GSoC: LALR Parsing: Midterms over!

We're halfway through the summer, now. I have not made nearly as much progress as I had previously hoped. Partly this is due to the difficulty of the problem, and partl\
y it is because I have not spent enough time on it. I think I'm currently at a point where I'm only a few mostly simple transformations away from producing a Determinis\
tic Push-Down Automaton which will be capable of parsing a LR(0) grammar. Whether implementing those last few mostly simple transformations will show this belief to be \
correct or not, I won't know until I do so.

Here are my next few goals:

  1. Make certain states in the DPDA push their names onto the stack when they are entered. This is essential in order for lookback to work.
  2. Eliminate epsilon-transitions by merging states. This is likely to be at least a bit tricky.
  3. Write more tests and fix bugs.
  4. Figure out what to do about the pop/output transitions. Specifically, decide what I want the output to mean.
  5. Write a DPDA interpreter.

My plan is to start with the last of these tasks, because it should be straightforward and it will be encouraging to be making progress again. After that, I'll tackle t\
he name-pushing; then the epsilon-transitions. Finally, I'll write more tests for the other steps of the parser generation process. Concurrently with all of this, I'll \
be thinking about how to handle pop/output transitions.

Time to go write an interpreter!