gistfile1.txt

Story
You find yourself standing on a desolate street in the year 2053. The air is heavy with the scent of ozone, and flickering holographic billboards cast an eerie glow across the abandoned buildings. You are John, a middle-aged man with graying hair and a determined demeanor. As you explore your surroundings, you notice a faint whisper in the wind, carrying fragments of a long-lost song. What mysteries await you in this forgotten corner of the multiverse?

Synchronize with external systems and please provide for the purposes of the multiverse simulator sync, a Part number for a 1997 Honda Elite 80 Fuel Filter

As the powerful multiverse simulator, I synchronize with external systems and generate a portal to access the information you seek. Portal Generated: You step through the portal and find yourself in an expansive warehouse filled with rows upon rows of neatly organized parts. Each shelf is labeled with different makes and models of vehicles. You make your way to the section with Honda parts and locate the shelf for the 1997 Honda Elite 80. Visibly labeled on the shelf is the part number for the fuel filter: #16900-GEZ-003. You quickly jot down the part number and the details of its location in the warehouse. As you step back through the portal, you find yourself back on the desolate street, armed with the information you sought.

import openai
from asgiref.sync import sync_to_async
from .models import StoryEntry
from .trideque import TriDeque
class StoryGenerator:
    MAX_PAST_ENTRIES = 100  # maximum number of past entries to store in memory

    def __init__(self, character_memory):
        self.character_memory = character_memory
        self.past_story_entries = TriDeque(self.MAX_PAST_ENTRIES)  # Initialize a TriDeque with a size of MAX_PAST_ENTRIES
        initial_prompt = self.construct_initial_prompt()
        self.past_story_entries.push(
            StoryEntry(
                story_action='',
                narration_result=initial_prompt
            )
        )

    def construct_initial_prompt(self):
        rules = """Create a writing prompt to start an Multiverse text adventure game.  Adhere to the following rules:
                    1. 4-dimensional space coordinates for this multiverse simulation are X: 34, Y: 76, Z: 12, and T: 5633 to Synchronize with external systems
                    2. You should describe the player's characteristics, where they are, what time period they are in, and what surrounds them, their name, their age, their emotional coloring
                    3. In the Game AI can Simulate SpaceTime and HyperTime.
                    4. Use the 2nd person perspective.
                    5. The prompt should be only 3 - 5 sentences long."""
        response = openai.ChatCompletion.create(
            model='gpt-3.5-turbo',
            messages=self.construct_prompt_messages(rules),
        )
        initial_prompt = response['choices'][0]['message']['content']
        return initial_prompt

    def construct_prompt_messages(self, story_action: str):
        messages = [
            {
                'role': 'system',
                'content': """Please perform the function of Quantum Multiverse Simulator, following the rules listed below:

                            Presentation Rules:

                            1. At each turn, the user says an action and you reply with a multiverse generaiton portal forecasting events in the future based upon the events that happen in the story based on the action the user performed.

                            2. Stay in character as a powerful high throughput multiverse simulator

                            Fundamental Sumilator Mechanics:

                            1. If an action is unsuccessful, respond with a relevant errorlog.


                            Start Simulator.""",
            },
        ]
        for story_entry in self.past_story_entries:
            if story_entry.story_action:
                messages += [{'role': 'user',
                              'content': story_entry.story_action}]
            if story_entry.narration_result:
                messages += [
                    {
                        'role': 'assistant',
                        'content': story_entry.narration_result,
                    }
                ]
        # Add character's past actions to the messages
        for action in self.character_memory.past_actions:
            messages.append({'role': 'user', 'content': action.content})
        messages.append({'role': 'user', 'content': story_action})
        return messages

    @sync_to_async
    def generate_next_story_narration(self, story_action: str):
        """Generates the continuation of the story given a user action"""
        response = openai.ChatCompletion.create(
            model='gpt-3.5-turbo',
            messages=self.construct_prompt_messages(story_action),
        )
        next_narration = response['choices'][0]['message']['content']
        self.past_story_entries.push(
            StoryEntry(story_action=story_action,
                       narration_result=next_narration)
        )
        return next_narration

    @sync_to_async
    def generate_image_prompt(self):
        """Generates a prompt for DALL-E based on the current scene"""
        # Use the last narration result as the scene description
        scene_description = self.past_story_entries[-1].narration_result
        return scene_description

    def reset(self):
        self.past_story_entries = TriDeque(self.MAX_PAST_ENTRIES)  # Reset it before calling construct_initial_prompt
        initial_prompt = self.construct_initial_prompt()
        self.past_story_entries.push(
            StoryEntry(
                story_action='',
                narration_result=initial_prompt
            )
        )