home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!haven.umd.edu!darwin.sura.net!sgiblab!troi!steve
- From: steve@dbaccess.com (Steve Suttles)
- Newsgroups: comp.os.vms
- Subject: Re: VAX C header file standards?
- Message-ID: <136@mccoy.dbaccess.com>
- Date: 8 Jan 93 23:26:29 GMT
- References: <1993Jan7.224528.13810@news2.cis.umn.edu>
- Organization: Cross Access Corp., Santa Clara, CA
- Lines: 57
- X-Newsreader: Tin 1.1 PL4
-
- rao@moose.cccs.umn.edu (Rao Akella) writes:
- :
- : Are there any standards for VAX C header files? How does Digital decide what
- : definition goes in which file?
-
- On a case-by-case basis. Usually, it is pretty obvious.
- :
- : For example, in V3.1 (-051, to be precise), the definition of caddr_t ("typedef
- : char * caddr_t;") is present in both DECW$INCLUDE:XOS.H and XRESOURCE.H. In
- : V3.2 (-044), the XOS.H definition was dropped. I'm feeling slightly bummed
- : about this because I recently made Gnu Xchess available for anon ftp at my
- : site, and I've since been informed that while V3.1 compiles Xchess correctly,
- : V3.2 returns compilation errors (because Xchess includes XOS.H but not
- : XRESOURCE.H).
-
- Cite the compiler version. Test new versions. You might be able to
- be compatible with both by including headers which are "extra" for some
- circumstances.
-
- :
- : Shouldn't all existing definitions be left in there in the interest of upward
- : compatibility? Or was there a good reason for Digital to change the header
- : files in between upgrades?
- :
- Perhaps...but you wouldn't like it. The reason they changed is for ANSI
- compatibility (well, approaching ANSI). In order to meet both goals, they
- would have to #ifdef on some indication that you were using/not_using ANSI
- compliant mode. The size of the headers would double, as would their
- complexity (which -does- have an effect--compile time and user comprehension).
-
- The "good" news is you probably won't have to deal with this again; VAX C
- gets virtually no support, since all the money is on DEC C. Of course, this
- is the bad news too--VAX C will fall off in popularity when there is any
- acceptable alternative (if you ported to VAX C, you already know that large
- parts of the runtime library--forget header files--are brain dead).
-
- sas
-
- PS: DEC in no way approves of my opinion of VAX C. I do not approve of their
- opinion, so I guess that's fair. They must think it's fair too, because they
- still won't fix the parts that are broken. Yes, I have a list, and yes, I gave
- it to them. The part that really bothers me is that I'm forced by circumstance
- to use and recommend VAX C, and that most of the problems could be fixed within
- the runtime libary, and those that can't, can be worked around...but even this
- information can't get distributed.
-
- For the record, just to be fair, the compiler is not at fault. The problem
- lies entirely within the runtime library, distributed with and as a part of
- VMS (the cause of the problem may have had something to do with
- interdepartmental politics at DEC--I've had my share of those to deal with,
- too [no, don't bother to ask]).
-
- --
- Steve Suttles Internet: steve@dbaccess.com Dr. DCL is IN!
- CROSS ACCESS Corporation UUCP: {uunet,sgiblab}!troi!steve Yo speako TECO!
- 2900 Gordon Ave, Suite 100 fax: (408) 735-0328 Talk data to me!
- Santa Clara, CA 95051-0718 vox: (408) 735-7545 HA! It's under 4 lines NOW!
-