I’ve just posted a copy of Borland’s Delphi grammar with my annotations, corrections, and tweaks. It’s not in EBNF form, but it’s close, and it’s a lot more accurate than Borland’s original grammar document.
This is a file that’s generated from a little Ruby script I wrote this morning. It’s not just a static document; it includes the current status of the Delphi parser as I’m working on it. It shows which language constructs I’ve fully implemented, which I’ve thrown away in favor of something else, and which I haven’t worked on yet. (As of now, out of 121 language constructs, I’ve got 21 finished.)
The colors and such aren’t all self-explanatory yet, but you can probably figure most of them out. I’ll be adding more explanatory text later. If you can’t figure something out, post a question here.
I expect the project to further accelerate as I go, partly because of the “Likely Targets” section I put into this status page. It probably took twenty minutes to add that to the report, but it’ll save me a lot of head-scratching and wondering what I can work on next without having to stop and finish something else first.
I’ll keep the project status page updated as I go (whenever my Net connection is actually working, that is). Go check it out.