[comp.unix.wizards] System V manuals

guy%gorodish@Sun.COM (Guy Harris) (09/30/87)

> First of all, the generic System 5 release 2 manual are arranged the same way
> as the SVID.

Not the on-line versions distributed with VAX S5R2, not the printed ones we got
with out VAX S5R2 distribution, not the printed ones somebody here had with an
Intel iAPX286 S5R2 distribution, and not the printed ones we got with S5R2 for
our 3B2s.  All of them put "fread" into section 3, and the system calls into
section 2, despite the fact that the SVID puts "fread" and the system calls all
into BA_OS.  Which allegedly "generic" version are you referring to?

> Other differences between read and fread (and write and fwrite) include:

All of which are the *intrinsically* significant differences (as opposed to
"'fread' drags in more code", which is significant in some cases and not in
others), and *none* of which have to do with "fread" being a library routine!
"read" could be a library routine (e.g., in Doug Gwyn's S5 emulation package
for 4BSD systems) and still have the same characteristics described.
	Guy Harris
	{ihnp4, decvax, seismo, decwrl, ...}!sun!guy
	guy@sun.com

peter@sugar.UUCP (10/03/87)

In article <29585@sun.uucp>, guy%gorodish@Sun.COM (Guy Harris) writes:
>>First of all, the generic System 5 release 2 manual are arranged the same way
>>as the SVID.
> Which allegedly "generic" version are you referring to?

"The UNIX System Users Manual". Copyright 1986 AT&T, published by Prentice-
Hall. ISBN 0-13-938242-9 025. Distributed to bookstores all over the country
as the official AT&T System V manual. It's spiral-bound with a white spine
and a red cover.

> > Other differences between read and fread (and write and fwrite) include:
> ...*none* of which have to do with "fread" being a library routine!
> "read" could be a library routine (e.g., in Doug Gwyn's S5 emulation package
> for 4BSD systems) and still have the same characteristics described.

"read" could not be a library routine and still have the desirable feature of
leaving the external file descriptor pointing at the last character read, unless
it's implemented by doing character-at-a-time I/O. Pretty inefficient. I don't
have the list with me, but at least this difference is related to read being
a system call.

PS: If you want to reply, send mail. I don't read comp.unix.wizards any more.
-- 
-- Peter da Silva  `-_-'  ...!hoptoad!academ!uhnix1!sugar!peter
-- Disclaimer: These U aren't mere opinions... these are *values*.