[Coco] Rainbow IDE status

[Roger Taylor]

At 12:14 AM 5/21/2007, you wrote:
>Hi Roger,
>
>Great work Roger! I can't wait to see the final result! Can I simply upgrade
>my existing copy?
>
>I must say I haven't done much development with my copy yet, because I was
>thrown back by it not wanting to compile my assembly source. However, I will
>soon give it a try again, and I hope that you can help me to get it to work.
>
>
>As an extra: One of the things I would really find useful is a disassembly
>capability. There are a lot of Windows command prompt 6809 disassemblers out
>there. Recently I have been using f9dasm (
>http://www.hermannseib.com/english/opensource.htm) with great results. Is it
>possible to interface Rainbow IDE with this and/or other assemblers?


Yes, you'll be able to install the Rainbow IDE 2.0 right over your 
existing installation without trashing your main .ini file that holds 
your key.  Even if you delete that file, you can use your existing 
key to register 2.0.  It would really benefit new users to purchase a 
copy now at the deal.

CCASM currently has a undocumented disassembler mode in it!  Well, 
sort of.  It's incomplete, but some output is 
generated.  Additionally, Rainbow has a hidden interactive mode that 
works along with CCASM to show you the current disassembly, where you 
should be able to navigate through the grid-listing and make changes, 
click 'disassemble' again, until the listing looks right.  This mode 
is incomplete as well so it is not accessible.  What's more: the 
final goal is for the IDE to take a binary program or ROM image, 
disassemble it once, bring the grid up, you make any chances 
necessary, retry, and the generated source code will be reassembled 
by CCASM, then the IDE will launch it like any other program.  It 
will be quite seamless if my plans go right.

Here's the idea... in ONE CLICK:

disassemble binary-> assembled generated source code again -> send to 
.DSK image or ROM Pak image -> launch in M.E.S.S. and run

So, if you run a binary through this process and it actually RUNS ok 
in the emulator, it was disassembled ok!

I'll definately be looking for one of the emulator-based 
disassemblers that pretend to run the code so as to separate code 
from data.  As long as you mark the beginning and end of the program 
(all code and data), a ROM image should disassemble pretty well.  I 
can't say the same for a multi-record based LOADMable program that 
loads code and data into different areas of RAM when you type 
LOADM.  I'm working on this with CCASM but it's on the back burner 
for now while I improve the Rainbow IDE.

CCASM produces a .dsm file which Rainbow knows how to read better 
than a mysterious .asm file.  The .dsm file holds markers for code, 
data, etc. already.

You can probably try this now if you have CCASM 4.0.  Just type:

cm someprogram.bin  and see if a .dsm file is created in your work 
directory.  Try the same for somegame.rom, etc.

It's been months since I've worked with this so I'll have to refresh 
my own memory when I get back into this project.  :)



-- 
Roger Taylor