dany/pdfio
1
0
Fork 0
mirror of https://github.com/michaelrsweet/pdfio.git synced 2025-08-29 07:12:02 +02:00
Code Issues Packages Projects Releases Wiki Activity

Fix decryption of UTF-16 strings (Issue #42)

Browse Source
This commit is contained in:
Michael R Sweet
2025-04-13 09:05:02 -04:00
parent 20dd2a6d28
commit 2dcef0936e
5 changed files with 110 additions and 121 deletions
Download Patch File Download Diff File Expand all files Collapse all files

123
pdfio-dict.c
View File

@@ -465,127 +465,12 @@ pdfioDictGetString(pdfio_dict_t *dict, // I - Dictionary
else if (value && value->type == PDFIO_VALTYPE_BINARY && value->value.binary.datalen < 4096)
{
// Convert binary string to regular string...
char temp[4096], // Temporary string
*tempptr; // Pointer into temporary string
unsigned char *dataptr; // Pointer into the data string
char temp[4096]; // Temporary UTF-8 string
if (!(value->value.binary.datalen & 1) && !memcmp(value->value.binary.data, "\376\377", 2))
if (!(value->value.binary.datalen & 1) && (!memcmp(value->value.binary.data, "\376\377", 2) || !memcmp(value->value.binary.data, "\377\376", 2)))
{
// Copy UTF-16 BE
int ch; // Unicode character
size_t remaining; // Remaining bytes
for (dataptr = value->value.binary.data + 2, remaining = value->value.binary.datalen - 2, tempptr = temp; remaining > 1 && tempptr < (temp + sizeof(temp) - 5); dataptr += 2, remaining -= 2)
{
ch = (dataptr[0] << 8) | dataptr[1];
if (ch >= 0xd800 && ch <= 0xdbff && remaining > 3)
{
// Multi-word UTF-16 char...
int lch; // Lower bits
lch = (dataptr[2] << 8) | dataptr[3];
if (lch < 0xdc00 || lch >= 0xdfff)
break;
ch = (((ch & 0x3ff) << 10) | (lch & 0x3ff)) + 0x10000;
dataptr += 2;
remaining -= 2;
}
else if (ch >= 0xfffe)
{
continue;
}
if (ch < 128)
{
// ASCII
*tempptr++ = (char)ch;
}
else if (ch < 4096)
{
// 2-byte UTF-8
*tempptr++ = (char)(0xc0 | (ch >> 6));
*tempptr++ = (char)(0x80 | (ch & 0x3f));
}
else if (ch < 65536)
{
// 3-byte UTF-8
*tempptr++ = (char)(0xe0 | (ch >> 12));
*tempptr++ = (char)(0x80 | ((ch >> 6) & 0x3f));
*tempptr++ = (char)(0x80 | (ch & 0x3f));
}
else
{
// 4-byte UTF-8
*tempptr++ = (char)(0xe0 | (ch >> 18));
*tempptr++ = (char)(0x80 | ((ch >> 12) & 0x3f));
*tempptr++ = (char)(0x80 | ((ch >> 6) & 0x3f));
*tempptr++ = (char)(0x80 | (ch & 0x3f));
}
}
*tempptr = '\0';
}
else if (!(value->value.binary.datalen & 1) && !memcmp(value->value.binary.data, "\377\376", 2))
{
// Copy UTF-16 LE
int ch; // Unicode character
size_t remaining; // Remaining bytes
for (dataptr = value->value.binary.data + 2, remaining = value->value.binary.datalen - 2, tempptr = temp; remaining > 1 && tempptr < (temp + sizeof(temp) - 5); dataptr += 2, remaining -= 2)
{
ch = (dataptr[1] << 8) | dataptr[0];
if (ch >= 0xd800 && ch <= 0xdbff && remaining > 3)
{
// Multi-word UTF-16 char...
int lch; // Lower bits
lch = (dataptr[3] << 8) | dataptr[2];
if (lch < 0xdc00 || lch >= 0xdfff)
break;
ch = (((ch & 0x3ff) << 10) | (lch & 0x3ff)) + 0x10000;
dataptr += 2;
remaining -= 2;
}
else if (ch >= 0xfffe)
{
continue;
}
if (ch < 128)
{
// ASCII
*tempptr++ = (char)ch;
}
else if (ch < 4096)
{
// 2-byte UTF-8
*tempptr++ = (char)(0xc0 | (ch >> 6));
*tempptr++ = (char)(0x80 | (ch & 0x3f));
}
else if (ch < 65536)
{
// 3-byte UTF-8
*tempptr++ = (char)(0xe0 | (ch >> 12));
*tempptr++ = (char)(0x80 | ((ch >> 6) & 0x3f));
*tempptr++ = (char)(0x80 | (ch & 0x3f));
}
else
{
// 4-byte UTF-8
*tempptr++ = (char)(0xe0 | (ch >> 18));
*tempptr++ = (char)(0x80 | ((ch >> 12) & 0x3f));
*tempptr++ = (char)(0x80 | ((ch >> 6) & 0x3f));
*tempptr++ = (char)(0x80 | (ch & 0x3f));
}
}
*tempptr = '\0';
// Copy UTF-16...
_pdfio_utf16cpy(temp, value->value.binary.data, value->value.binary.datalen, sizeof(temp));
}
else
{