Gennadiy Rozental wrote
Unfortunately I was not able to look with proper attention onto your submission during review. One thing I did though: all the major issues I mentioned in my original review looks resolved. I am sure will look into more details later, but for now it is enough for me to cast YES vote.
This library cover very important problem domain in software development and I think we all have now solid solution. Thanks and congratulation to Robert for being able to bring it to us.
I was waiting for a comment from you. Your incredibly insightful, exhaustive and focused review of the previous version was the main driving force behind this version. Looking back, it's still hard to for me to believe that in the six hours you said you spent anyone could really understand the implementation to the depth you did. Basically, all I did was to implement all your suggestions one at a time. Early on I switched over to your test library. This permitted me to add features one at a time and test at each stage. It did result in a lot of code being rewritten several times. But I think in system like this one can't really look far enough into the future implementation to avoid this. So it's not too much to say that if this library lives up to the hopes and expectations I have for it, it will be due to a collision between your previous review and my personality flaw that inspires me to throw myself at lost causes. I was going to suggest you might separate some time to review it. On the other hand, maybe I should quite while I'm ahead. Robert Ramey