Re: marking mystery code

jmccarty@spdmail.spd.dsccc.com (Mike McCarty)
14 Feb 1996 21:36:36 -0500

          From comp.compilers

Related articles
Re: Possible to write compiler to Java VM? ncohen@watson.ibm.com (1996-01-29)
Re: marking mystery code dewar@cs.nyu.edu (1996-02-13)
Re: marking mystery code dennison@escmail.orl.mmc.com (Ted Dennison) (1996-02-14)
Re: marking mystery code jmccarty@spdmail.spd.dsccc.com (1996-02-14)
Re: marking mystery code jgj@ssd.hcsc.com (1996-02-16)
Re: marking mystery code hagerman@ece.cmu.edu (1996-02-16)
Re: marking mystery code mnp@compass-da.com (Mitchell Perilstein) (1996-02-16)
Re: marking mystery code toon@moene.indiv.nluug.nl (Toon Moene) (1996-02-16)
Re: marking mystery code preston@tera.com (1996-02-17)
Re: marking mystery code rfg@monkeys.com (1996-02-19)
[2 later articles]
| List of all articles for this month |

From: jmccarty@spdmail.spd.dsccc.com (Mike McCarty)
Newsgroups: comp.compilers
Date: 14 Feb 1996 21:36:36 -0500
Organization: DSC Communications Corporation, Plano, Texas USA
References: 96-01-116 96-02-138
Keywords: C, GCC

)Gary says
)"Incidentally, I saw similar comments and code obfuscations when
)maintaining Fortran IV code 20 years ago. It is depressing that the
)FSF sponsored code products haven't been designed, implemented, and
)maintained at a higher level of professionalism than what was common
)in 1976."


Robert Dewar <dewar@cs.nyu.edu> wrote:
)Gary, are you speaking from personal experiences with the FSF sources,
)which many people have found easy to work with, or are you borrowing
)second or third hand comments without this personal experience?


I can speak from personal experience working with the FSF code. It is
generally bloated, difficult (read: impossible) to read, uncommented,
and much of it is totally unmaintainable.


We found a defect in grep. We tracked it down to a subroutine. The
subroutine was hundreds of lines long, completely uncommented, and had
nesting levels of ifs and switches to 8 or 9 deep. Furthermore, it had
gotos from such deeply nested stuff to similarly deeply nested stuff
several tens of lines away. I ran a McCabe extended complexity measure
on it, and it was over 2000 (10-20 is considered the absolute maximum
for reliable code). Do I need to state that we did not repair the
problem? We used a different grep. I -use- a different grep. The FSF
grep cannot work, and cannot be made to work.


Mike
--


Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.