Skip to content

Latest commit

 

History

History
37 lines (26 loc) · 1.71 KB

FINALREPORT.md

File metadata and controls

37 lines (26 loc) · 1.71 KB
Raw

Project Final Report - Pelican

Pelican is a static site generator written in Python. Unit testing was performed on this repository.

Unit Testing with Sufficient Coverage

Work was initially intended to test Pelican's linter. However, due to the complexity of the linting system within Pelican, I decided to write test suites to test Pelican's utility functionality.

Code coverage was conducted with coverage.py (https://coverage.readthedocs.io/en/coverage-5.1/#). However, I was not able to get the coverage report for the specific files I've altered due to the internal workings of Pelican and the vast number of dependencies involved which constantly led to errors with running coverage.py.

Link to Repository

https://github.com/csun-comp587-s20/pelican/blob/master/pelican/tests/test_functionality.py on the master branch of csun-comp-587-s20 organization.

Automated Testing

The initial intention outlined in my project proposal suggests the utilization of a grammar-based approach to test the linter system of Pelican. However, due to the convoluted nature of the program, automated testing was an unsuccessful task.

Lessons Learned

  • The Pelican project was found to be slightly too complicated for me to write tests for as the project already involved meticulously tested code by its developers/contributors.

  • Being able to write and execute efficient test suites requires deep understanding of the project and it's internal workings.

  • With my current knowledge of utilizing verification testing techniques with technologies alike Dafny, a verification testing approach could have been performed on this project.

  • Introducing property-based testing could also prove to be useful to test this framework.