Sschuhmann Helium Versions Save

• updated jupyter_client to v5.2.4
• updated jupyter_core to v4.6.1
• updated traitlets to v4.3.3
• updated decorator to v4.4.2
• updated ipython_genutils to v0.2.0

fixes #131

This release adds many quality of life features and fixes some bugs:

• control the size of inline plots (#89, #101, #108, #114, #121)
• transparent background for images (#73, #138)
• move the cursor after executing cell is the default (#137)
• selects the correct cell for execution (#122)
• bugfixes (#124)

Altogether, Helium should feel much smoother to use.

If you have any issues, please report them on github/issues

Hermes was renamed to Helium in this release!

• We have removed remote remote kernels to enable compatibility with Windows and Linux systems. However, this feature might be enabled again after the needed libraries are ported to the other systems

• This release introduces image scaling. Images which are generated as output are now scaled to the size of the view containing them.

• Fallback directory for kernel. A kernel launched from an open text view is started in the directory of the file, if no file is open during execution, the users home directory is used as a fallback option

• Test infrastructure @pykong has introduced proper test infrasrtructure to the plugin, which will help to ensure the quality of later releases

Fixed that text data in display_data were not shown as inline output. This has caused problem such as no execution results shown when we use IRkernel (#46).

This is update with big changes. Please read this release note. I had tackled with the known serious unstability issues, and large modification was required to solve it. Consequently it requires changes about the way to Jupyter kernel. The way to make connection with Jupyter kernels is changed as below.

1. The most basic way, start a kernelspec installed locally, as a subprocess of ST3 (the process stops when Sublime stops)

1. Run Hermes: connect kernel command.
2. Choose New kernel.
3. Choose the kernelspec you want to run.

2. Connect to the kernel already runnning and connected to Hermes

1. Run Hermes: connect kernel command.
2. Choose the kernel you want to connect.

3. Connect to a kernel already running under some other Jupyter app (such as Notebook)

1. Get connection info of the kernel. The way to get connection info differ among kernels, see the doc of each kernel (in ipython kernel, you can get it by %connect_info magic.)
2. Run Hermes: connect kernel command.
3. Choose New kernel.
4. Choose (Enter connection info).
5. Enter the connection info (Hermes accepts a path or connection info itself).

4. Connect to a kernel already running under some other Jupyter app, in a SSH server

1. Configure SSH servers in the setting file (opened by Hermes: Settings command.)
2. Get connection info of the kernel. The way to get connection info differ among kernels, see the doc of each kernel (in ipython kernel, you can get it by %connect_info magic.)
3. Run Hermes: connect kernel command.
4. Choose New kernel.
5. Choose (Connect remote kernel via SSH).
6. Choose the server, then enter the connection info.

I'm afraid that this change might break the workflow of some users, but I think it's more convenient for many people who works with locally running Jupyter. I ask you for your kind understanding about the changes.

New features

Thanks to @SamiPirbay, there are also new cool features, which are inline phantom mode and cell evaluation support.

Inline phantoms

You can get evaluation result as inline phantoms in the place where you executed code like Light Table or Hydrogen. It's enabled by setting the inline_output option True. Each phantom has a small x mark to close it.

Cell evaluation support

Regions surrounded by # %% or # <codecell> (you can configure it in cell_delimiter_pattern option item) are considered as "code cells".

You can execute a region by Hermes: Execute cell or Hermes: Execute Cell and Move command. Each cell has a clickable "Run Cell" phantom that appears next to the cell markers to run the cell.

Update in 0.4.3

Fix to reuse an established WebSocket connection to Jupyter kernel. This should solve some performance issues.

Fixed the bug that the editor hangs up when view is connected to a dead kernel.