Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!sdd.hp.com!spool.mu.edu!uunet!rbj
From: rbj@uunet.UU.NET (Root Boy Jim)
Newsgroups: comp.unix.questions
Subject: Re: docs for lex/yacc
Message-ID: <126666@uunet.UU.NET>
Date: 28 Mar 91 23:02:16 GMT
References: <1991Mar19.231319.1323@cbnewse.att.com> <126609@uunet.UU.NET> <7625@idunno.Princeton.EDU>
Distribution: usa
Organization: UUNET Communications Services, Falls Church, VA
Lines: 61

In article <7625@idunno.Princeton.EDU> rhl@grendel.Princeton.EDU (Robert Lupton (the Good)) writes:
>
>Oh yes, the X books betray an author who doesn't know much about C%,

Having thumped (perhaps unfairly) ORA, I owe them some kudos.  I have
them all, except for the Motif versions. Volumes 0, 2, and 5, are
pretty much collections of manual pages, with some added indices and
catalogues. It's pretty hard to do them wrong, so I doubt that your
quarrel is with them. Volume 3 is the user's guide, and doesn't get
into C.  Volume 7 is by Dan Heller, who is the comp.sources.x
moderator. I haven't looked at his book, but he seems like he knows
what he's doing in general. Volume 6 is who knows where?

That leaves Volumes 1 and 4. I have read all of 1, and about
half or 4, and learned a great deal from both. Both are by
Adrian Nye, who also edited the references (0 & 2), and
Tim O'Reilly himself, who edited 5 and contributed to 3.

Tim seems determined to keep this series high quality.

I have also looked at the Grey and Maroon book by RWS, Gettys,
and someone else. I'm sure everything is in there, but it's
very terse. And it's almost impossible to figure out how to
use the Intrinsics from the MIT documentation.

Now as to C style. I found several places where code could
have been written more succinctly, but this may have taken
away from explaining the point at hand. Writing programs
for X is not conducive to studly programming tricks, but is
rather plodding in style. There is so much to do, and the
motto seems to be "keep it simple, it's already complex enuf".

>and they are somewhat buggy.

The real question is "were they actually compiled and run
on a real machine before being published". I think they
probably were, especially since the source is distributed
in machine readable form (kudos for that, too).

Remember that X is somewhat buggy, has changed, and runs on
a variety of vendors hardware and software. Your mileage may very.
Also, I would think that some of these programs may have
deliberately left out subtle things in order to make their point.

Now as to ORA in general. I see them evolving from a company
that puts out books for beginners to one that is starting to
put out more sophisticated stuff. The perl book is excellent
and the one on system tuning look rather interesting. If you
want to learn X, this series is probably the best.

Of course, after you learn X, you might want to forget it,
but then that's another story :-)

>				Robert
>-----------------
>% While I am complaining, at least they know more C than the Authors of

of what? Lex & Yacc?
-- 
		[rbj@uunet 1] stty sane
		unknown mode: sane