Tonight I was working on one of the rooms and came across numerous problems related to multiple results. One of the issues I found was another array issue, which is the difference between $array + $array and array_merge(). After making a few other changes it appears room resolution has broken completely.
So now it's been two sessions where I've wasted multiple hours hunting down stupid bugs that unit testing would most likely have found. Because of this I am moving unit testing/refactoring to the number one priority. This will likely take a while.
The reason I haven't been doing unit testing is because I was working with the philosophy of get it out quick, regardless of code quality. So I've taken a lot of shortcuts in order to do the "MVP" philosophy. I now realize this is stupid. Anyone that isn't testing their code in the name of getting to market quickly is an idiot, and trust me, I've read it numerous times. If your product cannot be proven to work then it has no intrinsic value, no matter what the idea or presentation is.
The good news is writing my code in this way felt a little weird the whole time, and I am a skilled test designer and code architecht and am well equipped to refactor this into beautiful, well tested code.
Along the way I will probably knock out the deployment strategy, as building out debug, production, and testing environments will definitely be part of this process.
New list:
Unit testing
Deployment strategy (beta on sourceforge, production on my server)
Room testing/creation
Documentation, especially room creation manual
2 Hours
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Tonight I was working on one of the rooms and came across numerous problems related to multiple results. One of the issues I found was another array issue, which is the difference between $array + $array and array_merge(). After making a few other changes it appears room resolution has broken completely.
So now it's been two sessions where I've wasted multiple hours hunting down stupid bugs that unit testing would most likely have found. Because of this I am moving unit testing/refactoring to the number one priority. This will likely take a while.
The reason I haven't been doing unit testing is because I was working with the philosophy of get it out quick, regardless of code quality. So I've taken a lot of shortcuts in order to do the "MVP" philosophy. I now realize this is stupid. Anyone that isn't testing their code in the name of getting to market quickly is an idiot, and trust me, I've read it numerous times. If your product cannot be proven to work then it has no intrinsic value, no matter what the idea or presentation is.
The good news is writing my code in this way felt a little weird the whole time, and I am a skilled test designer and code architecht and am well equipped to refactor this into beautiful, well tested code.
Along the way I will probably knock out the deployment strategy, as building out debug, production, and testing environments will definitely be part of this process.
New list:
Unit testing
Deployment strategy (beta on sourceforge, production on my server)
Room testing/creation
Documentation, especially room creation manual
2 Hours