schwartz@gondor.psu.edu (Scott E. Schwartz) (11/14/87)
given:
main()
{
int a,b,signed;
a = b = signed = 1;
}
is this legal? PCC on a vax running 4.3BSD and a Sun running SunOS 3.2
doesn't complain. Neither does lint (at least on the sun), except to
report that a,b, and signed are set but not used. On the other hand
GCC doesn't want to see a variable named signed. Which compiler is
broken?
Disclaimer: I don't have a copy of the latest ANSI draft, or even an old
copy of K&R at hand.
-- Scott Schwartz schwartz@gondor.psu.edugwyn@brl-smoke.ARPA (Doug Gwyn ) (11/14/87)
In article <3086@psuvax1.psu.edu> schwartz@gondor.psu.edu (Scott E. Schwartz) writes: > int a,b,signed; > a = b = signed = 1; "signed" is now a C keyword, but it's a fairly recent addition to the language and therefore is not known to some older C compilers. Other even newer keywords are "const" and "volatile".
ftw@datacube.UUCP (11/16/87)
schwartz@gondor.psu.edu.UUCP writes: > given: > main() > { > int a,b,signed; > a = b = signed = 1; > } > is this legal? PCC on a vax running 4.3BSD and a Sun running SunOS 3.2 > doesn't complain. Neither does lint (at least on the sun), except to > report that a,b, and signed are set but not used. On the other hand > GCC doesn't want to see a variable named signed. Which compiler is > broken? > Disclaimer: I don't have a copy of the latest ANSI draft, or even an old > copy of K&R at hand. > -- Scott Schwartz schwartz@gondor.psu.edu According to dpANS, "signed" is a keyword, and therefore can't be used as a variable name. The answer to your question is that the Sun compiler is "broken". Farrell T. Woods Datacube Inc. Systems / Software Group 4 Dearborn Rd. Peabody, Ma 01960 VOICE: 617-535-6644; FAX: (617) 535-5643; TWX: (710) 347-0125 INTERNET: ftw@datacube.COM UUCP: {rutgers, ihnp4, mirror}!datacube!ftw
aeusesef@csun.UUCP (sean fagan) (11/16/87)
In article <3086@psuvax1.psu.edu> schwartz@gondor.psu.edu (Scott E. Schwartz) writes: >given: > main() > { > int a,b,signed; > > a = b = signed = 1; > } [is this legal? PCC doesn't complain, but gcc does] >Disclaimer: I don't have a copy of the latest ANSI draft, or even an old >copy of K&R at hand. The ANSI draft would work better; gcc is an ANSI C compiler, and ANSI created a modifier to be the opposite of unsigned, signed. gcc rightly complains, and pcc rightly doesn't. >-- Scott Schwartz schwartz@gondor.psu.edu ----- Sean Eric Fagan Office of Computing/Communications Resources (213) 852 5742 Suite 2600 1GTLSEF@CALSTATE.BITNET 5670 Wilshire Boulevard Los Angeles, CA 90036 {litvax, rdlvax, psivax, hplabs, ihnp4}!csun!aeusesef