";
import { handleEmbeddingsCore } from "open-sse/handlers/embeddingsCore.js";
import { errorResponse, unavailableResponse } from "open-sse/utils/error.js";
2026-03-12 05:20:46 -04:00
import { HTTP_STATUS } from "open-sse/config/runtimeConfig.js";
feat: add /v1/embeddings endpoint (OpenAI-compatible) (#146) * feat: implement /v1/embeddings endpoint (#117) Add OpenAI-compatible POST /v1/embeddings endpoint that routes through the existing provider credential + fallback infrastructure. Changes: - open-sse/handlers/embeddingsCore.js: core handler (handleEmbeddingsCore) * Validates input (string or array), encoding_format * Builds provider-specific URL and headers for openai, openrouter, and openai-compatible providers * Handles 401/403 token refresh via executor.refreshCredentials * Returns normalized OpenAI-format response { object: 'list', data, model, usage } - cloud/src/handlers/embeddings.js: cloud Worker handler (handleEmbeddings) * Auth + machineId resolution identical to handleChat * Provider credential fallback loop with rate-limit tracking - cloud/src/index.js: wire new routes * POST /v1/embeddings (new format — machineId from API key) * POST /{machineId}/v1/embeddings (old format — machineId from URL) * test: add unit tests for /v1/embeddings endpoint - Setup vitest as test framework (tests/ directory) - embeddingsCore.test.js (36 tests): - buildEmbeddingsBody: single string, array, encoding_format, default float - buildEmbeddingsUrl: openai, openrouter, openai-compatible-*, unsupported - buildEmbeddingsHeaders: per-provider headers, accessToken fallback - handleEmbeddingsCore: input validation, success path, provider errors, network errors, invalid JSON, token refresh 401 handling - embeddings.cloud.test.js (23 tests): - CORS OPTIONS preflight - Auth: missing/invalid/old-format/wrong key → 401/400 - Body validation: bad JSON, missing model, missing input, bad model → 400 - Happy path: single string, array, delegation, CORS header, machineId override - Rate limiting: all-rate-limited → 429 + Retry-After, no credentials → 400 - Error propagation: non-fallback errors, 429 exhausts accounts Total: 59/59 tests passing Framework: vitest v4.0.18, Node v22.22.0 * feat: add Next.js API route for /v1/embeddings endpoint Wire the embeddings handler into Next.js App Router. - src/app/api/v1/embeddings/route.js: Next.js API route (POST + OPTIONS) - src/sse/handlers/embeddings.js: SSE-layer handler mirroring chat.js pattern Uses handleEmbeddingsCore from open-sse/handlers/embeddingsCore.js with the same auth, credential fallback, and token refresh logic as the chat handler. Supports REQUIRE_API_KEY env var, provider fallback loop, and consistent logging.
2026-02-18 01:24:02 -05:00
import * as log from "../utils/logger.js";
import { updateProviderCredentials, checkAndRefreshToken } from "../services/tokenRefresh.js";
/**
* Handle embeddings request for the SSE/Next.js server.
* Follows the same auth + fallback pattern as handleChat.
*
* @param {Request} request
*/
export async function handleEmbeddings(request) {
let body;
try {
body = await request.json();
} catch {
log.warn("EMBEDDINGS", "Invalid JSON body");
return errorResponse(HTTP_STATUS.BAD_REQUEST, "Invalid JSON body");
}
const url = new URL(request.url);
const modelStr = body.model;
log.request("POST", `${url.pathname} | ${modelStr}`);
// Log API key (masked)
const apiKey = extractApiKey(request);
if (apiKey) {
log.debug("AUTH", `API Key: ${log.maskKey(apiKey)}`);
} else {
log.debug("AUTH", "No API key provided (local mode)");
}
// Enforce API key if enabled in settings
const settings = await getSettings();
if (settings.requireApiKey) {
feat: add /v1/embeddings endpoint (OpenAI-compatible) (#146) * feat: implement /v1/embeddings endpoint (#117) Add OpenAI-compatible POST /v1/embeddings endpoint that routes through the existing provider credential + fallback infrastructure. Changes: - open-sse/handlers/embeddingsCore.js: core handler (handleEmbeddingsCore) * Validates input (string or array), encoding_format * Builds provider-specific URL and headers for openai, openrouter, and openai-compatible providers * Handles 401/403 token refresh via executor.refreshCredentials * Returns normalized OpenAI-format response { object: 'list', data, model, usage } - cloud/src/handlers/embeddings.js: cloud Worker handler (handleEmbeddings) * Auth + machineId resolution identical to handleChat * Provider credential fallback loop with rate-limit tracking - cloud/src/index.js: wire new routes * POST /v1/embeddings (new format — machineId from API key) * POST /{machineId}/v1/embeddings (old format — machineId from URL) * test: add unit tests for /v1/embeddings endpoint - Setup vitest as test framework (tests/ directory) - embeddingsCore.test.js (36 tests): - buildEmbeddingsBody: single string, array, encoding_format, default float - buildEmbeddingsUrl: openai, openrouter, openai-compatible-*, unsupported - buildEmbeddingsHeaders: per-provider headers, accessToken fallback - handleEmbeddingsCore: input validation, success path, provider errors, network errors, invalid JSON, token refresh 401 handling - embeddings.cloud.test.js (23 tests): - CORS OPTIONS preflight - Auth: missing/invalid/old-format/wrong key → 401/400 - Body validation: bad JSON, missing model, missing input, bad model → 400 - Happy path: single string, array, delegation, CORS header, machineId override - Rate limiting: all-rate-limited → 429 + Retry-After, no credentials → 400 - Error propagation: non-fallback errors, 429 exhausts accounts Total: 59/59 tests passing Framework: vitest v4.0.18, Node v22.22.0 * feat: add Next.js API route for /v1/embeddings endpoint Wire the embeddings handler into Next.js App Router. - src/app/api/v1/embeddings/route.js: Next.js API route (POST + OPTIONS) - src/sse/handlers/embeddings.js: SSE-layer handler mirroring chat.js pattern Uses handleEmbeddingsCore from open-sse/handlers/embeddingsCore.js with the same auth, credential fallback, and token refresh logic as the chat handler. Supports REQUIRE_API_KEY env var, provider fallback loop, and consistent logging.
2026-02-18 01:24:02 -05:00
if (!apiKey) {
log.warn("AUTH", "Missing API key (requireApiKey=true)");
feat: add /v1/embeddings endpoint (OpenAI-compatible) (#146) * feat: implement /v1/embeddings endpoint (#117) Add OpenAI-compatible POST /v1/embeddings endpoint that routes through the existing provider credential + fallback infrastructure. Changes: - open-sse/handlers/embeddingsCore.js: core handler (handleEmbeddingsCore) * Validates input (string or array), encoding_format * Builds provider-specific URL and headers for openai, openrouter, and openai-compatible providers * Handles 401/403 token refresh via executor.refreshCredentials * Returns normalized OpenAI-format response { object: 'list', data, model, usage } - cloud/src/handlers/embeddings.js: cloud Worker handler (handleEmbeddings) * Auth + machineId resolution identical to handleChat * Provider credential fallback loop with rate-limit tracking - cloud/src/index.js: wire new routes * POST /v1/embeddings (new format — machineId from API key) * POST /{machineId}/v1/embeddings (old format — machineId from URL) * test: add unit tests for /v1/embeddings endpoint - Setup vitest as test framework (tests/ directory) - embeddingsCore.test.js (36 tests): - buildEmbeddingsBody: single string, array, encoding_format, default float - buildEmbeddingsUrl: openai, openrouter, openai-compatible-*, unsupported - buildEmbeddingsHeaders: per-provider headers, accessToken fallback - handleEmbeddingsCore: input validation, success path, provider errors, network errors, invalid JSON, token refresh 401 handling - embeddings.cloud.test.js (23 tests): - CORS OPTIONS preflight - Auth: missing/invalid/old-format/wrong key → 401/400 - Body validation: bad JSON, missing model, missing input, bad model → 400 - Happy path: single string, array, delegation, CORS header, machineId override - Rate limiting: all-rate-limited → 429 + Retry-After, no credentials → 400 - Error propagation: non-fallback errors, 429 exhausts accounts Total: 59/59 tests passing Framework: vitest v4.0.18, Node v22.22.0 * feat: add Next.js API route for /v1/embeddings endpoint Wire the embeddings handler into Next.js App Router. - src/app/api/v1/embeddings/route.js: Next.js API route (POST + OPTIONS) - src/sse/handlers/embeddings.js: SSE-layer handler mirroring chat.js pattern Uses handleEmbeddingsCore from open-sse/handlers/embeddingsCore.js with the same auth, credential fallback, and token refresh logic as the chat handler. Supports REQUIRE_API_KEY env var, provider fallback loop, and consistent logging.
2026-02-18 01:24:02 -05:00
return errorResponse(HTTP_STATUS.UNAUTHORIZED, "Missing API key");
}
const valid = await isValidApiKey(apiKey);
if (!valid) {
log.warn("AUTH", "Invalid API key (requireApiKey=true)");
feat: add /v1/embeddings endpoint (OpenAI-compatible) (#146) * feat: implement /v1/embeddings endpoint (#117) Add OpenAI-compatible POST /v1/embeddings endpoint that routes through the existing provider credential + fallback infrastructure. Changes: - open-sse/handlers/embeddingsCore.js: core handler (handleEmbeddingsCore) * Validates input (string or array), encoding_format * Builds provider-specific URL and headers for openai, openrouter, and openai-compatible providers * Handles 401/403 token refresh via executor.refreshCredentials * Returns normalized OpenAI-format response { object: 'list', data, model, usage } - cloud/src/handlers/embeddings.js: cloud Worker handler (handleEmbeddings) * Auth + machineId resolution identical to handleChat * Provider credential fallback loop with rate-limit tracking - cloud/src/index.js: wire new routes * POST /v1/embeddings (new format — machineId from API key) * POST /{machineId}/v1/embeddings (old format — machineId from URL) * test: add unit tests for /v1/embeddings endpoint - Setup vitest as test framework (tests/ directory) - embeddingsCore.test.js (36 tests): - buildEmbeddingsBody: single string, array, encoding_format, default float - buildEmbeddingsUrl: openai, openrouter, openai-compatible-*, unsupported - buildEmbeddingsHeaders: per-provider headers, accessToken fallback - handleEmbeddingsCore: input validation, success path, provider errors, network errors, invalid JSON, token refresh 401 handling - embeddings.cloud.test.js (23 tests): - CORS OPTIONS preflight - Auth: missing/invalid/old-format/wrong key → 401/400 - Body validation: bad JSON, missing model, missing input, bad model → 400 - Happy path: single string, array, delegation, CORS header, machineId override - Rate limiting: all-rate-limited → 429 + Retry-After, no credentials → 400 - Error propagation: non-fallback errors, 429 exhausts accounts Total: 59/59 tests passing Framework: vitest v4.0.18, Node v22.22.0 * feat: add Next.js API route for /v1/embeddings endpoint Wire the embeddings handler into Next.js App Router. - src/app/api/v1/embeddings/route.js: Next.js API route (POST + OPTIONS) - src/sse/handlers/embeddings.js: SSE-layer handler mirroring chat.js pattern Uses handleEmbeddingsCore from open-sse/handlers/embeddingsCore.js with the same auth, credential fallback, and token refresh logic as the chat handler. Supports REQUIRE_API_KEY env var, provider fallback loop, and consistent logging.
2026-02-18 01:24:02 -05:00
return errorResponse(HTTP_STATUS.UNAUTHORIZED, "Invalid API key");
}
}
if (!modelStr) {
log.warn("EMBEDDINGS", "Missing model");
return errorResponse(HTTP_STATUS.BAD_REQUEST, "Missing model");
}
if (!body.input) {
log.warn("EMBEDDINGS", "Missing input");
return errorResponse(HTTP_STATUS.BAD_REQUEST, "Missing required field: input");
}
const modelInfo = await getModelInfo(modelStr);
if (!modelInfo.provider) {
log.warn("EMBEDDINGS", "Invalid model format", { model: modelStr });
return errorResponse(HTTP_STATUS.BAD_REQUEST, "Invalid model format");
}
const { provider, model } = modelInfo;
if (modelStr !== `${provider}/${model}`) {
log.info("ROUTING", `${modelStr} → ${provider}/${model}`);
} else {
log.info("ROUTING", `Provider: ${provider}, Model: ${model}`);
}
// Credential + fallback loop (mirrors handleChat)
2026-03-13 22:37:29 -04:00
const excludeConnectionIds = new Set();
feat: add /v1/embeddings endpoint (OpenAI-compatible) (#146) * feat: implement /v1/embeddings endpoint (#117) Add OpenAI-compatible POST /v1/embeddings endpoint that routes through the existing provider credential + fallback infrastructure. Changes: - open-sse/handlers/embeddingsCore.js: core handler (handleEmbeddingsCore) * Validates input (string or array), encoding_format * Builds provider-specific URL and headers for openai, openrouter, and openai-compatible providers * Handles 401/403 token refresh via executor.refreshCredentials * Returns normalized OpenAI-format response { object: 'list', data, model, usage } - cloud/src/handlers/embeddings.js: cloud Worker handler (handleEmbeddings) * Auth + machineId resolution identical to handleChat * Provider credential fallback loop with rate-limit tracking - cloud/src/index.js: wire new routes * POST /v1/embeddings (new format — machineId from API key) * POST /{machineId}/v1/embeddings (old format — machineId from URL) * test: add unit tests for /v1/embeddings endpoint - Setup vitest as test framework (tests/ directory) - embeddingsCore.test.js (36 tests): - buildEmbeddingsBody: single string, array, encoding_format, default float - buildEmbeddingsUrl: openai, openrouter, openai-compatible-*, unsupported - buildEmbeddingsHeaders: per-provider headers, accessToken fallback - handleEmbeddingsCore: input validation, success path, provider errors, network errors, invalid JSON, token refresh 401 handling - embeddings.cloud.test.js (23 tests): - CORS OPTIONS preflight - Auth: missing/invalid/old-format/wrong key → 401/400 - Body validation: bad JSON, missing model, missing input, bad model → 400 - Happy path: single string, array, delegation, CORS header, machineId override - Rate limiting: all-rate-limited → 429 + Retry-After, no credentials → 400 - Error propagation: non-fallback errors, 429 exhausts accounts Total: 59/59 tests passing Framework: vitest v4.0.18, Node v22.22.0 * feat: add Next.js API route for /v1/embeddings endpoint Wire the embeddings handler into Next.js App Router. - src/app/api/v1/embeddings/route.js: Next.js API route (POST + OPTIONS) - src/sse/handlers/embeddings.js: SSE-layer handler mirroring chat.js pattern Uses handleEmbeddingsCore from open-sse/handlers/embeddingsCore.js with the same auth, credential fallback, and token refresh logic as the chat handler. Supports REQUIRE_API_KEY env var, provider fallback loop, and consistent logging.
2026-02-18 01:24:02 -05:00
let lastError = null;
let lastStatus = null;
while (true) {
2026-03-13 22:37:29 -04:00
const credentials = await getProviderCredentials(provider, excludeConnectionIds, model);
feat: add /v1/embeddings endpoint (OpenAI-compatible) (#146) * feat: implement /v1/embeddings endpoint (#117) Add OpenAI-compatible POST /v1/embeddings endpoint that routes through the existing provider credential + fallback infrastructure. Changes: - open-sse/handlers/embeddingsCore.js: core handler (handleEmbeddingsCore) * Validates input (string or array), encoding_format * Builds provider-specific URL and headers for openai, openrouter, and openai-compatible providers * Handles 401/403 token refresh via executor.refreshCredentials * Returns normalized OpenAI-format response { object: 'list', data, model, usage } - cloud/src/handlers/embeddings.js: cloud Worker handler (handleEmbeddings) * Auth + machineId resolution identical to handleChat * Provider credential fallback loop with rate-limit tracking - cloud/src/index.js: wire new routes * POST /v1/embeddings (new format — machineId from API key) * POST /{machineId}/v1/embeddings (old format — machineId from URL) * test: add unit tests for /v1/embeddings endpoint - Setup vitest as test framework (tests/ directory) - embeddingsCore.test.js (36 tests): - buildEmbeddingsBody: single string, array, encoding_format, default float - buildEmbeddingsUrl: openai, openrouter, openai-compatible-*, unsupported - buildEmbeddingsHeaders: per-provider headers, accessToken fallback - handleEmbeddingsCore: input validation, success path, provider errors, network errors, invalid JSON, token refresh 401 handling - embeddings.cloud.test.js (23 tests): - CORS OPTIONS preflight - Auth: missing/invalid/old-format/wrong key → 401/400 - Body validation: bad JSON, missing model, missing input, bad model → 400 - Happy path: single string, array, delegation, CORS header, machineId override - Rate limiting: all-rate-limited → 429 + Retry-After, no credentials → 400 - Error propagation: non-fallback errors, 429 exhausts accounts Total: 59/59 tests passing Framework: vitest v4.0.18, Node v22.22.0 * feat: add Next.js API route for /v1/embeddings endpoint Wire the embeddings handler into Next.js App Router. - src/app/api/v1/embeddings/route.js: Next.js API route (POST + OPTIONS) - src/sse/handlers/embeddings.js: SSE-layer handler mirroring chat.js pattern Uses handleEmbeddingsCore from open-sse/handlers/embeddingsCore.js with the same auth, credential fallback, and token refresh logic as the chat handler. Supports REQUIRE_API_KEY env var, provider fallback loop, and consistent logging.
2026-02-18 01:24:02 -05:00
// All accounts unavailable
if (!credentials || credentials.allRateLimited) {
if (credentials?.allRateLimited) {
const errorMsg = lastError || credentials.lastError || "Unavailable";
const status = lastStatus || Number(credentials.lastErrorCode) || HTTP_STATUS.SERVICE_UNAVAILABLE;
log.warn("EMBEDDINGS", `[${provider}/${model}] ${errorMsg} (${credentials.retryAfterHuman})`);
return unavailableResponse(status, `[${provider}/${model}] ${errorMsg}`, credentials.retryAfter, credentials.retryAfterHuman);
}
2026-03-13 22:37:29 -04:00
if (excludeConnectionIds.size === 0) {
feat: add /v1/embeddings endpoint (OpenAI-compatible) (#146) * feat: implement /v1/embeddings endpoint (#117) Add OpenAI-compatible POST /v1/embeddings endpoint that routes through the existing provider credential + fallback infrastructure. Changes: - open-sse/handlers/embeddingsCore.js: core handler (handleEmbeddingsCore) * Validates input (string or array), encoding_format * Builds provider-specific URL and headers for openai, openrouter, and openai-compatible providers * Handles 401/403 token refresh via executor.refreshCredentials * Returns normalized OpenAI-format response { object: 'list', data, model, usage } - cloud/src/handlers/embeddings.js: cloud Worker handler (handleEmbeddings) * Auth + machineId resolution identical to handleChat * Provider credential fallback loop with rate-limit tracking - cloud/src/index.js: wire new routes * POST /v1/embeddings (new format — machineId from API key) * POST /{machineId}/v1/embeddings (old format — machineId from URL) * test: add unit tests for /v1/embeddings endpoint - Setup vitest as test framework (tests/ directory) - embeddingsCore.test.js (36 tests): - buildEmbeddingsBody: single string, array, encoding_format, default float - buildEmbeddingsUrl: openai, openrouter, openai-compatible-*, unsupported - buildEmbeddingsHeaders: per-provider headers, accessToken fallback - handleEmbeddingsCore: input validation, success path, provider errors, network errors, invalid JSON, token refresh 401 handling - embeddings.cloud.test.js (23 tests): - CORS OPTIONS preflight - Auth: missing/invalid/old-format/wrong key → 401/400 - Body validation: bad JSON, missing model, missing input, bad model → 400 - Happy path: single string, array, delegation, CORS header, machineId override - Rate limiting: all-rate-limited → 429 + Retry-After, no credentials → 400 - Error propagation: non-fallback errors, 429 exhausts accounts Total: 59/59 tests passing Framework: vitest v4.0.18, Node v22.22.0 * feat: add Next.js API route for /v1/embeddings endpoint Wire the embeddings handler into Next.js App Router. - src/app/api/v1/embeddings/route.js: Next.js API route (POST + OPTIONS) - src/sse/handlers/embeddings.js: SSE-layer handler mirroring chat.js pattern Uses handleEmbeddingsCore from open-sse/handlers/embeddingsCore.js with the same auth, credential fallback, and token refresh logic as the chat handler. Supports REQUIRE_API_KEY env var, provider fallback loop, and consistent logging.
2026-02-18 01:24:02 -05:00
log.error("AUTH", `No credentials for provider: ${provider}`);
return errorResponse(HTTP_STATUS.BAD_REQUEST, `No credentials for provider: ${provider}`);
}
log.warn("EMBEDDINGS", "No more accounts available", { provider });
return errorResponse(lastStatus || HTTP_STATUS.SERVICE_UNAVAILABLE, lastError || "All accounts unavailable");
}
log.info("AUTH", `\x1b[32mUsing ${provider} account: ${credentials.connectionName}\x1b[0m`);
feat: add /v1/embeddings endpoint (OpenAI-compatible) (#146) * feat: implement /v1/embeddings endpoint (#117) Add OpenAI-compatible POST /v1/embeddings endpoint that routes through the existing provider credential + fallback infrastructure. Changes: - open-sse/handlers/embeddingsCore.js: core handler (handleEmbeddingsCore) * Validates input (string or array), encoding_format * Builds provider-specific URL and headers for openai, openrouter, and openai-compatible providers * Handles 401/403 token refresh via executor.refreshCredentials * Returns normalized OpenAI-format response { object: 'list', data, model, usage } - cloud/src/handlers/embeddings.js: cloud Worker handler (handleEmbeddings) * Auth + machineId resolution identical to handleChat * Provider credential fallback loop with rate-limit tracking - cloud/src/index.js: wire new routes * POST /v1/embeddings (new format — machineId from API key) * POST /{machineId}/v1/embeddings (old format — machineId from URL) * test: add unit tests for /v1/embeddings endpoint - Setup vitest as test framework (tests/ directory) - embeddingsCore.test.js (36 tests): - buildEmbeddingsBody: single string, array, encoding_format, default float - buildEmbeddingsUrl: openai, openrouter, openai-compatible-*, unsupported - buildEmbeddingsHeaders: per-provider headers, accessToken fallback - handleEmbeddingsCore: input validation, success path, provider errors, network errors, invalid JSON, token refresh 401 handling - embeddings.cloud.test.js (23 tests): - CORS OPTIONS preflight - Auth: missing/invalid/old-format/wrong key → 401/400 - Body validation: bad JSON, missing model, missing input, bad model → 400 - Happy path: single string, array, delegation, CORS header, machineId override - Rate limiting: all-rate-limited → 429 + Retry-After, no credentials → 400 - Error propagation: non-fallback errors, 429 exhausts accounts Total: 59/59 tests passing Framework: vitest v4.0.18, Node v22.22.0 * feat: add Next.js API route for /v1/embeddings endpoint Wire the embeddings handler into Next.js App Router. - src/app/api/v1/embeddings/route.js: Next.js API route (POST + OPTIONS) - src/sse/handlers/embeddings.js: SSE-layer handler mirroring chat.js pattern Uses handleEmbeddingsCore from open-sse/handlers/embeddingsCore.js with the same auth, credential fallback, and token refresh logic as the chat handler. Supports REQUIRE_API_KEY env var, provider fallback loop, and consistent logging.
2026-02-18 01:24:02 -05:00
const refreshedCredentials = await checkAndRefreshToken(provider, credentials);
const result = await handleEmbeddingsCore({
body: { ...body, model: `${provider}/${model}` },
modelInfo: { provider, model },
credentials: refreshedCredentials,
log,
onCredentialsRefreshed: async (newCreds) => {
await updateProviderCredentials(credentials.connectionId, {
accessToken: newCreds.accessToken,
refreshToken: newCreds.refreshToken,
providerSpecificData: newCreds.providerSpecificData,
testStatus: "active"
});
},
onRequestSuccess: async () => {
2026-02-27 22:04:57 -05:00
await clearAccountError(credentials.connectionId, credentials, model);
feat: add /v1/embeddings endpoint (OpenAI-compatible) (#146) * feat: implement /v1/embeddings endpoint (#117) Add OpenAI-compatible POST /v1/embeddings endpoint that routes through the existing provider credential + fallback infrastructure. Changes: - open-sse/handlers/embeddingsCore.js: core handler (handleEmbeddingsCore) * Validates input (string or array), encoding_format * Builds provider-specific URL and headers for openai, openrouter, and openai-compatible providers * Handles 401/403 token refresh via executor.refreshCredentials * Returns normalized OpenAI-format response { object: 'list', data, model, usage } - cloud/src/handlers/embeddings.js: cloud Worker handler (handleEmbeddings) * Auth + machineId resolution identical to handleChat * Provider credential fallback loop with rate-limit tracking - cloud/src/index.js: wire new routes * POST /v1/embeddings (new format — machineId from API key) * POST /{machineId}/v1/embeddings (old format — machineId from URL) * test: add unit tests for /v1/embeddings endpoint - Setup vitest as test framework (tests/ directory) - embeddingsCore.test.js (36 tests): - buildEmbeddingsBody: single string, array, encoding_format, default float - buildEmbeddingsUrl: openai, openrouter, openai-compatible-*, unsupported - buildEmbeddingsHeaders: per-provider headers, accessToken fallback - handleEmbeddingsCore: input validation, success path, provider errors, network errors, invalid JSON, token refresh 401 handling - embeddings.cloud.test.js (23 tests): - CORS OPTIONS preflight - Auth: missing/invalid/old-format/wrong key → 401/400 - Body validation: bad JSON, missing model, missing input, bad model → 400 - Happy path: single string, array, delegation, CORS header, machineId override - Rate limiting: all-rate-limited → 429 + Retry-After, no credentials → 400 - Error propagation: non-fallback errors, 429 exhausts accounts Total: 59/59 tests passing Framework: vitest v4.0.18, Node v22.22.0 * feat: add Next.js API route for /v1/embeddings endpoint Wire the embeddings handler into Next.js App Router. - src/app/api/v1/embeddings/route.js: Next.js API route (POST + OPTIONS) - src/sse/handlers/embeddings.js: SSE-layer handler mirroring chat.js pattern Uses handleEmbeddingsCore from open-sse/handlers/embeddingsCore.js with the same auth, credential fallback, and token refresh logic as the chat handler. Supports REQUIRE_API_KEY env var, provider fallback loop, and consistent logging.
2026-02-18 01:24:02 -05:00
}
});
if (result.success) return result.response;
const { shouldFallback } = await markAccountUnavailable(credentials.connectionId, result.status, result.error, provider, model);
if (shouldFallback) {
log.warn("AUTH", `Account ${credentials.connectionName} unavailable (${result.status}), trying fallback`);
2026-03-13 22:37:29 -04:00
excludeConnectionIds.add(credentials.connectionId);
feat: add /v1/embeddings endpoint (OpenAI-compatible) (#146) * feat: implement /v1/embeddings endpoint (#117) Add OpenAI-compatible POST /v1/embeddings endpoint that routes through the existing provider credential + fallback infrastructure. Changes: - open-sse/handlers/embeddingsCore.js: core handler (handleEmbeddingsCore) * Validates input (string or array), encoding_format * Builds provider-specific URL and headers for openai, openrouter, and openai-compatible providers * Handles 401/403 token refresh via executor.refreshCredentials * Returns normalized OpenAI-format response { object: 'list', data, model, usage } - cloud/src/handlers/embeddings.js: cloud Worker handler (handleEmbeddings) * Auth + machineId resolution identical to handleChat * Provider credential fallback loop with rate-limit tracking - cloud/src/index.js: wire new routes * POST /v1/embeddings (new format — machineId from API key) * POST /{machineId}/v1/embeddings (old format — machineId from URL) * test: add unit tests for /v1/embeddings endpoint - Setup vitest as test framework (tests/ directory) - embeddingsCore.test.js (36 tests): - buildEmbeddingsBody: single string, array, encoding_format, default float - buildEmbeddingsUrl: openai, openrouter, openai-compatible-*, unsupported - buildEmbeddingsHeaders: per-provider headers, accessToken fallback - handleEmbeddingsCore: input validation, success path, provider errors, network errors, invalid JSON, token refresh 401 handling - embeddings.cloud.test.js (23 tests): - CORS OPTIONS preflight - Auth: missing/invalid/old-format/wrong key → 401/400 - Body validation: bad JSON, missing model, missing input, bad model → 400 - Happy path: single string, array, delegation, CORS header, machineId override - Rate limiting: all-rate-limited → 429 + Retry-After, no credentials → 400 - Error propagation: non-fallback errors, 429 exhausts accounts Total: 59/59 tests passing Framework: vitest v4.0.18, Node v22.22.0 * feat: add Next.js API route for /v1/embeddings endpoint Wire the embeddings handler into Next.js App Router. - src/app/api/v1/embeddings/route.js: Next.js API route (POST + OPTIONS) - src/sse/handlers/embeddings.js: SSE-layer handler mirroring chat.js pattern Uses handleEmbeddingsCore from open-sse/handlers/embeddingsCore.js with the same auth, credential fallback, and token refresh logic as the chat handler. Supports REQUIRE_API_KEY env var, provider fallback loop, and consistent logging.
2026-02-18 01:24:02 -05:00
lastError = result.error;
lastStatus = result.status;
continue;
}
return result.response;
}
}