Douglas McClean

Take your time. I'm getting changes together for a bunch of the other things. I'm working on the stack shape analysis right now, so that will keep me busy for...

The machine-readable version (as a C header file) is in Partition VI, Annex C, Section C.2. It describes both the stack behavior and whether control goes straight through or potentially/unconditionally...

The file is opcodes.def, look in %ProgramFiles%\Microsoft SDKs\Windows\v7.0A\include Your mileage may vary depending on which OS you are on, on 64-bit it goes in the x86 program files directory, and...

I translated the opcodes.def file into Haskell. I'm working on the abstract analysis part for stack height only (not stack shape) because it is simpler to start with. The one...

That is a decent book. The excerpted definition of `methodDecl` looks like a simplified (by replacing abstracted non-terminals with their concrete representations) and slightly respelled version of the grammar in...

Yeah, I suppose it isn't that bad. Still, it seems more in line with the rest of things to have one alternative in OpCode (avoids duplicating analysis code) and two...

Agreed. I prototyped the type class approach to this problem and it doesn't work out well at all. It works sort-of-ok with the GHC overloaded-string-literals extension, but at a cost...

Is that really simpler? I think the local declarations are a fair bit easier to read in the monadic version, and you don't have to manage the commas, but I...

Sounds like a good plan. I have a side-by-side implementation that works. Later this evening I will do some renaming and pass it along.

I do have a best-of-both-worlds proposal. I'll do a draft when I get back to a full-sized keyboard.