[svanleent]

Some of you might have seen me in the D forum again. Mostly due to many things which where happening in the same time, I couldn't do much about DGrammar.

I began rewriting it, but instead of using C, C++ and YACC, I'm going to rewrite it using D.

My first try didn't work out as I expected, so I entered a rewriting schedule.

First I'm going to make some things at the far back-end of the machinery. The most important one is that I'm making a subproject, DMachine right now. DMachine contains right now only one type of machine (A PDA), and I'd like people to test it, before I'm making mistakes again. I'm going to nail a sticky thread so it is possible for everyone to add comments on the workings of it.

The next things I'm going to implement:

- Double stack machine (instead of [x X/Y] it is going to be [x X/Y A/B])
- DFA, Inferred DFA
- NDFA (Can be translated to a DFA and/or PDA)
- Regular Expressions

Only if the basis is their, I'm able to continue on DGrammar itself

The Unit is now under SVN, there are some minor glitches in it concerning lambda transitions. It is to strict now, I've got to solve this.

Regards,
Sjoerd
_________________
How C++ became ancient

[davidl]

it's a wonderful project.
with DGrammar we can implement a D compiler easily ~~