fix(patch): cherry-pick ee92db7 to release/v0.11.0-pr-11624 to patch version v0.11.0 and create version 0.11.1 (#12321)

Co-authored-by: Gaurav <39389231+gsquared94@users.noreply.github.com>

Author: gemini-cli-robot

packages/cli/src/ui/hooks/useQuotaAndFallback.test.ts

@@ -19,25 +19,15 @@ import {
   type FallbackModelHandler,
   UserTierId,
   AuthType,
-  isGenericQuotaExceededError,
-  isProQuotaExceededError,
+  TerminalQuotaError,
   makeFakeConfig,
+  type GoogleApiError,
+  RetryableQuotaError,
 } from '@google/gemini-cli-core';
 import { useQuotaAndFallback } from './useQuotaAndFallback.js';
 import type { UseHistoryManagerReturn } from './useHistoryManager.js';
 import { AuthState, MessageType } from '../types.js';
 
-// Mock the error checking functions from the core package to control test scenarios
-vi.mock('@google/gemini-cli-core', async (importOriginal) => {
-  const original =
-    await importOriginal<typeof import('@google/gemini-cli-core')>();
-  return {
-    ...original,
-    isGenericQuotaExceededError: vi.fn(),
-    isProQuotaExceededError: vi.fn(),
-  };
-});
-
 // Use a type alias for SpyInstance as it's not directly exported
 type SpyInstance = ReturnType<typeof vi.spyOn>;
 
@@ -47,12 +37,15 @@ describe('useQuotaAndFallback', () => {
   let mockSetAuthState: Mock;
   let mockSetModelSwitchedFromQuotaError: Mock;
   let setFallbackHandlerSpy: SpyInstance;
-
-  const mockedIsGenericQuotaExceededError = isGenericQuotaExceededError as Mock;
-  const mockedIsProQuotaExceededError = isProQuotaExceededError as Mock;
+  let mockGoogleApiError: GoogleApiError;
 
   beforeEach(() => {
     mockConfig = makeFakeConfig();
+    mockGoogleApiError = {
+      code: 429,
+      message: 'mock error',
+      details: [],
+    };
 
     // Spy on the method that requires the private field and mock its return.
     // This is cleaner than modifying the config class for tests.
@@ -72,9 +65,6 @@ describe('useQuotaAndFallback', () => {
 
     setFallbackHandlerSpy = vi.spyOn(mockConfig, 'setFallbackModelHandler');
     vi.spyOn(mockConfig, 'setQuotaErrorOccurred');
-
-    mockedIsGenericQuotaExceededError.mockReturnValue(false);
-    mockedIsProQuotaExceededError.mockReturnValue(false);
   });
 
   afterEach(() => {
@@ -140,51 +130,62 @@ describe('useQuotaAndFallback', () => {
     describe('Automatic Fallback Scenarios', () => {
       const testCases = [
         {
-          errorType: 'generic',
+          description: 'other error for FREE tier',
           tier: UserTierId.FREE,
+          error: new Error('some error'),
           expectedMessageSnippets: [
-            'Automatically switching from model-A to model-B',
+            'Automatically switching from model-A to model-B for faster responses',
             'upgrade to a Gemini Code Assist Standard or Enterprise plan',
           ],
         },
         {
-          errorType: 'generic',
-          tier: UserTierId.STANDARD, // Paid tier
+          description: 'other error for LEGACY tier',
+          tier: UserTierId.LEGACY, // Paid tier
+          error: new Error('some error'),
           expectedMessageSnippets: [
-            'Automatically switching from model-A to model-B',
+            'Automatically switching from model-A to model-B for faster responses',
             'switch to using a paid API key from AI Studio',
           ],
         },
         {
-          errorType: 'other',
+          description: 'retryable quota error for FREE tier',
           tier: UserTierId.FREE,
+          error: new RetryableQuotaError(
+            'retryable quota',
+            mockGoogleApiError,
+            5,
+          ),
           expectedMessageSnippets: [
-            'Automatically switching from model-A to model-B for faster responses',
-            'upgrade to a Gemini Code Assist Standard or Enterprise plan',
+            'Your requests are being throttled right now due to server being at capacity for model-A',
+            'Automatically switching from model-A to model-B',
+            'upgrading to a Gemini Code Assist Standard or Enterprise plan',
           ],
         },
         {
-          errorType: 'other',
+          description: 'retryable quota error for LEGACY tier',
           tier: UserTierId.LEGACY, // Paid tier
+          error: new RetryableQuotaError(
+            'retryable quota',
+            mockGoogleApiError,
+            5,
+          ),
           expectedMessageSnippets: [
-            'Automatically switching from model-A to model-B for faster responses',
+            'Your requests are being throttled right now due to server being at capacity for model-A',
+            'Automatically switching from model-A to model-B',
             'switch to using a paid API key from AI Studio',
           ],
         },
       ];
 
-      for (const { errorType, tier, expectedMessageSnippets } of testCases) {
-        it(`should handle ${errorType} error for ${tier} tier correctly`, async () => {
-          mockedIsGenericQuotaExceededError.mockReturnValue(
-            errorType === 'generic',
-          );
-
+      for (const {
+        description,
+        tier,
+        error,
+        expectedMessageSnippets,
+      } of testCases) {
+        it(`should handle ${description} correctly`, async () => {
           const handler = getRegisteredHandler(tier);
-          const result = await handler(
-            'model-A',
-            'model-B',
-            new Error('quota exceeded'),
-          );
+          const result = await handler('model-A', 'model-B', error);
 
           // Automatic fallbacks should return 'stop'
           expect(result).toBe('stop');
@@ -207,10 +208,6 @@ describe('useQuotaAndFallback', () => {
     });
 
     describe('Interactive Fallback (Pro Quota Error)', () => {
-      beforeEach(() => {
-        mockedIsProQuotaExceededError.mockReturnValue(true);
-      });
-
       it('should set an interactive request and wait for user choice', async () => {
         const { result } = renderHook(() =>
           useQuotaAndFallback({
@@ -229,7 +226,7 @@ describe('useQuotaAndFallback', () => {
         const promise = handler(
           'gemini-pro',
           'gemini-flash',
-          new Error('pro quota'),
+          new TerminalQuotaError('pro quota', mockGoogleApiError),
         );
 
         await act(async () => {});
@@ -268,7 +265,7 @@ describe('useQuotaAndFallback', () => {
         const promise1 = handler(
           'gemini-pro',
           'gemini-flash',
-          new Error('pro quota 1'),
+          new TerminalQuotaError('pro quota 1', mockGoogleApiError),
         );
         await act(async () => {});
 
@@ -278,7 +275,7 @@ describe('useQuotaAndFallback', () => {
         const result2 = await handler(
           'gemini-pro',
           'gemini-flash',
-          new Error('pro quota 2'),
+          new TerminalQuotaError('pro quota 2', mockGoogleApiError),
         );
 
         // The lock should have stopped the second request
@@ -297,10 +294,6 @@ describe('useQuotaAndFallback', () => {
   });
 
   describe('handleProQuotaChoice', () => {
-    beforeEach(() => {
-      mockedIsProQuotaExceededError.mockReturnValue(true);
-    });
-
     it('should do nothing if there is no pending pro quota request', () => {
       const { result } = renderHook(() =>
         useQuotaAndFallback({
@@ -336,7 +329,7 @@ describe('useQuotaAndFallback', () => {
       const promise = handler(
         'gemini-pro',
         'gemini-flash',
-        new Error('pro quota'),
+        new TerminalQuotaError('pro quota', mockGoogleApiError),
       );
       await act(async () => {}); // Allow state to update
 
@@ -367,7 +360,7 @@ describe('useQuotaAndFallback', () => {
       const promise = handler(
         'gemini-pro',
         'gemini-flash',
-        new Error('pro quota'),
+        new TerminalQuotaError('pro quota', mockGoogleApiError),
       );
       await act(async () => {}); // Allow state to update
 

packages/cli/src/ui/hooks/useQuotaAndFallback.ts

@@ -9,9 +9,9 @@ import {
   type Config,
   type FallbackModelHandler,
   type FallbackIntent,
-  isGenericQuotaExceededError,
-  isProQuotaExceededError,
+  TerminalQuotaError,
   UserTierId,
+  RetryableQuotaError,
 } from '@google/gemini-cli-core';
 import { useCallback, useEffect, useRef, useState } from 'react';
 import { type UseHistoryManagerReturn } from './useHistoryManager.js';
@@ -63,7 +63,7 @@ export function useQuotaAndFallback({
 
       let message: string;
 
-      if (error && isProQuotaExceededError(error)) {
+      if (error instanceof TerminalQuotaError) {
         // Pro Quota specific messages (Interactive)
         if (isPaidTier) {
           message = `⚡ You have reached your daily ${failedModel} quota limit.
@@ -76,31 +76,30 @@ export function useQuotaAndFallback({
 ⚡ Or you can utilize a Gemini API Key. See: https://goo.gle/gemini-cli-docs-auth#gemini-api-key
 ⚡ You can switch authentication methods by typing /auth`;
         }
-      } else if (error && isGenericQuotaExceededError(error)) {
-        // Generic Quota (Automatic fallback)
-        const actionMessage = `⚡ You have reached your daily quota limit.\n⚡ Automatically switching from ${failedModel} to ${fallbackModel} for the remainder of this session.`;
+      } else if (error instanceof RetryableQuotaError) {
+        // Short term quota retries exhausted (Automatic fallback)
+        const actionMessage = `⚡ Your requests are being throttled right now due to server being at capacity for ${failedModel}.\n⚡ Automatically switching from ${failedModel} to ${fallbackModel} for the remainder of this session.`;
 
         if (isPaidTier) {
           message = `${actionMessage}
-⚡ To continue accessing the ${failedModel} model today, consider using /auth to switch to using a paid API key from AI Studio at https://aistudio.google.com/apikey`;
+⚡ To continue accessing the ${failedModel} model, retry your request after some time or consider using /auth to switch to using a paid API key from AI Studio at https://aistudio.google.com/apikey`;
         } else {
           message = `${actionMessage}
-⚡ To increase your limits, upgrade to a Gemini Code Assist Standard or Enterprise plan with higher limits at https://goo.gle/set-up-gemini-code-assist
-⚡ Or you can utilize a Gemini API Key. See: https://goo.gle/gemini-cli-docs-auth#gemini-api-key
+⚡ Retry your requests after some time. Otherwise consider upgrading to a Gemini Code Assist Standard or Enterprise plan with higher limits at https://goo.gle/set-up-gemini-code-assist
 ⚡ You can switch authentication methods by typing /auth`;
         }
       } else {
-        // Consecutive 429s or other errors (Automatic fallback)
+        // Other errors (Automatic fallback)
         const actionMessage = `⚡ Automatically switching from ${failedModel} to ${fallbackModel} for faster responses for the remainder of this session.`;
 
         if (isPaidTier) {
           message = `${actionMessage}
-⚡ Possible reasons for this are that you have received multiple consecutive capacity errors or you have reached your daily ${failedModel} quota limit
-⚡ To continue accessing the ${failedModel} model today, consider using /auth to switch to using a paid API key from AI Studio at https://aistudio.google.com/apikey`;
+⚡ Your requests are being throttled temporarily due to server being at capacity for ${failedModel} or there is a service outage.
+⚡ To continue accessing the ${failedModel} model, you can retry your request after some time or consider using /auth to switch to using a paid API key from AI Studio at https://aistudio.google.com/apikey`;
         } else {
           message = `${actionMessage}
-⚡ Possible reasons for this are that you have received multiple consecutive capacity errors or you have reached your daily ${failedModel} quota limit
-⚡ To increase your limits, upgrade to a Gemini Code Assist Standard or Enterprise plan with higher limits at https://goo.gle/set-up-gemini-code-assist
+⚡ Your requests are being throttled temporarily due to server being at capacity for ${failedModel} or there is a service outage.
+⚡ To avoid being throttled, you can retry your request after some time or upgrade to a Gemini Code Assist Standard or Enterprise plan with higher limits at https://goo.gle/set-up-gemini-code-assist
 ⚡ Or you can utilize a Gemini API Key. See: https://goo.gle/gemini-cli-docs-auth#gemini-api-key
 ⚡ You can switch authentication methods by typing /auth`;
         }
@@ -119,7 +118,7 @@ export function useQuotaAndFallback({
       config.setQuotaErrorOccurred(true);
 
       // Interactive Fallback for Pro quota
-      if (error && isProQuotaExceededError(error)) {
+      if (error instanceof TerminalQuotaError) {
         if (isDialogPending.current) {
           return 'stop'; // A dialog is already active, so just stop this request.
         }

packages/core/index.ts

@@ -44,3 +44,5 @@ export { makeFakeConfig } from './src/test-utils/config.js';
 export * from './src/utils/pathReader.js';
 export { ClearcutLogger } from './src/telemetry/clearcut-logger/clearcut-logger.js';
 export { logModelSlashCommand } from './src/telemetry/loggers.js';
+export * from './src/utils/googleQuotaErrors.js';
+export type { GoogleApiError } from './src/utils/googleErrors.js';

packages/core/src/index.ts

@@ -45,6 +45,7 @@ export * from './utils/gitIgnoreParser.js';
 export * from './utils/gitUtils.js';
 export * from './utils/editor.js';
 export * from './utils/quotaErrorDetection.js';
+export * from './utils/googleQuotaErrors.js';
 export * from './utils/fileUtils.js';
 export * from './utils/retry.js';
 export * from './utils/shell-utils.js';