app.py

import requests
import gradio as gr
from ragatouille import RAGPretrainedModel
import logging
from pathlib import Path
from time import perf_counter
from sentence_transformers import CrossEncoder
from huggingface_hub import InferenceClient
from jinja2 import Environment, FileSystemLoader
import numpy as np
from os import getenv
from backend.query_llm import generate_hf, generate_qwen
from backend.semantic_search import table, retriever
from huggingface_hub import InferenceClient


# Bhashini API translation function
api_key = getenv('API_KEY')
user_id = getenv('USER_ID')

def bhashini_translate(text: str, from_code: str = "en", to_code: str = "hi") -> dict:
    """Translates text from source language to target language using the Bhashini API."""
    
    if not text.strip():
        print('Input text is empty. Please provide valid text for translation.')
        return {"status_code": 400, "message": "Input text is empty", "translated_content": None, "speech_content": None}
    else:
        print('Input text - ',text)
    print(f'Starting translation process from {from_code} to {to_code}...')
    print(f'Starting translation process from {from_code} to {to_code}...')
    gr.Warning(f'Translating to {to_code}...')
    
    url = 'https://meity-auth.ulcacontrib.org/ulca/apis/v0/model/getModelsPipeline'
    headers = {
        "Content-Type": "application/json",
        "userID": user_id,
        "ulcaApiKey": api_key
    }
    payload = {
        "pipelineTasks": [{"taskType": "translation", "config": {"language": {"sourceLanguage": from_code, "targetLanguage": to_code}}}],
        "pipelineRequestConfig": {"pipelineId": "64392f96daac500b55c543cd"}
    }
    
    print('Sending initial request to get the pipeline...')
    response = requests.post(url, json=payload, headers=headers)
    
    if response.status_code != 200:
        print(f'Error in initial request: {response.status_code}')
        return {"status_code": response.status_code, "message": "Error in translation request", "translated_content": None}

    print('Initial request successful, processing response...')
    response_data = response.json()
    service_id = response_data["pipelineResponseConfig"][0]["config"][0]["serviceId"]
    callback_url = response_data["pipelineInferenceAPIEndPoint"]["callbackUrl"]
    
    print(f'Service ID: {service_id}, Callback URL: {callback_url}')
    
    headers2 = {
        "Content-Type": "application/json",
        response_data["pipelineInferenceAPIEndPoint"]["inferenceApiKey"]["name"]: response_data["pipelineInferenceAPIEndPoint"]["inferenceApiKey"]["value"]
    }
    compute_payload = {
        "pipelineTasks": [{"taskType": "translation", "config": {"language": {"sourceLanguage": from_code, "targetLanguage": to_code}, "serviceId": service_id}}],
        "inputData": {"input": [{"source": text}], "audio": [{"audioContent": None}]}
    }
    
    print(f'Sending translation request with text: "{text}"')
    compute_response = requests.post(callback_url, json=compute_payload, headers=headers2)
    
    if compute_response.status_code != 200:
        print(f'Error in translation request: {compute_response.status_code}')
        return {"status_code": compute_response.status_code, "message": "Error in translation", "translated_content": None}
    
    print('Translation request successful, processing translation...')
    compute_response_data = compute_response.json()
    translated_content = compute_response_data["pipelineResponse"][0]["output"][0]["target"]
    
    print(f'Translation successful. Translated content: "{translated_content}"')
    return {"status_code": 200, "message": "Translation successful", "translated_content": translated_content}


# Existing chatbot functions
VECTOR_COLUMN_NAME = "vector"
TEXT_COLUMN_NAME = "text"
HF_TOKEN = getenv("HUGGING_FACE_HUB_TOKEN")
proj_dir = Path(__file__).parent

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1", token=HF_TOKEN)
env = Environment(loader=FileSystemLoader(proj_dir / 'templates'))

template = env.get_template('template.j2')
template_html = env.get_template('template_html.j2')

def add_text(history, text):
    history = [] if history is None else history
    history = history + [(text, None)]
    return history, gr.Textbox(value="", interactive=False)

def bot(history, cross_encoder):

    top_rerank = 25
    top_k_rank = 20
    query = history[-1][0] if history else ''

    if not query:
        gr.Warning("Please submit a non-empty string as a prompt")
        raise ValueError("Empty string was submitted")

    logger.warning('Retrieving documents...')
    
    if cross_encoder == '(HIGH ACCURATE) ColBERT':
        gr.Warning('Retrieving using ColBERT.. First time query will take a minute for model to load..pls wait')
        RAG = RAGPretrainedModel.from_pretrained("colbert-ir/colbertv2.0")
        RAG_db = RAG.from_index('.ragatouille/colbert/indexes/cbseclass10index')
        documents_full = RAG_db.search(query, k=top_k_rank)
        
        documents = [item['content'] for item in documents_full]
        prompt = template.render(documents=documents, query=query)
        prompt_html = template_html.render(documents=documents, query=query)
    
        generate_fn = generate_hf
    
        history[-1][1] = ""
        for character in generate_fn(prompt, history[:-1]):
            history[-1][1] = character
            yield history, prompt_html
    else:
        document_start = perf_counter()
    
        query_vec = retriever.encode(query)
        doc1 = table.search(query_vec, vector_column_name=VECTOR_COLUMN_NAME).limit(top_k_rank)
    
        documents = table.search(query_vec, vector_column_name=VECTOR_COLUMN_NAME).limit(top_rerank).to_list()
        documents = [doc[TEXT_COLUMN_NAME] for doc in documents]
    
        query_doc_pair = [[query, doc] for doc in documents]
        if cross_encoder == '(FAST) MiniLM-L6v2':
            cross_encoder1 = CrossEncoder('cross-encoder/ms-marco-MiniLM-L-6-v2')
        elif cross_encoder == '(ACCURATE) BGE reranker':
            cross_encoder1 = CrossEncoder('BAAI/bge-reranker-base')
        
        cross_scores = cross_encoder1.predict(query_doc_pair)
        sim_scores_argsort = list(reversed(np.argsort(cross_scores)))
        
        documents = [documents[idx] for idx in sim_scores_argsort[:top_k_rank]]
    
        document_time = perf_counter() - document_start
    
        prompt = template.render(documents=documents, query=query)
        prompt_html = template_html.render(documents=documents, query=query)
    
        #generate_fn = generate_hf
        generate_fn=generate_qwen
        # Create a new history entry instead of modifying the tuple directly
        new_history = history[:-1] + [ (query, "") ]
        for character in generate_fn(prompt, history[:-1]):
            new_history[-1] = (query, character) 
            print('Character',character)# Update the last tuple with new text
            yield new_history, prompt_html
        # history[-1][1] = ""
        # for character in generate_fn(prompt, history[:-1]):
        #     history[-1][1] = character
        #     yield history, prompt_html

#def translate_text(response_text, selected_language):
    
def translate_text(selected_language,history):
    
    iso_language_codes = {
        "Hindi": "hi",
        "Gom": "gom",
        "Kannada": "kn",
        "Dogri": "doi",
        "Bodo": "brx",
        "Urdu": "ur",
        "Tamil": "ta",
        "Kashmiri": "ks",
        "Assamese": "as",
        "Bengali": "bn",
        "Marathi": "mr",
        "Sindhi": "sd",
        "Maithili": "mai",
        "Punjabi": "pa",
        "Malayalam": "ml",
        "Manipuri": "mni",
        "Telugu": "te",
        "Sanskrit": "sa",
        "Nepali": "ne",
        "Santali": "sat",
        "Gujarati": "gu",
        "Odia": "or"
    }
    
    to_code = iso_language_codes[selected_language]
    response_text = history[-1][1] if history else ''
    translation = bhashini_translate(response_text, to_code=to_code)
    return translation['translated_content']
   

# Gradio interface
with gr.Blocks(theme='gradio/soft') as CHATBOT:
    history_state = gr.State([])
    with gr.Row():
        with gr.Column(scale=10):
            gr.HTML(value="""<div style="color: #FF4500;"><h1>ADWITIYA-</h1> <h1><span style="color: #008000">Custom Manual Chatbot and Quizbot</span></h1></div>""")
            gr.HTML(value=f"""<p style="font-family: sans-serif; font-size: 16px;">Using GenAI for CBIC Capacity Building - A free chat bot developed by National Customs Targeting Center using Open source LLMs for CBIC Officers</p>""")
            gr.HTML(value=f"""<p style="font-family: Arial, sans-serif; font-size: 14px;">Developed by NCTC,Mumbai. Suggestions may be sent to <a href="mailto:[email protected]" style="color: #00008B; font-style: italic;">[email protected]</a>.</p>""")

        with gr.Column(scale=3):
            gr.Image(value='logo.png', height=200, width=200)

    chatbot = gr.Chatbot(
        [],
        elem_id="chatbot",
        avatar_images=('https://aui.atlassian.com/aui/8.8/docs/images/avatar-person.svg',
                       'https://huggingface.co/datasets/huggingface/brand-assets/resolve/main/hf-logo.svg'),
        bubble_full_width=False,
        show_copy_button=True,
        show_share_button=True,
    )

    with gr.Row():
        txt = gr.Textbox(
            scale=3,
            show_label=False,
            placeholder="Enter text and press enter",
            container=False,
        )
        txt_btn = gr.Button(value="Submit text", scale=1)
    
    cross_encoder = gr.Radio(choices=['(FAST) MiniLM-L6v2', '(ACCURATE) BGE reranker', '(HIGH ACCURATE) ColBERT'], value='(ACCURATE) BGE reranker', label="Embeddings", info="Only First query to Colbert may take little time)")
    language_dropdown = gr.Dropdown(
        choices=[
            "Hindi", "Gom", "Kannada", "Dogri", "Bodo", "Urdu", "Tamil", "Kashmiri", "Assamese", "Bengali", "Marathi",
            "Sindhi", "Maithili", "Punjabi", "Malayalam", "Manipuri", "Telugu", "Sanskrit", "Nepali", "Santali",
            "Gujarati", "Odia"
        ],
        value="Hindi",  # default to Hindi
        label="Select Language for Translation"
    )
    
    prompt_html = gr.HTML()
    
    translated_textbox = gr.Textbox(label="Translated Response")
    def update_history_and_translate(txt, cross_encoder, history_state, language_dropdown):
        history = history_state
        history.append((txt, ""))
        #history_state.value=(history)
        
        # Call bot function
        bot_output = list(bot(history, cross_encoder))
        history, prompt_html = bot_output[-1]
        
        # Update the history state
        history_state[:] = history
        
        # Translate text
        translated_text = translate_text(language_dropdown, history)
        return history, prompt_html, translated_text

    txt_msg = txt_btn.click(update_history_and_translate, [txt, cross_encoder, history_state, language_dropdown], [chatbot, prompt_html, translated_textbox])
    txt_msg = txt.submit(update_history_and_translate, [txt, cross_encoder, history_state, language_dropdown], [chatbot, prompt_html, translated_textbox])

    examples = ['My transhipment cargo is missing','can u explain and tabulate difference between b 17 bond and a warehousing bond',
            'What are benefits of  the AEO Scheme and eligibility criteria?',
            'What are penalties for customs offences? ', 'what are penalties to customs officers misusing their powers under customs act?','What are eligibility criteria for exemption from cost recovery charges','list in detail what is procedure for obtaining new approval for openeing a CFS attached to an ICD']

    gr.Examples(examples, txt)


    
    # txt_msg = txt_btn.click(add_text, [chatbot, txt], [chatbot, txt], queue=False).then(
    #     bot, [chatbot, cross_encoder], [chatbot, prompt_html]).then(
    #     translate_text, [txt, language_dropdown], translated_textbox
    # )
    
    # txt_msg = txt.submit(add_text, [chatbot, txt], [chatbot, txt], queue=False).then(
    #     bot, [chatbot, cross_encoder], [chatbot, prompt_html]).then(
    #     translate_text, [txt, language_dropdown], translated_textbox
    # )

# Launch the Gradio application
CHATBOT.launch(share=True)