app.py

import streamlit as st
import pandas as pd
import datetime
import io
import nltk
import base64
from nltk.tokenize import sent_tokenize
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.decomposition import LatentDirichletAllocation

nltk.download('punkt')

def save_text_as_file(text, file_type):
    current_time = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
    file_name = f"text_file_{current_time}.{file_type}"
    with open(file_name, "w") as file:
        file.write(text)
    st.success(f"Text saved as {file_name}")
    return file_name

def save_list_as_excel(text):
    lines = text.split("\n")
    data = []
    for line in lines:
        if line.strip():
            parts = line.split(" - ", 1)
            if len(parts) == 2:
                data.append(parts)
            else:
                data.append([line.strip(), ""])
    df = pd.DataFrame(data, columns=["Character", "Description"])
    file_name = f"character_list_{datetime.datetime.now().strftime('%Y%m%d_%H%M%S')}.xlsx"
    df.to_excel(file_name, index=False)
    st.success(f"Character list saved as {file_name}")
    return file_name

def get_download_link(file_path):
    with open(file_path, 'rb') as f:
        data = f.read()
    b64 = base64.b64encode(data).decode()
    href = f'<a href="data:application/octet-stream;base64,{b64}" download="{file_path}">Download {file_path}</a>'
    return href

def perform_nlp(text):
    sentences = sent_tokenize(text)
    
    # Topic Modeling
    vectorizer = CountVectorizer(stop_words='english')
    X = vectorizer.fit_transform(sentences)
    lda = LatentDirichletAllocation(n_components=3, random_state=42)
    lda.fit(X)
    topics = lda.transform(X)
    
    # Display topics
    st.subheader("Topic Modeling")
    for i, topic in enumerate(topics):
        st.write(f"Topic {i+1}:")
        topic_words = ", ".join([vectorizer.get_feature_names_out()[i] for i in topic.argsort()[:-6:-1]])
        st.write(topic_words)
    
    # Word Frequency
    word_freq = pd.Series(" ".join(sentences).split()).value_counts()[:10]
    st.subheader("Word Frequency")
    st.bar_chart(word_freq)

def main():
    st.title("AI UI for Text Processing")
    
    text_input = st.text_area("Paste your text here")
    if st.button("Process Text"):
        if text_input.strip() == "":
            st.warning("Please paste some text.")
        else:
            file_name = None
            if text_input.strip().startswith(("1.", "1 -", "1 _")) and "\n" in text_input:
                file_name = save_list_as_excel(text_input)
            elif "." in text_input or "!" in text_input or "?" in text_input:
                file_name = save_text_as_file(text_input, "txt")
                perform_nlp(text_input)
            else:
                file_name = save_text_as_file(text_input, "txt")
            
            if file_name:
                try:
                    df = pd.read_excel(file_name)
                    st.subheader("Saved Data")
                    st.dataframe(df)
                    st.markdown(get_download_link(file_name), unsafe_allow_html=True)
                except:
                    pass

if __name__ == "__main__":
    main()