xiaxinhang/pr-agent
1
0
Fork 0
mirror of https://github.com/qodo-ai/pr-agent.git synced 2025-07-03 20:30:41 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #1808 from group-3-sPRinter/fix/retry-exclude-rate-limit-error

Browse Source 
fix: exclude RateLimitError from `@retry` in `AIHandler.chat_completion()`
This commit is contained in:
Tal
2025-05-24 16:14:25 +03:00
committed by GitHub
parent 4c98cffd37 466ec4ce90
commit 20b1a1f552
3 changed files with 34 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
pr_agent/algo/ai_handlers/langchain_ai_handler.py
View File

@ -6,8 +6,8 @@ except: # we don't enforce langchain as a dependency, so if it's not installed,
import functools import functools
from openai import APIError, RateLimitError, Timeout import openai
from retry import retry from tenacity import retry, retry_if_exception_type, retry_if_not_exception_type, stop_after_attempt
from pr_agent.algo.ai_handlers.base_ai_handler import BaseAiHandler from pr_agent.algo.ai_handlers.base_ai_handler import BaseAiHandler
from pr_agent.config_loader import get_settings from pr_agent.config_loader import get_settings
@ -36,8 +36,10 @@ class LangChainOpenAIHandler(BaseAiHandler):
""" """
return get_settings().get("OPENAI.DEPLOYMENT_ID", None) return get_settings().get("OPENAI.DEPLOYMENT_ID", None)
@retry(exceptions=(APIError, Timeout, AttributeError, RateLimitError), @retry(
tries=OPENAI_RETRIES, delay=2, backoff=2, jitter=(1, 3)) retry=retry_if_exception_type(openai.APIError) & retry_if_not_exception_type(openai.RateLimitError),
stop=stop_after_attempt(OPENAI_RETRIES),
)
async def chat_completion(self, model: str, system: str, user: str, temperature: float = 0.2): async def chat_completion(self, model: str, system: str, user: str, temperature: float = 0.2):
try: try:
messages = [SystemMessage(content=system), HumanMessage(content=user)] messages = [SystemMessage(content=system), HumanMessage(content=user)]
@ -47,9 +49,15 @@ class LangChainOpenAIHandler(BaseAiHandler):
finish_reason = "completed" finish_reason = "completed"
return resp.content, finish_reason return resp.content, finish_reason
except (Exception) as e: except openai.RateLimitError as e:
get_logger().error("Unknown error during OpenAI inference: ", e) get_logger().error(f"Rate limit error during LLM inference: {e}")
raise e raise
except openai.APIError as e:
get_logger().warning(f"Error during LLM inference: {e}")
raise
except Exception as e:
get_logger().warning(f"Unknown error during LLM inference: {e}")
raise openai.APIError from e
def _create_chat(self, deployment_id=None): def _create_chat(self, deployment_id=None):
try: try:

12
pr_agent/algo/ai_handlers/litellm_ai_handler.py
View File

@ -3,7 +3,7 @@ import litellm
import openai import openai
import requests import requests
from litellm import acompletion from litellm import acompletion
from tenacity import retry, retry_if_exception_type, stop_after_attempt from tenacity import retry, retry_if_exception_type, retry_if_not_exception_type, stop_after_attempt
from pr_agent.algo import CLAUDE_EXTENDED_THINKING_MODELS, NO_SUPPORT_TEMPERATURE_MODELS, SUPPORT_REASONING_EFFORT_MODELS, USER_MESSAGE_ONLY_MODELS from pr_agent.algo import CLAUDE_EXTENDED_THINKING_MODELS, NO_SUPPORT_TEMPERATURE_MODELS, SUPPORT_REASONING_EFFORT_MODELS, USER_MESSAGE_ONLY_MODELS
from pr_agent.algo.ai_handlers.base_ai_handler import BaseAiHandler from pr_agent.algo.ai_handlers.base_ai_handler import BaseAiHandler
@ -274,8 +274,8 @@ class LiteLLMAIHandler(BaseAiHandler):
return get_settings().get("OPENAI.DEPLOYMENT_ID", None) return get_settings().get("OPENAI.DEPLOYMENT_ID", None)
@retry( @retry(
retry=retry_if_exception_type((openai.APIError, openai.APIConnectionError, openai.APITimeoutError)), # No retry on RateLimitError retry=retry_if_exception_type(openai.APIError) & retry_if_not_exception_type(openai.RateLimitError),
stop=stop_after_attempt(OPENAI_RETRIES) stop=stop_after_attempt(OPENAI_RETRIES),
) )
async def chat_completion(self, model: str, system: str, user: str, temperature: float = 0.2, img_path: str = None): async def chat_completion(self, model: str, system: str, user: str, temperature: float = 0.2, img_path: str = None):
try: try:
@ -371,13 +371,13 @@ class LiteLLMAIHandler(BaseAiHandler):
get_logger().info(f"\nUser prompt:\n{user}") get_logger().info(f"\nUser prompt:\n{user}")
response = await acompletion(**kwargs) response = await acompletion(**kwargs)
except (openai.RateLimitError) as e: except openai.RateLimitError as e:
get_logger().error(f"Rate limit error during LLM inference: {e}") get_logger().error(f"Rate limit error during LLM inference: {e}")
raise raise
except (openai.APIError, openai.APITimeoutError) as e: except openai.APIError as e:
get_logger().warning(f"Error during LLM inference: {e}") get_logger().warning(f"Error during LLM inference: {e}")
raise raise
except (Exception) as e: except Exception as e:
get_logger().warning(f"Unknown error during LLM inference: {e}") get_logger().warning(f"Unknown error during LLM inference: {e}")
raise openai.APIError from e raise openai.APIError from e
if response is None or len(response["choices"]) == 0: if response is None or len(response["choices"]) == 0:

24
pr_agent/algo/ai_handlers/openai_ai_handler.py
View File

@ -1,8 +1,8 @@
from os import environ from os import environ
from pr_agent.algo.ai_handlers.base_ai_handler import BaseAiHandler from pr_agent.algo.ai_handlers.base_ai_handler import BaseAiHandler
import openai import openai
from openai import APIError, AsyncOpenAI, RateLimitError, Timeout from openai import AsyncOpenAI
from retry import retry from tenacity import retry, retry_if_exception_type, retry_if_not_exception_type, stop_after_attempt
from pr_agent.algo.ai_handlers.base_ai_handler import BaseAiHandler from pr_agent.algo.ai_handlers.base_ai_handler import BaseAiHandler
from pr_agent.config_loader import get_settings from pr_agent.config_loader import get_settings
@ -38,8 +38,10 @@ class OpenAIHandler(BaseAiHandler):
""" """
return get_settings().get("OPENAI.DEPLOYMENT_ID", None) return get_settings().get("OPENAI.DEPLOYMENT_ID", None)
@retry(exceptions=(APIError, Timeout, AttributeError, RateLimitError), @retry(
tries=OPENAI_RETRIES, delay=2, backoff=2, jitter=(1, 3)) retry=retry_if_exception_type(openai.APIError) & retry_if_not_exception_type(openai.RateLimitError),
stop=stop_after_attempt(OPENAI_RETRIES),
)
async def chat_completion(self, model: str, system: str, user: str, temperature: float = 0.2): async def chat_completion(self, model: str, system: str, user: str, temperature: float = 0.2):
try: try:
get_logger().info("System: ", system) get_logger().info("System: ", system)
@ -57,12 +59,12 @@ class OpenAIHandler(BaseAiHandler):
get_logger().info("AI response", response=resp, messages=messages, finish_reason=finish_reason, get_logger().info("AI response", response=resp, messages=messages, finish_reason=finish_reason,
model=model, usage=usage) model=model, usage=usage)
return resp, finish_reason return resp, finish_reason
except (APIError, Timeout) as e: except openai.RateLimitError as e:
get_logger().error("Error during OpenAI inference: ", e) get_logger().error(f"Rate limit error during LLM inference: {e}")
raise raise
except (RateLimitError) as e: except openai.APIError as e:
get_logger().error("Rate limit error during OpenAI inference: ", e) get_logger().warning(f"Error during LLM inference: {e}")
raise
except (Exception) as e:
get_logger().error("Unknown error during OpenAI inference: ", e)
raise raise
except Exception as e:
get_logger().warning(f"Unknown error during LLM inference: {e}")
raise openai.APIError from e