winamp/Src/auth/decode.cpp

78 lines
1.4 KiB
C++
Raw Normal View History

2024-09-24 12:54:57 +00:00
#include <stdio.h>
#include <bfc/platform/types.h>
#include <strsafe.h>
static uint8_t quickhex(char c)
{
int hexvalue = c;
if (hexvalue & 0x10)
hexvalue &= ~0x30;
else
{
hexvalue &= 0xF;
hexvalue += 9;
}
return hexvalue;
}
static uint8_t DecodeEscape(const char *&str)
{
uint8_t a = quickhex(*++str);
uint8_t b = quickhex(*++str);
str++;
return a * 16 + b;
}
static void DecodeEscapedUTF8(char *&output, const char *&input)
{
uint8_t utf8_data[1024] = {0}; // hopefully big enough!!
int num_utf8_words=0;
bool error=false;
while (input && *input && *input == '%' && num_utf8_words < sizeof(utf8_data))
{
if (isxdigit(input[1]) && isxdigit(input[2]))
{
utf8_data[num_utf8_words++]=DecodeEscape(input);
}
else if (input[1] == '%')
{
input+=2;
utf8_data[num_utf8_words++]='%';
}
else
{
error = true;
break;
}
}
memcpy(output, utf8_data, num_utf8_words);
output+=num_utf8_words;
//StringCchCopyExA(output, num_utf8_words, (char *)utf8_data, &output, 0, 0);
if (error)
{
*output++ = *input++;
}
}
// benski> We have the luxury of knowing that decoding will ALWAYS produce smaller strings
// so we can do it in-place
void UrlDecode(char *str)
{
const char *itr = str;
while (itr && *itr)
{
switch (*itr)
{
case '%':
DecodeEscapedUTF8(str, itr);
break;
default:
*str++ = *itr++;
break;
}
}
*str = 0;
}