Right now, my procedure for testing the code for a tool is very informal. I have written some short sample pages for each tool that include basic cases, so I run the tool on them and visually skim through the result to see if it's working as intended. This was quick to set up but carries two issues:
- it means that I don't test for cases that the sample pages don't cover.
- I may miss some errors that I introduced elsewhere while skimming through the result.
For whichever tools that the team deems consistently useful/applicable enough, it may be worth looking into a more rigid testing framework. One option is writing tests into test functions that explicitly run each individual tool function on various applicable cases, and return the tests that fail. For this, I would try looking into unit testing in javascript.
Right now, my procedure for testing the code for a tool is very informal. I have written some short sample pages for each tool that include basic cases, so I run the tool on them and visually skim through the result to see if it's working as intended. This was quick to set up but carries two issues:
For whichever tools that the team deems consistently useful/applicable enough, it may be worth looking into a more rigid testing framework. One option is writing tests into test functions that explicitly run each individual tool function on various applicable cases, and return the tests that fail. For this, I would try looking into unit testing in javascript.