Hi, Etienne!
Thanks for the prompt reply and feedback.
I am doing well, thank you!
It had been a while since I did some parsing. The last project I worked one was a SQLBase to MySQL translator. Now I am putting together a transpiler for Pascal to Go. I have standard Pascal code that I need to translate to Go, and it is faster to put together a translator than to convert the code manually. There are certainly some things that will need some manual adjustments, but the core will be much faster with a transpiler.
I actually took some time to build it and play around with it. I noticed all the test grammars are no longer valid, which I guess was used in an earlier syntax. That made me play around with it a bit, by fixing them based on the SableCC 4 grammar.
After fixing the code, I was hoping to see some code generated, but t my surprise, I see only a message saying that SableCC successfully compiled the grammar. When I looked at the code, I saw that the code generation section was commented out with TODO messages.
I liked the fact that SableCC 4 switched to an LR(K) parser generator, but now that you mention that it isn't usable I had no choice other than switch back to SableCC 4.
Regarding your feedback, is it just the Lexer generator that isn't complete, or the parser as well? I noticed that you have a different AST structure and the new Walkers (Visitors). If the parser code is complete, is there a way of us hooking a hand generated scanner that uses the new Token class structure, or even adapting a scanner generated by SableCC 3? Or in terms of features, it lacks both the Lexer and Parser generator?
I will look into the paper you referred to me about ObjectMacro. Thanks for the pointer!
I look forward to hearing from you again, and take the opportunity to wish you and the rest of the SableCC community a wonderful Christmas Eve.
Take care!
Fidel H Viegas