Accept criteria

• package can be installed via awesome-panel-cli
• README readable on pypi page
• pn package commands added.

I would like other packages like awesome-panel, awesome-panel-extensions, panel-chemistry, pn-highcharts etc. to be able to *plug intotheawesome-panel-cli`.

Accept Criteria

• [] Concept: Other packages can plug in and
  • add their own app, notebook and component templates to pn create app, pn create notebook. and pn create component.
  • add thir own commands. For example pn awesome-panel serve.
• [] Tested
• [] Documented
• [] Showcased

When running pn create widget and similar the user should be able to specify a name. For example

pn create widget color_input

Accept Criteria

• [] view
• [] viewer
• [] widget
• [] pane
• [] layout

The current implementation of the create command assumes the user cds into the folder where he/ she wants to create the files.

It can be useful to be able to specify the target via a --cwd or --target flag.

Github provides Releases and Packages.

So far I have not used these but I would like to learn and starting using as it is super useful.

Accept Criteria

• [] How to documented or automated
• [] Used in practice

I would like to add a docker reference template and pn docker commands.

Commands

• [] pn docker build
• [] pn create docker-file
• [] pn docker export
• [] pn docker push
• [] pn docker run
• [] pn docker system-prune
• [] pn docker test

Accept Criteria

• [] Tested
• [] Documented

I would like to make it easy to deploy to azure via a pn deploy azure command.

Accept Criteria

• [] Secure
• [] Tested
• [] Documented

Inspiration

• django-simple-deploy

When you run pn hello apps you should be able to go to https://localhost:5006 and see a nice gallery page.

Currently this does not work due to holoviz/panel#3970. When this gets fixed there will be a simple index page.

I would like though a nice gallery page with more info as found at https://www.awesome-panel.org/gallery.

The template can be found here https://github.com/awesome-panel/awesome-panel-extensions/tree/master/awesome_panel_extensions/frameworks/fast/templates/fast_gallery_template

Accept Criteria

• [] The FastGalleryTemplate carved out into its own project awesome-panel-gallery created and managed using the awesome-panel-cli.
  • [] Should support both to be panel served and to be .saved to a standalone html template.
• [] Added to the reference apps at https://github.com/awesome-panel/awesome-panel-cli/tree/main/src/awesome_panel_cli/templates/apps
• [] served automatically by pn hello apps as an index page at /.

Make a conda package available on conda-forge.

Inspiration might be found in the panel-chemistry repository which is already on conda-forge.

Tasks

• [] Create conda-forge recipe
• [] Publish the package (manually)
• [] Document conda package in README
• [] Document conda package in Docs
• [] Setup automated build and release in via github actions

It would be super powerful to add a --dryrun to all pn commands to just show which shell commands will be run.

Accept Criteria

• [] All commands
• [] Tested
• [] Documented

There is a longer list of project related to awesome-panel. They should all be converted to use the awesome-panel-cli.

• [] awesome-panel
• [] awesome-panel-extensions
• [] awesome-panel-introduction
• [] awesome-panel-lightning
• [] examples
• [] panel-chemistry
• [] panel-highcharts
• [] panel-vegafusion

I would like the pn create project concept to create repositories that support both sphinx documentation and pn convert apps deployed via github pages with ease.

The issue I see is that they would both use the docs folder

• Sphinx uses the docs folder by default
• Github pages only supports the docs folder for deployment to for example awesome-panel.github.com/awesome-panel-cli.

Accept Criteria

• [] Technical solution(s) found
• [] Implemented
• [] Tested
• [] Documented.

When you run pn --help you see something like this

I've maked some commands that do not have documentation. I would like that added.

The commands that don't have documentation are coming from the .add_typer commands marked below.

So the task is to figure out how to add docs to those and add it.

You can find the .add_typer code here.

Accept Criteria

• [] Technical solution found
• [] Docs added. Also, for nested commands.

The cli tool has been developed on Linux in a bash shell.

The cli tool should be tested and improved for other combinations of os and terminal:

• os: linux, mac os, windows
• terminal: bash, cmd.exe, powershell
• dev environment. vs code, pycharm, jupyter, spyder

Test procedures

Usage

1. Install for usage
2. Try the different commands manually including pn create project.

Development

1. Install for development
2. Run pn test all

Please report any issues found as new issues. Thanks.

I would like to make it easy to also support webassembly apps converted via pn convert.

Accept Criteria

• [] All reference apps can be pn converted to webassembly.
  • Might require some refactoring.
• [] pn serve --webassembly serves the apps as web assembly apps
• [] pn create examples includes the pn converted apps and an examples/wapps folder.

The module, class and function docstrings should be reviewed and updated where appropriate.

Accept Criteria

• [] Contains examples where relevant
• [] Useful for users when used from the terminal via pn.
• [] Useful for developers when used in IDE like VS Code.

Add reference apps that can be created via pn create app.

• hello_world
• streaming_indicators
• cross_filtering
• altair_starter
• Any of the DataViz examples below from awesome-panel.org.

Accept Criteria

• [] Works in default and dark mode.
• [] Works with pn convert to webassembly.
• [] Looks and works great
• [] Code is simple to understand and repurpose.

The introduction.ipynb notebook is just a copy of the getting_started.ipynb notebook.

I would like a separate introduction notebook.

Accept Criteria

• [] Illustrates why Panel and data apps is useful
• [] Provides useful links for more information

Description

This does not work

Reproduction steps

1. Go to '...'
2. Click on '....'
3. Scroll down to '....'
4. See error

Code

import panel as pn

Logs

remote: 
remote: Create a pull request for 'feature/template' on GitHub by visiting:
remote:      https://github.com/awesome-panel/awesome-panel-cli/pull/new/feature/template
remote: 
To https://github.com/awesome-panel/awesome-panel-cli.git

Requirements

panel==0.13.1

When you run for example pn create view both a view and a test_view test files should be created in the right folders.

Accept Criteria

• [] view
• [] widget
• [] pane
• [] layout

I would like to have the process for building, testing and deploying the awesome-panel-cli package automated via Github actions to enable more contributors and to professionalize the project.

• [] github actions added.
  • It should use pn commands for every step such that they can be repeated locally with ease.
• [] documented
• [] badges for tests and test coverage etc added to README.

I would like to make it easy to deploy to azure via a pn deploy amazon command.

Accept Criteria

• [] Secure
• [] Tested
• [] Documented

Inspiration

• django-simple-deploy

The purpose is to quickly illustrate what this project is and why its useful.

Accept Criteria

• Added to README
• [] Added to DOCS
• [] Released on youtube
• Works on twitter and linkedin

Some pn commands will report success even when they fail.

• [] pn build package
• [] pn release package

Reproducible Example

Try to upload a package twice to test pypi

$ pn release package 0.1.0 --test
[09:33:31] INFO     running: python -m twine upload dist/*0.1.0* --repository testpypi                                                                                                                  shared.py:33
Uploading distributions to https://test.pypi.org/legacy/
Enter your username: MarcSkovMadsen
WARNING  Error getting password from keyring                                                                                                                                                                        
         Traceback (most recent call last):                                                                                                                                                                         
           File "/home/jovyan/repos/private/awesome-panel/awesome-panel-cli/.venv/lib/python3.9/site-packages/twine/auth.py", line 74, in get_password_from_keyring                                                 
             return cast(str, keyring.get_password(system, username))                                                                                                                                               
           File "/home/jovyan/repos/private/awesome-panel/awesome-panel-cli/.venv/lib/python3.9/site-packages/keyring/core.py", line 55, in get_password                                                            
             return get_keyring().get_password(service_name, username)                                                                                                                                              
           File "/home/jovyan/repos/private/awesome-panel/awesome-panel-cli/.venv/lib/python3.9/site-packages/keyring/backends/fail.py", line 25, in get_password                                                   
             raise NoKeyringError(msg)                                                                                                                                                                              
         keyring.errors.NoKeyringError: No recommended backend was available. Install a recommended 3rd party backend package; or, install the keyrings.alt package if you want to use the non-recommended backends.
         See https://pypi.org/project/keyring for details.                                                                                                                                                          
Enter your password: 
Uploading awesome_panel_cli-0.1.0-py3-none-any.whl
100% ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 37.8/37.8 kB • 00:00 • 50.9 MB/s
WARNING  Error during upload. Retry with the --verbose option for more details.                                                                                                                                     
ERROR    HTTPError: 400 Bad Request from https://test.pypi.org/legacy/                                                                                                                                              
         File already exists. See https://test.pypi.org/help/#file-name-reuse for more information.                                                                                                                 
[09:33:42] ERROR    failed: python -m twine upload dist/*0.1.0* --repository testpypi                                                                                                                   shared.py:37
           INFO     Package build finished with success 

I would like to have docs available on read the docs to help users of this package.

Accept Criteria

• [] Sphinx Docs
• [] Furo theme
• [] Basic docs files written
• [] Docs deployed to read the docs
• [] The relevant sphinx commands added to the awesome-panel-cli
• [] docs folder added to project template
• [] General concept for docs as commands in the awesome-panel-cli, for example pn docs build, pn docs rebuild, pn docs serve.
• [] General concept for testing docs added as pn test docs or similar.
• [] Automated testing of docs added to github actions.

I would like to be able easily create the following reference components using pn create component

• [] view (i.e. just a function taking some arguments)
• [] viewer_widget
• [] viewer_pane
• [] viewer_layout
• [] reactive_html_widget
• [] reactive_html_pane
• [] reactive_html_layout
• [] bokeh_widget
• [] bokeh_pane
• [] bokeh_layout
• [] template
• [] data_service (Extract, Validate, Transform, Validate) component.

Ideally, I would like to have some concept for creating great reusable components that separates the concerns of the

• view
• logic
• state

See for example High level view and logic separation in React.

The concept must be useful for beginning, low code data app developers too, i.e. not overly abstracted or complicated.

References

• Viewer Component
• ReactiveHTML Component
• Bokeh Components and Awesome Panel Bokeh Components.