import gradio as gr
import joblib
import PyPDF2
import nltk
from collections import Counter

nltk.download("punkt")
nltk.download("punkt_tab")
nltk.download("stopwords")

from nltk.corpus import stopwords
from nltk.tokenize import word_tokenize

model = joblib.load("pdf_model.pkl")
vectorizer = joblib.load("vectorizer.pkl")


def extract_text(file):

    text = ""

    reader = PyPDF2.PdfReader(file)

    for page in reader.pages:
        page_text = page.extract_text()
        if page_text:
            text += page_text

    return text


def extract_keywords(text):

    words = word_tokenize(text.lower())

    filtered = [
        w for w in words
        if w.isalpha() and w not in stopwords.words("english")
    ]

    counts = Counter(filtered)

    keywords = [w for w,_ in counts.most_common(5)]

    return keywords


def summarize(text):

    sentences = text.split(".")
    return ".".join(sentences[:3])


def analyze_pdf(file):

    text = extract_text(file)

    keywords = extract_keywords(text)

    summary = summarize(text)

    X = vectorizer.transform([text])

    pred = model.predict(X)[0]

    category = {
        0: "Finance / Banking Document",
        1: "Technology / Cloud / Machine Learning"
    }

    return f"""
Category: {category[pred]}

Keywords: {", ".join(keywords)}

Summary:
{summary}
"""


iface = gr.Interface(
    fn=analyze_pdf,
    inputs=gr.File(),
    outputs="text",
    title="AI PDF Analyzer",
    description="Upload a PDF to analyze its content, keywords and summary."
)

iface.launch()