mirrors/recycled-ni-neutrino
1
0
Fork 0
mirror of https://github.com/tuxbox-fork-migrations/recycled-ni-neutrino.git synced 2025-09-07 05:38:28 +02:00
Code Issues Packages Projects Releases Wiki Activity

src/eitd/edvbstring: unbreak isUTF8()

Browse Source 
Origin commit data
------------------
Branch: ni/coolstream
Commit: 9ead2b459b
Author: martii <m4rtii@gmx.de>
Date: 2013-12-15 (Sun, 15 Dec 2013)


------------------
No further description and justification available within origin commit message!

------------------
This commit was generated by Migit
This commit is contained in:
martii
2013-12-15 14:44:26 +01:00
committed by Jacek Jendrzej
parent 242c52b53a
commit 51a912ff3b
1 changed files with 27 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

44
src/eitd/edvbstring.cpp
View File

@@ -2301,30 +2301,40 @@ int isUTF8(const std::string &string)
{ {
unsigned int len=string.size(); unsigned int len=string.size();
for (unsigned int i=0; i < len; ++i) for (unsigned int i=0; i < len;)
{ {
if (!(string[i]&0x80)) // normal ASCII int trailing = 0;
if (string[i] >> 7 == 0) // 0xxxxxxx
{
i++;
continue; continue;
if ((string[i] & 0xE0) == 0xC0) // one char following. }
if (string[i] >> 5 == 6) // 110xxxxx 10xxxxxx
{ {
// first, length check: if (++i >= len)
if (i+1 >= len) return 0;
return 0; // certainly NOT utf-8 trailing = 1;
i++; }
if ((string[i]&0xC0) != 0x80) else if (string[i] >> 4 == 14) // 1110xxxx 10xxxxxx 10xxxxxx
return 0; // no, not UTF-8.
} else if ((string[i] & 0xF0) == 0xE0)
{ {
if ((i+1) >= len) if (++i >= len)
return 0; return 0;
trailing = 2;
}
else if ((string[i] >> 3) == 30) // 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
{
if (++i >= len)
return 0;
trailing = 3;
} else
return 0;
while (trailing) {
if (i >= len || string[i] >> 6 != 2)
return 0;
trailing--;
i++; i++;
if ((string[i]&0xC0) != 0x80)
return 0;
i++;
if ((string[i]&0xC0) != 0x80)
return 0;
} }
} }
return 1; // can be UTF8 (or pure ASCII, at least no non-UTF-8 8bit characters) return 1; // can be UTF8 (or pure ASCII, at least no non-UTF-8 8bit characters)
} }