# Import necessary libraries
import streamlit as st
import re
import nltk
import os
from nltk.corpus import stopwords
from nltk import FreqDist
from graphviz import Digraph
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.cluster import KMeans

# Set page configuration with a title and favicon
st.set_page_config(
    page_title="📺Transcript📜EDA🔍NLTK",
    page_icon="🌠",
    layout="wide",
    initial_sidebar_state="expanded",
    menu_items={
        'Get Help': 'https://huggingface.co/awacke1',
        'Report a bug': "https://huggingface.co/spaces/awacke1/WebDataDownload",
        'About': "# Midjourney: https://discord.com/channels/@me/997514686608191558"
    }
)

st.markdown('''🔍 **Exploratory Data Analysis (EDA)** 📊:     - Dive deep into the sea of data with our EDA feature, unveiling hidden patterns 🕵️‍♂️ and insights 🧠 in your transcripts. Transform raw data into a treasure trove of information 🏆.
📜 **Natural Language Toolkit (NLTK)** 🛠️:     - Harness the power of NLTK to process and understand human language 🗣️. From tokenization to sentiment analysis, our toolkit is your compass 🧭 in the vast landscape of natural language processing (NLP).
📺 **Transcript Analysis** 📈:     - Elevate your text analysis with our advanced transcript analysis tools. Whether it's speech recognition 🎙️ or thematic extraction 🌐, turn your audiovisual content into actionable insights 🔑.''')

# Download NLTK resources
nltk.download('punkt')
nltk.download('stopwords')

def remove_timestamps(text):
    return re.sub(r'\d{1,2}:\d{2}\n.*\n', '', text)

def extract_high_information_words(text, top_n=10):
    words = nltk.word_tokenize(text)
    words = [word.lower() for word in words if word.isalpha()]
    stop_words = set(stopwords.words('english'))
    filtered_words = [word for word in words if word not in stop_words]
    freq_dist = FreqDist(filtered_words)
    return [word for word, _ in freq_dist.most_common(top_n)]

def create_relationship_graph(words):
    graph = Digraph()
    for index, word in enumerate(words):
        graph.node(str(index), word)
        if index > 0:
            graph.edge(str(index - 1), str(index), label=str(index))
    return graph

def display_relationship_graph(words):
    graph = create_relationship_graph(words)
    st.graphviz_chart(graph)

def extract_context_words(text, high_information_words):
    words = nltk.word_tokenize(text)
    context_words = []
    for index, word in enumerate(words):
        if word.lower() in high_information_words:
            before_word = words[index - 1] if index > 0 else None
            after_word = words[index + 1] if index < len(words) - 1 else None
            context_words.append((before_word, word, after_word))
    return context_words

def create_context_graph(context_words):
    graph = Digraph()
    for index, (before_word, high_info_word, after_word) in enumerate(context_words):
        #graph.node(f'before{index}', before_word, shape='box') if before_word else None
        if before_word:  graph.node(f'before{index}', before_word, shape='box') # else None
        graph.node(f'high{index}', high_info_word, shape='ellipse')
        #graph.node(f'after{index}', after_word, shape='diamond') if after_word else None
        if after_word: graph.node(f'after{index}', after_word, shape='diamond') # else None
        if before_word:
            graph.edge(f'before{index}', f'high{index}')
        if after_word:
            graph.edge(f'high{index}', f'after{index}')
    return graph

def display_context_graph(context_words):
    graph = create_context_graph(context_words)
    st.graphviz_chart(graph)

def display_context_table(context_words):
    table = "| Before | High Info Word | After |\n|--------|----------------|-------|\n"
    for before, high, after in context_words:
        table += f"| {before if before else ''} | {high} | {after if after else ''} |\n"
    st.markdown(table)

def load_example_files():
    # Exclude specific files
    excluded_files = {'freeze.txt', 'requirements.txt', 'packages.txt', 'pre-requirements.txt'}
    
    # List all .txt files excluding the ones in excluded_files
    example_files = [f for f in os.listdir() if f.endswith('.txt') and f not in excluded_files]
    
    # Check if there are any files to select from
    if example_files:
        selected_file = st.selectbox("📄 Select an example file:", example_files)
        if st.button(f"📂 Load {selected_file}"):
            with open(selected_file, 'r', encoding="utf-8") as file:
                return file.read()
    else:
        st.write("No suitable example files found.")
    
    return None

def cluster_sentences(sentences, num_clusters):
    # Filter sentences with length over 10 characters
    sentences = [sentence for sentence in sentences if len(sentence) > 10]

    # Check if the number of sentences is less than the desired number of clusters
    if len(sentences) < num_clusters:
        # If so, adjust the number of clusters to match the number of sentences
        num_clusters = len(sentences)

    # Vectorize the sentences
    vectorizer = TfidfVectorizer()
    X = vectorizer.fit_transform(sentences)

    # Perform k-means clustering
    kmeans = KMeans(n_clusters=num_clusters, random_state=42)
    kmeans.fit(X)

    # Get the cluster labels for each sentence
    labels = kmeans.labels_

    # Group sentences by cluster
    clustered_sentences = [[] for _ in range(num_clusters)]
    for i, label in enumerate(labels):
        clustered_sentences[label].append((i, sentences[i]))

    return clustered_sentences

# Main code for UI
uploaded_file = st.file_uploader("📁 Choose a .txt file", type=['txt'])

example_text = load_example_files()

if example_text:
    file_text = example_text
elif uploaded_file:
    file_text = uploaded_file.read().decode("utf-8")
else:
    file_text = ""

if file_text:
    text_without_timestamps = remove_timestamps(file_text)
    sentences = [sentence.strip() for sentence in text_without_timestamps.split('.') if len(sentence.strip()) > 10]

    top_words = extract_high_information_words(text_without_timestamps, 10)

    with st.expander("📊 Top 10 High Information Words"):
        st.write(top_words)

    with st.expander("📈 Relationship Graph"):
        display_relationship_graph(top_words)

    context_words = extract_context_words(text_without_timestamps, top_words)

    with st.expander("🔗 Context Graph"):
        display_context_graph(context_words)

    with st.expander("📑 Context Table"):
        display_context_table(context_words)

    # with st.expander("Innovation Outlines"):
    #    showInnovationOutlines()

    with st.expander("📝 Sentence Clustering"):
        num_clusters = st.slider("Number of Clusters", min_value=2, max_value=10, value=5)
        clustered_sentences = cluster_sentences(sentences, num_clusters)

        output_text = ""
        for i, cluster in enumerate(clustered_sentences):
            output_text += f"## 🌟 Cluster {i+1}:\n"
            for original_index, sentence in cluster:
                output_text += f"- Original Line {original_index+1}: {sentence}\n"
            output_text += "\n"

        st.markdown(output_text)

st.markdown("https://cdn-uploads.huggingface.co/production/uploads/620630b603825909dcbeba35/Id9kntHFHZf_oFFrEmGh5.png")