Re: [bug-gnulib] [PATCH]: mbchar.h and iswcntrl

[Bruno Haible]

Yoann Vandoorselaere wrote:
> > Just for info: Which system is this which has a <wchar.h> and a <wctype.h>
> > header file but no iswcntrl function? The isw* functions are the main
> > contents of <wctype.h>. I don't expect a system to have <wctype.h>
> > but lack the functions.
> 
> FreeBSD 4.x got the wctype.h header, however, the header start with #if
> 0 and end with #endif directive. :-)

How insane! And it affects all releases from FreeBSD 4.4 to 4.11. [1] [2]
I'm committing the appended fix. (There's not only iswcntrl which needs
a substitute, but also all others, because of mb_isalpha..mb_isxdigit.)

[1] http://www.freebsd.org/cgi/cvsweb.cgi/src/include/wctype.h
[2] http://cvsup.pt.freebsd.org/cgi-bin/cvsweb/cvsweb.cgi/src/include/wctype.h

2006-07-28  Bruno Haible  <address@hidden>

        * m4/mbchar.m4 (gl_MBCHAR): Also test for iswcntrl.
        * lib/mbchar.h (iswalnum, iswalpha, iswblank, iswcntrl, iswdigit, 
iswgraph,
        iswlower, iswprint, iswpunct, iswspace, iswupper, iswxdigit): Define
        fallbacks.
        Avoids link error on FreeBSD 4.x.
        Reported by Yoann Vandoorselaere <address@hidden>.

        * lib/wcwidth.h (iswprint): Assume an ASCII compatible wide character
        encoding.
        * lib/mbswidth.c (iswcntrl): Likewise.

*** m4/mbchar.m4        26 Sep 2005 13:58:51 -0000      1.2
--- m4/mbchar.m4        28 Jul 2006 15:14:39 -0000
***************
*** 1,5 ****
! # mbchar.m4 serial 2
! dnl Copyright (C) 2005 Free Software Foundation, Inc.
  dnl This file is free software; the Free Software Foundation
  dnl gives unlimited permission to copy and/or distribute it,
  dnl with or without modifications, as long as this notice is preserved.
--- 1,5 ----
! # mbchar.m4 serial 3
! dnl Copyright (C) 2005-2006 Free Software Foundation, Inc.
  dnl This file is free software; the Free Software Foundation
  dnl gives unlimited permission to copy and/or distribute it,
  dnl with or without modifications, as long as this notice is preserved.
***************
*** 16,20 ****
--- 16,22 ----
    dnl Compile mbchar.c only if HAVE_WCHAR_H && HAVE_WCTYPE_H.
    if test $ac_cv_header_wchar_h = yes && test $ac_cv_header_wctype_h = yes; 
then
      AC_LIBOBJ([mbchar])
+     dnl Prerequisites of mbchar.h and mbchar.c.
+     AC_CHECK_FUNCS([iswcntrl])
    fi
  ])
*** lib/mbchar.h        28 Jun 2006 13:11:03 -0000      1.3
--- lib/mbchar.h        28 Jul 2006 15:14:39 -0000
***************
*** 157,162 ****
--- 157,271 ----
  #include <wchar.h>
  
  #include <wctype.h>
+ /* FreeBSD 4.4 to 4.11 has <wctype.h> but lacks the functions.
+    Assume all 12 functions are implemented the same way, or not at all.  */
+ #if !defined iswalnum && !HAVE_ISWCNTRL
+ ststic inline int
+ iswalnum (wint_t wc)
+ {
+   return (wc >= 0 && wc < 128
+         ? (wc >= '0' && wc <= '9') || ((wc & ~0x20) >= 'A' && (wc & ~0x20) <= 
'Z')
+         : 0);
+ }
+ #endif
+ #if !defined iswalpha && !HAVE_ISWCNTRL
+ ststic inline int
+ iswalpha (wint_t wc)
+ {
+   return (wc >= 0 && wc < 128
+         ? (wc & ~0x20) >= 'A' && (wc & ~0x20) <= 'Z'
+         : 0);
+ }
+ #endif
+ #if !defined iswblank && !HAVE_ISWCNTRL
+ ststic inline int
+ iswblank (wint_t wc)
+ {
+   return (wc >= 0 && wc < 128
+         ? wc == ' ' || wc == '\t'
+         : 0);
+ }
+ #endif
+ #if !defined iswcntrl && !HAVE_ISWCNTRL
+ ststic inline int
+ iswcntrl (wint_t wc)
+ {
+   return (wc >= 0 && wc < 128
+         ? (wc & ~0x1f) == 0 || wc == 0x7f
+         : 0);
+ }
+ #endif
+ #if !defined iswdigit && !HAVE_ISWCNTRL
+ ststic inline int
+ iswdigit (wint_t wc)
+ {
+   return (wc >= '0' && wc <= '9');
+ }
+ #endif
+ #if !defined iswgraph && !HAVE_ISWCNTRL
+ ststic inline int
+ iswgraph (wint_t wc)
+ {
+   return (wc >= 0 && wc < 128
+         ? wc >= '!' && wc <= '~'
+         : 1);
+ }
+ #endif
+ #if !defined iswlower && !HAVE_ISWCNTRL
+ ststic inline int
+ iswlower (wint_t wc)
+ {
+   return (wc >= 0 && wc < 128
+         ? wc >= 'a' && wc <= 'z'
+         : 0);
+ }
+ #endif
+ #if !defined iswprint && !HAVE_ISWCNTRL
+ ststic inline int
+ iswprint (wint_t wc)
+ {
+   return (wc >= 0 && wc < 128
+         ? wc >= ' ' && wc <= '~'
+         : 1);
+ }
+ #endif
+ #if !defined iswpunct && !HAVE_ISWCNTRL
+ ststic inline int
+ iswpunct (wint_t wc)
+ {
+   return (wc >= 0 && wc < 128
+         ? wc >= '!' && wc <= '~'
+           && !((wc >= '0' && wc <= '9')
+                || ((wc & ~0x20) >= 'A' && (wc & ~0x20) <= 'Z'))
+         : 1);
+ }
+ #endif
+ #if !defined iswspace && !HAVE_ISWCNTRL
+ ststic inline int
+ iswspace (wint_t wc)
+ {
+   return (wc >= 0 && wc < 128
+         ? wc == ' ' || wc == '\t'
+           || wc == '\n' || wc == '\v' || wc == '\f' || wc == '\r'
+         : 0);
+ }
+ #endif
+ #if !defined iswupper && !HAVE_ISWCNTRL
+ ststic inline int
+ iswupper (wint_t wc)
+ {
+   return (wc >= 0 && wc < 128
+         ? wc >= 'A' && wc <= 'Z'
+         : 0);
+ }
+ #endif
+ #if !defined iswxdigit && !HAVE_ISWCNTRL
+ ststic inline int
+ iswxdigit (wint_t wc)
+ {
+   return (wc >= '0' && wc <= '9') || ((wc & ~0x20) >= 'A' && (wc & ~0x20) <= 
'F');
+ }
+ #endif
  
  #include "wcwidth.h"
  
*** lib/wcwidth.h       28 Jun 2006 17:03:53 -0000      1.2
--- lib/wcwidth.h       28 Jul 2006 15:14:39 -0000
***************
*** 36,42 ****
  #  include <wctype.h>
  # endif
  # if !defined iswprint && !HAVE_ISWPRINT
! #  define iswprint(wc) 1
  # endif
  
  # ifndef HAVE_DECL_WCWIDTH
--- 36,48 ----
  #  include <wctype.h>
  # endif
  # if !defined iswprint && !HAVE_ISWPRINT
! ststic inline int
! iswprint (wint_t wc)
! {
!   return (wc >= 0 && wc < 128
!         ? wc >= ' ' && wc <= '~'
!         : 1);
! }
  # endif
  
  # ifndef HAVE_DECL_WCWIDTH
*** lib/mbswidth.c      4 Jul 2006 16:46:23 -0000       1.17
--- lib/mbswidth.c      28 Jul 2006 15:14:39 -0000
***************
*** 51,57 ****
  # include <wctype.h>
  #endif
  #if !defined iswcntrl && !HAVE_ISWCNTRL
! # define iswcntrl(wc) 0
  #endif
  
  #ifndef mbsinit
--- 51,57 ----
  # include <wctype.h>
  #endif
  #if !defined iswcntrl && !HAVE_ISWCNTRL
! # define iswcntrl(wc) (((wc) & ~0x1f) == 0 || (wc) == 0x7f)
  #endif
  
  #ifndef mbsinit