* feat: apply local user changes and fixes * ;D * Implement OpenAI support, model-specific rate limiting, and robustness fixes * docs: update pr title * feat: ensure unique openai models endpoint * fix: startup banner alignment and removed duplicates * feat: add model fallback system with --fallback flag * fix: accounts cli hanging after completion * feat: add exit option to accounts cli menu * fix: remove circular dependency warning for fallback flag * feat: show active modes in banner and hide their flags * Remove OpenAI compatibility and fallback features from PR #35 Cherry-picked selective fixes from PR #35 while removing: - OpenAI-compatible API endpoints (/openai/v1/*) - Model fallback system (fallback-config.js) - Thinking block skip for Gemini models - Unnecessary files (pullrequest.md, test-fix.js, test-openai.js) Retained improvements: - Network error handling with retry logic - Model-specific rate limiting - Enhanced health check with quota info - CLI fixes (exit option, process.exit) - Startup banner alignment (debug mode only) 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com> * banner alignment fix * Refactor: Model-specific rate limits and cleanup deprecated code - Remove global rate limit fields (isRateLimited, rateLimitResetTime) in favor of model-specific limits (modelRateLimits[modelId]) - Remove deprecated wrapper functions (is429Error, isAuthInvalidError) from handlers - Filter fetchAvailableModels to only return Claude and Gemini models - Fix getCurrentStickyAccount() to pass model param after waiting - Update /account-limits endpoint to show model-specific limits - Remove multi-account OAuth flow to avoid state mismatch errors 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com> * fix: show (x/y) limited status in account-limits table - Status is now "ok" only when all models are available - Shows "(x/y) limited" when x out of y models are exhausted - Provides better visibility into partial rate limiting 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com> * docs: update CLAUDE.md with model-specific rate limiting - Document modelRateLimits[modelId] for per-model rate tracking - Add isNetworkError() helper to utilities section 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com> --------- Co-authored-by: M1noa <minoa@minoa.cat> Co-authored-by: Minoa <altgithub@minoa.cat> Co-authored-by: Claude <noreply@anthropic.com>
This commit is contained in:
Badri Narayanan S
committed by
GitHub
GitHub
parent
2d05dd5b62
commit
9c4a712a9a
@@ -11,27 +11,12 @@ import {
|
||||
MAX_WAIT_BEFORE_ERROR_MS
|
||||
} from '../constants.js';
|
||||
import { isRateLimitError, isAuthError } from '../errors.js';
|
||||
import { formatDuration, sleep } from '../utils/helpers.js';
|
||||
import { formatDuration, sleep, isNetworkError } from '../utils/helpers.js';
|
||||
import { logger } from '../utils/logger.js';
|
||||
import { parseResetTime } from './rate-limit-parser.js';
|
||||
import { buildCloudCodeRequest, buildHeaders } from './request-builder.js';
|
||||
import { streamSSEResponse } from './sse-streamer.js';
|
||||
|
||||
/**
|
||||
* Check if an error is a rate limit error (429 or RESOURCE_EXHAUSTED)
|
||||
* @deprecated Use isRateLimitError from errors.js instead
|
||||
*/
|
||||
function is429Error(error) {
|
||||
return isRateLimitError(error);
|
||||
}
|
||||
|
||||
/**
|
||||
* Check if an error is an auth-invalid error (credentials need re-authentication)
|
||||
* @deprecated Use isAuthError from errors.js instead
|
||||
*/
|
||||
function isAuthInvalidError(error) {
|
||||
return isAuthError(error);
|
||||
}
|
||||
|
||||
/**
|
||||
* Send a streaming request to Cloud Code with multi-account support
|
||||
@@ -56,7 +41,7 @@ export async function* sendMessageStream(anthropicRequest, accountManager) {
|
||||
|
||||
for (let attempt = 0; attempt < maxAttempts; attempt++) {
|
||||
// Use sticky account selection for cache continuity
|
||||
const { account: stickyAccount, waitMs } = accountManager.pickStickyAccount();
|
||||
const { account: stickyAccount, waitMs } = accountManager.pickStickyAccount(model);
|
||||
let account = stickyAccount;
|
||||
|
||||
// Handle waiting for sticky account
|
||||
@@ -64,19 +49,19 @@ export async function* sendMessageStream(anthropicRequest, accountManager) {
|
||||
logger.info(`[CloudCode] Waiting ${formatDuration(waitMs)} for sticky account...`);
|
||||
await sleep(waitMs);
|
||||
accountManager.clearExpiredLimits();
|
||||
account = accountManager.getCurrentStickyAccount();
|
||||
account = accountManager.getCurrentStickyAccount(model);
|
||||
}
|
||||
|
||||
// Handle all accounts rate-limited
|
||||
if (!account) {
|
||||
if (accountManager.isAllRateLimited()) {
|
||||
const allWaitMs = accountManager.getMinWaitTimeMs();
|
||||
if (accountManager.isAllRateLimited(model)) {
|
||||
const allWaitMs = accountManager.getMinWaitTimeMs(model);
|
||||
const resetTime = new Date(Date.now() + allWaitMs).toISOString();
|
||||
|
||||
// If wait time is too long (> 2 minutes), throw error immediately
|
||||
if (allWaitMs > MAX_WAIT_BEFORE_ERROR_MS) {
|
||||
throw new Error(
|
||||
`RESOURCE_EXHAUSTED: Rate limited. Quota will reset after ${formatDuration(allWaitMs)}. Next available: ${resetTime}`
|
||||
`RESOURCE_EXHAUSTED: Rate limited on ${model}. Quota will reset after ${formatDuration(allWaitMs)}. Next available: ${resetTime}`
|
||||
);
|
||||
}
|
||||
|
||||
@@ -85,7 +70,7 @@ export async function* sendMessageStream(anthropicRequest, accountManager) {
|
||||
logger.warn(`[CloudCode] All ${accountCount} account(s) rate-limited. Waiting ${formatDuration(allWaitMs)}...`);
|
||||
await sleep(allWaitMs);
|
||||
accountManager.clearExpiredLimits();
|
||||
account = accountManager.pickNext();
|
||||
account = accountManager.pickNext(model);
|
||||
}
|
||||
|
||||
if (!account) {
|
||||
@@ -153,7 +138,7 @@ export async function* sendMessageStream(anthropicRequest, accountManager) {
|
||||
return;
|
||||
|
||||
} catch (endpointError) {
|
||||
if (is429Error(endpointError)) {
|
||||
if (isRateLimitError(endpointError)) {
|
||||
throw endpointError; // Re-throw to trigger account switch
|
||||
}
|
||||
logger.warn(`[CloudCode] Stream error at ${endpoint}:`, endpointError.message);
|
||||
@@ -166,19 +151,19 @@ export async function* sendMessageStream(anthropicRequest, accountManager) {
|
||||
// If all endpoints returned 429, mark account as rate-limited
|
||||
if (lastError.is429) {
|
||||
logger.warn(`[CloudCode] All endpoints rate-limited for ${account.email}`);
|
||||
accountManager.markRateLimited(account.email, lastError.resetMs);
|
||||
accountManager.markRateLimited(account.email, lastError.resetMs, model);
|
||||
throw new Error(`Rate limited: ${lastError.errorText}`);
|
||||
}
|
||||
throw lastError;
|
||||
}
|
||||
|
||||
} catch (error) {
|
||||
if (is429Error(error)) {
|
||||
if (isRateLimitError(error)) {
|
||||
// Rate limited - already marked, continue to next account
|
||||
logger.info(`[CloudCode] Account ${account.email} rate-limited, trying next...`);
|
||||
continue;
|
||||
}
|
||||
if (isAuthInvalidError(error)) {
|
||||
if (isAuthError(error)) {
|
||||
// Auth invalid - already marked, continue to next account
|
||||
logger.warn(`[CloudCode] Account ${account.email} has invalid credentials, trying next...`);
|
||||
continue;
|
||||
@@ -187,10 +172,17 @@ export async function* sendMessageStream(anthropicRequest, accountManager) {
|
||||
// UNLESS it's a 500 error, then we treat it as a "soft" failure for this account and try the next one
|
||||
if (error.message.includes('API error 5') || error.message.includes('500') || error.message.includes('503')) {
|
||||
logger.warn(`[CloudCode] Account ${account.email} failed with 5xx stream error, trying next...`);
|
||||
accountManager.pickNext(); // Force advance to next account
|
||||
accountManager.pickNext(model); // Force advance to next account
|
||||
continue;
|
||||
}
|
||||
|
||||
if (isNetworkError(error)) {
|
||||
logger.warn(`[CloudCode] Network error for ${account.email} (stream), trying next account... (${error.message})`);
|
||||
await sleep(1000); // Brief pause before retry
|
||||
accountManager.pickNext(model); // Advance to next account
|
||||
continue;
|
||||
}
|
||||
|
||||
throw error;
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user