|
DesertDragon
Junior Member
Russia
18 Posts |
 Posted - Oct 17 2015 : 9:38:43 PM
|
VA_X.dll file version 10.9.2076.0 built 2015.09.15
DevEnv.exe version 14.0.23107.0 Community
msenv.dll version 14.0.23107.0
Comctl32.dll version 6.10.10240.16384
Windows 10 10.0 Build 10240
4 processors (x86-64, WOW64)
Language info: 1251, 0x419
There is bug with decoding UTF-8 in VA.
In VA Show Find References and Rename windows the cyrillic and other not ANSI symbols in comments from UTF-8 files saved with BOM showing without decoding.
UTF-8 strings from files already opened in Visual Studio is showing well.
I attach a VS debugger to another VS windows with VA running and set breakpoint to CreateFile, MultiByteToWideChar and WideCharToMultiByte functions on VA_X.dll threads. And reveal what when you open each files during finding referenes you convert UTF-8 file to wide char buffer with codepage 65001 and after convert wide char buffer to multibyte buffer again with codepage 65001 unlike windows ANSI code page. |
|
UTF-8 source code files decoding bug
