20240122

 Fell asleep around 7 and woke up around 1:30. Got up half an hour later (waited for electronic heater) to do some research. I had already decided to adopt the idea I had came out last Friday. It may not reduce much work as I had expected, but finally I could see how the architecture should be. I had been refactoring in the past several weeks, partly for I liked improving my code, partly for I couldn't see the picture. And the architecture was actually pretty old one, though I could't use it due to some minor differences. Bought SICP and dragon's book for reference. I was quite sure I had read the latter when I was in university. Though theories about compiler could be helpful, I didn't have to go very deep, glad I didn't have to handle LLVM and asembly language. Watched a football game around 4 and then had breakfast at 7. Read book then and returned to my study room to read English articles for a while. Learned Rust on Educative afterwards. Played piano after lunch, then started to work. Instead of focusing on backend part, I decided to write date types in the first place, then I could develop both ends in parallel, front end for generating and editing it, backend for saving, loading and doing some calculation with it. Got stuck in the end, had to refer to the books I had bought to write a clear AST, but they were still not delivered. Completed 16 pomodoroes today - one for English, two for reading, three for learning, and five each for research and work.