/*
Copyright (C) 2004, 2005 John E. Davis

This file is part of the S-Lang Library.

The S-Lang Library is free software; you can redistribute it and/or
modify it under the terms of the GNU General Public License as
published by the Free Software Foundation; either version 2 of the
License, or (at your option) any later version.

The S-Lang Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
General Public License for more details.

You should have received a copy of the GNU General Public License
along with this library; if not, write to the Free Software
Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307,
USA.  
*/

#define _GNU_SOURCE
#include "slinclud.h"

#include <ctype.h>

#include "slang.h"
#include "_slang.h"


char *SLmake_string(char *str)
{
   return SLmake_nstring(str, strlen (str));
}

char *SLmake_nstring (char *str, unsigned int n)
{
   char *ptr;

   if (NULL == (ptr = SLmalloc(n + 1)))
     {
	return NULL;
     }
   SLMEMCPY (ptr, str, n);
   ptr[n] = 0;
   return(ptr);
}

void SLmake_lut (unsigned char *lut, unsigned char *range, unsigned char reverse)
{
   /* register unsigned char *l = lut, *lmax = lut + 256; */
   int i, r1, r2;

   memset ((char *)lut, reverse, 256);
   /* while (l < lmax) *l++ = reverse; */
   reverse = !reverse;

   r1 = *range++;
   while (r1)
     {
	r2 = *range++;
	if ((r2 == '-') && (*range != 0))
	  {
	     r2 = *range++;
	     for (i = r1; i <= r2; i++) lut[i] = reverse;
	     r1 = *range++;
	     continue;
	  }
	lut[r1] = reverse;
	r1 = r2;
     }
}


char *_pSLskip_whitespace (char *s)
{
   while (isspace (*s))
     s++;
   
   return s;
}

/*
 * This function assumes that the initial \ char has been removed.
 */
char *_pSLexpand_escaped_char(char *p, SLwchar_Type *ch, int *isunicodep)
{
   int i = 0;
   SLwchar_Type max = 0;
   SLwchar_Type num, base = 0;
   SLwchar_Type ch1;
   int isunicode;

   ch1 = *p++;
   isunicode = 0;

   switch (ch1)
     {
      default: num = ch1; break;
      case 'n': num = '\n'; break;
      case 't': num = '\t'; break;
      case 'v': num = '\v'; break;
      case 'b': num = '\b'; break;
      case 'r': num = '\r'; break;
      case 'f': num = '\f'; break;
      case 'E': case 'e': num = 27; break;
      case 'a': num = 7;
	break;

	/* octal */
      case '0': case '1': case '2': case '3':
      case '4': case '5': case '6': case '7':
	max = '7';
	base = 8; i = 2; num = ch1 - '0';
	break;

      case 'd':			       /* decimal -- S-Lang extension */
	base = 10;
	i = 3;
	max = '9';
	num = 0;
	break;

      case 'x':			       /* hex */
	base = 16;
	max = '9';
	i = 2;
	num = 0;
	
	if (*p == '{')
	  {
	     p++;
	     i = 0;
	     while (p[i] && (p[i] != '}'))
	       i++;
	     if (p[i] != '}')
	       {
		  SLang_verror (SL_UNICODE_ERROR, "Escaped unicode character missing closing }.");
		  return NULL;
	       }
	     isunicode = 1;
	  }
	break;
     }

   while (i)
     {
	ch1 = *p;

	i--;

	if ((ch1 <= max) && (ch1 >= '0'))
	  {
	     num = base * num + (ch1 - '0');
	  }
	else if (base == 16)
	  {
	     ch1 |= 0x20;
	     if ((ch1 < 'a') || ((ch1 > 'f'))) break;
	     num = base * num + 10 + (ch1 - 'a');
	  }
	else break;
	p++;
     }

   if (isunicode)
     {
	if (*p != '}')
	  {
	     SLang_verror (SL_UNICODE_ERROR, "Malformed Escaped unicode character.");
	     return NULL;
	  }
	p++;
     }

   if (isunicodep != NULL)
     *isunicodep = isunicode;

   *ch = num;
   return p;
}

/* s and t could represent the same space.  Note: the space for s is
 * assumed to be at least tmax-t bytes.
 */
int SLexpand_escaped_string (char *s, char *t, char *tmax, int utf8_encode)
{
   if (utf8_encode == -1)
     utf8_encode = _pSLinterp_UTF8_Mode;

   while (t < tmax)
     {
	SLwchar_Type wch;
	char *s1;
	int isunicode;
	char ch = *t++;

	if (ch != '\\')
	  {
	     *s++ = ch;
	     continue;
	  }
	
	if (NULL == (t = _pSLexpand_escaped_char (t, &wch, &isunicode)))
	  {
	     *s = 0;
	     return -1;
	  }

	/* As discussed above, the escaped representation is larger than the
	 * encoding.
	 */
	if ((isunicode == 0)
#if 0
	    && ((wch < 127)
		|| (utf8_encode == 0))
#endif
	    )
	  {
	     *s++ = (char) wch;
	     continue;
	  }

	s1 = (char *)SLutf8_encode (wch, (SLuchar_Type *)s, 6);
	if (s1 == NULL)
	  {
	     SLang_verror (SL_INVALID_UTF8, "Unable to UTF-8 encode 0x%lX\n", (unsigned long)wch);
	     *s = 0;
	     return -1;
	  }
	s = s1;
     }
   *s = 0;
   return 0;
}

int SLextract_list_element (char *list, unsigned int nth, char delim,
			    char *elem, unsigned int buflen)
{
   char *el, *elmax;
   char ch;

   while (nth > 0)
     {
	while ((0 != (ch = *list)) && (ch != delim))
	  list++;

	if (ch == 0) return -1;

	list++;
	nth--;
     }

   el = elem;
   elmax = el + (buflen - 1);

   while ((0 != (ch = *list)) && (ch != delim) && (el < elmax))
     *el++ = *list++;
   *el = 0;

   return 0;
}

int SLvsnprintf (char *buf, unsigned int buflen, char *fmt, va_list ap)
{
#ifdef HAVE_VSNPRINTF
   int status;

   status = vsnprintf(buf, buflen, fmt, ap);
   if ((unsigned int)status >= buflen) 
     status = -1;
   return status;
#else
   unsigned int len;

   /* On some systems vsprintf returns useless information.  So, punt */
   vsprintf (buf, fmt, ap);
   len = strlen (buf);
   if (len >= buflen)
     {
	SLang_exit_error ("\
Your system lacks the vsnprintf system call and vsprintf overflowed a buffer.\n\
The integrity of this program has been violated.\n");
	return EOF;		       /* NOT reached */
     }
   return (int)len;
#endif
}

int SLsnprintf (char *buf, unsigned int buflen, char *fmt, ...)
{
   int status;
   va_list ap;

   va_start (ap, fmt);
   status = SLvsnprintf (buf, buflen, fmt, ap);
   va_end (ap);

   return status;
}

typedef struct _Cleanup_Function_Type
{
   struct _Cleanup_Function_Type *next;
   void (*f)(void);
}
Cleanup_Function_Type;

static Cleanup_Function_Type *Cleanup_Function_List;

static void cleanup_slang (void)
{
   while (Cleanup_Function_List != NULL)
     {
	Cleanup_Function_Type *next = Cleanup_Function_List->next;
	(*Cleanup_Function_List->f)();
	SLfree ((char *) Cleanup_Function_List);
	Cleanup_Function_List = next;
     }
}

#ifndef HAVE_ATEXIT
# ifdef HAVE_ON_EXIT
static void on_exit_cleanup_slang (int arg_unused)
{
   (void) arg_unused;
   cleanup_slang ();
}
# endif
#endif

int SLang_add_cleanup_function (void (*f)(void))
{
   Cleanup_Function_Type *l;

   l = (Cleanup_Function_Type *) SLmalloc (sizeof (Cleanup_Function_Type));
   if (l == NULL)
     return -1;

   l->f = f;
   l->next = Cleanup_Function_List;

   if (Cleanup_Function_List == NULL)
     {
#ifdef HAVE_ATEXIT
	(void) atexit (cleanup_slang);
#else
# ifdef HAVE_ON_EXIT
	(void) on_exit (on_exit_cleanup_slang, 0);
# endif
#endif
     }
   Cleanup_Function_List = l;
   return 0;
}


int SLang_guess_type (char *t)
{
   char *p;
   register char ch;
   int modifier;
#define MODIFIER_H	0x01
#define MODIFIER_L	0x02
#define MODIFIER_U	0x04
#define MODIFIER_LL	0x08
#define MODIFIER_SIZE_MASK 0x0F
#define MODIFIER_HEX	0x10

   modifier = 0;
   if (*t == '-') t++;
   p = t;

#if SLANG_HAS_FLOAT
   if (*p != '.')
     {
#endif
	while ((*p >= '0') && (*p <= '9')) p++;
	if (t == p) return (SLANG_STRING_TYPE);
	if ((*p == 'x') && (p == t + 1))   /* 0x?? */
	  {
	     modifier |= MODIFIER_HEX;
	     p++;
	     while (ch = *p,
		    ((ch >= '0') && (ch <= '9'))
		    || (((ch | 0x20) >= 'a') && ((ch | 0x20) <= 'f'))) p++;
	  }

	/* Now look for U, H, L, LL, UH, UL, ULL, modifiers */
	ch = *p | 0x20;
	if (ch == 'u')
	  {
	     modifier |= MODIFIER_U;
	     p++;
	     ch = *p | 0x20;
	  }
	if (ch == 'h')
	  {
	     modifier |= MODIFIER_H;
	     p++;
	     ch = *p | 0x20;
	  }
	else if (ch == 'l')
	  {
	     p++;
	     ch = *p | 0x20;
	     if (ch == 'l')
	       {
		  modifier |= MODIFIER_LL;
		  p++;
		  ch = *p | 0x20;
	       }
	     else
	       modifier |= MODIFIER_L;
	  }
	if ((ch == 'u') && (0 == (modifier & MODIFIER_U)))
	  {
	     modifier |= MODIFIER_U;
	     p++;
	  }

	if (*p == 0) switch (modifier & MODIFIER_SIZE_MASK)
	  {
	   case 0:
	     return SLANG_INT_TYPE;
	   case MODIFIER_H:
	     return SLANG_SHORT_TYPE;
	   case MODIFIER_L:
	     return SLANG_LONG_TYPE;
	   case MODIFIER_U:
	     return SLANG_UINT_TYPE;
	   case MODIFIER_U|MODIFIER_H:
	     return SLANG_USHORT_TYPE;
	   case MODIFIER_U|MODIFIER_L:
	     return SLANG_ULONG_TYPE;
	   case MODIFIER_LL:
	     return SLANG_LLONG_TYPE;
	   case MODIFIER_U|MODIFIER_LL:
	     return SLANG_ULLONG_TYPE;
	   default:
	     return SLANG_STRING_TYPE;
	  }
	     
	if (modifier)
	  return SLANG_STRING_TYPE;
#if SLANG_HAS_FLOAT
     }

   /* now down to double case */
   if (*p == '.')
     {
	p++;
	while ((*p >= '0') && (*p <= '9')) p++;
     }
   if (*p == 0) return(SLANG_DOUBLE_TYPE);
   if ((*p != 'e') && (*p != 'E'))
     {
# if SLANG_HAS_COMPLEX
	if (((*p == 'i') || (*p == 'j'))
	    && (p[1] == 0))
	  return SLANG_COMPLEX_TYPE;
# endif
	if (((*p | 0x20) == 'f') && (p[1] == 0))
	  return SLANG_FLOAT_TYPE;

	return SLANG_STRING_TYPE;
     }

   p++;
   if ((*p == '-') || (*p == '+')) p++;
   while ((*p >= '0') && (*p <= '9')) p++;
   if (*p != 0)
     {
# if SLANG_HAS_COMPLEX
	if (((*p == 'i') || (*p == 'j'))
	    && (p[1] == 0))
	  return SLANG_COMPLEX_TYPE;
# endif
	if (((*p | 0x20) == 'f') && (p[1] == 0))
	  return SLANG_FLOAT_TYPE;

	return SLANG_STRING_TYPE;
     }
   return SLANG_DOUBLE_TYPE;
#else
   return SLANG_STRING_TYPE;
#endif				       /* SLANG_HAS_FLOAT */
}

static int hex_atoul (unsigned char *s, unsigned long *ul)
{
   unsigned char ch;
   unsigned long value;
   int base;
   unsigned int num_processed;

   num_processed = 0;
   if (*s != '0')
     base = 10;
   else
     {
	s++;				       /* skip the leading 0 */

	/* look for 'x' which indicates hex */
	if ((*s | 0x20) == 'x')
	  {
	     base = 16;
	     s++;
	     if (*s == 0)
	       {
		  SLang_set_error (SL_SYNTAX_ERROR);
		  return -1;
	       }
	  }
	else 
	  {
	     base = 8;
	     num_processed++;
	  }
     }

   value = 0;
   while ((ch = *s++) != 0)
     {
	char ch1 = ch | 0x20;
	switch (ch1)
	  {
	   default:
	     SLang_set_error (SL_SYNTAX_ERROR);
	     return -1;

	   case 'u':
	   case 'l':
	   case 'h':
	     if (num_processed == 0)
	       {
		  SLang_set_error (SL_SYNTAX_ERROR);
		  return -1;
	       }
	     *ul = value;
	     return 0;

	   case '8':
	   case '9':
	     if (base == 8) 
	       {
		  SLang_verror (SL_SYNTAX_ERROR, "8 or 9 are not permitted in an octal number");
		  return -1;
	       }
	     /* drop */
	   case '0':
	   case '1':
	   case '2':
	   case '3':
	   case '4':
	   case '5':
	   case '6':
	   case '7':
	     ch1 -= '0';
	     break;

	   case 'a':
	   case 'b':
	   case 'c':
	   case 'd':
	   case 'e':
	   case 'f':
	     if (base != 16) 
	       {
		  SLang_verror (SL_SYNTAX_ERROR, "Only digits may appear in an octal or decimal number");
		  return -1;
	       }
	     ch1 = (ch1 - 'a') + 10;
	     break;
	  }
	value = value * base + ch1;
	num_processed++;
     }
   *ul = value;
   return 0;
}

#ifdef HAVE_LONG_LONG
static int hex_atoull (unsigned char *s, unsigned long long *ul)
{
   unsigned char ch;
   unsigned long long value;
   int base;

   if (*s != '0')
     base = 10;
   else
     {
	s++;				       /* skip the leading 0 */

	/* look for 'x' which indicates hex */
	if ((*s | 0x20) == 'x')
	  {
	     base = 16;
	     s++;
	     if (*s == 0)
	       {
		  SLang_set_error (SL_SYNTAX_ERROR);
		  return -1;
	       }
	  }
	else base = 8;
     }

   value = 0;
   while ((ch = *s++) != 0)
     {
	char ch1 = ch | 0x20;
	switch (ch1)
	  {
	   default:
	     SLang_set_error (SL_SYNTAX_ERROR);
	     return -1;

	   case 'u':
	   case 'l':
	     *ul = value;
	     return 0;

	   case '8':
	   case '9':
	     if (base == 8) 
	       {
		  SLang_verror (SL_SYNTAX_ERROR, "8 or 9 are not permitted in an octal number");
		  return -1;
	       }
	     /* drop */
	   case '0':
	   case '1':
	   case '2':
	   case '3':
	   case '4':
	   case '5':
	   case '6':
	   case '7':
	     ch1 -= '0';
	     break;

	   case 'a':
	   case 'b':
	   case 'c':
	   case 'd':
	   case 'e':
	   case 'f':
	     if (base != 16) 
	       {
		  SLang_verror (SL_SYNTAX_ERROR, "Only digits may appear in an octal or decimal number");
		  return -1;
	       }
	     ch1 = (ch1 - 'a') + 10;
	     break;
	  }
	value = value * base + ch1;
     }
   *ul = value;
   return 0;
}
#endif				       /* HAVE_LONG_LONG */
   
/* Note: These routines do not check integer overflow.  I would use the C
 * library functions atol and atoul but some implementations check overflow
 * and some do not.  The following implementations provide a consistent
 * behavior.
 */
static unsigned char *get_sign (unsigned char *s, int *signp)
{
   s = (unsigned char *) _pSLskip_whitespace ((char *)s);

   if (*s == '-') 
     {
	*signp = -1;
	s++;
     }
   else
     {
	*signp = 1;
	if (*s == '+') s++;
     }
   return s;
}

unsigned long SLatoul (unsigned char *s)
{
   int sign;
   unsigned long value;

   s = get_sign (s, &sign);
   if (-1 == hex_atoul (s, &value))
     return (unsigned long) -1;

   if (sign == -1)
     value = (unsigned long)-1L * value;

   return value;
}

long SLatol (unsigned char *s)
{
   int sign;
   unsigned long value;

   s = get_sign (s, &sign);
   if (-1 == hex_atoul (s, &value))
     return -1L;

   if (sign == -1)
     return (long)(-1L * value);

   return (long) value;
}

int SLatoi (unsigned char *s)
{
   return (int) SLatol (s);
}

#if HAVE_LONG_LONG
long long SLatoll (unsigned char *s)
{
   int sign;
   unsigned long long value;

   s = get_sign (s, &sign);
   if (-1 == hex_atoull (s, &value))
     return -1LL;

   if (sign == -1)
     return (long long)(-1LL * value);

   return (long long) value;
}

unsigned long long SLatoull (unsigned char *s)
{
   int sign;
   unsigned long long value;

   s = get_sign (s, &sign);
   if (-1 == hex_atoull (s, &value))
     return (unsigned long long) -1;

   if (sign == -1)
     return (unsigned long long)(-1LL * value);

   return value;
}

#endif