int CTortoiseGitBlameData::GetEncode(int *bomoffset)
{
int encoding = 0;
BYTE_VECTOR rawAll;
for (const auto& rawBytes : m_RawLines)
rawAll.append(rawBytes.data(), rawBytes.size());
encoding = GetEncode(rawAll.data(), static_cast<int>(rawAll.size()), bomoffset);
return encoding;
}
int CTortoiseGitBlameData::GetEncode(int *bomoffset)
{
int encoding = 0;
BYTE_VECTOR rawAll;
for (auto it = m_RawLines.begin(), it_end = m_RawLines.end(); it != it_end; ++it)
{
rawAll.append(&(*it)[0], it->size());
}
encoding = GetEncode(&rawAll[0], (int)rawAll.size(), bomoffset);
return encoding;
}
int CTortoiseGitBlameData::GetEncode(int *bomoffset)
{
int encoding = 0;
BYTE_VECTOR rawAll;
for (const auto& rawBytes : m_RawLines)
{
rawAll.append(&rawBytes[0], rawBytes.size());
}
encoding = GetEncode(&rawAll[0], (int)rawAll.size(), bomoffset);
return encoding;
}
int CTortoiseGitBlameData::UpdateEncoding(int encode)
{
int encoding = encode;
int bomoffset = 0;
if (encoding==0)
{
BYTE_VECTOR all;
for (auto it = m_RawLines.begin(); it != m_RawLines.end(); ++it)
{
if (!it->empty())
all.append(&(*it)[0], it->size());
}
encoding = GetEncode(&all[0], (int)all.size(), &bomoffset);
}
if (encoding != m_encode)
{
m_encode = encoding;
m_Utf8Lines.resize(m_RawLines.size());
for (size_t i_Lines = 0; i_Lines < m_RawLines.size(); ++i_Lines)
{
const BYTE_VECTOR& rawLine = m_RawLines[i_Lines];
int bomoffset = 0;
CStringA lineUtf8;
lineUtf8.Empty();
if (!rawLine.empty())
{
if (encoding == 1201)
{
CString line;
int size = (int)((rawLine.size() - bomoffset)/2);
TCHAR *buffer = line.GetBuffer(size);
memcpy(buffer, &rawLine[bomoffset], sizeof(TCHAR)*size);
// swap the bytes to little-endian order to get proper strings in wchar_t format
wchar_t * pSwapBuf = buffer;
for (int i = 0; i < size; ++i)
{
*pSwapBuf = WideCharSwap2(*pSwapBuf);
++pSwapBuf;
}
line.ReleaseBuffer();
lineUtf8 = CUnicodeUtils::GetUTF8(line);
}
else if (encoding == 1200)
{
CString line;
// the first bomoffset is 2, after that it's 1 (see issue #920)
// also: don't set bomoffset if called from Encodings menu (i.e. start == 42 and bomoffset == 0); bomoffset gets only set if autodetected
if (bomoffset == 0 && i_Lines != 0)
{
bomoffset = 1;
}
int size = (int)((rawLine.size() - bomoffset)/2);
TCHAR *buffer = line.GetBuffer(size);
memcpy(buffer, &rawLine[bomoffset], sizeof(TCHAR) * size);
line.ReleaseBuffer();
lineUtf8 = CUnicodeUtils::GetUTF8(line);
}
else if (encoding == CP_UTF8)
lineUtf8 = CStringA((LPCSTR)&rawLine[bomoffset], (int)(rawLine.size() - bomoffset));
else
{
CString line = CUnicodeUtils::GetUnicode(CStringA((LPCSTR)&rawLine[bomoffset], (int)(rawLine.size() - bomoffset)), encoding);
lineUtf8 = CUnicodeUtils::GetUTF8(line);
}
}
m_Utf8Lines[i_Lines] = lineUtf8;
bomoffset = 0;
}
}
return encoding;
}
