view src/s/sunos4-0.h @ 4407:4ee73bbe4f8e

Always use boyer_moore in ASCII or Latin-1 buffers with ASCII search strings. 2007-12-26 Aidan Kehoe <kehoea@parhasard.net> * casetab.c: Extend and correct some case table documentation. * search.c (search_buffer): Correct a bug where only the first entry for a character in the case equivalence table was examined in determining if the Boyer-Moore search algorithm is appropriate. If there are case mappings outside of the charset and row of the characters specified in the search string, those case mappings can be safely ignored (and Boyer-Moore search can be used) if we know from the buffer statistics that the corresponding characters cannot occur. * search.c (boyer_moore): Assert that we haven't been passed a string with varying characters sets or rows within character sets. That's what simple_search is for. In the very rare event that a character in the search string has a canonical case mapping that is not in the same character set and row, don't try to search for the canonical character, search for some other character that is in the the desired character set and row. Assert that the case table isn't corrupt. Do not search for any character case mappings that cannot possibly occur in the buffer, given the buffer metadata about its contents.
author Aidan Kehoe <kehoea@parhasard.net>
date Wed, 26 Dec 2007 17:30:16 +0100
parents e9a3f8b4de53
children
line wrap: on
line source

/* Synched up with: FSF 19.31. */

/* For building XEmacs under SunOS 4.1.* with static libraries. */

#ifndef _S_SUNOS4_H_
#define _S_SUNOS4_H_

#include "bsd4-2.h"

#ifndef SUNOS4
#define SUNOS4
#endif

#if 0  /* This may have been needed for an earlier version of Sun OS 4.
	  It seems to cause warnings in 4.0.3 and 4.1.  */
#define O_NDELAY        FNDELAY /* Non-blocking I/O (4.2 style) */
#endif

#ifdef NOT_C_CODE
  /* The new sunOS unexec eliminates the need for a custom crt0.o, so we
     can just let the compiler invoke the linker and don't have to guess
     what options it might have passed it. */
# define ORDINARY_LINK
# define START_FILES
# define LD_CMD $(CC)
# ifndef LD_SWITCH_SYSTEM
# define LD_SWITCH_SYSTEM "-Bstatic"
# endif
# define UNEXEC "unexsunos4.o"
#endif /* NOT_C_CODE */

#define RUN_TIME_REMAP

/* these don't matter, but we have to define something to keep
   sysdep.c from introducing bogus symbols */
#define TEXT_START 0
#define DATA_START 0

/* #### XEmacs: #define of SYSTEM_MALLOC removed.  Is this OK?  FSF says:

   In SunOS 4.1, a static function called by tzsetwall reportedly
   clears the byte just past an eight byte region it mallocs, corrupting
   GNU malloc's memory pool.  But Sun's malloc doesn't seem to mind. */

/* XEmacs: additions for proper prototyping. */
#ifndef NOT_C_CODE
#ifdef __STDC__
/* Sun's headers are categorically losing.
   Mly uses broken-sun.h to get the protos for this, but lcc provides all
   of the prototypes for the ANSI routines.  So I'm just going to put the
   protos of the non-ANSI routines that we use here (I guess that would
   be things that are Posix but not ANSI?)  You're in a maze of twisty
   little standards, all alike...
 */
/* Since lcc is not going to be heavily used anymore if it ever was, I'm
   putting broken-sun.h back in. */
/* Since Gcc 2.8 appears to have fixed the problem, I'm conditionalizing */
/* this ugly hack. */

#if defined (__GNUC__)
#if defined (__GNUC_MINOR__)
#if ((__GNUC__ == 2) && (__GNUC_MINOR__ > 7)) || ((__GNUC__ > 2))
/* Don't include for gcc 2.8.0*/
#else
#include "../broken-sun.h"
#endif

#else /* __GNUC_MINOR__ is undefined */
#include "../broken-sun.h"
#endif

#else
/* Not GNU C */
#endif

extern char *strdup ();
extern char *ttyname (int);
extern void tzsetwall (void);
extern int getpagesize (void);

#include <memory.h>
#ifdef __SUNPRO_C
/* Suppress zillions of warnings from outdated SunOS4 prototypes */
/* Bother! Sun can't even get the arg types right. */
#include <string.h>
#define memset(a,b,c) memset((char*) (a), b, c)
#define memcpy(a,b,c) memcpy((char*) (a), (char*) (b), c)
#define memcmp(a,b,c) memcmp((char*) (a), (char*) (b), c)
#define memchr(a,b,c) memchr((char*) (a), b, c)
void * __builtin_alloca(int);
#ifdef HAVE_X_WINDOWS
#include <X11/Xlib.h>
#define XFree(p) XFree((char*)(p))
#endif /* X Windows */
#endif /* __SUNPRO_C */

#endif /* __STDC__ */

# ifdef __GNUC__
  /* XEmacs addition: */
  /* gcc has the bug that it claims to conform to the ANSI C standard
     (which is what setting __STDC__ to 1 means) but does not necessarily
     provide all of the library routines which the standard requires of a
     conforming compiler -- such as memmove.  The other Sun ANSI compilers
     (Sun's acc and Lucid's lcc) do not have this bug. */
#  define memmove(to, from, size) bcopy ((char *) (from), (char *) (to), (size))
/* We must define mkdir with this arg prototype
   to match GCC's fixed stat.h.  */
#  define MKDIR_PROTOTYPE \
  int mkdir (const char *dpath, unsigned short dmode)
# endif /* __GNUC__ */

/* ANSI C requires that realloc accept a null pointer argument,
   but ancient implementations such as SunOS 4 don't allow this.
   We redefine realloc here so that the source code can be written to
   use the ANSI C API. */
#include <sys/types.h>
#ifdef __GNUC__
inline	/* Suppress warning: realloc_accepting_nullptr defined but not used */
#endif
static void*
realloc_accepting_nullptr (void *ptr, size_t size)
{
  extern char *realloc ();
  extern char *malloc ();
  return ptr ? (void *) realloc (ptr, size) : (void *) malloc (size);
}
#define realloc(ptr, size) realloc_accepting_nullptr (ptr, size)

#endif /* C_CODE */

#endif /* _S_SUNOS4_H_ */