app.py

import gradio as gr
from transformers import pipeline
from PIL import Image
from fastapi import FastAPI
from starlette.responses import RedirectResponse

# 🔥 Fix for Pydantic v2 compatibility with Gradio
import gradio.context
from pydantic import BaseModel
if not hasattr(BaseModel, "model_fields"):  # model_fields was renamed from __fields__ in Pydantic v1 → v2
    BaseModel.model_fields = BaseModel.__fields__

# 🔁 Load Hugging Face Pipelines
summarizer = pipeline("summarization", model="sshleifer/distilbart-cnn-12-6")
image_captioner = pipeline("image-to-text", model="nlpconnect/vit-gpt2-image-captioning")

# 🚀 Create FastAPI App
app = FastAPI()

def analyze_input(file, question=None):
    if file is None:
        return "Please upload a document or image."

    filename = file.name.lower()

    # 🖼️ Image
    if filename.endswith((".png", ".jpg", ".jpeg")):
        image = Image.open(file)
        caption = image_captioner(image)[0]['generated_text']
        return f"📷 Image Interpretation:\n{caption}"

    # 📄 Document
    elif filename.endswith((".pdf", ".docx", ".pptx", ".xlsx")):
        import pdfplumber
        import docx
        import pptx
        import pandas as pd

        try:
            text = ""

            if filename.endswith(".pdf"):
                with pdfplumber.open(file) as pdf:
                    text = "\n".join([page.extract_text() for page in pdf.pages if page.extract_text()])

            elif filename.endswith(".docx"):
                doc = docx.Document(file)
                text = "\n".join([p.text for p in doc.paragraphs if p.text.strip()])

            elif filename.endswith(".pptx"):
                prs = pptx.Presentation(file)
                for slide in prs.slides:
                    for shape in slide.shapes:
                        if hasattr(shape, "text"):
                            text += shape.text + "\n"

            elif filename.endswith(".xlsx"):
                df = pd.read_excel(file, sheet_name=None)
                text = "\n".join([df[sheet].to_string() for sheet in df])

            if not text.strip():
                return "❌ Could not extract meaningful text from the document."

            summary = summarizer(text[:3000], max_length=200, min_length=30, do_sample=False)
            return f"📄 Document Summary:\n{summary[0]['summary_text']}"

        except Exception as e:
            return f"❌ Error processing document: {str(e)}"

    else:
        return "❌ Unsupported file type. Please upload a valid image or document."

# 🎛️ Gradio UI
iface = gr.Interface(
    fn=analyze_input,
    inputs=gr.File(label="Upload Document or Image"),
    outputs=gr.Textbox(label="Result", lines=10),
    title="Document & Image Analysis Web Service",
    description="Upload a document (PDF, DOCX, PPTX, XLSX) or image to get a summary or caption. CPU-friendly."
)

# ⌨️ Wrap in Tabbed UI
demo = gr.TabbedInterface([iface], ["Docs and Images"])

# 🔗 Mount Gradio to FastAPI
app = gr.mount_gradio_app(app, demo, path="/")

# 🏠 Base redirect
@app.get("/")
def home():
    return RedirectResponse(url="/")