/****************************************************************************
**
*F GetStr() . . . . . . . . . . . . . . . . . . . . . get a string, local
**
** 'GetStr' reads a string from the current input file into the variable
** 'STATE(ValueObj)' and sets 'Symbol' to 'S_STRING'. The opening double
** quote '"' of the string is the current character pointed to by 'In'.
**
** A string is a sequence of characters delimited by double quotes '"'. It
** must not include '"' or <newline> characters, but the escape sequences
** '\"' or '\n' can be used instead. The escape sequence '\<newline>' is
** ignored, making it possible to split long strings over multiple lines.
**
** An error is raised if the string includes a <newline> character or if the
** file ends before the closing '"'.
*/
static void GetStr(void)
{
Obj string = 0;
Char buf[1024];
UInt i = 0;
Char c = PEEK_CURR_CHAR();
while (c != '"' && c != '\n' && c != '\377') {
if (c == '\\') {
c = GetEscapedChar();
}
i = AddCharToBuf(&string, buf, sizeof(buf), i, c);
// read the next character
c = GET_NEXT_CHAR();
}
// append any remaining data to STATE(ValueObj)
STATE(ValueObj) = AppendBufToString(string, buf, i);
if (c == '\n')
SyntaxError("String must not include <newline>");
if (c == '\377') {
*STATE(In) = '\0';
SyntaxError("String must end with \" before end of file");
}
}
/****************************************************************************
**
*F GetChar() . . . . . . . . . . . . . . . . . get a single character, local
**
** 'GetChar' reads the next character from the current input file into the
** variable 'STATE(Value)' and sets 'Symbol' to 'S_CHAR'. The opening single quote
** '\'' of the character is the current character pointed to by 'In'.
**
** A character is a single character delimited by single quotes '\''. It
** must not be '\'' or <newline>, but the escape sequences '\\\'' or '\n'
** can be used instead.
*/
static void GetChar(void)
{
/* skip '\'' */
Char c = GET_NEXT_CHAR();
/* handle escape equences */
if ( c == '\n' ) {
SyntaxError("Character literal must not include <newline>");
} else {
if ( c == '\\' ) {
STATE(Value)[0] = GetEscapedChar();
} else {
/* put normal chars into 'STATE(Value)' */
STATE(Value)[0] = c;
}
/* read the next character */
c = GET_NEXT_CHAR();
/* check for terminating single quote, and skip */
if ( c == '\'' ) {
c = GET_NEXT_CHAR();
} else {
SyntaxError("Missing single quote in character constant");
}
}
}
unsigned short
LinkObjectStore::GetHashIdx(const char* str, unsigned short max_idx, unsigned int &full_index)
{
if (*str)
{
// Calculate a hashing index based on the input string, unescaping
// any URL escapes as we go along
unsigned int val = 0;
register unsigned int val1;
register const unsigned char* tmp = (const unsigned char*) str;
while (*tmp)
{
int len1;
register unsigned char val2;
register unsigned char val_a, val_b, val_c, val_d;
val1 = 0;
val_a = tmp[0];
val_b = tmp[1];
val_c = (val_b ? tmp[2] : 0);
val_d = (val_c ? tmp[3] : 0);
if(val_a != '%' && val_b != '%' && val_c != '%' && val_d != '%')
{
// Simple hashing if there are no escaped characters in
// this chunk.
val1 = (((unsigned int) val_a) << 24) | (((unsigned int) val_b) << 16) | (((unsigned int) val_c) << 8) | val_d;
if(val_d)
{
tmp+=4;
}
else
{
val = 33*val ^ val1;
break;
}
}
else
{
// Hash each character by itself when there are escaped
// characters.
for(len1 = 4; len1 > 0 && *tmp;len1--,tmp++)
{
val1 <<= 8;
val2 = *tmp;
if(val2 =='%' && tmp[1] && tmp[2])
{
// Unescape only if it is a proper hexadecimal escape.
if (uni_isxdigit(tmp[1]) && uni_isxdigit(tmp[2]))
{
val2 = GetEscapedChar(tmp[1], tmp[2]);
tmp += 2;
}
}
val1 |= val2;
}
if(len1)
{
val1 <<= (len1 << 3); // multiply by 8
}
}
val = 33*val ^ val1;
}
full_index = val;
return (unsigned short)(val % max_idx);
}
else
{
full_index = 0;
return 0;
}
}
