7.0.0

7.0.0

6.1.0

This is a minor release which contains a fix for dashboard URL construction logic in the presence of query parameters. Now query parameters reported by a Cluster implementation in the dashboard URL are correctly propagated to individual dashboard panes. This is relevant for cases where things like authentication tokens are included in a URL. Thanks @ntabris for the contribution in #258.

6.0.0

This is a major release that includes:

1. A new system for configuring default dashboard layouts (#247). This is accessible from the File menu as well as the command palette, and allows you to launch a set of dashboard panels for a cluster with a single command. For more information see this explanation.
2. A bugfix that prevented Dask Dashboard charts from rendering properly when using JupyterLab's dark themes (#243)

5.3.0

This minor release for dask-labextension includes a few new features and bugfixes.

1. The extension now uses the new Dask logo and branding. For more discussion, see this issue.
2. There is a new configuration option browserDashboardCheck which can force the extension to check for a Dask dashboard using the frontend browser session (rather than a request made on the server side). This can be useful in cases where a browser cookie is needed to authenticate to the dashboard. Thanks to new contributor @viniciusdc for implementing the feature!
3. There is an error message which suggests the user might have a misconfigured/not-installed serverextension if it is not found. This was a little overeager, resulting in some false positives. In this release it should be a little slower to show the message. (cf #237)

5.2.0

This is release for dask-labextension contains a few bugfixes and minor enhancements:

1. Fixes a bug where polling the dashboard wouldn't back off if the extension loses its connection (#220)
2. Adds an option to hide the cluster manager user interface (#219) for deployments where it does not make sense. Note: this does not remove the underlying REST API, it only hides the frontend UI elements.
3. Fixes a layout restoration bug where, if an active cluster is not found, a dashboard panel is closed. This prevents pre-created layouts from being distributed in the absence of a running cluster (#217).

5.0.2

Contains a fix for an issue in which continued polling of the dask dashboard could result in unbounded memory increases on the scheduler.

5.0.0

This is the first release of dask-labextension that supports JupyterLab 3.0. There are no changes in functionality, but installation should be significantly easier. No more nodejs, no more rebuilding the application. You can install with just pip:

pip install jupyterlab
pip install dask-labextension