Merge pull request #206 from gitautoai/wes

Fix: write_pr_body encountered an BadRequestError: Error code: 400 - {'error': {'message': "This model's maximum context length is 128000 tokens. However, your messages resulted in 142586 tokens. Please reduce the length of the messages.", 'type': 'invalid_request_error', 'param': 'messages', 'code': 'context_length_exceeded'}}

config.py

@@ -36,6 +36,7 @@ def get_env_var(name: str) -> str:
 OPENAI_FINAL_STATUSES: list[str] = ["cancelled", "completed", "expired", "failed"]
 OPENAI_MAX_ARRAY_LENGTH = 32  # https://community.openai.com/t/assistant-threads-create-400-messages-array-too-long/754574/1
 OPENAI_MAX_STRING_LENGTH = 256000  # https://community.openai.com/t/assistant-threads-create-400-messages-array-too-long/754574/5
+OPENAI_MAX_CONTEXT_TOKENS = 128000  # https://gitauto-ai.sentry.io/issues/5582421515/events/9a09416e714c4a66bf1bd86916702be2/?project=4506865231200256&referrer=issue_details.related_trace_issue
 OPENAI_MAX_TOKENS = 4096
 OPENAI_MODEL_ID = "gpt-4o"
 OPENAI_ORG_ID: str = get_env_var(name="OPENAI_ORG_ID")

requirements.txt

@@ -1,61 +1,61 @@
 annotated-types==0.7.0
 anyio==4.4.0
-astroid==3.2.2  # required by pylint
+astroid==3.2.2
 black==24.4.2
 certifi==2024.6.2
 cffi==1.16.0
-cfgv==3.4.0  # required by pre-commit
+cfgv==3.4.0
 charset-normalizer==3.3.2
 click==8.1.7
 colorama==0.4.6
 cryptography==42.0.8
 deprecation==2.1.0
-dill==0.3.8  # required by pylint
-distlib==0.3.8  # required by virtualenv
+dill==0.3.8
+distlib==0.3.8
 distro==1.9.0
-dnspython==2.6.1  # required by email-validator
-email_validator==2.2.0  # required by fastapi
+dnspython==2.6.1
+email_validator==2.2.0
 exceptiongroup==1.2.1
 fastapi==0.111.0
-fastapi-cli==0.0.4  # required by fastapi
-filelock==3.15.4  # required by virtualenv
+fastapi-cli==0.0.4
+filelock==3.15.4
 gotrue==2.5.4
 h11==0.14.0
 httpcore==1.0.5
 httpx==0.27.0
-identify==2.5.36  # required by pre-commit
+identify==2.5.36
 idna==3.7
-isort==5.13.2  # required by pylint
-Jinja2==3.1.4  # required by fastapi
+isort==5.13.2
+Jinja2==3.1.4
 mangum==0.17.0
-markdown-it-py==3.0.0  # required by rich
-MarkupSafe==2.1.5  # required by Jinja2
-mccabe==0.7.0  # required by pylint
-mdurl==0.1.2  # required by markdown-it-py
-mypy-extensions==1.0.0  # required by black
-nodeenv==1.9.1  # required by pre-commit
+markdown-it-py==3.0.0
+MarkupSafe==2.1.5
+mccabe==0.7.0
+mdurl==0.1.2
+mypy-extensions==1.0.0
+nodeenv==1.9.1
 openai==1.35.7
-orjson==3.10.5  # required by fastapi
+orjson==3.10.5
 packaging==24.1
-pathspec==0.12.1  # required by black
-platformdirs==4.2.2  # required by pylint
+pathspec==0.12.1
+platformdirs==4.2.2
 postgrest==0.16.8
 pre-commit==3.7.1
 pycparser==2.22
-pydantic==2.8.0  # required by fastapi, gotrue, openai, postgrest
-pydantic_core==2.20.0  # required by pydantic
-Pygments==2.18.0  # required by rich
+pydantic==2.8.0
+pydantic_core==2.20.0
+Pygments==2.18.0
 PyJWT==2.8.0
 pylint==3.2.5
 python-dateutil==2.9.0.post0
 python-dotenv==1.0.1
-python-multipart==0.0.9  # required by fastapi
-PyYAML==6.0.1  # required by pre-commit
+python-multipart==0.0.9
+PyYAML==6.0.1
 realtime==1.0.6
 requests==2.32.3
-rich==13.7.1  # required by typer
+rich==13.7.1
 sentry-sdk==2.7.1
-shellingham==1.5.4  # required by typer
+shellingham==1.5.4
 six==1.16.0
 sniffio==1.3.1
 starlette==0.37.2
@@ -64,12 +64,13 @@ StrEnum==0.4.15
 stripe==10.1.0
 supabase==2.5.1
 supafunc==0.4.6
-tomlkit==0.12.5  # required by pylint
+tiktoken==0.7.0
+tomlkit==0.12.5
 tqdm==4.66.4
-typer==0.12.3  # required by fastapi-cli
+typer==0.12.3
 typing_extensions==4.12.2
-ujson==5.10.0  # required by fastapi
+ujson==5.10.0
 urllib3==2.2.2
 uvicorn==0.30.1
-virtualenv==20.26.3  # required by pre-commit
+virtualenv==20.26.3
 websockets==12.0

services/openai/agent.py

@@ -255,7 +255,7 @@ def submit_message(
 
 def wait_on_run(run: Run, thread: Thread, token: str, run_name: str) -> tuple[Run, str]:
     """https://cookbook.openai.com/examples/assistants_api_overview_python"""
-    print(f"Run `{run_name}` status before loop: { run.status}")
+    print(f"Run `{run_name}` status before loop: {run.status}")
     client: OpenAI = create_openai_client()
     input_data = ""
     while run.status not in OPENAI_FINAL_STATUSES:

services/openai/chat.py

@@ -6,17 +6,19 @@
 from config import OPENAI_MODEL_ID, OPENAI_TEMPERATURE
 from services.openai.init import create_openai_client
 from services.openai.instructions import SYSTEM_INSTRUCTION_FOR_WRITING_PR
+from services.openai.truncate import truncate_message
 from utils.handle_exceptions import handle_exceptions
 
 
 @handle_exceptions(raise_on_error=True)
 def write_pr_body(input_message: str) -> str:
     """https://platform.openai.com/docs/api-reference/chat/create"""
     client: OpenAI = create_openai_client()
+    truncated_msg: str = truncate_message(input_message=input_message)
     completion: ChatCompletion = client.chat.completions.create(
         messages=[
             {"role": "system", "content": SYSTEM_INSTRUCTION_FOR_WRITING_PR},
-            {"role": "user", "content": input_message},
+            {"role": "user", "content": truncated_msg if truncated_msg else input_message},
         ],
         model=OPENAI_MODEL_ID,
         n=1,

services/openai/truncate.py

@@ -0,0 +1,15 @@
+import tiktoken
+from config import OPENAI_MAX_CONTEXT_TOKENS, OPENAI_MODEL_ID
+from utils.handle_exceptions import handle_exceptions
+
+
+@handle_exceptions(default_return_value="", raise_on_error=False)
+def truncate_message(input_message: str) -> str:
+    encoding: tiktoken.Encoding = tiktoken.encoding_for_model(
+        model_name=OPENAI_MODEL_ID
+    )
+    tokens: list[int] = encoding.encode(text=input_message)
+    if len(tokens) > OPENAI_MAX_CONTEXT_TOKENS:
+        tokens = tokens[:OPENAI_MAX_CONTEXT_TOKENS]
+        truncated_message: str = encoding.decode(tokens=tokens)
+    return truncated_message