path: root/lib/util/util_str.c

/* 
   Unix SMB/CIFS implementation.
   Samba utility functions
   
   Copyright (C) Andrew Tridgell 1992-2001
   Copyright (C) Simo Sorce      2001-2002
   Copyright (C) Martin Pool     2003
   Copyright (C) James Peach	 2005
   
   This program 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 3 of the License, or
   (at your option) any later version.
   
   This program 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 program.  If not, see <http://www.gnu.org/licenses/>.
*/

#include "includes.h"
#include "system/locale.h"
#undef strncasecmp
#undef strcasemp

/**
 * @file
 * @brief String utilities.
 **/


/**
 Trim the specified elements off the front and back of a string.
**/
_PUBLIC_ bool trim_string(char *s, const char *front, const char *back)
{
	bool ret = false;
	size_t front_len;
	size_t back_len;
	size_t len;

	/* Ignore null or empty strings. */
	if (!s || (s[0] == '\0'))
		return false;

	front_len	= front? strlen(front) : 0;
	back_len	= back? strlen(back) : 0;

	len = strlen(s);

	if (front_len) {
		while (len && strncmp(s, front, front_len)==0) {
			/* Must use memmove here as src & dest can
			 * easily overlap. Found by valgrind. JRA. */
			memmove(s, s+front_len, (len-front_len)+1);
			len -= front_len;
			ret=true;
		}
	}
	
	if (back_len) {
		while ((len >= back_len) && strncmp(s+len-back_len,back,back_len)==0) {
			s[len-back_len]='\0';
			len -= back_len;
			ret=true;
		}
	}
	return ret;
}

/**
 Find the number of 'c' chars in a string
**/
_PUBLIC_ _PURE_ size_t count_chars(const char *s, char c)
{
	size_t count = 0;

	while (*s) {
		if (*s == c) count++;
		s ++;
	}

	return count;
}



/**
 Safe string copy into a known length string. maxlength does not
 include the terminating zero.
**/
_PUBLIC_ char *safe_strcpy(char *dest,const char *src, size_t maxlength)
{
	size_t len;

	if (!dest) {
		DEBUG(0,("ERROR: NULL dest in safe_strcpy\n"));
		return NULL;
	}

#ifdef DEVELOPER
	/* We intentionally write out at the extremity of the destination
	 * string.  If the destination is too short (e.g. pstrcpy into mallocd
	 * or fstring) then this should cause an error under a memory
	 * checker. */
	dest[maxlength] = '\0';
	if (PTR_DIFF(&len, dest) > 0) {  /* check if destination is on the stack, ok if so */
		log_suspicious_usage("safe_strcpy", src);
	}
#endif

	if (!src) {
		*dest = 0;
		return dest;
	}  

	len = strlen(src);

	if (len > maxlength) {
		DEBUG(0,("ERROR: string overflow by %u (%u - %u) in safe_strcpy [%.50s]\n",
			 (uint_t)(len-maxlength), (unsigned)len, (unsigned)maxlength, src));
		len = maxlength;
	}
      
	memmove(dest, src, len);
	dest[len] = 0;
	return dest;
}  

/**
 Safe string cat into a string. maxlength does not
 include the terminating zero.
**/
_PUBLIC_ char *safe_strcat(char *dest, const char *src, size_t maxlength)
{
	size_t src_len, dest_len;

	if (!dest) {
		DEBUG(0,("ERROR: NULL dest in safe_strcat\n"));
		return NULL;
	}

	if (!src)
		return dest;
	
#ifdef DEVELOPER
	if (PTR_DIFF(&src_len, dest) > 0) {  /* check if destination is on the stack, ok if so */
		log_suspicious_usage("safe_strcat", src);
	}
#endif
	src_len = strlen(src);
	dest_len = strlen(dest);

	if (src_len + dest_len > maxlength) {
		DEBUG(0,("ERROR: string overflow by %d in safe_strcat [%.50s]\n",
			 (int)(src_len + dest_len - maxlength), src));
		if (maxlength > dest_len) {
			memcpy(&dest[dest_len], src, maxlength - dest_len);
		}
		dest[maxlength] = 0;
		return NULL;
	}
	
	memcpy(&dest[dest_len], src, src_len);
	dest[dest_len + src_len] = 0;
	return dest;
}

/**
 Routine to get hex characters and turn them into a 16 byte array.
 the array can be variable length, and any non-hex-numeric
 characters are skipped.  "0xnn" or "0Xnn" is specially catered
 for.

 valid examples: "0A5D15"; "0x15, 0x49, 0xa2"; "59\ta9\te3\n"


**/
_PUBLIC_ size_t strhex_to_str(char *p, size_t p_len, const char *strhex, size_t strhex_len)
{
	size_t i;
	size_t num_chars = 0;
	uint8_t   lonybble, hinybble;
	const char     *hexchars = "0123456789ABCDEF";
	char           *p1 = NULL, *p2 = NULL;

	for (i = 0; i < strhex_len && strhex[i] != 0; i++) {
		if (strncasecmp(hexchars, "0x", 2) == 0) {
			i++; /* skip two chars */
			continue;
		}

		if (!(p1 = strchr(hexchars, toupper((unsigned char)strhex[i]))))
			break;

		i++; /* next hex digit */

		if (!(p2 = strchr(hexchars, toupper((unsigned char)strhex[i]))))
			break;

		/* get the two nybbles */
		hinybble = PTR_DIFF(p1, hexchars);
		lonybble = PTR_DIFF(p2, hexchars);

		if (num_chars >= p_len) {
			break;
		}

		p[num_chars] = (hinybble << 4) | lonybble;
		num_chars++;

		p1 = NULL;
		p2 = NULL;
	}
	return num_chars;
}

/** 
 * Parse a hex string and return a data blob. 
 */
_PUBLIC_ _PURE_ DATA_BLOB strhex_to_data_blob(TALLOC_CTX *mem_ctx, const char *strhex) 
{
	DATA_BLOB ret_blob = data_blob_talloc(mem_ctx, NULL, strlen(strhex)/2+1);

	ret_blob.length = strhex_to_str((char *)ret_blob.data, ret_blob.length,
					strhex,
					strlen(strhex));

	return ret_blob;
}


/**
 * Routine to print a buffer as HEX digits, into an allocated string.
 */
_PUBLIC_ void hex_encode(const unsigned char *buff_in, size_t len, char **out_hex_buffer)
{
	int i;
	char *hex_buffer;

	*out_hex_buffer = malloc_array_p(char, (len*2)+1);
	hex_buffer = *out_hex_buffer;

	for (i = 0; i < len; i++)
		slprintf(&hex_buffer[i*2], 3, "%02X", buff_in[i]);
}

/**
 * talloc version of hex_encode()
 */
_PUBLIC_ char *hex_encode_talloc(TALLOC_CTX *mem_ctx, const unsigned char *buff_in, size_t len)
{
	int i;
	char *hex_buffer;

	hex_buffer = talloc_array(mem_ctx, char, (len*2)+1);

	for (i = 0; i < len; i++)
		slprintf(&hex_buffer[i*2], 3, "%02X", buff_in[i]);

	return hex_buffer;
}

/**
 Unescape a URL encoded string, in place.
**/

_PUBLIC_ void rfc1738_unescape(char *buf)
{
	char *p=buf;

	while ((p=strchr(p,'+')))
		*p = ' ';

	p = buf;

	while (p && *p && (p=strchr(p,'%'))) {
		int c1 = p[1];
		int c2 = p[2];

		if (c1 >= '0' && c1 <= '9')
			c1 = c1 - '0';
		else if (c1 >= 'A' && c1 <= 'F')
			c1 = 10 + c1 - 'A';
		else if (c1 >= 'a' && c1 <= 'f')
			c1 = 10 + c1 - 'a';
		else {p++; continue;}

		if (c2 >= '0' && c2 <= '9')
			c2 = c2 - '0';
		else if (c2 >= 'A' && c2 <= 'F')
			c2 = 10 + c2 - 'A';
		else if (c2 >= 'a' && c2 <= 'f')
			c2 = 10 + c2 - 'a';
		else {p++; continue;}
			
		*p = (c1<<4) | c2;

		memmove(p+1, p+3, strlen(p+3)+1);
		p++;
	}
}

#ifdef VALGRIND
size_t valgrind_strlen(const char *s)
{
	size_t count;
	for(count = 0; *s++; count++)
		;
	return count;
}
#endif


/**
  format a string into length-prefixed dotted domain format, as used in NBT
  and in some ADS structures
**/
_PUBLIC_ const char *str_format_nbt_domain(TALLOC_CTX *mem_ctx, const char *s)
{
	char *ret;
	int i;
	if (!s || !*s) {
		return talloc_strdup(mem_ctx, "");
	}
	ret = talloc_array(mem_ctx, char, strlen(s)+2);
	if (!ret) {
		return ret;
	}
	
	memcpy(ret+1, s, strlen(s)+1);
	ret[0] = '.';

	for (i=0;ret[i];i++) {
		if (ret[i] == '.') {
			char *p = strchr(ret+i+1, '.');
			if (p) {
				ret[i] = p-(ret+i+1);
			} else {
				ret[i] = strlen(ret+i+1);
			}
		}
	}

	return ret;
}

/**
 * Add a string to an array of strings.
 *
 * num should be a pointer to an integer that holds the current 
 * number of elements in strings. It will be updated by this function.
 */
_PUBLIC_ bool add_string_to_array(TALLOC_CTX *mem_ctx,
			 const char *str, const char ***strings, int *num)
{
	char *dup_str = talloc_strdup(mem_ctx, str);

	*strings = talloc_realloc(mem_ctx,
				    *strings,
				    const char *, ((*num)+1));

	if ((*strings == NULL) || (dup_str == NULL))
		return false;

	(*strings)[*num] = dup_str;
	*num += 1;

	return true;
}



/**
  varient of strcmp() that handles NULL ptrs
**/
_PUBLIC_ int strcmp_safe(const char *s1, const char *s2)
{
	if (s1 == s2) {
		return 0;
	}
	if (s1 == NULL || s2 == NULL) {
		return s1?-1:1;
	}
	return strcmp(s1, s2);
}


/**
return the number of bytes occupied by a buffer in ASCII format
the result includes the null termination
limited by 'n' bytes
**/
_PUBLIC_ size_t ascii_len_n(const char *src, size_t n)
{
	size_t len;

	len = strnlen(src, n);
	if (len+1 <= n) {
		len += 1;
	}

	return len;
}

/**
 Set a boolean variable from the text value stored in the passed string.
 Returns true in success, false if the passed string does not correctly 
 represent a boolean.
**/

_PUBLIC_ bool set_boolean(const char *boolean_string, bool *boolean)
{
	if (strwicmp(boolean_string, "yes") == 0 ||
	    strwicmp(boolean_string, "true") == 0 ||
	    strwicmp(boolean_string, "on") == 0 ||
	    strwicmp(boolean_string, "1") == 0) {
		*boolean = true;
		return true;
	} else if (strwicmp(boolean_string, "no") == 0 ||
		   strwicmp(boolean_string, "false") == 0 ||
		   strwicmp(boolean_string, "off") == 0 ||
		   strwicmp(boolean_string, "0") == 0) {
		*boolean = false;
		return true;
	}
	return false;
}

/**
 * Parse a string containing a boolean value.
 *
 * val will be set to the read value.
 *
 * @retval true if a boolean value was parsed, false otherwise.
 */
_PUBLIC_ bool conv_str_bool(const char * str, bool * val)
{
	char *	end = NULL;
	long	lval;

	if (str == NULL || *str == '\0') {
		return false;
	}

	lval = strtol(str, &end, 10 /* base */);
	if (end == NULL || *end != '\0' || end == str) {
		return set_boolean(str, val);
	}

	*val = (lval) ? true : false;
	return true;
}

/**
 * Convert a size specification like 16K into an integral number of bytes. 
 **/
_PUBLIC_ bool conv_str_size(const char * str, uint64_t * val)
{
	char *		    end = NULL;
	unsigned long long  lval;

	if (str == NULL || *str == '\0') {
		return false;
	}

	lval = strtoull(str, &end, 10 /* base */);
	if (end == NULL || end == str) {
		return false;
	}

	if (*end) {
		if (strwicmp(end, "K") == 0) {
			lval *= 1024ULL;
		} else if (strwicmp(end, "M") == 0) {
			lval *= (1024ULL * 1024ULL);
		} else if (strwicmp(end, "G") == 0) {
			lval *= (1024ULL * 1024ULL * 1024ULL);
		} else if (strwicmp(end, "T") == 0) {
			lval *= (1024ULL * 1024ULL * 1024ULL * 1024ULL);
		} else if (strwicmp(end, "P") == 0) {
			lval *= (1024ULL * 1024ULL * 1024ULL * 1024ULL * 1024ULL);
		} else {
			return false;
		}
	}

	*val = (uint64_t)lval;
	return true;
}

/**
 * Parse a uint64_t value from a string
 *
 * val will be set to the value read.
 *
 * @retval true if parsing was successful, false otherwise
 */
_PUBLIC_ bool conv_str_u64(const char * str, uint64_t * val)
{
	char *		    end = NULL;
	unsigned long long  lval;

	if (str == NULL || *str == '\0') {
		return false;
	}

	lval = strtoull(str, &end, 10 /* base */);
	if (end == NULL || *end != '\0' || end == str) {
		return false;
	}

	*val = (uint64_t)lval;
	return true;
}

/**
return the number of bytes occupied by a buffer in CH_UTF16 format
the result includes the null termination
**/
_PUBLIC_ size_t utf16_len(const void *buf)
{
	size_t len;

	for (len = 0; SVAL(buf,len); len += 2) ;

	return len + 2;
}

/**
return the number of bytes occupied by a buffer in CH_UTF16 format
the result includes the null termination
limited by 'n' bytes
**/
_PUBLIC_ size_t utf16_len_n(const void *src, size_t n)
{
	size_t len;

	for (len = 0; (len+2 < n) && SVAL(src, len); len += 2) ;

	if (len+2 <= n) {
		len += 2;
	}

	return len;
}

_PUBLIC_ size_t ucs2_align(const void *base_ptr, const void *p, int flags)
{
	if (flags & (STR_NOALIGN|STR_ASCII))
		return 0;
	return PTR_DIFF(p, base_ptr) & 1;
}

/**
Do a case-insensitive, whitespace-ignoring string compare.
**/
_PUBLIC_ int strwicmp(const char *psz1, const char *psz2)
{
	/* if BOTH strings are NULL, return TRUE, if ONE is NULL return */
	/* appropriate value. */
	if (psz1 == psz2)
		return (0);
	else if (psz1 == NULL)
		return (-1);
	else if (psz2 == NULL)
		return (1);

	/* sync the strings on first non-whitespace */
	while (1) {
		while (isspace((int)*psz1))
			psz1++;
		while (isspace((int)*psz2))
			psz2++;
		if (toupper((unsigned char)*psz1) != toupper((unsigned char)*psz2) 
		    || *psz1 == '\0'
		    || *psz2 == '\0')
			break;
		psz1++;
		psz2++;
	}
	return (*psz1 - *psz2);
}

/**
 String replace.
**/
_PUBLIC_ void string_replace(char *s, char oldc, char newc)
{
	while (*s) {
		if (*s == oldc) *s = newc;
		s++;
	}
}

/**
 * Compare 2 strings.
 *
 * @note The comparison is case-insensitive.
 **/
_PUBLIC_ bool strequal(const char *s1, const char *s2)
{
	if (s1 == s2)
		return true;
	if (!s1 || !s2)
		return false;
  
	return strcasecmp(s1,s2) == 0;
}