Spaces:

Omartificial-Intelligence-Space
/

context-caching-gemini-pdf-qa

Sleeping

App Files Files Community

Omartificial-Intelligence-Space commited on Jul 12

Commit

5f30414

verified ·

1 Parent(s): 3228f67

Create app.py

Browse files

Files changed (1) hide show

app.py +795 -0

app.py ADDED Viewed

	@@ -0,0 +1,795 @@

+from flask import Flask, request, jsonify, render_template_string
+from flask_cors import CORS
+from google import genai
+from google.genai import types
+import os
+import io
+import httpx
+import uuid
+from datetime import datetime, timezone, timedelta
+from dotenv import load_dotenv
+import json
+# Load environment variables
+load_dotenv()
+app = Flask(__name__)
+CORS(app)
+# Initialize Gemini client
+client = genai.Client(api_key=os.getenv('GOOGLE_API_KEY'))
+# In-memory storage for demo (in production, use a database)
+document_caches = {}
+user_sessions = {}
+# HTML template for the web interface
+HTML_TEMPLATE = """
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>Smart Document Analysis Platform</title>
+    <style>
+        * {
+            margin: 0;
+            padding: 0;
+            box-sizing: border-box;
+        }
+        body {
+            font-family: 'Inter', -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, sans-serif;
+            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+            min-height: 100vh;
+            color: #333;
+        }
+        .container {
+            max-width: 1400px;
+            margin: 0 auto;
+            padding: 20px;
+            min-height: 100vh;
+        }
+        .header {
+            text-align: center;
+            margin-bottom: 30px;
+            color: white;
+        }
+        .header h1 {
+            font-size: 2.8em;
+            font-weight: 700;
+            margin-bottom: 10px;
+            text-shadow: 0 2px 4px rgba(0,0,0,0.3);
+        }
+        .header p {
+            font-size: 1.2em;
+            opacity: 0.9;
+            font-weight: 300;
+        }
+        .main-content {
+            display: grid;
+            grid-template-columns: 1fr 1fr;
+            gap: 30px;
+            height: calc(100vh - 200px);
+        }
+        .left-panel {
+            background: white;
+            border-radius: 20px;
+            padding: 30px;
+            box-shadow: 0 20px 40px rgba(0,0,0,0.1);
+            overflow-y: auto;
+        }
+        .right-panel {
+            background: white;
+            border-radius: 20px;
+            padding: 30px;
+            box-shadow: 0 20px 40px rgba(0,0,0,0.1);
+            display: flex;
+            flex-direction: column;
+        }
+        .panel-title {
+            font-size: 1.5em;
+            font-weight: 600;
+            margin-bottom: 20px;
+            color: #2d3748;
+            display: flex;
+            align-items: center;
+            gap: 10px;
+        }
+        .upload-section {
+            margin-bottom: 30px;
+        }
+        .upload-area {
+            border: 2px dashed #667eea;
+            border-radius: 15px;
+            padding: 40px;
+            text-align: center;
+            background: #f8fafc;
+            transition: all 0.3s ease;
+            margin-bottom: 20px;
+        }
+        .upload-area:hover {
+            border-color: #764ba2;
+            background: #f0f2ff;
+            transform: translateY(-2px);
+        }
+        .upload-area.dragover {
+            border-color: #764ba2;
+            background: #e8f0ff;
+            transform: scale(1.02);
+        }
+        .upload-icon {
+            font-size: 3em;
+            color: #667eea;
+            margin-bottom: 15px;
+        }
+        .file-input {
+            display: none;
+        }
+        .upload-btn {
+            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+            color: white;
+            border: none;
+            padding: 15px 30px;
+            border-radius: 25px;
+            cursor: pointer;
+            font-size: 1.1em;
+            font-weight: 500;
+            transition: all 0.3s ease;
+            margin: 10px;
+        }
+        .upload-btn:hover {
+            transform: translateY(-2px);
+            box-shadow: 0 10px 20px rgba(102, 126, 234, 0.3);
+        }
+        .url-input {
+            width: 100%;
+            padding: 15px;
+            border: 2px solid #e2e8f0;
+            border-radius: 10px;
+            font-size: 1em;
+            margin-bottom: 15px;
+            transition: border-color 0.3s ease;
+        }
+        .url-input:focus {
+            outline: none;
+            border-color: #667eea;
+        }
+        .btn {
+            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+            color: white;
+            border: none;
+            padding: 12px 25px;
+            border-radius: 20px;
+            cursor: pointer;
+            font-size: 1em;
+            font-weight: 500;
+            transition: all 0.3s ease;
+        }
+        .btn:hover {
+            transform: translateY(-1px);
+            box-shadow: 0 5px 15px rgba(102, 126, 234, 0.3);
+        }
+        .btn:disabled {
+            opacity: 0.6;
+            cursor: not-allowed;
+            transform: none;
+        }
+        .chat-container {
+            flex: 1;
+            border: 1px solid #e2e8f0;
+            border-radius: 15px;
+            overflow-y: auto;
+            padding: 20px;
+            background: #f8fafc;
+            margin-bottom: 20px;
+        }
+        .message {
+            margin-bottom: 15px;
+            padding: 15px;
+            border-radius: 12px;
+            max-width: 85%;
+            animation: fadeIn 0.3s ease;
+        }
+        @keyframes fadeIn {
+            from { opacity: 0; transform: translateY(10px); }
+            to { opacity: 1; transform: translateY(0); }
+        }
+        .user-message {
+            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+            color: white;
+            margin-left: auto;
+            box-shadow: 0 4px 12px rgba(102, 126, 234, 0.3);
+        }
+        .ai-message {
+            background: white;
+            color: #333;
+            border: 1px solid #e2e8f0;
+            box-shadow: 0 2px 8px rgba(0,0,0,0.1);
+        }
+        .input-group {
+            display: flex;
+            gap: 10px;
+        }
+        .question-input {
+            flex: 1;
+            padding: 15px;
+            border: 2px solid #e2e8f0;
+            border-radius: 12px;
+            font-size: 1em;
+            transition: border-color 0.3s ease;
+        }
+        .question-input:focus {
+            outline: none;
+            border-color: #667eea;
+        }
+        .cache-info {
+            background: linear-gradient(135deg, #48bb78 0%, #38a169 100%);
+            border-radius: 12px;
+            padding: 20px;
+            margin-bottom: 20px;
+            color: white;
+            box-shadow: 0 4px 12px rgba(72, 187, 120, 0.3);
+        }
+        .cache-info h3 {
+            margin-bottom: 10px;
+            font-weight: 600;
+        }
+        .loading {
+            text-align: center;
+            padding: 40px;
+            color: #666;
+        }
+        .loading-spinner {
+            border: 3px solid #f3f3f3;
+            border-top: 3px solid #667eea;
+            border-radius: 50%;
+            width: 40px;
+            height: 40px;
+            animation: spin 1s linear infinite;
+            margin: 0 auto 20px;
+        }
+        @keyframes spin {
+            0% { transform: rotate(0deg); }
+            100% { transform: rotate(360deg); }
+        }
+        .error {
+            background: linear-gradient(135deg, #f56565 0%, #e53e3e 100%);
+            border-radius: 12px;
+            padding: 15px;
+            color: white;
+            margin-bottom: 20px;
+            box-shadow: 0 4px 12px rgba(245, 101, 101, 0.3);
+        }
+        .success {
+            background: linear-gradient(135deg, #48bb78 0%, #38a169 100%);
+            border-radius: 12px;
+            padding: 15px;
+            color: white;
+            margin-bottom: 20px;
+            box-shadow: 0 4px 12px rgba(72, 187, 120, 0.3);
+        }
+        @media (max-width: 768px) {
+            .main-content {
+                grid-template-columns: 1fr;
+                gap: 20px;
+            }
+            .header h1 {
+                font-size: 2em;
+            }
+        }
+    </style>
+</head>
+<body>
+    <div class="container">
+        <div class="header">
+            <h1>📚 Smart Document Analysis Platform</h1>
+            <p>Upload PDF documents once, ask questions forever with Gemini API caching</p>
+        </div>
+        <div class="main-content">
+            <!-- Left Panel - Upload Section -->
+            <div class="left-panel">
+                <div class="panel-title">
+                    📤 Upload PDF Document
+                </div>
+                <div class="upload-section">
+                    <div class="upload-area" id="uploadArea">
+                        <div class="upload-icon">📄</div>
+                        <p>Drag and drop your PDF file here, or click to select</p>
+                        <input type="file" id="fileInput" class="file-input" accept=".pdf">
+                        <button class="upload-btn" onclick="document.getElementById('fileInput').click()">
+                            Choose PDF File
+                        </button>
+                    </div>
+                    <div style="margin-top: 20px;">
+                        <h3>Or provide a URL:</h3>
+                        <input type="url" id="urlInput" class="url-input" placeholder="https://example.com/document.pdf">
+                        <button class="btn" onclick="uploadFromUrl()">Upload from URL</button>
+                    </div>
+                </div>
+                <div id="loading" class="loading" style="display: none;">
+                    <div class="loading-spinner"></div>
+                    <p id="loadingText">Processing your PDF... This may take a moment.</p>
+                </div>
+                <div id="error" class="error" style="display: none;"></div>
+                <div id="success" class="success" style="display: none;"></div>
+            </div>
+            <!-- Right Panel - Chat Section -->
+            <div class="right-panel">
+                <div class="panel-title">
+                    💬 Ask Questions
+                </div>
+                <div id="cacheInfo" class="cache-info" style="display: none;">
+                    <h3>✅ Document Cached Successfully!</h3>
+                    <p>Your PDF has been cached using Gemini API. You can now ask multiple questions without re-uploading.</p>
+                    <p><strong>Cache ID:</strong> <span id="cacheId"></span></p>
+                    <p><strong>Tokens Cached:</strong> <span id="tokenCount"></span></p>
+                </div>
+                <div class="chat-container" id="chatContainer">
+                    <div class="message ai-message">
+                        👋 Hello! I'm ready to analyze your PDF documents. Upload a document to get started!
+                    </div>
+                </div>
+                <div class="input-group">
+                    <input type="text" id="questionInput" class="question-input" placeholder="Ask a question about your document...">
+                    <button class="btn" onclick="askQuestion()" id="askBtn">Ask</button>
+                </div>
+            </div>
+        </div>
+    </div>
+    <script>
+        let currentCacheId = null;
+        // File upload handling
+        const uploadArea = document.getElementById('uploadArea');
+        const fileInput = document.getElementById('fileInput');
+        uploadArea.addEventListener('dragover', (e) => {
+            e.preventDefault();
+            uploadArea.classList.add('dragover');
+        });
+        uploadArea.addEventListener('dragleave', () => {
+            uploadArea.classList.remove('dragover');
+        });
+        uploadArea.addEventListener('drop', (e) => {
+            e.preventDefault();
+            uploadArea.classList.remove('dragover');
+            const files = e.dataTransfer.files;
+            if (files.length > 0) {
+                uploadFile(files[0]);
+            }
+        });
+        fileInput.addEventListener('change', (e) => {
+            if (e.target.files.length > 0) {
+                uploadFile(e.target.files[0]);
+            }
+        });
+        async function uploadFile(file) {
+            if (!file.type.includes('pdf')) {
+                showError('Please select a PDF file.');
+                return;
+            }
+            showLoading('Uploading PDF...');
+            const formData = new FormData();
+            formData.append('file', file);
+            try {
+                const response = await fetch('/upload', {
+                    method: 'POST',
+                    body: formData
+                });
+                const result = await response.json();
+                if (result.success) {
+                    currentCacheId = result.cache_id;
+                    document.getElementById('cacheId').textContent = result.cache_id;
+                    document.getElementById('tokenCount').textContent = result.token_count;
+                    document.getElementById('cacheInfo').style.display = 'block';
+                    showSuccess('PDF uploaded and cached successfully!');
+                    // Add initial message
+                    addMessage('I\'ve analyzed your PDF document. What would you like to know about it?', 'ai');
+                } else {
+                    showError(result.error);
+                }
+            } catch (error) {
+                showError('Error uploading file: ' + error.message);
+            } finally {
+                hideLoading();
+            }
+        }
+        async function uploadFromUrl() {
+            const url = document.getElementById('urlInput').value;
+            if (!url) {
+                showError('Please enter a valid URL.');
+                return;
+            }
+            showLoading('Uploading PDF from URL...');
+            try {
+                const response = await fetch('/upload-url', {
+                    method: 'POST',
+                    headers: {
+                        'Content-Type': 'application/json'
+                    },
+                    body: JSON.stringify({ url: url })
+                });
+                const result = await response.json();
+                if (result.success) {
+                    currentCacheId = result.cache_id;
+                    document.getElementById('cacheId').textContent = result.cache_id;
+                    document.getElementById('tokenCount').textContent = result.token_count;
+                    document.getElementById('cacheInfo').style.display = 'block';
+                    showSuccess('PDF uploaded and cached successfully!');
+                    // Add initial message
+                    addMessage('I\'ve analyzed your PDF document. What would you like to know about it?', 'ai');
+                } else {
+                    showError(result.error);
+                }
+            } catch (error) {
+                showError('Error uploading from URL: ' + error.message);
+            } finally {
+                hideLoading();
+            }
+        }
+        async function askQuestion() {
+            const question = document.getElementById('questionInput').value;
+            if (!question.trim()) return;
+            if (!currentCacheId) {
+                showError('Please upload a PDF document first.');
+                return;
+            }
+            // Add user message to chat
+            addMessage(question, 'user');
+            document.getElementById('questionInput').value = '';
+            // Show loading state
+            const askBtn = document.getElementById('askBtn');
+            const originalText = askBtn.textContent;
+            askBtn.textContent = 'Generating...';
+            askBtn.disabled = true;
+            try {
+                const response = await fetch('/ask', {
+                    method: 'POST',
+                    headers: {
+                        'Content-Type': 'application/json'
+                    },
+                    body: JSON.stringify({
+                        question: question,
+                        cache_id: currentCacheId
+                    })
+                });
+                const result = await response.json();
+                if (result.success) {
+                    addMessage(result.answer, 'ai');
+                } else {
+                    addMessage('Error: ' + result.error, 'ai');
+                }
+            } catch (error) {
+                addMessage('Error: ' + error.message, 'ai');
+            } finally {
+                askBtn.textContent = originalText;
+                askBtn.disabled = false;
+            }
+        }
+        function addMessage(text, sender) {
+            const chatContainer = document.getElementById('chatContainer');
+            const messageDiv = document.createElement('div');
+            messageDiv.className = `message ${sender}-message`;
+            messageDiv.textContent = text;
+            chatContainer.appendChild(messageDiv);
+            chatContainer.scrollTop = chatContainer.scrollHeight;
+        }
+        function showLoading(text = 'Processing...') {
+            document.getElementById('loadingText').textContent = text;
+            document.getElementById('loading').style.display = 'block';
+        }
+        function hideLoading() {
+            document.getElementById('loading').style.display = 'none';
+        }
+        function showError(message) {
+            const errorDiv = document.getElementById('error');
+            errorDiv.textContent = message;
+            errorDiv.style.display = 'block';
+            setTimeout(() => {
+                errorDiv.style.display = 'none';
+            }, 5000);
+        }
+        function showSuccess(message) {
+            const successDiv = document.getElementById('success');
+            successDiv.textContent = message;
+            successDiv.style.display = 'block';
+            setTimeout(() => {
+                successDiv.style.display = 'none';
+            }, 5000);
+        }
+        // Enter key to ask question
+        document.getElementById('questionInput').addEventListener('keypress', (e) => {
+            if (e.key === 'Enter') {
+                askQuestion();
+            }
+        });
+    </script>
+</body>
+</html>
+"""
+@app.route('/')
+def index():
+    return render_template_string(HTML_TEMPLATE)
+@app.route('/upload', methods=['POST'])
+def upload_file():
+    try:
+        if 'file' not in request.files:
+            return jsonify({'success': False, 'error': 'No file provided'})
+        file = request.files['file']
+        if file.filename == '':
+            return jsonify({'success': False, 'error': 'No file selected'})
+        # Read file content
+        file_content = file.read()
+        file_io = io.BytesIO(file_content)
+        # Upload to Gemini File API
+        document = client.files.upload(
+            file=file_io,
+            config=dict(mime_type='application/pdf')
+        )
+        # Create cache with system instruction
+        try:
+            system_instruction = "You are an expert document analyzer. Provide detailed, accurate answers based on the uploaded document content. Always be helpful and thorough in your responses."
+            # Use the correct model format as per documentation
+            model = 'models/gemini-2.0-flash-001'
+            cache = client.caches.create(
+                model=model,
+                config=types.CreateCachedContentConfig(
+                    display_name='pdf document cache',
+                    system_instruction=system_instruction,
+                    contents=[document],
+                    ttl="3600s",  # 1 hour TTL
+                )
+            )
+            # Store cache info
+            cache_id = str(uuid.uuid4())
+            document_caches[cache_id] = {
+                'cache_name': cache.name,
+                'document_name': file.filename,
+                'created_at': datetime.now().isoformat()
+            }
+            return jsonify({
+                'success': True,
+                'cache_id': cache_id,
+                'token_count': getattr(cache.usage_metadata, 'cached_token_count', 'Unknown')
+            })
+        except Exception as cache_error:
+            # If caching fails due to small content, provide alternative approach
+            if "Cached content is too small" in str(cache_error):
+                return jsonify({
+                    'success': False,
+                    'error': 'PDF is too small for caching. Please upload a larger document (minimum 4,096 tokens required).',
+                    'suggestion': 'Try uploading a longer document or combine multiple documents.'
+                })
+            else:
+                raise cache_error
+    except Exception as e:
+        return jsonify({'success': False, 'error': str(e)})
+@app.route('/upload-url', methods=['POST'])
+def upload_from_url():
+    try:
+        data = request.get_json()
+        url = data.get('url')
+        if not url:
+            return jsonify({'success': False, 'error': 'No URL provided'})
+        # Download file from URL
+        response = httpx.get(url)
+        response.raise_for_status()
+        file_io = io.BytesIO(response.content)
+        # Upload to Gemini File API
+        document = client.files.upload(
+            file=file_io,
+            config=dict(mime_type='application/pdf')
+        )
+        # Create cache with system instruction
+        try:
+            system_instruction = "You are an expert document analyzer. Provide detailed, accurate answers based on the uploaded document content. Always be helpful and thorough in your responses."
+            # Use the correct model format as per documentation
+            model = 'models/gemini-2.0-flash-001'
+            cache = client.caches.create(
+                model=model,
+                config=types.CreateCachedContentConfig(
+                    display_name='pdf document cache',
+                    system_instruction=system_instruction,
+                    contents=[document],
+                    ttl="3600s",  # 1 hour TTL
+                )
+            )
+            # Store cache info
+            cache_id = str(uuid.uuid4())
+            document_caches[cache_id] = {
+                'cache_name': cache.name,
+                'document_name': url,
+                'created_at': datetime.now().isoformat()
+            }
+            return jsonify({
+                'success': True,
+                'cache_id': cache_id,
+                'token_count': getattr(cache.usage_metadata, 'cached_token_count', 'Unknown')
+            })
+        except Exception as cache_error:
+            # If caching fails due to small content, provide alternative approach
+            if "Cached content is too small" in str(cache_error):
+                return jsonify({
+                    'success': False,
+                    'error': 'PDF is too small for caching. Please upload a larger document (minimum 4,096 tokens required).',
+                    'suggestion': 'Try uploading a longer document or combine multiple documents.'
+                })
+            else:
+                raise cache_error
+    except Exception as e:
+        return jsonify({'success': False, 'error': str(e)})
+@app.route('/ask', methods=['POST'])
+def ask_question():
+    try:
+        data = request.get_json()
+        question = data.get('question')
+        cache_id = data.get('cache_id')
+        if not question or not cache_id:
+            return jsonify({'success': False, 'error': 'Missing question or cache_id'})
+        if cache_id not in document_caches:
+            return jsonify({'success': False, 'error': 'Cache not found'})
+        cache_info = document_caches[cache_id]
+        # Generate response using cached content with correct model format
+        response = client.models.generate_content(
+            model='models/gemini-2.0-flash-001',
+            contents=question,
+            config=types.GenerateContentConfig(
+                cached_content=cache_info['cache_name']
+            )
+        )
+        return jsonify({
+            'success': True,
+            'answer': response.text
+        })
+    except Exception as e:
+        return jsonify({'success': False, 'error': str(e)})
+@app.route('/caches', methods=['GET'])
+def list_caches():
+    try:
+        caches = []
+        for cache_id, cache_info in document_caches.items():
+            caches.append({
+                'cache_id': cache_id,
+                'document_name': cache_info['document_name'],
+                'created_at': cache_info['created_at']
+            })
+        return jsonify({'success': True, 'caches': caches})
+    except Exception as e:
+        return jsonify({'success': False, 'error': str(e)})
+@app.route('/cache/<cache_id>', methods=['DELETE'])
+def delete_cache(cache_id):
+    try:
+        if cache_id not in document_caches:
+            return jsonify({'success': False, 'error': 'Cache not found'})
+        cache_info = document_caches[cache_id]
+        # Delete from Gemini API
+        client.caches.delete(cache_info['cache_name'])
+        # Remove from local storage
+        del document_caches[cache_id]
+        return jsonify({'success': True, 'message': 'Cache deleted successfully'})
+    except Exception as e:
+        return jsonify({'success': False, 'error': str(e)})
+if __name__ == '__main__':
+    import os
+    port = int(os.environ.get("PORT", 7860))
+    app.run(debug=True, host='0.0.0.0', port=port)