This app aims to provide a graphical interface for customization of project_info.jsonconfig file used by shesmu. The interface requires flask installation so it can run in a web browser locally. Some settings also need to be specified, see INSTALL for that.
Launch the virtual environment and install Flask:
source venv/bin/activate
pip install -r requirements.txt
Configure the application's settings file:
CONFIG_FILE_DIRECTORY=/home/USERNAME/secrets
cp ui-config.cfg.example "${CONFIG_FILE_DIRECTORY}"/ui-config.cfg
Modify the new file to point to the locations of the shesmu configuration files (currently in spb-seqware-production).
Normally, you would update your production branch of the repo with shesmu configuration files (currently it is spb-seqware-production) and go into your flask-ui directory where config-ui.py script resides. Before running the app, some environment variables need to be set:
export FLASK_APP=config_ui.py
export FLASK_ENV=development
export UICONFIG_SETTINGS="/home/USERNAME/secrets/ui_config.cfg"
Assuming that you have your virtual environment configured, you also need to run
source venv/bin/activate
To make things easier you may edit the pre_flight.sh script and start your app with
source pre_flight.sh
====================================================
the prompt should change after that. After everything is initialized, type
flask run
The message should look similar to this:
* Serving Flask app 'config_ui.py' (lazy loading)
* Environment: development
* Debug mode: on
* Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)
* Restarting with stat
* Debugger is active!
* Debugger PIN: 874-681-120
Crl+C on your console is sufficient, flask will stop serving the app at 127.0.0.1:5000 You may also type deactivate to terminate the python virtual environment used by flask
The UI shows a couple of drop-down lists, one for projects and the other one for presets Steps for a session may include some or all of the following:
- Select a project, review the configuration. Make your selection of parameters and enable or disable individual components of computational pipelines
- Reset if not happy with your changes, this will restore configuration from the disk
- Apply a preset if applicable
- Create a configuration for a new project by cloning, then applying a preset or changing individual parameters
- Click Apply, that will update the project_info.jsonconfig on disk
- Note that if you check/uncheck some boxes you will need to click Apply otherwise your changes will be lost if you select another project
- Go to your local directory with shesmu config files, create a branch, review and commit your changes. Push to the repo and create a Pull Request
The UI app relies on project_presets.conf file which defines a number of settings for standard pipelines configured by GSI. Occasionally, settings may be removed or added to the project_info.jsonconfig. flask UI app will detect such changes and auomatically update the preset file. Pipeline Leads should review these changes and ensure that they were applied correctly (especially important when a complex setting as ##_pipeline is added). The app takes it's best guess what to use to configure a new setting but it always needs to be verified. The proper procedure for this would be to use git diff followed by manual edits in something like vim or other editor of choice.
Developed using Python 3.8, Pycharm IDE v.2020.1 and Flask 2.0.2