Test Report
Testing Scope
The scope of testing is to verify fitment to the specification from the perspective of
Functionality
Deploy ability
Configurability
Customizability
Verification is performed not only from the end user perspective but also from the System Integrator (SI) point of view. Hence Configurability and Extensibility of the software is also assessed. This ensures the readiness of software for use in multiple countries. MOSIP is an “API First” product platform.
Testing scope has been focused on the below features:
Authenticate user credentials
Credential type
Download VC
Durian data storage integration
Error Handling
Inji Home page
Issuer selection in the “Home Page”
Issuers support
Latest draft changes support for OpenID4VCI
Multi languages support
OpenID4VP Implementation
PDF and CBOR
PDF Generation and auto download
PDF View
QR Code
Responsive view of the web
Retrieve credential types
Retrieve issuers list
Search bar functionality
Success/Failure notifier after downloading the VC PDF
Theme update
UI Automation
Access violations
Test Approach
Persona based approach has been adopted to perform the IV&V, by simulating test scenarios that resemble a real-time implementation.
A Persona is a fictional character/user profile created to represent a user type that might use a product/or a service in a similar way. Persona based testing is a software testing technique that puts software testers in the customer's shoes, assesses their needs from the software and thereby determines use cases/scenarios that the customers will execute. The persona needs may be addressed through any of the following.
Functionality
Deploy ability
Configurability
Customizability
The verification methods may differ based on how the need was addressed.
Verified configuration
Verification is performed on various configurations as mentioned below
Default configuration - with 6 Lang
English
French
Arabic
Hindi
Tamil
Kannada
Feature Health
Test execution statistics
Functional test results
Below are the test metrics by performing functional testing. The process followed was black box testing which based its test cases on the specifications of the software component under test. Functional test was performed in combination of individual module testing as well as integration testing. Test data were prepared in line with the user stories. Expected results were monitored by examining the user interface. The coverage includes GUI testing, System testing, End-To-End flows across multiple configurations. The testing cycle included simulation of multiple identity schema and respective UI schema configurations.
Here is the detailed breakdown of metrics for each module:
UI Automation and API Test Rig test results
Below are the test metrics by performing UI Automation and API Test rig.
Here is the detailed breakdown of metrics for each module:
Functional and test rig code base branch which is used for the above metrics is:
Hash Tag: 58e77d26fc1b98884c11638bba70c128d27994e3
Detailed Test metrics
Below are the detailed test metrics by performing manual/automation testing. The project metrics are derived from Defect density, Test coverage, Test execution coverage, test tracking and efficiency.
The various metrics that assist in test tracking and efficiency are as follows:
Passed Test Cases Coverage: It measures the percentage of passed test cases. (Number of passed tests / Total number of tests executed) x 100
Failed Test Case Coverage: It measures the percentage of all the failed test cases. (Number of failed tests / Total number of test cases executed) x 100
Git hub link for the xls file:
<TBD>
Last updated