char* unescape_string(char* string) {
/*
* Description: given an input stream converts escaped characters into their appropriate char representation and returns it as a new string
* Parameters: string to be converted
* Modifies: nothing
* Returns: resulting string with all the escaped characters with their appropriate values
*
*/
//printf("unecaping: %s\n", string);
char* unescaped_string = (char*)malloc(strlen(string) * sizeof(char) + 1);
int current_position = 0;
int unescaped_string_position = 0;
unsigned char escape_character = 0;
for(current_position = 0; current_position < strlen(string); current_position++) {
escape_character = *(string + current_position);
if(*(string + current_position) == '\\') {
if(*(string + current_position + 1) == 'x') {
current_position++;
int hex_count;
escape_character = 0;
for(hex_count = 1; hex_count <= MAX_HEX_CHARS; hex_count++) {
if(!isxdigit(*(string + current_position + hex_count))) {
break;
}
escape_character = escape_character * 16 + char_to_hex(*(string + current_position + hex_count));
}
hex_count--;
current_position += hex_count;
} else if(is_oct_digit((*(string + current_position + 1))) == 1) {
int oct_count;
escape_character = 0;
for(oct_count = 1; oct_count <= MAX_OCT_CHARS; oct_count++) {
if(is_oct_digit(*(string + current_position + oct_count)) == 0) {
break;
}
escape_character = escape_character * 8 + char_to_oct(*(string + current_position + oct_count));
}
current_position += oct_count;
} else {
escape_character = is_escape_character(*(string + current_position + 1));
if(escape_character == 0) {
escape_character = *(string + current_position);
} else {
current_position++;
}
}
}
*(unescaped_string + unescaped_string_position) = escape_character;
unescaped_string_position++;
escape_character = 0;
}
*(unescaped_string + unescaped_string_position + 1) = '\0';
return unescaped_string;
}
Long VG_(strtoll8) ( Char* str, Char** endptr )
{
Bool neg = False;
Long n = 0, digit = 0;
// Skip leading whitespace.
while (VG_(isspace)(*str)) str++;
// Allow a leading '-' or '+'.
if (*str == '-') { str++; neg = True; }
else if (*str == '+') { str++; }
while (is_oct_digit(*str, &digit)) {
n = 8*n + digit;
str++;
}
if (neg) n = -n;
if (endptr) *endptr = str; // Record first failing character.
return n;
}
int regex_parser::process_escape(const char *re, int ptr, int_set *chars){
if (ptr==strlen(re)){
return (++ptr);
//fatal("regex_parser:: process_escape: \\ in last position.");
}
char c=re[ptr];
int next;
if(is_x(c)){
if(ptr>strlen(re)-3)
fatal("regex_parser::process_escape: invalid hex escape sequence.");
else if (!is_hex_digit(re[ptr+1]) || !is_hex_digit(re[ptr+2]))
fatal("regex_parser::process_escape: invalid hex escape sequence.");
else{
char tmp[5];
tmp[0]='0';tmp[1]=c;tmp[2]=re[ptr+1];tmp[3]=re[ptr+2]; tmp[4]='\0';
sscanf(tmp,"0x%x", &next);
chars->insert(next);
ptr=ptr+3;
}
}else if (is_oct_digit(c)){
if(ptr>strlen(re)-3)
{next=escaped(c);ptr++;chars->insert(next);} //normal escape sequence
else if (!is_oct_digit(re[ptr+1]) || !is_oct_digit(re[ptr+2]))
{next=escaped(c);ptr++;chars->insert(next);} //normal escape sequence
else{
//really an octal sequence!
char tmp[5];
tmp[0]='0';tmp[1]=c;tmp[2]=re[ptr+1];tmp[3]=re[ptr+2]; tmp[4]='\0';
sscanf(tmp,"0%o", &next);
chars->insert(next);
ptr=ptr+3;
}
}else if(c=='s'){
chars->insert('\t');
chars->insert('\n');
chars->insert('\r');
chars->insert('\x0C');
chars->insert('\x20');
ptr++;
}else if(c=='S'){
chars->insert('\t');
chars->insert('\n');
chars->insert('\r');
chars->insert('\x0C');
chars->insert('\x20');
chars->negate();
ptr++;
}else if(c=='d'){
chars->insert('0');chars->insert('1');chars->insert('2');
chars->insert('3');chars->insert('4');chars->insert('5');
chars->insert('6');chars->insert('7');chars->insert('8');
chars->insert('9');
ptr++;
}else if(c=='D'){
chars->insert('0');chars->insert('1');chars->insert('2');
chars->insert('3');chars->insert('4');chars->insert('5');
chars->insert('6');chars->insert('7');chars->insert('8');
chars->insert('9');
chars->negate();
ptr++;
}else if(c=='w'){
chars->insert('_');
chars->insert('0');chars->insert('1');chars->insert('2');
chars->insert('3');chars->insert('4');chars->insert('5');
chars->insert('6');chars->insert('7');chars->insert('8');
chars->insert('9');
chars->insert('a');chars->insert('b');chars->insert('c');
chars->insert('d');chars->insert('e');chars->insert('f');
chars->insert('g');chars->insert('h');chars->insert('i');
chars->insert('j');chars->insert('k');chars->insert('l');
chars->insert('m');chars->insert('n');chars->insert('o');
chars->insert('p');chars->insert('q');chars->insert('r');
chars->insert('s');chars->insert('t');chars->insert('u');
chars->insert('v');chars->insert('w');chars->insert('x');
chars->insert('y');chars->insert('z');
chars->insert('A');chars->insert('B');chars->insert('C');
chars->insert('D');chars->insert('E');chars->insert('F');
chars->insert('G');chars->insert('H');chars->insert('I');
chars->insert('J');chars->insert('K');chars->insert('L');
chars->insert('M');chars->insert('N');chars->insert('O');
chars->insert('P');chars->insert('Q');chars->insert('R');
chars->insert('S');chars->insert('T');chars->insert('U');
chars->insert('V');chars->insert('W');chars->insert('X');
chars->insert('Y');chars->insert('Z');
ptr++;
}else if(c=='W'){
chars->insert('_');
chars->insert('0');chars->insert('1');chars->insert('2');
chars->insert('3');chars->insert('4');chars->insert('5');
chars->insert('6');chars->insert('7');chars->insert('8');
chars->insert('9');
chars->insert('a');chars->insert('b');chars->insert('c');
chars->insert('d');chars->insert('e');chars->insert('f');
chars->insert('g');chars->insert('h');chars->insert('i');
chars->insert('j');chars->insert('k');chars->insert('l');
chars->insert('m');chars->insert('n');chars->insert('o');
chars->insert('p');chars->insert('q');chars->insert('r');
chars->insert('s');chars->insert('t');chars->insert('u');
chars->insert('v');chars->insert('w');chars->insert('x');
chars->insert('y');chars->insert('z');
chars->insert('A');chars->insert('B');chars->insert('C');
chars->insert('D');chars->insert('E');chars->insert('F');
chars->insert('G');chars->insert('H');chars->insert('I');
chars->insert('J');chars->insert('K');chars->insert('L');
chars->insert('M');chars->insert('N');chars->insert('O');
chars->insert('P');chars->insert('Q');chars->insert('R');
chars->insert('S');chars->insert('T');chars->insert('U');
chars->insert('V');chars->insert('W');chars->insert('X');
chars->insert('Y');chars->insert('Z');
chars->negate();
ptr++;
}else{
next=escaped(c);
chars->insert(next);
ptr++;
}
return ptr;
}
long parse_integer(char *str, char **endptr,
int len, int base, int *err){
if(len == 0){
len = strlen(str);
}
//since signed overflow is undefined and I don't really
//want to use gmp for this, use an unsigned long to hold the value
uint64_t num = 0;
int i = 0, negitive = 0, overflow = 0;
//ignore leading space
while(isspace(str[i]) && ++i < len);
if(i < len){
if(str[i] == '-'){
negitive = 1;
i++;
}
}
if(i == len){
if(endptr){*endptr = str;}
return 0;
}
#define TO_NUMBER(start, end, base) \
({uint64_t number = 0, next = 0; \
int j; \
for(j = start; j < end; j++){ \
next = (number * base) + char_to_number(str[j]); \
if(next < number){/*overflow*/ \
overflow = j; \
break; \
} else { \
number = next; \
} \
} \
number;})
//I'm pretty sure this will cause an error if the string is something like
//0x abcdefg. It should be read as a 0, but I'm not sure what will happen
if(base == 0 && ((i+1) < len)){
if(str[i] == '0' && str[i+1] == 'x'){
base = 16;
i+=2;
} else if(str[i] == '0' && str[i+1] == 'o'){
base = 8;
i+=2;
} else if(str[i] == '0' && str[i+1] == 'b'){
base = 2;
i+=2;
} else {
base = 10;
}
}
while(str[i] == '0' && ++i < len);//read leading zeros
if(i == len){
if(endptr){*endptr = (str+i);}
return 0;
}
/*
Use special cases for 2, 8, 10, and 16 to speed them up.
Multiplies in base 2, 8, or 16 become shifts, and for x86 at least
a multiply by 10 becomes two lea instructions.
*/
#define DO_CASE(base) \
if(i <= max_length_per_base[base]){ \
num = TO_NUMBER(start, i, base); \
} \
if(i > max_length_per_base[base] || overflow > 0 || num>LONG_MAX){ \
*err = errno = ERANGE; \
*endptr = (str+i); \
return (negitive ? LONG_MIN : LONG_MAX); \
}
int start = i;
if(base == 10){
while(isdigit(str[i]) && ++i < len);
DO_CASE(base);
} else if(base == 16){
while(isxdigit(str[i]) && ++i < len);
DO_CASE(base);
} else if(base == 8){
while(is_oct_digit(str[i]) && ++i < len);
DO_CASE(base);
} else if(base == 2){
while(is_binary_digit(str[i]) && ++i < len);
DO_CASE(base);
} else if(base < 10){
i = memspn((uint8_t*)str + i, len - i, valid_digits, base);
DO_CASE(base);
} else {
i = memspn((uint8_t*)str + i, len -i, valid_digits, 10 + (base-10)*2);
DO_CASE(base);
}
if(endptr){*endptr = (str + i);}
return (long)(negitive ? -num : num);
}
