Introduce rca_accelerator_chatbot package

Containerfile

@@ -6,12 +6,16 @@ RUN groupadd -g 65532 chatgroup && \
 
 WORKDIR /app
 
-COPY src/ .
-COPY pdm.lock pyproject.toml Makefile .
+COPY pdm.lock pyproject.toml Makefile LICENSE README.md .
+COPY src/ src/
 RUN make install-pdm install-global
+
 RUN chown -R chatuser:chatgroup /app
 
 USER chatuser
 EXPOSE 8000
 
-CMD ["chainlit", "run", "app.py", "--host",  "0.0.0.0"]
+RUN cp -r src/rca_accelerator_chatbot/data/* . && \
+    cp -r src/rca_accelerator_chatbot/data/.chainlit .
+
+CMD ["chainlit", "run", "src/rca_accelerator_chatbot/app.py", "--host",  "0.0.0.0"]

Containerfile.api

@@ -9,12 +9,14 @@ RUN dnf install -y krb5-workstation krb5-libs
 WORKDIR /app
 RUN chown -R chatuser:chatgroup /app
 
-COPY src/ .
-COPY pdm.lock pyproject.toml Makefile api-entrypoint.sh .
+COPY src/ src/
+COPY pdm.lock pyproject.toml Makefile api-entrypoint.sh LICENSE README.md .
 RUN make install-pdm install-global
 RUN chmod +x api-entrypoint.sh
 
 USER chatuser
 EXPOSE 8001
 
+RUN cp -r src/rca_accelerator_chatbot/data/.chainlit .
+
 ENTRYPOINT ["/app/api-entrypoint.sh"]

api-entrypoint.sh

@@ -6,4 +6,4 @@ if [ -n "$KRB_USER" ] && [ -n "$KRB_PASS" ]; then
   echo "$KRB_PASS" | kinit "$KRB_USER"
 fi
 
-exec uvicorn api:app --host 0.0.0.0 --port 8001
+exec uvicorn rca_accelerator_chatbot.api:app --host 0.0.0.0 --port 8001

pyproject.toml

@@ -2,6 +2,7 @@
 name = "rca-accelerator-chatbot"
 version = "0.1.0"
 description = "Useful chatbot that can be used to do RCA of CI errors."
+readme = "README.md"
 authors = []
 dependencies = [
     "openai==1.68.2",
@@ -19,6 +20,7 @@ dependencies = [
     "beautifulsoup4>=4.13.4",
 ]
 requires-python = "==3.12.*"
+license = "Apache-2.0"
 
 [tool.pdm.dev-dependencies]
 dev = [
@@ -27,3 +29,19 @@ dev = [
     "fastapi==0.115.8",
     "ruff==0.9.0",
 ]
+
+[build-system]
+requires = ["setuptools>=61.0"]
+build-backend = "setuptools.build_meta"
+
+[tool.setuptools]
+include-package-data = true
+
+[tool.setuptools.package-data]
+rca_accelerator_chatbot = ["data/**", "data/.chainlit/**"]
+
+[tool.setuptools.packages.find]
+where = ["src"]
+
+[project.scripts]
+rca_accelerator_chatbot = "rca_accelerator_chatbot.main:main"

src/api.py → src/rca_accelerator_chatbot/api.py

@@ -4,17 +4,21 @@
 import asyncio
 from typing import Dict, Any, List
 import re
+
 import httpx
 from httpx_gssapi import HTTPSPNEGOAuth, OPTIONAL
 from bs4 import BeautifulSoup
 from fastapi import Depends, FastAPI, HTTPException
 from pydantic import BaseModel, Field, HttpUrl
-from constants import CI_LOGS_PROFILE, DOCS_PROFILE, RCA_FULL_PROFILE
-from chat import handle_user_message_api
-from config import config
-from settings import ModelSettings
-from generation import discover_generative_model_names
-from embeddings import discover_embeddings_model_names
+
+from rca_accelerator_chatbot.constants import (
+    CI_LOGS_PROFILE, DOCS_PROFILE, RCA_FULL_PROFILE
+)
+from rca_accelerator_chatbot.chat import handle_user_message_api
+from rca_accelerator_chatbot.config import config
+from rca_accelerator_chatbot.settings import ModelSettings
+from rca_accelerator_chatbot.generation import discover_generative_model_names
+from rca_accelerator_chatbot.embeddings import discover_embeddings_model_names
 
 app = FastAPI(title="RCAccelerator API")
 

src/app.py → src/rca_accelerator_chatbot/app.py

@@ -5,12 +5,12 @@
 import chainlit as cl
 from chainlit.input_widget import Select, Switch, Slider
 
-from config import config
-import constants
-from chat import handle_user_message
-from auth import authentification
-from generation import discover_generative_model_names
-from embeddings import discover_embeddings_model_names
+from rca_accelerator_chatbot.config import config
+from rca_accelerator_chatbot import constants
+from rca_accelerator_chatbot.chat import handle_user_message
+from rca_accelerator_chatbot.auth import authentification
+from rca_accelerator_chatbot.generation import  discover_generative_model_names
+from rca_accelerator_chatbot.embeddings import discover_embeddings_model_names
 
 
 @cl.set_chat_profiles

src/auth.py → src/rca_accelerator_chatbot/auth.py

@@ -8,7 +8,7 @@
 from bcrypt import checkpw
 import chainlit as cl
 
-from config import config
+from rca_accelerator_chatbot.config import config
 
 
 # pylint: disable=too-few-public-methods

src/chat.py → src/rca_accelerator_chatbot/chat.py

@@ -5,21 +5,21 @@
 import httpx
 from openai.types.chat import ChatCompletionAssistantMessageParam
 
-import constants
-from prompt import build_prompt
-from vectordb import vector_store
-from generation import get_response
-from embeddings import (
+from rca_accelerator_chatbot import constants
+from rca_accelerator_chatbot.prompt import build_prompt
+from rca_accelerator_chatbot.vectordb import vector_store
+from rca_accelerator_chatbot.generation import get_response
+from rca_accelerator_chatbot.embeddings import (
     get_num_tokens, generate_embedding,
     get_rerank_score, get_default_embeddings_model_name
 )
-from settings import ModelSettings, HistorySettings, ThreadMessages
-from config import config
-from constants import (
+from rca_accelerator_chatbot.settings import ModelSettings, HistorySettings, ThreadMessages
+from rca_accelerator_chatbot.config import config
+from rca_accelerator_chatbot.constants import (
     DOCS_PROFILE,
     RCA_FULL_PROFILE,
     TEXT_UPLOAD_TEMPLATE,
-    )
+)
 
 
 # Create mock message and response objects

src/config.py → src/rca_accelerator_chatbot/config.py

@@ -3,7 +3,7 @@
 from dataclasses import dataclass
 import os
 
-from constants import (
+from rca_accelerator_chatbot.constants import (
     CI_LOGS_SYSTEM_PROMPT,
     DOCS_SYSTEM_PROMPT,
     WELCOME_MESSAGE,

src/embeddings.py → src/rca_accelerator_chatbot/embeddings.py

@@ -7,8 +7,8 @@
 import httpx
 from openai import AsyncOpenAI, OpenAIError
 
-from config import config
-from generation import extract_model_ids
+from rca_accelerator_chatbot.config import config
+from rca_accelerator_chatbot.generation import extract_model_ids
 
 # Initialize embedding LLM client
 emb_llm = AsyncOpenAI(

src/generation.py → src/rca_accelerator_chatbot/generation.py

@@ -3,9 +3,11 @@
 import chainlit as cl
 from openai import AsyncOpenAI, OpenAIError
 
-from settings import ModelSettings, ThreadMessages
-from config import config
-from constants import DOCS_PROFILE, RCA_FULL_PROFILE, CI_LOGS_PROFILE
+from rca_accelerator_chatbot.settings import ModelSettings, ThreadMessages
+from rca_accelerator_chatbot.config import config
+from rca_accelerator_chatbot.constants import (
+    DOCS_PROFILE, RCA_FULL_PROFILE, CI_LOGS_PROFILE
+)
 
 # Initialize generative LLM client
 gen_llm = AsyncOpenAI(

src/rca_accelerator_chatbot/main.py

@@ -0,0 +1,13 @@
+"""Main package entrypoint"""
+from pathlib import Path
+import sys
+import subprocess
+
+def main():
+    """Main entrypoint for the chatbot"""
+    package_dir = Path(__file__).parent
+    data_dir = package_dir.joinpath("data").as_posix()
+    app_path = package_dir.joinpath("app.py").as_posix()
+
+    subprocess.run(["chainlit", "run", app_path, *sys.argv[1:]],
+                   cwd=data_dir, check=True)

src/prompt.py → src/rca_accelerator_chatbot/prompt.py

@@ -4,12 +4,12 @@
     ChatCompletionSystemMessageParam,
 )
 
-from config import config
-from constants import (
+from rca_accelerator_chatbot.config import config
+from rca_accelerator_chatbot.constants import (
     NO_RESULTS_FOUND, SEARCH_RESULTS_TEMPLATE, SEARCH_RESULT_TRUNCATED_CHUNK
 )
-from settings import HistorySettings, ThreadMessages
-from generation import get_system_prompt_per_profile
+from rca_accelerator_chatbot.settings import HistorySettings, ThreadMessages
+from rca_accelerator_chatbot.generation import get_system_prompt_per_profile
 
 
 def search_result_to_str(search_result: dict) -> str:

src/vectordb.py → src/rca_accelerator_chatbot/vectordb.py

@@ -5,7 +5,7 @@
 from qdrant_client import QdrantClient
 from qdrant_client.http.exceptions import ApiException
 
-from config import config
+from rca_accelerator_chatbot.config import config
 
 
 class VectorStore: