import gradio as gr
from transformers import pipeline
import chardet

# Initialize the question-answering pipeline
#qa_pipeline = pipeline("question-answering",model="deepset/roberta-base-squad2")
qa_pipeline = pipeline("question-answering",model="distilbert-base-cased-distilled-squad")
def answer_question(context, question):
    result = qa_pipeline(question=question, context=context)
    return result['answer']


def process(context_file, question):
    # Read the context from the uploaded file
    
    #with open(context_file.name, 'r') as file:
        #context = file.read()
    with open(context_file.name, 'rb') as file:
        raw_data = file.read()
        result = chardet.detect(raw_data)
        encoding = result['encoding']
    
        # Fallback to a default encoding if detection fails
        if encoding is None:
            encoding = 'utf-8'  # You can change this to another default encoding
    
        context = raw_data.decode(encoding, errors='replace')  # Replace errors with a placeholder
    

    answer = answer_question(context, question)
    return answer


# Gradio interface
demo = gr.Interface(
    fn=process,
    inputs=[gr.File(label="Upload Context File"), gr.Textbox(label="Question")],
    outputs=[gr.Textbox(label="Answer")],
    title="Question Answering",
    description="Upload a file with context and ask a question. The answer will be displayed."
)

if __name__ == "__main__":
    demo.launch()