# Browser Extension Guide

The DocuElevate Browser Extension enables users to clip web pages and send files from their web browser directly to DocuElevate for processing.

## Overview

The browser extension provides a seamless way to process files and capture web content without manually downloading or copying them first. Users can send file URLs or clip entire web pages with a single click, and DocuElevate will process them automatically.

## Features

### Core Functionality

- **Web Page Clipping**: Capture full pages or selected content as PDF documents
- **One-Click File Sending**: Send file URLs from the browser to DocuElevate
- **Dual Mode Interface**: Toggle between "Send URL" and "Clip Page" modes
- **Context Menu Integration**: Right-click on links, pages, or selections for quick actions
- **PDF Conversion**: Automatic conversion of clipped pages to PDF format
- **Popup Interface**: Simple configuration and file/page submission UI
- **Status Notifications**: Immediate feedback on submission success or failure

### Security Features

- **Minimal Permissions**: Only requests necessary browser permissions
- **Secure Storage**: Configuration stored locally in browser extension storage
- **Direct Communication**: All requests go directly to your DocuElevate server
- **Session-Based Auth**: Supports DocuElevate authentication via session cookies
- **Local PDF Generation**: Pages converted to PDF in your browser before upload

### Cross-Browser Support

The extension is compatible with:
- Google Chrome
- Microsoft Edge
- Chromium-based browsers (Brave, Opera, etc.)
- Mozilla Firefox (full support including PDF conversion)

## Installation

### For End Users

See the [Browser Extension README](../browser-extension/README.md) for detailed installation instructions.

Quick steps:
1. Load the extension from the `browser-extension` folder
2. Configure your DocuElevate server URL
3. Optionally add authentication (session cookie)
4. Start sending files or clipping pages!

### For Administrators

#### Prerequisites

- DocuElevate server running with URL upload API enabled
- File upload API accessible at `/api/files/upload`
- Server accessible from users' browsers (not blocked by firewall/CORS)
- Optional: Authentication configured if required

#### Deployment Options

**Option 1: Direct Distribution**
- Share the `browser-extension` folder with users
- Users load it as an unpacked extension

**Option 2: Internal Extension Store**
- Package the extension as a .zip file
- Distribute via internal Chrome Web Store or Firefox Add-ons for Enterprise

**Option 3: Public Store** (requires additional steps)
- Submit to Chrome Web Store
- Submit to Firefox Add-ons

## Configuration

### Extension Settings

Users need to configure two settings:

1. **DocuElevate Server URL** (required)
   - Format: `https://docuelevate.example.com`
   - Must be accessible from user's browser
   - Should not include trailing slashes or API paths

2. **Session Cookie** (optional, if auth enabled)
   - Format: `session=cookie_value_here`
   - Obtained by logging into DocuElevate and copying session cookie
   - Expires when the session expires (users need to update it)

### Server Configuration

No server-side configuration is required. The extension uses existing API endpoints:

```
POST /api/process-url      # For URL mode
POST /api/files/upload     # For clip mode
```

Ensure these endpoints are:
- Accessible from users' browsers
- Not blocked by CORS policies (if different domain)
- Properly secured with authentication if needed

## Usage

### Mode Selection

The extension has two modes accessible via the popup:

1. **Send URL Mode** (default): Send file URLs to DocuElevate
2. **Clip Page Mode**: Capture and convert web pages to PDF

Toggle between modes by clicking the mode buttons in the popup.

### Sending Files via Popup (URL Mode)

1. Click the DocuElevate extension icon
2. Select "Send URL" mode
3. The current page URL is displayed
4. Optionally enter a custom filename
5. Click "Send to DocuElevate"
6. Status message shows success or error

### Clipping Pages via Popup (Clip Mode)

1. Click the DocuElevate extension icon
2. Select "Clip Page" mode
3. The current page title is displayed
4. Choose one of:
   - **Clip Full Page**: Captures entire page content
   - **Clip Selection**: Captures only selected text (select first)
5. Optionally enter a custom filename
6. Page is converted to PDF and uploaded
7. Status message shows success or error

### Sending URLs via Context Menu

1. Right-click on any link or the current page
2. Select "Send URL to DocuElevate"
3. A notification appears with the result

### Clipping via Context Menu

1. **Full Page**: Right-click on any page and select "Clip Full Page to DocuElevate"
2. **Selection**: Select text, right-click, and select "Clip Selection to DocuElevate"
3. A notification appears with the result

### Supported Content

**URL Mode** - DocuElevate will process these file types:

**Document URLs**:
- PDFs: `https://example.com/document.pdf`
- Office: `https://example.com/report.docx`
- Spreadsheets: `https://example.com/data.xlsx`
- Text files: `https://example.com/notes.txt`

**Image URLs**:
- `https://example.com/image.jpg`
- `https://example.com/scan.png`
- `https://example.com/diagram.svg`

**Clip Mode** - Any web page can be clipped:
- Articles, blogs, documentation
- Forms, receipts, confirmations
- Social media posts, comments
- Any HTML content with styling

## How It Works

### Architecture

```
                                    URL Mode
┌─────────────┐         ┌──────────────────┐         ┌──────────────┐
│   Browser   │         │  Browser Ext.    │         │ DocuElevate  │
│     Tab     │────────▶│  (popup.js)      │────────▶│   Server     │
│             │  URL    │                  │  API    │ /process-url │
└─────────────┘         └──────────────────┘         └──────────────┘

                                   Clip Mode
┌─────────────┐         ┌──────────────────┐         ┌──────────────┐
│   Browser   │ Capture │  Browser Ext.    │ Convert │    Browser   │
│     Tab     │────────▶│  (content.js)    │────────▶│ printToPDF() │
│   (HTML)    │         │                  │         │              │
└─────────────┘         └──────────────────┘         └──────────────┘
                               │                            │
                               │                            │ PDF
                               ▼                            ▼
                        ┌──────────────────┐         ┌──────────────┐
                        │ background.js    │────────▶│ DocuElevate  │
                        │                  │  Upload │   Server     │
                        └──────────────────┘         │ /files/upload│
                               │                     └──────────────┘
                               │ Stores config
                               ▼
                        ┌──────────────────┐
                        │ Browser Storage  │
                        │ (chrome.storage) │
                        └──────────────────┘
```

### Data Flow - URL Mode

1. **User initiates send**: Via popup or context menu
2. **Extension gets current URL**: From active tab
3. **Extension loads config**: Server URL and session cookie from storage
4. **API request sent**: POST to `/api/process-url` with URL and optional filename
5. **DocuElevate processes**:
   - Downloads file from URL
   - Validates file type and size
   - Enqueues for processing (OCR, metadata extraction)
6. **Response returned**: Task ID and status
7. **User notified**: Success or error message displayed

### Data Flow - Clip Mode

1. **User initiates clip**: Via popup or context menu
2. **Extension captures page**: 
   - Content script extracts HTML with styles
   - For selection: captures only selected range
   - For full page: captures entire document body
3. **HTML to PDF conversion**:
   - Background script creates temporary tab with HTML
   - Browser's printToPDF API converts to PDF
   - Temporary tab is closed
4. **PDF upload**:
   - Extension loads config from storage
   - FormData created with PDF blob
   - POST to `/api/files/upload` with authentication
5. **DocuElevate processes**:
   - Receives PDF file
   - Validates and stores
   - Enqueues for OCR and metadata extraction
6. **Response returned**: Task ID and status
7. **User notified**: Success or error notification

### Security Flow

The extension implements several security measures:

1. **No direct file access**: Extension only sends URLs or generated PDFs
2. **Local PDF generation**: Pages converted to PDF in user's browser, not server-side
3. **User-controlled config**: Server URL and auth stored per-user
4. **HTTPS recommended**: Encourages secure communication
5. **Minimal permissions**: Only requests necessary browser APIs
6. **Server-side validation**: DocuElevate validates all uploads
7. **Content isolation**: Captured HTML processed in isolated context

## Technical Details

### Extension Structure

**manifest.json**: Extension metadata and configuration
- Manifest v3 format (latest standard)
- Minimal permissions requested
- Compatible with Chrome, Edge, and Firefox
- Version 1.1.0 with web clipping support

**popup/**: User interface files
- `popup.html`: Extension popup interface with mode toggle
- `popup.css`: Styling with modern UI design
- `popup.js`: Configuration and file sending logic

**scripts/**: Background functionality
- `background.js`: Service worker for context menu and notifications
- `content.js`: Content script for page interaction (minimal)

**icons/**: Extension icons in multiple sizes

### API Integration

The extension communicates with DocuElevate via two API endpoints:

**URL Mode - Request Format**:
```javascript
POST /api/process-url
Content-Type: application/json
Cookie: session=<session_value>  // if auth enabled

{
  "url": "https://example.com/file.pdf",
  "filename": "optional-custom-name.pdf"  // optional
}
```

**URL Mode - Response Format**:
```javascript
{
  "task_id": "abc-123-def",
  "status": "queued",
  "message": "File downloaded from URL and queued for processing",
  "filename": "file.pdf",
  "size": 1024
}
```

**Clip Mode - Request Format**:
```javascript
POST /api/files/upload
Content-Type: multipart/form-data
Cookie: session=<session_value>  // if auth enabled

FormData:
  file: <PDF Blob> (page-title.pdf)
```

**Clip Mode - Response Format**:
```javascript
{
  "task_id": "def-456-ghi",
  "status": "processing",
  "message": "File uploaded and queued for processing",
  "filename": "page-title.pdf"
}
```

**Error Response** (both modes):
```javascript
{
  "detail": "Error message explaining what went wrong"
}
```

### Browser Permissions

The extension requests these permissions:

- **activeTab**: Get URL and content of current tab
- **storage**: Save configuration (server URL, session cookie)
- **contextMenus**: Add context menu options for sending/clipping
- **notifications**: Show success/error notifications
- **scripting**: Inject content capture code into web pages
- **host_permissions**: Access page content for clipping (restricted to active tab)

All permissions are used only for stated purposes. No data is collected or transmitted to third parties.

## Troubleshooting

### Common Issues

**"Failed to connect to DocuElevate server"**
- Check server URL is correct and accessible
- Verify server is running
- Check firewall/network settings
- Test API directly: `curl https://your-server/api/process-url`

**"Authentication required" (401 error)**
- User needs to add session cookie
- Session may have expired (log in again)
- Check AUTH_ENABLED setting in DocuElevate

**"Unsupported file type" (400 error)**
- URL must point to a supported file type
- Check file extension and Content-Type header
- See DocuElevate supported file types

**"File too large" (413 error)**
- File exceeds MAX_UPLOAD_SIZE setting
- Contact admin to increase limit or use smaller file

**Extension not appearing**
- Ensure developer mode is enabled
- Reload the extension
- Check browser console for errors

### Debug Mode

To debug the extension:

1. **Open Extension Console**:
   - Chrome: Right-click extension icon → "Inspect popup"
   - Or go to `chrome://extensions/` → Click "Inspect views: service worker"

2. **Check Console Logs**:
   - Look for error messages
   - Verify API requests are being sent
   - Check response status codes

3. **Test API Directly**:
   ```bash
   curl -X POST https://your-server/api/process-url \
     -H "Content-Type: application/json" \
     -H "Cookie: session=your_session" \
     -d '{"url": "https://example.com/test.pdf"}'
   ```

## Best Practices

### For Users

- Keep your session cookie secure and don't share it
- Verify the DocuElevate server URL is correct before saving
- Only send files from trusted sources
- Check file size limits before sending large files

### For Administrators

- Configure HTTPS for production servers
- Set appropriate MAX_UPLOAD_SIZE limits
- Enable authentication for security
- Monitor API usage and set rate limits if needed
- Provide clear documentation to users on getting session cookies

## Future Enhancements

Possible future improvements:

- **OAuth2 authentication** instead of session cookies
  - Current session cookie approach has limitations:
    - Session cookies expire and need manual refresh
    - Users must manually copy cookie from browser DevTools
    - No automatic token refresh mechanism
  - OAuth2 would provide:
    - Automatic token refresh
    - Better security with short-lived access tokens
    - Easier user experience (login flow instead of cookie copying)
- File preview before sending
- Batch processing multiple URLs
- Progress indication for large files
- History of sent files
- Custom processing options (OCR language, metadata fields, etc.)

### Session Cookie Security Best Practices

For users of the current implementation:

1. **Keep session cookies secure**: Never share your session cookie value
2. **Refresh regularly**: Session cookies expire; update the extension when you log in again
3. **Use HTTPS**: Always use HTTPS for your DocuElevate server
4. **Clear on logout**: Remove the session cookie from extension when logging out
5. **Private browsing**: Session cookies from private/incognito windows have shorter lifetimes

## Related Documentation

- [DocuElevate API Documentation](./API.md)
- [DocuElevate Configuration Guide](./ConfigurationGuide.md)
- [DocuElevate Security Guide](../SECURITY_AUDIT.md)
- [Browser Extension README](../browser-extension/README.md)

## Support

For issues or questions:
- Check the troubleshooting section above
- Review the [main documentation](./README.md)
- Open an issue on [GitHub](https://github.com/christianlouis/DocuElevate/issues)