Feature/add pytest workflow

langchain-crash-course/4_rag/logs/rag_application.log

@@ -139,4 +139,4 @@
 2025-08-26 23:24:09,222 - 8_rag_web_scrape_firecrawl.py - INFO - Querying vector store 'chroma_db_web_scrape_firecrawl'...
 2025-08-26 23:24:09,222 - 8_rag_web_scrape_firecrawl.py - INFO - Vector store 'chroma_db_web_scrape_firecrawl' already exists. No need to initialize.
 2025-08-26 23:24:10,098 - 8_rag_web_scrape_firecrawl.py - INFO - Relevant documents retrieved successfully.
-2025-08-26 23:45:35,451 - 8_rag_web_scrape_firecrawl.py - INFO - User exited conversation
+2025-08-26 23:45:35,451 - 8_rag_web_scrape_firecrawl.py - INFO - User exited conversation2025-11-04 21:31:41,119 - rag_with_metadata.py - INFO - ==================================================

langchain-crash-course/4_rag/utils/logger.py

@@ -42,7 +42,7 @@ def setup(module_name: str) -> logging.Logger:
 
         # Create formatters and handlers
         formatter: logging.Formatter = logging.Formatter(
-            "%(asctime)s - %(name)s - %(levelname)s - %(message)s"
+            "%(asctime)s - %(levelname)s - %(module)s - %(message)s"
         )
 
         # Rotating file handler

langchain-crash-course/5_agents_tools/agent_react_chat.py

@@ -80,7 +80,7 @@
 
 
 # Module path
-module_path: Path = Path(__file__).resolve().parent
+module_path: Path = Path(__file__).resolve()
 
 # Set logger
 logger: Logger = RAGLogger.get_logger(module_name=module_path.name)
@@ -178,6 +178,7 @@ async def main() -> None:
                 input={"input": query},
                 config={"configurable": {"session_id": session_id}},
             )
+            logger.info(msg=f"Agent: {response['output'][:100]}.....")
             print(f"Agent: {response['output']}")
 
         except (KeyboardInterrupt, EOFError):

langchain-crash-course/5_agents_tools/agent_react_rag_context.py

@@ -0,0 +1,203 @@
+# agent_react_rag_context.py
+
+# Import standard libraries
+import sys
+from logging import Logger
+from pathlib import Path
+from typing import Any
+
+# Import environment variables
+from dotenv import load_dotenv
+
+# Import langchain modules
+from langchain.chains import (
+    create_history_aware_retriever,
+    create_retrieval_chain,
+)
+from langchain.chains.combine_documents import create_stuff_documents_chain
+from langchain_chroma import Chroma
+from langchain_core.messages import AIMessage, HumanMessage
+from langchain_core.messages.base import BaseMessage
+from langchain_core.prompts.chat import ChatPromptTemplate, MessagesPlaceholder
+from langchain_core.runnables.base import Runnable
+from langchain_core.vectorstores.base import VectorStoreRetriever
+from langchain_ollama import ChatOllama
+from langchain_ollama.embeddings import OllamaEmbeddings
+
+# Import custom logger
+from utils.logger import RAGLogger
+
+# Load environment variables
+load_dotenv()
+
+# Module path
+module_path: Path = Path(__file__).resolve()
+
+# Set logger
+logger: Logger = RAGLogger.get_logger(module_name=module_path.name)
+
+# Log application startup
+logger.info(msg="=" * 50)
+logger.info(msg="Starting Agent ReAct RAG Context Application")
+logger.info(msg="=" * 50)
+
+# Define directories and paths
+rag_dir: Path = Path(__file__).parents[1] / "4_rag"
+books_dir: Path = rag_dir / "books"
+db_dir: Path = rag_dir / "db"
+store_name: str = "chroma_db_with_metadata"
+persistent_directory: Path = db_dir / store_name
+
+# Define embeddings models
+ollama_embeddings = OllamaEmbeddings(
+    model="nomic-embed-text:latest",
+)
+
+# Define LLM
+llm = ChatOllama(model="gemma3:4b")
+
+# Check vector store existence
+if not persistent_directory.exists():
+    logger.error(
+        msg=f"Vector store '{store_name}' does not exist. Please check the path."
+    )
+    sys.exit(1)
+
+# Load vector store and create retriever
+try:
+    logger.info(msg=f"Loading vector store '{store_name}'...")
+    # Load the Chroma vector store
+    db: Chroma = Chroma(
+        persist_directory=str(persistent_directory),
+        embedding_function=ollama_embeddings,
+    )
+    # Create a retriever
+    retriever: VectorStoreRetriever = db.as_retriever(
+        search_type="similarity",
+        search_kwargs={"k": 3},
+    )
+    logger.info(msg=f"Created retriever from vector store '{store_name}' successfully.")
+
+except Exception as e:
+    logger.error(msg=f"Unexpected error querying vector store '{store_name}': {e}")
+    sys.exit(1)
+
+# Contextualize question prompt
+# System prompt helps the AI understand that it should reformulate the question
+# based on the chat history to make it a standalone question
+contextualize_q_system_prompt = """
+Given a chat history and the latest user question, this prompt helps the AI reformulate 
+the question to be standalone. The reformulated question should be understandable 
+without relying on prior chat context. The AI should not answer the question—only 
+rephrase it if necessary, or return it unchanged if already standalone.
+"""
+
+# Create contextualize question prompt template
+contextualize_q_prompt_template: ChatPromptTemplate = ChatPromptTemplate.from_messages(
+    messages=[
+        ("system", contextualize_q_system_prompt),
+        MessagesPlaceholder(variable_name="chat_history"),
+        ("human", "{input}"),
+    ]
+)
+
+# Create a history-aware retriever
+# this users the LLM to help reformulate the question based on chat history
+history_aware_retriever: VectorStoreRetriever = create_history_aware_retriever(
+    llm,
+    retriever,
+    contextualize_q_prompt_template,
+)
+
+# Answer question prompt
+# This system prompt helps the AI understand that it should provide concise answers
+# based on the retrieved context and indicates what to do if the answer is unknown
+qa_system_prompt = """
+You are an assistant for question-answering tasks.
+Use the following pieces of retrieved context to answer the question. 
+If you don't know the answer, just say that you don't know. 
+Limit your response to a maximum of ten sentences and keep the answer concise.
+\n\n
+{context}
+"""
+
+# Create answer question prompt template
+qa_prompt_template: ChatPromptTemplate = ChatPromptTemplate.from_messages(
+    messages=[
+        ("system", qa_system_prompt),
+        MessagesPlaceholder(variable_name="chat_history"),
+        ("human", "{input}"),
+    ]
+)
+
+# Create a chain to combine documents for question answering
+# `create_stuff_documents_chain` feeds all retrieved context into the LLM
+question_answering_chain: Runnable[dict[str, Any], Any] = create_stuff_documents_chain(
+    llm=llm, prompt=qa_prompt_template
+)
+
+# Create a RAG chain that combines the history-aware retriever and the question answering chain
+rag_chain: Runnable[dict[str, Any], Any] = create_retrieval_chain(
+    history_aware_retriever, question_answering_chain
+)
+
+
+# Run RAG LLM conversation
+def main() -> None:
+    """
+    Runs the main conversational loop for the RAG-based chat application.
+
+    This function initializes the chat history and enters an infinite loop to
+    continuously accept user input. It processes the user's query through the
+    RAG chain, prints the AI's response, and updates the chat history.
+    The loop can be exited by typing 'exit', or by sending a
+    KeyboardInterrupt (Ctrl+C) or EOFError (Ctrl+D).
+    """
+    print("\nStart chatting with AI! Type 'exit' to end the conversation.")
+
+    # Initialize chat history
+    chat_history: list[BaseMessage] = []
+
+    while True:
+        try:
+            # User query
+            query: str = input("You: ").strip()
+
+            if not query:
+                continue
+
+            if query.lower() == "exit":
+                logger.info(msg="User exited conversation")
+                print("Exiting...")
+                break
+
+            # Process user query through RAG chain
+            logger.info(msg="Processing user query through RAG chain...")
+            result: Any = rag_chain.invoke(
+                input={"input": query, "chat_history": chat_history}
+            )
+
+            # Display AI response
+            if result:
+                logger.info(msg="AI response generated successfully")
+                print(f"AI: {result['answer']}")
+
+                # Update chat history
+                chat_history.append(HumanMessage(content=query))
+                chat_history.append(AIMessage(content=result["answer"]))
+                logger.info(msg="Chat history updated successfully")
+
+        except (KeyboardInterrupt, EOFError):
+            logger.info(msg="Keyboard interrupt or EOF error")
+            print("Exiting...")
+            break
+
+        except Exception as e:
+            logger.error(msg=f"Unexpected error: {e}")
+            print("Exiting...")
+            break
+
+
+# Main entry point
+if __name__ == "__main__":
+    main()

langchain-crash-course/5_agents_tools/agent_tools_basic.py

@@ -73,7 +73,7 @@
 
 
 # Module path
-module_path: Path = Path(__file__).resolve().parent
+module_path: Path = Path(__file__).resolve()
 
 # Set logger
 logger: Logger = RAGLogger.get_logger(module_name=module_path.name)
@@ -161,7 +161,7 @@ async def main() -> None:
 
             # Run agent executor
             response: Any = await agent_executor.ainvoke(input={"input": query})
-            logger.info(msg="Agent response generated successfully")
+            logger.info(msg=f"Agent: {response['output']}")
             print(f"Agent: {response['output']}")
 
         except (KeyboardInterrupt, EOFError):

langchain-crash-course/5_agents_tools/logs/agent_tools.log

@@ -1,8 +1,22 @@
-2025-08-31 22:51:29,350 - 5_agents_tools - INFO - Start Agent Tools Basic Application...
-2025-08-31 22:52:03,279 - 5_agents_tools - INFO - Agent response generated successfully
-2025-08-31 22:52:11,062 - 5_agents_tools - INFO - User exited conversation
-2025-09-04 21:00:41,977 - 5_agents_tools - INFO - Start Agent React Chat Application...
-2025-09-04 21:01:15,876 - 5_agents_tools - INFO - Getting current time: 2025-09-04 21:01:15
-2025-09-04 21:01:56,097 - 5_agents_tools - ERROR - Error getting Wikipedia summary: Page id "michael jacks" does not match any pages. Try another id!
-2025-09-04 21:02:05,819 - 5_agents_tools - INFO - Getting Wikipedia summary: Michael Joseph Jackson (August 29, 1958 – June 25, 2009) was an American singer, songwriter, dancer,.....
-2025-09-04 21:30:59,868 - 5_agents_tools - INFO - User exited conversation
+2025-09-25 15:26:55,633 - agent_tools_basic.py - INFO - Start Agent Tools Basic Application...
+2025-09-25 15:27:19,931 - agent_tools_basic.py - INFO - Agent response generated successfully
+2025-09-25 15:27:35,672 - agent_tools_basic.py - INFO - User exited conversation
+2025-09-25 15:52:42,338 - agent_react_chat.py - INFO - Start Agent React Chat Application...
+2025-09-25 15:53:17,567 - agent_react_chat.py - INFO - Getting Wikipedia summary: Microsoft Corporation is an American multinational corporation and technology conglomerate headquart.....
+2025-09-25 15:53:29,552 - agent_react_chat.py - INFO - Getting Wikipedia summary: Microsoft Corporation is an American multinational corporation and technology conglomerate headquart.....
+2025-09-25 15:53:50,242 - agent_react_chat.py - INFO - Agent: Microsoft is a massive technology company with a rich history. It was founded in 1975 by Bill Gates .....
+2025-09-25 15:54:34,323 - agent_react_chat.py - ERROR - Error getting Wikipedia summary: Page id "bill games" does not match any pages. Try another id!
+2025-09-25 15:54:45,004 - agent_react_chat.py - INFO - Agent: Bill Gates is a prominent American business magnate, investor, and philanthropist. He co-founded Mic.....
+2025-09-25 15:55:00,779 - agent_react_chat.py - INFO - User exited conversation
+2025-11-04 22:52:56,368 - INFO - agent_react_rag_context - ==================================================
+2025-11-04 22:52:56,368 - INFO - agent_react_rag_context - Starting Agent ReAct RAG Context Application
+2025-11-04 22:52:56,368 - INFO - agent_react_rag_context - ==================================================
+2025-11-04 22:52:58,444 - INFO - agent_react_rag_context - Loading vector store 'chroma_db_with_metadata'...
+2025-11-04 22:52:58,589 - INFO - agent_react_rag_context - Created retriever from vector store 'chroma_db_with_metadata' successfully.
+2025-11-04 22:54:46,933 - INFO - agent_react_rag_context - Processing user query through RAG chain...
+2025-11-04 22:55:14,195 - INFO - agent_react_rag_context - AI response generated successfully
+2025-11-04 22:55:14,198 - INFO - agent_react_rag_context - Chat history updated successfully
+2025-11-04 22:57:53,311 - INFO - agent_react_rag_context - Processing user query through RAG chain...
+2025-11-04 22:58:41,439 - INFO - agent_react_rag_context - AI response generated successfully
+2025-11-04 22:58:41,466 - INFO - agent_react_rag_context - Chat history updated successfully
+2025-11-04 22:59:12,397 - INFO - agent_react_rag_context - User exited conversation

langchain-crash-course/5_agents_tools/utils/logger.py

@@ -42,7 +42,7 @@ def setup(module_name: str) -> logging.Logger:
 
         # Create formatters and handlers
         formatter: logging.Formatter = logging.Formatter(
-            "%(asctime)s - %(name)s - %(levelname)s - %(message)s"
+            "%(asctime)s - %(levelname)s - %(module)s - %(message)s"
         )
 
         # Rotating file handler