zsh-workers
 help / color / mirror / code / Atom feed
* Beta 18 Compilation Problem Under Nextstep 3.3
@ 1996-05-21 21:47 Irving Wolfe
  1996-05-21 23:36 ` Mark Borges
  0 siblings, 1 reply; 3+ messages in thread
From: Irving Wolfe @ 1996-05-21 21:47 UTC (permalink / raw)
  To: zsh-workers; +Cc: i-a-o

cc -c -I.. -I. -I.  -DHAVE_CONFIG_H -Wall -Wno-implicit -Wmissing-prototypes -O -pipe builtin.c
builtin.c: In function `bin_kill':
builtin.c:662: `sigs' undeclared (first use this function)
builtin.c:662: (Each undeclared identifier is reported only once
builtin.c:662: for each function it appears in.)
builtin.c:662: warning: format argument is not a pointer (arg 2)
builtin.c:664: warning: format argument is not a pointer (arg 2)
builtin.c: In function `bin_trap':
builtin.c:5257: `sigs' undeclared (first use this function)
builtin.c:5257: warning: format argument is not a pointer (arg 2)
builtin.c:5260: warning: format argument is not a pointer (arg 2)
builtin.c: In function `getsignum':
builtin.c:5507: `sigs' undeclared (first use this function)
*** Exit 1
Stop.




^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Beta 18 Compilation Problem Under Nextstep 3.3
  1996-05-21 21:47 Beta 18 Compilation Problem Under Nextstep 3.3 Irving Wolfe
@ 1996-05-21 23:36 ` Mark Borges
  1996-05-22  0:45   ` Johan Danielsson
  0 siblings, 1 reply; 3+ messages in thread
From: Mark Borges @ 1996-05-21 23:36 UTC (permalink / raw)
  To: Irving Wolfe; +Cc: zsh-workers

>> On Tue, 21 May 1996 14:47:13 -0700 (PDT),
>> Irving Wolfe(IW) wrote:
IW> cc -c -I.. -I. -I.  -DHAVE_CONFIG_H -Wall -Wno-implicit -Wmissing-prototypes -O -pipe builtin.c
IW> builtin.c: In function `bin_kill':
IW> builtin.c:662: `sigs' undeclared (first use this function)

I had the same problem under Solaris-2.5 (but it's not an OS bug).

I tracked it down to the change made to signames.awk, which on my
system is interpreted by mawk,

  $ mawk -Wv
  mawk 1.2.2 Jan 1996, Copyright (C) Michael D. Brennan

which generated a bogus signames.h file.

Anyway, I backed out the change (why is it necessary to escape the
`#else'?) and it worked fine.

The only thing SUNWspro's cc complains about now is stuff like:

----------------------------------------
"hashtable.c", line 233: warning: argument #1 is incompatible with prototype:
	prototype: pointer to const uchar : "./utils.pro", line 85
	argument : pointer to char
----------------------------------------

Good luck.

*** signames.awk	Mon May 20 06:40:00 1996
--- ../../zsh-2.6-beta17/Src/signames.awk	Mon May  6 08:08:58 1996
***************
*** 1,5 ****
  # 
! # $Id: signames.awk,v 2.1 1996/05/20 01:24:24 hzoli Exp $
  #
  # {g,n}awk script to generate signals.h
  # provided by Geoff Wing <mason@werple.apana.org.au>
--- 1,5 ----
  # 
! # $Id: signames.awk,v 2.0 1996/05/02 22:57:04 hzoli Exp $
  #
  # {g,n}awk script to generate signals.h
  # provided by Geoff Wing <mason@werple.apana.org.au>
***************
*** 52,58 ****
  
  END {
      ps = "%s"
!     ifdstr = sprintf("#ifdef USE_SUSPENDED\n\t%csuspended%s%c,\n\#else\n\t%cstopped%s%c,\n#endif\n", 034, ps, 034, 034, ps, 034)
  
      printf("%s\n%s\n\n%s\t%d\n\n%s\n\n%s\n\t%c%s%c,\n", "/** signals.h                                 **/", "/** architecture-customized signals.h for zsh **/", "#define SIGCOUNT", max, "#ifdef GLOBALS", "char *sigmsg[SIGCOUNT+2] = {", 034, "done", 034)
  
--- 52,58 ----
  
  END {
      ps = "%s"
!     ifdstr = sprintf("#ifdef USE_SUSPENDED\n\t%csuspended%s%c,\n#else\n\t%cstopped%s%c,\n#endif\n", 034, ps, 034, 034, ps, 034)
  
      printf("%s\n%s\n\n%s\t%d\n\n%s\n\n%s\n\t%c%s%c,\n", "/** signals.h                                 **/", "/** architecture-customized signals.h for zsh **/", "#define SIGCOUNT", max, "#ifdef GLOBALS", "char *sigmsg[SIGCOUNT+2] = {", 034, "done", 034)
  
***************
*** 87,93 ****
      print "\tNULL"
      print "};"
      print ""
!     print "\#else"
      print "extern char *sigs[SIGCOUNT+4],*sigmsg[SIGCOUNT+2];"
      print "#endif"
  }
--- 87,93 ----
      print "\tNULL"
      print "};"
      print ""
!     print "#else"
      print "extern char *sigs[SIGCOUNT+4],*sigmsg[SIGCOUNT+2];"
      print "#endif"
  }

-- 
  -mb-



^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Beta 18 Compilation Problem Under Nextstep 3.3
  1996-05-21 23:36 ` Mark Borges
@ 1996-05-22  0:45   ` Johan Danielsson
  0 siblings, 0 replies; 3+ messages in thread
From: Johan Danielsson @ 1996-05-22  0:45 UTC (permalink / raw)
  To: Mark Borges; +Cc: zsh-workers

Mark Borges <mdb@cdc.noaa.gov> writes:

> Anyway, I backed out the change (why is it necessary to escape the
> `#else'?) and it worked fine.

Hmm, it was my patch for UNICOS, that somehow has a bug in awk.

$ uname -a
kallsup kallsup 9.0.2.0 roo.2 CRAY J90
$ echo | awk '{ print "#else" }'
bad switch yylook 79

Older awks printed a literal backslash when it wasn't followed by a
"legal" character, newer awks removes the backslash.

$ uname -a
SunOS sunsite.nada.kth.se 5.4 Generic_101945-32 sun4d sparc
$ echo | awk '{ print "\z" }' 
\z
$ echo | nawk '{ print "\z" }' 
z
$ echo | gawk '{ print "\z" }'
z

/Johan



^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~1996-05-22  1:04 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
1996-05-21 21:47 Beta 18 Compilation Problem Under Nextstep 3.3 Irving Wolfe
1996-05-21 23:36 ` Mark Borges
1996-05-22  0:45   ` Johan Danielsson

Code repositories for project(s) associated with this public inbox

	https://git.vuxu.org/mirror/zsh/

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).