FViewer: Quickstart Guide

Welcome to FViewer, a modern, browser-based astronomical FITS file viewer designed to replace traditional desktop tools like fv and ds9.

FViewer provides a fast, hardware-accelerated web UI combined with a powerful bidirectional Python API, allowing you to control the viewer and extract data directly from your Jupyter Notebooks.

Installation & Setup

Ensure you have FViewer installed in your current Python environment:

pip install fviewer

Start the FViewer backend server and UI in your terminal (or it may be auto-started by your JupyterHub environment):

fviewer

This will automatically open a browser window to the application at http://127.0.0.1:8000.

You specify custom host and port by passing them to the command line.

Calling fviewer with a file name will load it automatically.

---

Python API Tutorial

The easiest way to use FViewer is alongside a Jupyter Notebook. This allows you to load files, adjust image parameters, and retrieve drawn regions programmatically.

1. Connect to the Viewer

First, import the client and connect it to your open browser tab.

from fviewer import FViewer

# Initialize the client (defaults to localhost:8000)
viewer = FViewer()

# Block execution until the browser tab is open and connected
viewer.wait_for_ready(timeout=15)

2. Load a FITS File

Use the load_file method to open a FITS file. Note: For security reasons, the file path must be relative to the directory where you started the FViewer server.

# Loads the file into the UI and automatically displays the Primary HDU
viewer.load_file("data/m51_chandra.fits")

3. Adjust Image Settings

You can easily adjust the image scaling and colormap to highlight different features in your astronomical data.

# Check current settings
print("Current Colormap:", viewer.get_colormap())

# Apply a new stretch and colormap
viewer.set_stretch("log")
viewer.set_colormap("plasma") # Options: 'gray', 'heat', 'cool', 'plasma'

4. Programmatic Regions

You can push mathematical regions from your Python code directly to the FViewer canvas.

# Draw a red circle at Pixel X=500, Y=500 with a radius of 50
viewer.add_circle(x=500, y=500, radius=50, color="#ff0000")

# Draw an elliptical background region
viewer.add_ellipse(
    x=450, y=450, rx=100, ry=50, 
    angle=45, color="#00ffff", 
    is_background=True
)

5. Extracting Data Back to Jupyter

If you draw regions manually using your mouse in the FViewer UI, you can instantly pull those coordinates back into Python for analysis.

# Retrieve all drawn regions in World Coordinate System (RA/Dec)
regions = viewer.get_regions(format="fk5")

for reg in regions:
    print(f"Type: {reg['type']}, RA: {reg['ra']:.5f}, Dec: {reg['dec']:.5f}")

# Save the current UI regions to a DS9-compatible .reg file on disk
viewer.save_regions("my_science_regions.reg", format="fk5")

---

User Interface Overview

If you prefer to work manually, the FViewer UI is designed like a compact Desktop IDE:

• Left Sidebar (HDU List): Shows all extensions in your FITS file. Clicking an IMAGE extension opens the image viewer. Clicking a BINTABLE extension opens the massive-scale Virtual Table.
• Image Viewer:
  • Pan/Zoom: Use your mouse wheel to zoom in/out natively. Click and drag to pan around the image.
  • Regions: Use the Regions dropdown to select drawing tools (Circle, Box, Ellipse, Annulus). Click and drag on the image to draw. Select a region and press Delete to remove it.
  • Status Bar: Look at the bottom of the screen to see real-time Pixel coordinates, Flux values, and WCS (RA/Dec) coordinates.
• Virtual Table: Double-click any cell to edit it. The table uses lazy-loading, meaning you can scroll through millions of rows instantly without crashing your browser.
• Right Sidebar (Plotter): Toggle the plotter using the top menu to create interactive 1D Histograms and 2D Scatter plots based on your table data or image regions.