npm:pywwt 1.4.0 (2022-12-01)

• Require version 0.12 of the research app, to get the latest improvements
  including the Vue 3 update (#342, @pkgw).

pypa:pywwt 0.16.2 (2022-10-26)

• Update a construct in setupbase.py to fix installation with Python 3.11
  (#340, @dhomeier).
• Update the export of interactive figure data to work with named
  colormaps (#339, @Carifio24).

The DOI of this release is 10.5281/zenodo.7255701.

pypa:pywwt 0.16.1 (2022-10-12)

• Fix loading of HDULists as image layers (#337, @pkgw)

The DOI of this release is 10.5281/zenodo.7187355.

pypa:pywwt 0.16.0 (2022-10-08)

This release improves the support for viewing arbitrarily-sized FITS files in
pywwt, taking advantage of the new support in the Toasty library for all-sky
datasets! Version 0.18 of Toasty, or newer, is now required.

Another cool new change is that pywwt releases are now automatically deposited
with Zenodo and assigned DOIs in the process. This makes it possible to cite
the pywwt software, indicating exactly which version you were using, in
scholarly contexts!

• Handle all-sky FITS files with TOAST rather than hipsgen, using the
  newest Toasty features (#335, @imbasimba).
• Allow toggling whether or not a table layer is selectable (#334, @Carifio24).
• Allow the time_decay attribute of layers to be serialized (#332, @Carifio24).
• Make it possible to customize the app URL used by the Jupter widget (#331,
  @pkgw).
• Implement automated Zenodo deposit with DOI registration (@pkgw)

The DOI of this release is 10.5281/zenodo.7164148.

pypa:pywwt 0.15.2 (2022-01-19)

• Update the bundled version of the research app to 0.9.2, to ensure we have
  support for naming imageset layers (#326, @pkgw)

npm:pywwt 1.3.3 (2022-01-19)

• Require version 0.9.2 of the research app, to ensure we have support for
  naming imageset layers (#326, @pkgw)

pypa:pywwt 0.15.1 (2022-01-17)

• Fix installation on Python 3.10 (#325, @pkgw)
• Fix a small logic error with imageset naming that prevented some imagesets
  from loading in the app (#325, @pkgw)

pypa:pywwt 0.15.0 (2022-01-14)

This release adds support for viewing arbitrarily-sized FITS files in pywwt,
with automatic tiling provided by the Toasty library!

• Extend the existing FITS viewing support to automatically tile FITS images if
  needed, using either Toasty or CDS' hipsgen depending on the angular size of
  the image (#316, @imbasimba). While the Python interfaces don't look any
  different, there is now a lot more machinery under the hood so that pywwt can
  perform well whether your FITS file is a megabyte or a gigabyte in size.
• Use the new WWT Kernel Data Relay (KDR) for serving up kernel-side data
  in Jupyter (#316, @pkgw, @imbasimba). You will now need to install the Python
  package wwt_kernel_data_relay as a Jupyter Server extension for FITS viewing
  to work in pywwt. This extension is necessary for pywwt to be able to send
  FITS data to the WWT web app on-demand, as you navigate an image. The KDR
  extension should be useful for other similar visualization tools as well.
• Lots of other under-the-hood work to get asynchronous tiling and data service
  working smoothly.
• Automatically use better default names and data cuts when viewing FITS files
  (#323, @imbasimba).
• Substantially clean up and (hopefully) improve the documentation (#318, #324,
  @pkgw). This includes a new documentation theme based on the Astropy docs.

npm:pywwt 1.3.2 (2022-01-14)

• Fix up license annotations
• Require the latest research-app, version 0.9

pypa:pywwt 0.14.0 (2021-11-01)

• When pywwt is installed as a Jupyter server extension, have it provide its
  static resource files, most notably the bundled copy of the research app, from
  the additional URL path $jupyterBaseUrl/wwtstatic/.... Coupled with the
  recent releases of the WWT JupyterLab extension to make the app URL
  configurable, this should give us a future-proof way to make the WWT
  JupyterLab experience fully functional even in Jupyter(Lab) installations that
  reject all cross-origin data requests (#315, @pkgw).

pypa:pywwt 0.13.1 (2021-10-14)

• Update the Jupyter(Lab) frontend code to fix a dumb bug that broke the widget
  when the "baseUrl" was non-trivial (#314, @pkgw).

npm:pywwt 1.3.1 (2021-10-14)

• Fix a dumb string-processing bug in the URL canonicalization code that broke
  the widget in Jupyter environments where the "base URL" is nontrivial (#314,
  @pkgw).

pypa:pywwt 0.13.0 (2021-09-24)

This is an important release, updating all variations of the pywwt UI ­— the Qt
widget, the ipywidgets widget, and the JupyterLab integration — to use the WWT
“research app”, instead of the old hand-coded HTML/JS wrapper that was specific
to pywwt. This enables us to take advantage of the app's much more sophisticated
UI and robust development framework, and to provide a homogeneous user interface
across these different variations. (#301, #302, @pkgw)

• This release bundles version 0.7.1 of the research app, which includes support
  for many new features, including catalog HiPS datasets, tiled FITS rendering,
  and more.
• Building in the capabilities of this new framework, the WWT widget has gained
  most_recent_source and selected_sources properties that allow code to
  interact with user selections of sources in the UI. Python code can be
  notified of selection actions by using the set_selection_change_callback()
  method. (#311, @Carifio24)
• Also building on the new framework, the WWT widget now knows about HiPS
  progressive catalogs (#308, @imbasimba, @pkgw). You can query available
  catalogs (via the available_hips_catalog_names) property, add them to the UI
  programmatically with wwt.layers.add_hips_catalog_layer, and obtain the data
  currently visible in the WWT view with the layer.refresh() method.
• To support the above, pywwt has started adding asynchronous processing support
  (#308, @pkgw). When pywwt is used inside a Jupyter kernel, the internals of
  the kernel message queue processing are modified to allow "expedited" message
  handling, which is needed to properly support certain pywwt interactive
  functionalities. This hack is relatively self-contained so hopefully it won't
  be too fragile, but time will have to tell.
• The wwt.load_image_collection() method now has a recursive for nested WTML
  files (#308, @imbasimba).
• The WWT widget constructors have a new hide_all_chrome option that hides all
  of the research app’s “chrome” (UI features), to enable a user experience that
  looks similar to the old pre-app style. This mode is the default in the Qt
  widget, so as to preserve the UX of the Glue app.
• Fix pywwt to be able to parse ImageSet XML data even when the ThumbnailUrl
  attribute is missing (#312, @Carifio24).
• Fixes for Astropy 4.3 (#310, @pkgw).
• The type of the wwt.solar_system.scale setting was corrected.
• The Jupyter server was failing to serve files with unguessable content-types;
  this is now fixed.
• Yet more improvements to the CI, test suite, and API documentation.

Tutorial and how-to documentation for all of this new functionality hasn't been
prepared ... yet.

npm:pywwt 1.3.0 (2021-09-24)

• Bundle the research app in the pywwt distribution; currently requiring version
  0.7.1 (#301, #313, @pkgw). Compared to the previous release of this package,
  this includes a ton of new features: HiPS catalogs with data retrieval, tiled
  FITS, and lots of UI polish!
• Port the ipywidgets frontend to use the research app
• Modify the internal APIs to allow the app to come and go, not just become
  ready once

pypa:pywwt 0.12.0 (2021-06-14)

• A variety of internal cleanups relating to the HTML and JS files that are
  bundled with pywwt. These shouldn't affect anything user-visible, yet,
  but we want to publish the cleaned-up code to validate that everything is
  still working correctly.
• Some improvements to the documentation and test suite.

npm:pywwt 1.2.0 (2021-06-14)

• New release to track internal cleanups in the pywwt Python code; the built-in
  widget HTML is now served from $baseurl/wwt/widget/, instead of
  $baseurl/wwt/wwt.html.
• Internal cleanups.

pypa:pywwt 0.11.3 (2021-06-09)

• Another attempt to get ReadTheDocs to update the "stable" version of the
  documentation. We need to automate updating of the stable Git branch.

pypa:pywwt 0.11.2 (2021-06-09)

• Tidy up and modernize the installation instructions. Our ReadTheDocs defaults
  to the "stable" version of the docs, so we need to make a release to get them
  to show up for most folks.

pypa:pywwt 0.11.1 (2021-06-03)

• Add compatibility with the change in how the engine handles the
  transparentBlack setting for FITS datasets. The code is written to maintain
  compatibility with older versions of the engine as well (@imbasimba).
• Update some test-suite images to account for the changes in the new
  WebGL-based fits rendering (@pkgw).
• Temporarily disable checking of outgoing links in the docs, since
  Matplotlib currently has a problem that causes the check to fail (@pkgw).

pypa:pywwt 0.11.0 (2021-02-09)

• First-draft support for the JupyterLab application. Rather then providing the
  WWT view as a widget tied to a specific notebook cell, the application
  provides it as a freestanding window that can exist separately from any one
  notebook or cell. This app is developed independently as a Vue app, so it's a
  lot easier to add native UI to it, as compared to the HTML widget that we have
  to provide in our plugin here.