"Paul Robinson" <paul@[EMAIL PROTECTED]
> wrote in message
news:BMuKd.9$iD.7@[EMAIL PROTECTED]
> JerryMouse wrote:
> > Paul Robinson wrote:
> >
> >>I am considering the development of a Cobol compiler itself written in
> >>Cobol. Please don't flame me over the perceived stupidity of such a
> >>choice, I think it is a better idea than a language that cannot
> >>compile itself. I do believe obviously that there is enough power in
> >>the language to implement a COBOL compiler using the COBOL language.
After
> >>all, a compiler is simply a program to translate text from a
> >>source into a target. We're not talking rocket science or brain
surgery
> >>here.
> >>
> >>I believe there is money to be made with a good inexpensive Cobol
> >>compiler which is available for several different targets, which would
> >>allow applications to be used unchanged on different environments,
> >>such as IBM Mainframe, DOS, Windows, Linux and others. Having done
> >>maintenance on three other language compilers and written one from
> >>scratch for Fortran in three months, I decided to focus on Cobol as I
> >>think it is an underserved market and I think there is both room for
> >>another product and sufficient market need to merit the time and
> >>effort in developing one.
> >>
> >>I am wondering if there was interest in such a thing. What I am
> >>thinking about is something that did not require run-time licensing
> >>fees, would not be hugely expensive as most Cobol compilers seem to
> >>be, and by writing the compiler in Cobol itself, the people who use
> >>the compiler, if they wanted an additional feature the compiler did
> >>not supply, could write the code for it themselves if they chose to
> >>do so.
> >
> >
> > The Realia compiler is written in COBOL. I saw the source listing once
-
a
> > binder as tall as a basketball player (well, maybe a SHORT basketball
> > player). As I recall it was 800,000 lines of code.
> >
>
> Maybe I'm missing something - or deluding myself - but I think that
> seems extremely high. 800,000 lines of code seems to indicate an
> application that is far too complex or possibly much of the code is
> generated by other programs from specifications instead of written to
> handle compilation of one or more Cobol source program files. Or maybe
> I'm just a little bit brighter than I thought I was.
I don't think that's all that high. The monolithic COBOL74 compiler for
Unisys MCP systems is in an ALGOL dialect (NEWP), and it's about 133k
lines
(sup****ting lots of Unisys extensions); the predecessor COBOL(68) compiler
in ALGOL was just under 60K.
Given COBOL's infamous wordiness, and the standards features that
'85-and-later COBOL has that '74 did not, 800K for a modern COBOL compiler
written in COBOL doesn't seem all that excessive.
The COBOL85 compiler for the Unisys MCP environment is modular and part of
a
system that accommodates both multiple code generators and multiple
languages, so line counts for it (in Pascal83) aren't really comparable.
-Chuck Stevens
|
