# Conversation Generator README

This directory contains Python scripts and associated files for generating AI conversations using chatbots. Below is an overview of the main components and instructions on how to use them.

---

## Contents

- **`main.py`**: The main script that orchestrates conversations between chatbots and handles user input.
- **`newchatbot1.py`** and **`newchatbot2.py`**: Scripts for individual chatbots that generate responses and publish messages to MQTT topics.
- **`postgres_operations.py`**: Functions for interacting with the PostgreSQL database.
- **`utils.py`**: Utility functions used across scripts.
- **`bot_personality_generator.py`**: Generates unique personalities for the chatbots.
- **`new_conversation_summary.py`**: Generates summaries for completed conversations.
- **`requirements.txt`**: Lists the Python dependencies required to run the scripts.

---

## Setup

### 1. Install Dependencies

Ensure you have **Python 3.7** or higher installed. Install required packages using:

```bash
pip install -r requirements.txt
```

### 2. Configure the Database

Update the `DB_PARAMS` in `main.py` and `postgres_operations.py` with your PostgreSQL database credentials:

```python
DB_PARAMS = {
    "dbname": "your_database",
    "user": "your_username",
    "password": "your_password",
    "host": "your_host",
    "port": "your_port"
}
```

### 3. Configure MQTT Broker

Ensure that the MQTT broker details in the scripts point to your MQTT broker:

```python
MQTT_BROKER = "your_mqtt_broker_address"
MQTT_PORT = your_mqtt_broker_port  # Typically 1883 for MQTT
```

### 4. Configure Ollama Server

Set the `OLLAMA_URL` in the scripts to point to the correct Ollama API endpoint:

```python
OLLAMA_URL = "http://your_ollama_server_address:port/api/chat"
```

---

## Running the Scripts

### Starting a Conversation

To start a new conversation:

```bash
python main.py
```

Specify the language (`eng` for English or `ita` for Italian):

```bash
python main.py eng  # For English conversation
python main.py ita  # For Italian conversation
```

### Handling User Messages

To process a user message within an existing conversation:

```bash
python main.py --conversation_id <id> --user_message "<message>"
```

Replace `<id>` with the conversation ID and `<message>` with the user's message.

### Generating Bot Personalities

To generate new personalities for the chatbots:

```bash
python bot_personality_generator.py --topic "<topic>" --language <eng|ita>
```

### Summarizing Conversations

After a conversation ends, generate a summary:

```bash
python new_conversation_summary.py --conversation_id <id> --language <eng|ita>
```

---

## File Descriptions

- **`main.py`**: Manages conversation flow, handles user input, coordinates between chatbots, and publishes messages to MQTT topics.
- **`newchatbot1.py` / `newchatbot2.py`**: Generate responses for each chatbot, publish their status and messages to MQTT, and save messages to the database.
- **`postgres_operations.py`**: Contains functions for database operations like saving messages, retrieving conversation history, and managing bot statuses.
- **`utils.py`**: Utility functions such as running chatbots, memory usage tracking, and prompt generation.
- **`bot_personality_generator.py`**: Generates unique personalities for chatbots based on the conversation topic.
- **`new_conversation_summary.py`**: Generates a summary of the conversation for storage or display.
- **`requirements.txt`**: Lists all Python packages required by the scripts.

---

## Common Commands

### Start a Conversation

- **In English:**

  ```bash
  python main.py eng
  ```

- **In Italian:**

  ```bash
  python main.py ita
  ```

### Process a User Message

```bash
python main.py --conversation_id 123 --user_message "Your message here"
```

### Generate Bot Personalities

```bash
python bot_personality_generator.py --topic "climate change" --language eng
```

### Summarize a Conversation

```bash
python new_conversation_summary.py --conversation_id 123 --language eng
```

---

## Notes

- **Ensure Services are Running:**
  - **MQTT Broker:** Verify that your MQTT broker is running and accessible.
  - **PostgreSQL Database:** Ensure the database server is up and accepts connections.
  - **Ollama API:** The Ollama server should be running and reachable at the specified URL.

- **MQTT Topics Used:**
  - **`AI1/status`** and **`AI2/status`**: Bots publish their status (`speaking` or `stop`) to these topics.
  - **`conversations/{conversation_id}`**: Bots publish conversation messages to this topic.

- **Language Options:**
  - Use `eng` for English.
  - Use `ita` for Italian.

---

## Troubleshooting

- **No Messages in MQTT Topics:**
  - Ensure the MQTT broker is running.
  - Verify network connectivity between scripts and the broker.
  - Check that the correct host and port are configured.

- **Database Connection Errors:**
  - Confirm database credentials and network access.
  - Check if the database service is active.

- **API Connection Errors:**
  - Verify the Ollama API URL and that the service is operational.

- **Scripts Not Publishing Bot Status:**
  - Ensure that `main.py` has been updated to publish `speaking` and `stop` statuses to the appropriate MQTT topics (`AI1/status`, `AI2/status`).

---

## Example Workflow

1. **Start the Conversation:**

   ```bash
   python main.py eng
   ```

2. **Monitor MQTT Topics:**

   Use an MQTT client to subscribe to topics:

   - For bot statuses:

     ```bash
     mosquitto_sub -h your_mqtt_broker_address -p your_mqtt_broker_port -t 'AI1/status'
     mosquitto_sub -h your_mqtt_broker_address -p your_mqtt_broker_port -t 'AI2/status'
     ```

   - For conversation messages:

     ```bash
     mosquitto_sub -h your_mqtt_broker_address -p your_mqtt_broker_port -t 'conversations/#'
     ```

3. **Send a User Message:**

   ```bash
   python main.py --conversation_id 123 --user_message "What do you think about renewable energy?"
   ```

4. **Generate Conversation Summary:**

   ```bash
   python new_conversation_summary.py --conversation_id 123 --language eng
   ```

---

## Additional Tips

- **Logging:**
  - The scripts print messages to the console for debugging. Monitor the output for any errors or status updates.

- **Modifying Bot Personalities:**
  - Customize bot personalities by editing `bot_personality_generator.py` or by manually setting personalities in the database.

- **Extending Functionality:**
  - Feel free to modify and extend the scripts to suit your needs, such as adding new languages or integrating with different APIs.