import torch
import re
import gradio as gr
import pandas as pd
import matplotlib.pyplot as plt
from pydantic import BaseModel, PydanticUserError, ConfigDict
from pydantic import BaseModel, ConfigDict

class MyModel(BaseModel):
    request: 'starlette.requests.Request'
    model_config = ConfigDict(arbitrary_types_allowed=True)
from pydantic_core import core_schema
from starlette.requests import Request

def get_pydantic_core_schema(request_type, handler):
    return core_schema.any_schema()

Request.__get_pydantic_core_schema__ = get_pydantic_core_schema


from transformers import pipeline

pipe = pipeline("text-classification", 
                model="distilbert/distilbert-base-uncased-finetuned-sst-2-english")



analyzer=pipeline(task="text-classification", model="distilbert/distilbert-base-uncased-finetuned-sst-2-english")


def sentiment_analyzer(review):
    sentiment = analyzer(review)
    return sentiment[0]['label']

def sentiment_bar_chart(df):
    sentiment_counts=df['Sentiment'].value_counts()

    fig, ax=plt.subplots()
    sentiment_counts.plot(kind='pie', ax=ax, autopct='%1.1f%%', color=['green', 'red'])
    ax.set_title('Review Sentiment Counts')
    ax.set_xlabel('Sentiment')
    ax.set_ylabel('Count')
    #ax.set_xticklabels(['Positive', 'Negative'], rotation=0)

    return fig
    

def get_sentiment(review):
    from textblob import TextBlob
    analysis=TextBlob(review)
    return analysis.sentiment.polarity

def read_reviews_and_analyze_sentiment(file_object):
    print("file_obj received:", file_object)
    
    # Handle file path or file object
    if hasattr(file_object, 'name'):
        file_name = file_object.name
    else:
        file_name = file_object

    # Read file based on extension
    if str(file_name).lower().endswith('.csv'):
        df = pd.read_csv(file_object)
    elif str(file_name).lower().endswith(('.xlsx', '.xls', '.xlsm', '.xlsb', '.ods', '.odt')):
        df = pd.read_excel(file_object)
    else:
        raise ValueError("Unsupported file type. Please upload a CSV or Excel file.")
    
    df['Sentiment']=df['Reviews'].apply(sentiment_analyzer)
    chart_object=sentiment_bar_chart(df)

    return df, chart_object


gr.close_all()
demo = gr.Interface(
    fn=read_reviews_and_analyze_sentiment,
    inputs=gr.File(file_types=['.xlsx','csv'], label='Upload your review comment file'),
    outputs=[gr.Dataframe(label='Sentiments'), gr.Plot(label='Sentiment Analysis')],
    title='KS Sentiment Analyzer',
    description='This application will classify the sentiment of reviews based on an uploaded Excel file. The file must contain a column named "Reviews".'
)

demo.launch(share=True)