Retrieving data from the current widget’s view

So far, we’ve seen how to send information (tables, MOCs, …) into the widget. The other way also works! This notebook contains a list of methods to extract diverse information about the current view. However, here are several information about retrieving data from the widget:

• when the view is modified programmatically (i.e. not by clicking on the buttons), the update of its properties is done between cell execution. This means that you’ll see a WidgetNotReadyError when you try to modify the view and retrieve information about it in the same cell. Simply switch the property-reading to the next cell and everything will work as intended!

• if you generate different views of the same Aladin() instances – either by calling display multiple times, or by using the Create new view for cell output button in Jupyter, the information contained in wcs and fov_xy will always correspond to the most recently created view.

from astropy.coordinates import SkyCoord
import astropy.units as u
from astropy.wcs import WCS
from astroquery.simbad import Simbad
from astroquery.vizier import Vizier
import matplotlib.pyplot as plt
import numpy as np

from ipyaladin import Aladin

aladin = Aladin(fov=5, height=600, target="M31")
aladin

Getting the current WCS

The World Coordinates System (WCS) describes the field of view, the projection, and it’s rotation. It is returned as an astropy.coordinates.WCS object.

aladin.wcs  # Recover the current WCS

If you edit the view either by modifiing the widget through its interface, or programmatically:

aladin.height = 800
aladin.survey = "CDS/P/PLANCK/R2/HFI/color"
aladin.target = "LMC"
aladin.frame = "Galactic"
aladin.fov = 50

The wcs is updated and you can print its new value in the next cell:

aladin.wcs

If you try to recover the value in the same cell, you’ll get a WidgetCommunicationError error. This is because the calculation of the WCS is done by Aladin Lite between cell executions.

Getting the field of view

The field of view is printed in the bottom left corner of the view. You can grab the two values with:

aladin.fov_xy  # Recover the current field of view for the x and y axis

Getting the selection

First, load two catalogs:

m1 = SkyCoord.from_name("m1")

# if these don't work, do `pip install astroquery -U --pre`
simbad = Simbad.query_region(m1, radius=0.1 * u.deg)
usno = Vizier(catalog="I/284/out", row_limit=-1).query_region(m1, radius=0.05 * u.deg)[
    0
]

aladin.add_table(simbad, name="simbad", color="purple", shape="circle", source_size=20)
aladin.add_table(usno, name="usno", color="teal", shape="square", source_size=20)
aladin.target = m1
aladin.fov = 0.3
aladin.survey = "CDS/P/PanSTARRS/DR1/g"

After that, select a region in the Aladin Lite widget (you can also trigger a selection with right click then select sources):

aladin.selection("circle")

Then, you can recover the selection as a list of astropy.Table objects with:

aladin.selected_objects

Getting the view as a fits file

The following method allow you to retrieve the current view as a fits file.

fits = aladin.get_view_as_fits()

fits[0].header

Display the fits file using matplotlib:

wcs = WCS(fits[0].header)

plt.subplot(projection=wcs)
plt.imshow(fits[0].data, cmap="binary_r", norm="asinh", vmin=0.001)

Saving the view as an image file

save_view_as_image takes a screenshot of the widget. The first argument is the path where the file should be written, the second is the format (“png”, “jpeg”, “webp”), and the third is a boolean to indicate if you want to include the Aladin Lite logo in the image. These data are reflecting the pixels on the screen. For scientific use cases, opt for get_view_as_fits that is a bit longer since it retrieves the actual survey from a server.

screenshot_wcs = aladin.wcs
aladin.save_view_as_image(path="./crab.png", image_format="png", with_logo=True)

img = plt.imread("./crab.png")
plt.imshow(img)

And with the WCS that we stored at the same time than the screenshot

img = plt.imread("./crab.png")
plt.subplot(projection=screenshot_wcs)
# in FITS, the lines of pixels are flipped, so we have to mimic this
plt.imshow(np.flipud(img))