| No Comments

One thing I found that worked really well was having test cases to run constantly. After much frustration I figured out how to debug the executable [the functions in classes have to have the class name in front for breakpoints, i.e. break Scanner::scan] and then was more efficient in finding errors. Although making the test cases was a pain, they were definitely worth making. I will try to continue this trend into future iterations.
Another thing that was useful was the repository. Once I figured out how to use it, sharing code was extremely easy and could be accessed from anywhere. The merge function is also very nice. It eliminates the possibility of missing a crucial new line of code when merging by hand.
One thing that wasn't all that great for my team was the division of coding. We never had a set plan on who was doing what pieces. I was trying to figure out exactly what the functions were supposed to do and ended up writing all the main functions, only leaving the regular expressions and test functions for my partner.
I will try and plan the coding better next time by meeting with my partner before we start off and looking over the code to analyze what needs to be done. The first iteration was successful for us so I guess the only thing I would change is planning and write more test cases early on. I had some test cases at the beginning but they weren't exhaustive and we ended up finding odd errors later on when we added more tests.

Recent Entries

Find recent content on the main index or look in the archives to find all content.