Just a quick update today.  The Plan is done (yay!), and I’ve started on the Interpreter and Code Generation layer.  I’ve got basic Unicode support working, and I’m now implementing the other new Lexer features.

I’ve decided to put streaming capabilities into the system, as they might prove useful and don’t require a lot of extra work.  In the past, the parser has read in the entire source file before doing anything with it, then processed it in one go.  Streaming will allow it to be read in as it is available/needed, and allow elements early in the parse to be discarded via action code.  This will be useful if you are building an interpreter, for instance, as it means you can use the parser to process the source in chunks, after each of which you take the necessary actions and then move on.  The parser will, at that point, not error-correct back into the discarded code, nor will it keep the discarded code in memory.

Until next time . . . .