Skip to contents

Administration

The framework provides administration capabilities that mostly (but not only) cover data model tasks.

Admin console

An administration console is delivered as a standalone Shiny app within the package.

The main reason is that in most cases, it’s not recommended to have the data model(s) management accessible from within the application. Another reason is that administration tasks are not expected to be performed on regular basis. Therefore there is no need to have the main module server and the session overloaded with reactives & observers that won’t be called in the daily scenarios.

To run the app, use the admin() function:

kitems::admin(path = "path/to/data")

The admin console gives access to all the data models belonging to the path argument in one place. One tab per data model will be displayed inside the UI.

Features

Data model

The data model section displays a table of the attributes:

From here, it’s possible to create or update (after row selection) an attribute thanks to the attribute wizard:

It is meant to guide you through the attribute creation and perform checks (in particular at the default value step).

The danger zone toggle allows access to actions that cannot be undone:

Caution

  • Deleting an argument will delete the corresponding column in the item table!

  • Deleting a data model will delete the corresponding item table!

These actions cannot be undone…

Raw table

The raw table section displays the item table as-is (hidden attributes are displayed).

It’s possible to reorder the attributes (the items columns) from here.

View

The view section displays the item table as it will be available from the filtered_view_widget() function:

  • only attributes with display = TRUE are visible

  • mask is applied on column names (replace . and _ with space, capitalize first letter)

You can add or remove attributes to be hidden from this view (or update the attribute in the data model section).

Initialization

When the console detects an empty folder inside the provided path, it will allow you to create a new data model:

You can create one from scratch or import existing data (see next section).

Data import

It is possible to import existing data (and create the corresponding data model) through a wizard.

  • Select the file to import (.csv)

  • Check the items preview

  • Check the extracted data model

  • Validate import

Important

At this stage, the import process is not very flexible as it does not allow to tune the extracted data model, or select specific columns of the data. This needs to be done manually in a second step.

There is a known bug / problem after the select file step:
In case the data does not contain an id column, unique ids will be generated but performances are poor when data contains many rows.

See issue Generate unique ids is slow during data import

Migration

As of version v0.7.1, data models now carry a version number that corresponds to the latest data model structure upgrade.

Whenever the module server detects either a data model without any version or with an obsolete version number, it will raise a warning dialog and ask for the app to be closed.

Data model migration is performed through the admin console. It will detect when a migration is required and display a button to run the migration procedure.

Tip

It is recommended to run the admin console after a package upgrade to check whether a migration is needed or not.

It is also recommended to check the changelog page.

Behind the scene

Because the package is meant to be as flexible as possible, it is possible to access the admin components from exported functions.

Caution

That being said, it is not recommended to wrap the admin console inside your app as it should be kept away from the users.

Admin module server

The module server function is delivered as kitems_admin() .

Admin UI

The module UI component is wrapped in the admin_widget() function.