Fix: Refactored configuration_variables.py and logger.py

Makefile

@@ -93,7 +93,7 @@ migrate:
 	poetry run python labs/manage.py migrate
 
 createuser:
-	DJANGO_SUPERUSER_PASSWORD=admin poetry run python labs/manage.py createsuperuser --noinput --username=admin --email=a@b.com
+	DJANGO_SUPERUSER_PASSWORD=admin poetry run python labs/manage.py createsuperuser --noinput --username=admin --email=admin@example.com
 
 load_fixtures:
 	python labs/manage.py loaddata $(wildcard labs/fixtures/*.json)

labs/api/schemas/codemonkey.py

@@ -1,6 +1,7 @@
 from typing import List, Optional
 
 from api.schemas.github import GithubSchema
+from django.conf import settings
 from pydantic import BaseModel
 
 
@@ -18,7 +19,9 @@ class VectorizeRepositorySchema(BaseModel):
     repository_path: str
 
 
-class FindEmbeddingsSchema(LocalRepositoryShema): ...
+class FindEmbeddingsSchema(LocalRepositoryShema):
+    similarity_threshold: float = settings.EMBEDDINGS_SIMILARITY_TRESHOLD
+    max_results: int = settings.EMBEDDINGS_MAX_RESULTS
 
 
 class PreparePromptContextSchema(BaseModel):

labs/config/celery.py

@@ -1,10 +1,10 @@
 import logging
 import os
 
-import config.configuration_variables as settings
 import redis
 from celery import Celery
 from celery.signals import task_failure
+from django.conf import settings
 from kombu import Queue
 from redbeat import RedBeatSchedulerEntry, schedulers
 

labs/config/logger.py

@@ -0,0 +1,19 @@
+from datetime import datetime
+
+from django.conf import settings
+from pythonjsonlogger import jsonlogger
+
+
+class CustomJsonFormatter(jsonlogger.JsonFormatter):
+    def add_fields(self, log_record, record, message_dict):
+        super(CustomJsonFormatter, self).add_fields(log_record, record, message_dict)
+
+        if not log_record.get("timestamp"):
+            log_record["timestamp"] = datetime.now().strftime(settings.LOGGING_DATETIME_FORMAT)
+
+        if log_record.get("level"):
+            log_record["level"] = log_record["level"].upper()
+        else:
+            log_record["level"] = record.levelname
+
+        log_record["project"] = "codemonkey"

labs/config/settings.py

@@ -87,6 +87,50 @@
 }
 
 
+# Logging
+def create_logging_directory() -> Path:
+    logs_path = BASE_DIR.parent / "logs"
+    if not os.path.exists(logs_path):
+        os.makedirs(logs_path)
+
+    return logs_path / "debug.log"
+
+
+LOGGING_DATETIME_FORMAT = "%Y-%m-%d %H:%M:%S,%f"
+LOGGING = {
+    "version": 1,
+    "disable_existing_loggers": False,
+    "formatters": {
+        "standard": {
+            "format": "[%(asctime)s][%(levelname)s][%(name)s]: %(message)s",
+            "datefmt": LOGGING_DATETIME_FORMAT,
+        },
+        "json": {"()": "config.logger.CustomJsonFormatter"},
+    },
+    "handlers": {
+        "console": {
+            "level": "DEBUG",
+            "class": "logging.StreamHandler",
+            "formatter": "standard",
+        },
+        "file": {
+            "level": "DEBUG",
+            "class": "logging.handlers.RotatingFileHandler",
+            "filename": create_logging_directory(),
+            "maxBytes": 10000000,
+            "backupCount": 5,
+            "formatter": "json",
+        },
+    },
+    "loggers": {
+        "labs": {
+            "level": "DEBUG",
+            "handlers": ["console", "file"],
+            "propagate": False,
+        }
+    },
+}
+
 # Password validation
 # https://docs.djangoproject.com/en/5.1/ref/settings/#auth-password-validators
 
@@ -127,3 +171,20 @@
 # https://docs.djangoproject.com/en/5.1/ref/settings/#default-auto-field
 
 DEFAULT_AUTO_FIELD = "django.db.models.BigAutoField"
+
+# Celery settings
+CELERY_BROKER_URL = os.environ.get("CELERY_BROKER_URL")
+CELERY_BACKEND_URL = os.environ.get("CELERY_BACKEND_URL")
+
+# Redis settings
+REDIS_HOST = os.environ.get("REDIS_HOST")
+REDIS_PORT = os.environ.get("REDIS_PORT")
+
+# Custom settings
+GITHUB_API_BASE_URL = "https://api.github.com"
+
+LOCAL_LLM_HOST = os.environ.get("LOCAL_LLM_HOST", "http://ollama:11434")
+
+CLONE_DESTINATION_DIR = os.getenv("CLONE_DESTINATION_DIR", "/tmp/")
+EMBEDDINGS_SIMILARITY_TRESHOLD = 0.7
+EMBEDDINGS_MAX_RESULTS = 10

labs/embeddings/embedder.py

@@ -1,6 +1,7 @@
 from dataclasses import dataclass
 from typing import Any, Dict, List, Optional, Union
 
+from django.conf import settings
 from embeddings.models import Embedding
 from pgvector.django import CosineDistance
 
@@ -20,7 +21,11 @@ def embed(self, prompt, *args, **kwargs) -> Embeddings:
         return self.embedder.embed(prompt, *args, **kwargs)
 
     def retrieve_embeddings(
-        self, query: str, repository: str, similarity_threshold: int = 0.7, number_of_results: int = 10
+        self,
+        query: str,
+        repository: str,
+        similarity_threshold: float = settings.EMBEDDINGS_SIMILARITY_TRESHOLD,
+        max_results: int = settings.EMBEDDINGS_MAX_RESULTS,
     ) -> List[Embedding]:
         query = query.replace("\n", "")
         embedded_query = self.embed(prompt=query).embeddings
@@ -29,7 +34,7 @@ def retrieve_embeddings(
 
         return Embedding.objects.annotate(distance=CosineDistance("embedding", embedded_query[0])).filter(
             repository=repository, distance__lt=similarity_threshold
-        )[:number_of_results]
+        )[:max_results]
 
     def reembed_code(
         self,

labs/embeddings/ollama.py

@@ -1,4 +1,4 @@
-from config.configuration_variables import LOCAL_LLM_HOST
+from django.conf import settings
 from embeddings.embedder import Embeddings
 from ollama import Client
 
@@ -7,7 +7,7 @@ class OllamaEmbedder:
     def __init__(self, model):
         self._model_name = model
 
-        self._client = Client(LOCAL_LLM_HOST)
+        self._client = Client(settings.LOCAL_LLM_HOST)
 
     def embed(self, prompt, *args, **kwargs) -> Embeddings:
         result = self._client.embed(self._model_name, prompt, *args, **kwargs)

labs/github/github.py

@@ -3,9 +3,9 @@
 import os
 from dataclasses import dataclass
 
-import config.configuration_variables as settings
 import git
 import requests
+from django.conf import settings
 
 logger = logging.getLogger(__name__)
 

labs/llm/ollama.py

@@ -1,12 +1,12 @@
-from config.configuration_variables import LOCAL_LLM_HOST
+from django.conf import settings
 from ollama import Client
 
 
 class OllamaRequester:
     def __init__(self, model):
         self._model_name = model
 
-        self._client = Client(LOCAL_LLM_HOST)
+        self._client = Client(settings.LOCAL_LLM_HOST)
 
     def completion_without_proxy(self, messages, *args, **kwargs):
         """

labs/tasks/llm.py

@@ -1,9 +1,9 @@
 import json
 import logging
 
-import config.configuration_variables as settings
 from config.celery import app
 from core.models import Model, VectorizerModel
+from django.conf import settings
 from embeddings.embedder import Embedder
 from embeddings.vectorizers.vectorizer import Vectorizer
 from llm.requester import Requester
@@ -85,11 +85,19 @@ def vectorize_repository_task(prefix="", repository_path=""):
 
 
 @app.task
-def find_embeddings_task(prefix="", issue_body="", repository_path=""):
+def find_embeddings_task(
+    prefix="",
+    issue_body="",
+    repository_path="",
+    similarity_threshold=settings.EMBEDDINGS_SIMILARITY_TRESHOLD,
+    max_results=settings.EMBEDDINGS_MAX_RESULTS,
+):
     embedder_class, *embeder_args = Model.get_active_embedding_model()
     embeddings_results = Embedder(embedder_class, *embeder_args).retrieve_embeddings(
         redis_client.get(RedisVariable.ISSUE_BODY, prefix=prefix, default=issue_body),
         redis_client.get(RedisVariable.REPOSITORY_PATH, prefix=prefix, default=repository_path),
+        similarity_threshold,
+        max_results,
     )
     similar_embeddings = [
         (embedding.repository, embedding.file_path, embedding.text) for embedding in embeddings_results
@@ -102,7 +110,10 @@ def find_embeddings_task(prefix="", issue_body="", repository_path=""):
 
 
 @app.task
-def prepare_prompt_and_context_task(prefix="", issue_body="", embeddings=[]):
+def prepare_prompt_and_context_task(prefix="", issue_body="", embeddings=None):
+    if not embeddings:
+        embeddings = []
+
     prompt = get_prompt(redis_client.get(RedisVariable.ISSUE_BODY, prefix=prefix, default=issue_body))
     redis_client.set(RedisVariable.PROMPT, prefix=prefix, value=prompt)
 
@@ -116,7 +127,10 @@ def prepare_prompt_and_context_task(prefix="", issue_body="", embeddings=[]):
 
 
 @app.task
-def get_llm_response_task(prefix="", context={}):
+def get_llm_response_task(prefix="", context=None):
+    if not context:
+        context = {}
+
     context = json.loads(redis_client.get(RedisVariable.CONTEXT, prefix=prefix, default=context))
     llm_response = get_llm_response(context)
 

labs/tasks/repository.py

@@ -1,8 +1,8 @@
 import json
 
-import config.configuration_variables as settings
 from config.celery import app
 from decorators import time_and_log_function
+from django.conf import settings
 from github.github import GithubRequests
 from parsers.response import create_file, modify_file, parse_llm_output
 from tasks.redis_client import RedisStrictClient, RedisVariable

labs/tasks/run.py

@@ -1,8 +1,8 @@
 import os.path
 
-import config.configuration_variables as settings
 from celery import chain
 from config.celery import app
+from django.conf import settings
 from tasks import (
     apply_code_changes_task,
     clone_repository_task,