Initital deploy

main.py

@@ -0,0 +1,127 @@
+from fastapi import FastAPI, UploadFile, File, HTTPException
+from fastapi.middleware.cors import CORSMiddleware
+from pydantic import BaseModel
+import os
+import logging
+from parser import parse_pdf, parse_text
+from rag import RAG
+import requests
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+app = FastAPI(title="Simple NotebookLM-like AI Tool")
+
+# Allow frontend at http://localhost:5500 to access this backend
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],  # Replace with ["http://localhost:5500"] for tighter security
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+rag = RAG()
+documents = {}  # {file_id: {"file_path": str, "file_name": str}}
+
+class QueryRequest(BaseModel):
+    question: str
+    file_id: str
+
+@app.post("/upload")
+async def upload_file(file: UploadFile = File(...)):
+    try:
+        logger.info(f"Received file: {file.filename}, type: {file.content_type}")
+
+        if file.content_type not in ["application/pdf", "text/plain", "application/octet-stream"]:
+            raise HTTPException(status_code=400, detail="Only PDF or text files allowed")
+
+        file_id = str(len(documents) + 1)
+        safe_name = file.filename.replace("/", "_").replace("\\", "_")
+        file_path = f"uploads/{file_id}_{safe_name}"
+
+        os.makedirs("uploads", exist_ok=True)
+        file_bytes = await file.read()
+
+        with open(file_path, "wb") as f:
+            f.write(file_bytes)
+
+        text = parse_pdf(file_bytes) if file.filename.endswith(".pdf") else parse_text(file_bytes)
+        logger.info(f"Extracted text length: {len(text.strip())}")
+
+        if not text.strip():
+            logger.warning("No text extracted. Assigning default placeholder text.")
+            text = "No extractable text found in this file."
+
+        documents[file_id] = {"file_path": file_path, "file_name": file.filename}
+
+        try:
+            await rag.embed_document(file_id, text)
+        except Exception as e:
+            logger.error(f"Embedding failed: {e}")
+            raise HTTPException(status_code=500, detail="Embedding failed")
+
+        return {"file_id": file_id, "file_name": file.filename}
+
+    except HTTPException as http_err:
+        raise http_err
+    except Exception as e:
+        logger.error(f"Upload failed: {e}")
+        raise HTTPException(status_code=500, detail="Failed to process file")
+
+@app.post("/query")
+async def query_file(request: QueryRequest):
+    if request.file_id not in documents:
+        raise HTTPException(status_code=404, detail="File not found")
+
+    try:
+        context = await rag.query_document(request.question, request.file_id)
+        if not context:
+            return {"answer": "No relevant information found in the document."}
+
+        context_text = "\n".join(context)
+        prompt = f"Using the following context, answer the question: {request.question}\n\nContext: {context_text[:5000]}"
+
+        response = requests.post(
+            "http://localhost:11434/api/generate",
+            json={"model": "qwen:7b", "prompt": prompt, "stream": False}
+        )
+        response.raise_for_status()
+        data = response.json()
+        answer = data.get("response", "Error generating answer")
+
+        logger.info(f"Question: {request.question}, Answer: {answer[:50]}...")
+        return {"answer": answer}
+
+    except Exception as e:
+        logger.error(f"Query failed: {e}")
+        raise HTTPException(status_code=500, detail="Error answering question")
+
+@app.post("/summarize")
+async def summarize_file(file_id: str):
+    if file_id not in documents:
+        raise HTTPException(status_code=404, detail="File not found")
+
+    try:
+        with open(documents[file_id]["file_path"], "rb") as f:
+            content = f.read()
+            text = parse_pdf(content) if documents[file_id]["file_name"].endswith(".pdf") else parse_text(content)
+
+        if not text.strip():
+            raise ValueError("No text found to summarize.")
+
+        prompt = f"Summarize this text in 100 words or less: {text[:5000]}"
+        response = requests.post(
+            "http://localhost:11434/api/generate",
+            json={"model": "qwen:7b", "prompt": prompt, "stream": False}
+        )
+        response.raise_for_status()
+        data = response.json()
+        summary = data.get("response", "Error generating summary")
+
+        logger.info(f"Summary for file {file_id}: {summary[:50]}...")
+        return {"summary": summary}
+
+    except Exception as e:
+        logger.error(f"Summarization failed: {e}")
+        raise HTTPException(status_code=500, detail="Error generating summary")

parser.py

@@ -0,0 +1,28 @@
+from PyPDF2 import PdfReader
+from io import BytesIO
+import logging
+
+logger = logging.getLogger(__name__)
+
+def parse_pdf(file_content):
+    try:
+        reader = PdfReader(BytesIO(file_content))
+        text = ""
+        for page in reader.pages:
+            extracted = page.extract_text()
+            if extracted:
+                text += extracted
+        return text
+    except Exception as e:
+        logger.error(f"PDF parse error: {e}")
+        return ""
+
+def parse_text(file_content):
+    try:
+        return file_content.decode("utf-8")
+    except UnicodeDecodeError:
+        try:
+            return file_content.decode("latin1")
+        except Exception as e:
+            logger.error(f"Text parse error: {e}")
+            return ""

rag.py

@@ -0,0 +1,50 @@
+from sentence_transformers import SentenceTransformer
+import faiss
+import numpy as np
+import textwrap
+import pickle
+import os
+import logging
+
+logger = logging.getLogger(__name__)
+
+class RAG:
+    def __init__(self):
+        self.embedder = SentenceTransformer("all-MiniLM-L6-v2")
+        self.index = faiss.IndexFlatL2(384)
+        self.metadata = {}
+        self.index_file = "faiss_index/index.faiss"
+        self.metadata_file = "faiss_index/metadata.pkl"
+        os.makedirs("faiss_index", exist_ok=True)
+
+        if os.path.exists(self.index_file):
+            self.index = faiss.read_index(self.index_file)
+            with open(self.metadata_file, "rb") as f:
+                self.metadata = pickle.load(f)
+
+    async def embed_document(self, file_id: str, text: str):
+        chunks = textwrap.wrap(text, 500)
+        if not chunks:
+            logger.warning(f"Empty document provided for file_id {file_id}")
+            raise ValueError("Document contains no text to embed.")
+
+        embeddings = self.embedder.encode(chunks)
+        ids = np.array([i for i in range(len(self.metadata), len(self.metadata) + len(chunks))])
+        self.index.add(embeddings.astype("float32"))
+
+        for i, chunk in enumerate(chunks):
+            self.metadata[ids[i]] = {"file_id": file_id, "text": chunk}
+
+        faiss.write_index(self.index, self.index_file)
+        with open(self.metadata_file, "wb") as f:
+            pickle.dump(self.metadata, f)
+
+    async def query_document(self, question: str, file_id: str) -> list:
+        query_embedding = self.embedder.encode([question])[0].astype("float32")
+        distances, indices = self.index.search(np.array([query_embedding]), k=3)
+
+        return [
+            self.metadata[idx]["text"]
+            for idx in indices[0]
+            if idx in self.metadata and self.metadata[idx]["file_id"] == file_id
+        ]

requirement.txt

@@ -0,0 +1,6 @@
+fastapi==0.103.0
+uvicorn==0.23.2
+PyPDF2==3.0.1
+sentence-transformers==2.2.2
+faiss-cpu-windows==1.7.4
+requests==2.31.0

static/index.html

@@ -0,0 +1,177 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0"/>
+    <title>Simple NotebookLM-like Tool</title>
+    <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.3/dist/css/bootstrap.min.css" rel="stylesheet">
+    <style>
+        body {
+            background-color: #f8f9fa;
+            display: flex;
+            justify-content: center;
+            align-items: center;
+            min-height: 100vh;
+            margin: 0;
+            font-family: Arial, sans-serif;
+        }
+        .container {
+            max-width: 600px;
+            background: white;
+            padding: 20px;
+            border-radius: 8px;
+            box-shadow: 0 2px 10px rgba(0, 0, 0, 0.1);
+        }
+        h1 {
+            font-size: 1.8rem;
+            text-align: center;
+            margin-bottom: 20px;
+        }
+        .form-section {
+            margin-bottom: 20px;
+        }
+        #result, #summary {
+            white-space: pre-wrap;
+            background: #f1f1f1;
+            padding: 10px;
+            border-radius: 4px;
+            min-height: 50px;
+        }
+        .btn {
+            width: 100%;
+        }
+        input[type="file"], input[type="text"] {
+            margin-bottom: 10px;
+        }
+    </style>
+</head>
+<body>
+    <div class="container">
+        <h1>NotebookLM-like Tool</h1>
+
+        <!-- File Upload Section -->
+        <div class="form-section">
+            <h3>Upload File</h3>
+            <input type="file" id="fileInput" accept=".pdf,.txt" class="form-control">
+            <button onclick="uploadFile()" class="btn btn-primary mt-2">Upload</button>
+            <div id="uploadStatus" class="mt-2"></div>
+        </div>
+
+        <!-- Question Input Section -->
+        <div class="form-section">
+            <h3>Ask a Question</h3>
+            <input type="text" id="questionInput" placeholder="Enter your question" class="form-control">
+            <button onclick="askQuestion()" class="btn btn-primary mt-2">Ask</button>
+            <div id="result" class="mt-2"></div>
+        </div>
+
+        <!-- Summary Section -->
+        <div class="form-section">
+            <h3>Summary</h3>
+            <button onclick="getSummary()" class="btn btn-secondary">Get Summary</button>
+            <div id="summary" class="mt-2"></div>
+        </div>
+    </div>
+
+    <!-- Bootstrap JS Bundle -->
+    <script src="https://cdn.jsdelivr.net/npm/bootstrap@5.3.3/dist/js/bootstrap.bundle.min.js"></script>
+    <script>
+        let fileId = null;
+
+        async function uploadFile() {
+            const fileInput = document.getElementById('fileInput');
+            const uploadStatus = document.getElementById('uploadStatus');
+            const resultDiv = document.getElementById('result');
+            const summaryDiv = document.getElementById('summary');
+
+            // Reset old states
+            resultDiv.innerText = '';
+            summaryDiv.innerText = '';
+            uploadStatus.className = "mt-2 text-info";
+
+            if (!fileInput.files[0]) {
+                uploadStatus.innerText = 'Please select a file.';
+                uploadStatus.className = "mt-2 text-danger";
+                return;
+            }
+
+            const formData = new FormData();
+            formData.append('file', fileInput.files[0]);
+
+            try {
+                uploadStatus.innerText = 'Uploading...';
+                const response = await fetch('http://localhost:8000/upload', {
+                    method: 'POST',
+                    body: formData
+                });
+                const data = await response.json();
+
+                if (response.ok) {
+                    fileId = data.file_id;
+                    uploadStatus.innerText = `✅ File uploaded: ${data.file_name} (ID: ${fileId})`;
+                    uploadStatus.className = "mt-2 text-success";
+                } else {
+                    uploadStatus.innerText = `❌ Error: ${data.detail}`;
+                    uploadStatus.className = "mt-2 text-danger";
+                }
+            } catch (error) {
+                uploadStatus.innerText = '❌ Upload failed due to server/network error.';
+                uploadStatus.className = "mt-2 text-danger";
+                console.error(error);
+            }
+        }
+
+        async function askQuestion() {
+            const questionInput = document.getElementById('questionInput');
+            const resultDiv = document.getElementById('result');
+            const question = questionInput.value.trim();
+
+            if (!fileId) {
+                resultDiv.innerText = '⚠️ Please upload a file first.';
+                return;
+            }
+            if (!question) {
+                resultDiv.innerText = '⚠️ Please enter a question.';
+                return;
+            }
+
+            try {
+                resultDiv.innerText = 'Fetching answer...';
+                const response = await fetch('http://localhost:8000/query', {
+                    method: 'POST',
+                    headers: { 'Content-Type': 'application/json' },
+                    body: JSON.stringify({ question, file_id: fileId })
+                });
+                const data = await response.json();
+                resultDiv.innerText = response.ok ? data.answer : `❌ Error: ${data.detail}`;
+            } catch (error) {
+                resultDiv.innerText = '❌ Error fetching answer.';
+                console.error(error);
+            }
+        }
+
+        async function getSummary() {
+            const summaryDiv = document.getElementById('summary');
+
+            if (!fileId) {
+                summaryDiv.innerText = '⚠️ Please upload a file first.';
+                return;
+            }
+
+            try {
+                summaryDiv.innerText = 'Fetching summary...';
+                const response = await fetch('http://localhost:8000/summarize', {
+                    method: 'POST',
+                    headers: { 'Content-Type': 'application/json' },
+                    body: JSON.stringify({ file_id: fileId })
+                });
+                const data = await response.json();
+                summaryDiv.innerText = response.ok ? data.summary : `❌ Error: ${data.detail}`;
+            } catch (error) {
+                summaryDiv.innerText = '❌ Error fetching summary.';
+                console.error(error);
+            }
+        }
+    </script>
+</body>
+</html>