Case Study: Query Classifier Tool

Client

Our client is a business and technology consulting firm.

Business Need

• Refactor the front-end code of the web application on Angular, fix bugs.
• users’ web page error handling.
• Add administrator interface.
• Configure rights system based on Azure Active Directory;
• Implement UI search for items in the table using the Elasticsearch component from MS Azure.
• Set up a secure HTTPS connection using Let’s Encrypt.
• Refactor the server code on the Node.js, configure the the logging of errors.

Solution

• Updated Angular from version 5 to the latest version 7, refactored the front-end code according to Google’s guidelines, fixed bugs, replaced the self-written code with the use of built-in functions of Angular and open-source libraries (WS, @angular/material, rxjs, adal-angular5, ngx-permissions, fontawesome).
• Rewritten the UI from the old version of Bootstrap to the latest version of Material Ui, leaving only the grid from Bootstrap.
• Simplified the interface with the application, instead of navigating to individual pages implemented interactive dialogs.
• Implemented user error handling and displaying error messages to the user.
• Designed the administrator interface from zero point according to the general style of the application.
• Implemented a system of rights based on Azure Active Directory.
• Implemented UI search for items in the table using the Microsoft Azure Search service, configured the correct displaying of paging.
• Set up a secure connection via HTTPS using the service Let’s Encrypt.
• Refactored the server code on the Node.js, set up logging of errors.
• Rewrote the parsing code of the CSV-file with a set of customer data (about 100 thousand records), implemented the ability to notify the user about the parsing progress using the WebSocket Protocol.

Benefits and Results

• Our client got a well-structured front-end code, written in accordance with Google guidelines.
• We reduced the amount of front-end and back-end code. The application’s stability and speed were increased.
• The client’s app runs on the latest versions of the popular open-source libraries, which simplifies its support.
• The application has a modern, user-friendly and intuitive design.
• Our client highly appreciated Akvelon’s expertise and took our developers to its next project.

Technology Used

Angular, Node.js, Material UI, Azure Active Directory, Azure Search, Let’s Encrypt, WebSocket