Sensotheka - Management of Inventory, Files & Users - Shiny Contest Submission

Sensotheka - Management of Inventory, Files & Users

Authors: David Barkemeyer, Johannes Sturm

Abstract: Sensotheka supports the Department of Vehicle Propulsion Systems at the TU Berlin in managing its sensors. A SQLite database manages users, sensors and transactions. Sensors can be borrowed and returned via QR codes.

Full Description:

See also my other contributions:

Contents

  1. Background
  2. User Management
  3. Sensor Management
  4. File Management
  5. QR Codes
  6. Internationalization
  7. Cookies
  8. Deployment

1. Background

Every semester, the Department of Vehicle Propulsion Systems at TU Berlin offers the so-called "Propulsion Systems Project", in which students can work on one of the department's various projects. In the winter semester 2020/21, the development of a web application to manage the sensors of the department was tendered. Since no particular framework was given and Johannes and I both have some experience working with Shiny, we decided to take on this project and develop the Shiny application presented here.

2. User Management

There are three different user roles: User, Moderator and Administrator. Depending on the user role, different tabs are visible in the sidebar and many operations differ depending on the user role. For example, an Administrator can borrow and return sensors for any user, while a User can only perform transactions for themselves.

3. Sensor Management

Relationships between sensors are mapped using a three-part hierarchy. A given sensor is a Subtype. Subtypes that have the same functionality but differ in size or manufacturer are all part of the same Type. Types that form an experimental setup can be grouped into a Group. Administrators can add, modify, and remove entities at any level of the introduced hierarchy.

Users may borrow Subtypes up to the available quantity and return them when they finished working with them. Additionally, they can browse all associated Groups so that operating on different Subtypes of an experimental setup is accelerated. Furthermore, all previously attached files on each step of the hierachy are shown.

4. File Management

Administrators can upload PDF files for each step of the hierarchy. These are all stored in separate subdirectories of the 'files' directory. All users can then download these files. File contents are e.g. data sheets for specific sensors or theoretical backgrounds for experimental setups.

5. QR Codes

Administrators can create QR codes for Types. These QR codes are printed and stored in the sensor warehouse. If a user scans one of them with a QR reader, they will be redirected to the Borrowing & Returning tab and the scanned Type will already be selected. For this functionality, the query string of the URL is processed.

6. Internationalization

The user interface of the application we developed for the project was exclusively in German. For our contribution to the contest, we decided to internationalize all the code and localize it in English, Spanish and French - languages that we or friends of ours know and whose grammatical structure has similarities with German. We first tried the {shiny.i18n} package, but missed some features like interpolation and parameterized keys, so we decided to build our own customised fork that takes some ideas from the above package and develops them further. The language selector in the upper right corner allows a quick and live language change.

7. Cookies

There are two cookies that store client data:

  • user: the encrypted username, so that a user can be logged in automatically as long as this cookie does not expire.
  • dark-mode: the dark mode setting, so that the dark mode is set accordingly when the application is started.

8. Deployment

The original application is deployed at the department of the TU Berlin. We initially deployed the contest version on Shinyapps.io. However, we noticed that for now persistent data storage with SQLite is not supported on Shinyapps.io. Therefore, we decided to link to our private Shiny server.


Keywords: stock-management, user-management, file-management, internationalization, database-integration
Shiny app: https://sensotheka.davidbarke.com
Repo: GitHub - PFA-WebApp/App: Shiny App for PFA (Projekt Fahrzeugantriebe)
RStudio Cloud: Posit Cloud

Thumbnail:
image

Full image:

2 Likes