import gradio as gr
from openai import OpenAI
from langchain_community.document_loaders import TextLoader
from langchain_text_splitters import CharacterTextSplitter
from langchain_community.vectorstores import FAISS
from langchain_community.embeddings import FakeEmbeddings
import os

# Load and split the text
loader = TextLoader("mindmate.txt")
documents = loader.load()
text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
split_docs = text_splitter.split_documents(documents)
vector_db = FAISS.from_documents(split_docs, FakeEmbeddings(size=100))

# Set up OpenRouter with DeepSeek R1
client = OpenAI(
    base_url="https://openrouter.ai/api/v1",
    api_key="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"
)

# Define system prompt
SYSTEM_PROMPT = (
    "You are a warm and emotionally intelligent mental health companion 🧠💛. "
    "You deeply understand the user's problems and respond with empathy and clarity. "
    "Provide comforting, short fixes as bullet points (•). "
    "Keep responses clean – do not use markdown like ** or * anywhere. "
    "Add emojis (🌟💪🌈🫶) to make it emotionally expressive. "
    "Be soothing, friendly, and non-judgmental. Be on the user's side always. "
    "Make sure the advice is helpful, practical, and to the point."
)

def chatbot(name, issue):
    full_prompt = f"{name} is feeling emotionally low. Reason: {issue}. Please help."
    docs = vector_db.similarity_search(issue, k=2)
    context = "\n\n".join([doc.page_content for doc in docs])

    response = client.chat.completions.create(
        model="deepseek/deepseek-r1:free",
        messages=[
            {"role": "system", "content": SYSTEM_PROMPT},
            {"role": "user", "content": f"Context:\n{context}\n\nUser: {full_prompt}"}
        ]
    )

    return response.choices[0].message.content

# Gradio UI
with gr.Blocks(theme=gr.themes.Soft()) as demo:
    gr.Markdown("## 🧠 MindMate – Your Mental Health Companion 💛\nShare your feelings and get comforting support 💙")
    with gr.Row():
        with gr.Column():
            name_input = gr.Textbox(label="Your Name", placeholder="e.g., Dhruvil")
            issue_input = gr.Textbox(lines=3, placeholder="What's troubling you today?", label="What’s bothering you?")
            send_button = gr.Button("🪄 Get Support")
        with gr.Column():
            chatbot_output = gr.Textbox(lines=12, label="MindMate's Response")

    send_button.click(fn=chatbot, inputs=[name_input, issue_input], outputs=chatbot_output)

demo.launch()