abbyswag/qwen-code
1
0
Fork 0
mirror of https://github.com/QwenLM/qwen-code.git synced 2025-12-20 16:57:46 +00:00
Code Issues Packages Projects Releases Wiki Activity

Remove auto-execution on Flash in the event of a 429/Quota failover (#3662)

Browse Source 
Co-authored-by: Jenna Inouye <jinouye@google.com>
This commit is contained in:
Bryan Morgan
2025-07-09 13:55:56 -04:00
committed by GitHub
parent 01e756481f
commit 8a6509ffeb
14 changed files with 292 additions and 86 deletions
Download Patch File Download Diff File Expand all files Collapse all files

64
packages/cli/src/ui/utils/errorParsing.test.ts
View File

@@ -39,7 +39,7 @@ describe('parseAndFormatApiError', () => {
);
expect(result).toContain('[API Error: Rate limit exceeded');
expect(result).toContain(
'Slow response times detected. Switching to the gemini-2.5-flash model',
'Possible quota limitations in place or slow response times detected. Switching to the gemini-2.5-flash model',
);
});
@@ -55,7 +55,7 @@ describe('parseAndFormatApiError', () => {
);
expect(result).toContain('[API Error: Rate limit exceeded');
expect(result).toContain(
'Slow response times detected. Switching to the gemini-2.5-flash model',
'Possible quota limitations in place or slow response times detected. Switching to the gemini-2.5-flash model',
);
});
@@ -169,7 +169,7 @@ describe('parseAndFormatApiError', () => {
);
expect(result).toContain('[API Error: Rate limit exceeded');
expect(result).toContain(
'Slow response times detected. Switching to the gemini-2.5-flash model',
'Possible quota limitations in place or slow response times detected. Switching to the gemini-2.5-flash model',
);
expect(result).not.toContain(
'You have reached your daily gemini-2.5-pro quota limit',
@@ -262,21 +262,17 @@ describe('parseAndFormatApiError', () => {
);
});
it('should handle different Gemini version strings in Pro quota exceeded errors', () => {
const errorMessage15 =
'got status: 429 Too Many Requests. {"error":{"code":429,"message":"Quota exceeded for quota metric \'Gemini 1.5 Pro Requests\' and limit \'RequestsPerDay\' of service \'generativelanguage.googleapis.com\' for consumer \'project_number:123456789\'.","status":"RESOURCE_EXHAUSTED"}}';
it('should handle different Gemini 2.5 version strings in Pro quota exceeded errors', () => {
const errorMessage25 =
'got status: 429 Too Many Requests. {"error":{"code":429,"message":"Quota exceeded for quota metric \'Gemini 2.5 Pro Requests\' and limit \'RequestsPerDay\' of service \'generativelanguage.googleapis.com\' for consumer \'project_number:123456789\'.","status":"RESOURCE_EXHAUSTED"}}';
const errorMessagePreview =
'got status: 429 Too Many Requests. {"error":{"code":429,"message":"Quota exceeded for quota metric \'Gemini 2.5-preview Pro Requests\' and limit \'RequestsPerDay\' of service \'generativelanguage.googleapis.com\' for consumer \'project_number:123456789\'.","status":"RESOURCE_EXHAUSTED"}}';
const errorMessageBeta =
'got status: 429 Too Many Requests. {"error":{"code":429,"message":"Quota exceeded for quota metric \'Gemini beta-3.0 Pro Requests\' and limit \'RequestsPerDay\' of service \'generativelanguage.googleapis.com\' for consumer \'project_number:123456789\'.","status":"RESOURCE_EXHAUSTED"}}';
const errorMessageExperimental =
'got status: 429 Too Many Requests. {"error":{"code":429,"message":"Quota exceeded for quota metric \'Gemini experimental-v2 Pro Requests\' and limit \'RequestsPerDay\' of service \'generativelanguage.googleapis.com\' for consumer \'project_number:123456789\'.","status":"RESOURCE_EXHAUSTED"}}';
const result15 = parseAndFormatApiError(
errorMessage15,
const result25 = parseAndFormatApiError(
errorMessage25,
AuthType.LOGIN_WITH_GOOGLE,
undefined,
'gemini-1.5-pro',
'gemini-2.5-pro',
DEFAULT_GEMINI_FLASH_MODEL,
);
const resultPreview = parseAndFormatApiError(
@@ -286,45 +282,19 @@ describe('parseAndFormatApiError', () => {
'gemini-2.5-preview-pro',
DEFAULT_GEMINI_FLASH_MODEL,
);
const resultBeta = parseAndFormatApiError(
errorMessageBeta,
AuthType.LOGIN_WITH_GOOGLE,
undefined,
'gemini-beta-3.0-pro',
DEFAULT_GEMINI_FLASH_MODEL,
);
const resultExperimental = parseAndFormatApiError(
errorMessageExperimental,
AuthType.LOGIN_WITH_GOOGLE,
undefined,
'gemini-experimental-v2-pro',
DEFAULT_GEMINI_FLASH_MODEL,
);
expect(result15).toContain(
'You have reached your daily gemini-1.5-pro quota limit',
expect(result25).toContain(
'You have reached your daily gemini-2.5-pro quota limit',
);
expect(resultPreview).toContain(
'You have reached your daily gemini-2.5-preview-pro quota limit',
);
expect(resultBeta).toContain(
'You have reached your daily gemini-beta-3.0-pro quota limit',
);
expect(resultExperimental).toContain(
'You have reached your daily gemini-experimental-v2-pro quota limit',
);
expect(result15).toContain(
expect(result25).toContain(
'upgrade to a Gemini Code Assist Standard or Enterprise plan',
);
expect(resultPreview).toContain(
'upgrade to a Gemini Code Assist Standard or Enterprise plan',
);
expect(resultBeta).toContain(
'upgrade to a Gemini Code Assist Standard or Enterprise plan',
);
expect(resultExperimental).toContain(
'upgrade to a Gemini Code Assist Standard or Enterprise plan',
);
});
it('should not match non-Pro models with similar version strings', () => {
@@ -339,16 +309,6 @@ describe('parseAndFormatApiError', () => {
"Quota exceeded for quota metric 'Gemini 2.5-preview Flash Requests' and limit",
),
).toBe(false);
expect(
isProQuotaExceededError(
"Quota exceeded for quota metric 'Gemini beta-3.0 Flash Requests' and limit",
),
).toBe(false);
expect(
isProQuotaExceededError(
"Quota exceeded for quota metric 'Gemini experimental-v2 Flash Requests' and limit",
),
).toBe(false);
// Test other model types
expect(

6
packages/cli/src/ui/utils/errorParsing.ts
View File

@@ -19,7 +19,7 @@ import {
const getRateLimitErrorMessageGoogleFree = (
fallbackModel: string = DEFAULT_GEMINI_FLASH_MODEL,
) =>
`\nSlow response times detected. Switching to the ${fallbackModel} model for the rest of this session.`;
`\nPossible quota limitations in place or slow response times detected. Switching to the ${fallbackModel} model for the rest of this session.`;
const getRateLimitErrorMessageGoogleProQuotaFree = (
currentModel: string = DEFAULT_GEMINI_MODEL,
@@ -34,7 +34,7 @@ const getRateLimitErrorMessageGoogleGenericQuotaFree = () =>
const getRateLimitErrorMessageGooglePaid = (
fallbackModel: string = DEFAULT_GEMINI_FLASH_MODEL,
) =>
`\nSlow response times detected. Switching to the ${fallbackModel} model for the rest of this session. We appreciate you for choosing Gemini Code Assist and the Gemini CLI.`;
`\nPossible quota limitations in place or slow response times detected. Switching to the ${fallbackModel} model for the rest of this session. We appreciate you for choosing Gemini Code Assist and the Gemini CLI.`;
const getRateLimitErrorMessageGoogleProQuotaPaid = (
currentModel: string = DEFAULT_GEMINI_MODEL,
@@ -53,7 +53,7 @@ const RATE_LIMIT_ERROR_MESSAGE_VERTEX =
const getRateLimitErrorMessageDefault = (
fallbackModel: string = DEFAULT_GEMINI_FLASH_MODEL,
) =>
`\nSlow response times detected. Switching to the ${fallbackModel} model for the rest of this session.`;
`\nPossible quota limitations in place or slow response times detected. Switching to the ${fallbackModel} model for the rest of this session.`;
function getRateLimitMessage(
authType?: AuthType,