This is the mail archive of the gdb-patches@sourceware.cygnus.com mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

[eliz@is.elta.co.il: Problem with gnu-regex.c when wctype.h is available]


------- Start of forwarded message -------
Date: Sun, 27 Jun 1999 18:49:48 +0300 (IDT)
From: Eli Zaretskii <eliz@is.elta.co.il>
X-Sender: eliz@is
To: gdb-patches@sourceware.cygnus.com
cc: DJ Delorie <dj@delorie.com>
Subject: Problem with gnu-regex.c when wctype.h is available
Content-Type: TEXT/PLAIN; charset=US-ASCII

If wctype.h and wchar.h are available, but btowc is not (DJGPP is a case 
in point), gnu-regex.c causes link errors.  Here's a patch:

1999-06-26  Eli Zaretskii  <eliz@is.elta.co.il>

	* gnu-regex.c (CHAR_CLASS_MAX_LENGTH): Don't use wide characters
	unless HAVE_BTOWC is defined.

*** cvs/gdb/gnu-regex.c	Fri Apr 16 04:34:00 1999
- --- ./gdb/gnu-regex.c	Mon May 17 17:21:56 1999
***************
*** 1700,1706 ****
         } 								\
      }
  
! #if defined _LIBC || (defined HAVE_WCTYPE_H && defined HAVE_WCHAR_H)
  /* The GNU C library provides support for user-defined character classes
     and the functions from ISO C amendement 1.  */
  # ifdef CHARCLASS_NAME_MAX
- --- 1700,1710 ----
         } 								\
      }
  
! /* Use this only if they have btowc(), since wctype() is used below
!    together with btowc().  btowc() is defined in the 1994 Amendment 1
!    to ISO C and may not be present on systems where we have wchar.h
!    and wctype.h.   */
! #if defined _LIBC || (defined HAVE_WCTYPE_H && defined HAVE_WCHAR_H && defined HAVE_BTOWC)
  /* The GNU C library provides support for user-defined character classes
     and the functions from ISO C amendement 1.  */
  # ifdef CHARCLASS_NAME_MAX
------- End of forwarded message -------

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]