[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: LYNX-DEV Solaris 2.5 curses problem compiling lynx 2.6
From: |
Eli's redistribution point |
Subject: |
Re: LYNX-DEV Solaris 2.5 curses problem compiling lynx 2.6 |
Date: |
Wed, 29 Jan 97 15:46 EST |
Jeff Zucker <address@hidden> wrote:
> I'm having troubles compiling Lynx 2.6 on Solaris 2.5 with gcc.
> /opt/gnu/bin/gcc -g -DFANCY_CURSES -DUNIX -DSVR4 -DSOLARIS2 -DCURS_PERFOR
> MANCE -DUTMPX_FOR_UTMP -DUSE_DIRENT -DLOCALE -DHAVE_TERMIOS_H -I../WWW
> /Library/Implementation -DDIRED_SUPPORT -DOK_TAR -DOK_GZIP -DOK_UUDECO
> DE -DOK_OVERRIDE -DOK_PERMIT -I.. -c LYClean.c
> In file included from LYCurses.h:50,
> from LYClean.c:3:
> /usr/include/curses.h:321: parse error before `__va_list'
> /usr/include/curses.h:322: parse error before `__va_list'
Either add an #include for <stdarg.h> (or <varargs.h>) OR change the
function prototypes to use void* for the last argument on those
two lines.
The void * hack has the advantage of being a more general solution,
as some curses programs may use a varargs.h which doesn't agree
with the stdarg.h stuff.
It is times like this that strict type checking really gets on my
nerves. K&R C is so much better at DWIMNWIS (Do what I mean, not
what I say). There was a case at work where I added a single C
function to an other entirely C++ project because I couldn't figure
out the proper cast to get a really ugly memory reference to work.
(It was something like a char* -- which should have been a void* --
pointing to an array of unions of structures of which I wanted a
char** member from one of the structures. Like I said, ugly.)
Elijah
------
please do not CC me when replying to the list
;
; To UNSUBSCRIBE: Send a mail message to address@hidden
; with "unsubscribe lynx-dev" (without the
; quotation marks) on a line by itself.
;