Re: gnubol: random thoughts

[Tim Josling <tej@melbpc.org.au>]

Chad,

I think we are on the same wavelength. We implement a subset of cobol in
C, then use cobol for other parts (eg parts of the runtime) using that
subset.

We could make the subset a separate "exe" file to make it easy for
people to know if they are compliant with the subset.

If you mess this up it can be complex. Look at the GCC installation
procedure, as documented in "Using and Porting GCC".

Tim Josling

Chad Slaughter wrote:
> Tim Josling wrote :>
> >I good idea I think. But it does create a botstrapping issue. How do you
> >compile a cobol compiler written in cobol?
> No. dont confuse what I said. I didnt say write a cobol compiler in cobol
> that wont work. cobol isnt a system language.
> this idea i am going for is what features of cobol could be implemented
> as thru copy/replace on top of a smaller subset?

I don't understand this?

> could we implement nucleus level 2 in cobol using only nucleus level 1?

I think the parser and lexer have to be in C/Java but the rest can be
largely COBOL, especially if we allow access to the C runtime from
COBOL.

> Chad Slaughter  -- slaught at advancenet.net

--
This message was sent through the gnu-cobol mailing list.  To remove yourself
from this mailing list, send a message to majordomo@lusars.net with the
words "unsubscribe gnu-cobol" in the message body.  For more information on
the GNU COBOL project, send mail to gnu-cobol-owner@lusars.net.