mewin/nana
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

add a namespace nana::utf for UTF-8 character processing

Browse Source
This commit is contained in:
Jinhao 2016-01-01 12:30:18 +08:00
parent a42ebe19b4
commit 0a4f04267e
4 changed files with 199 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
include/nana/charset.hpp
View File

@ -18,6 +18,24 @@
namespace nana
{
namespace utf
{
/// Attempt to get a pointer to a character of UTF-8 string by a specifed character index.
/// @param text_utf8 A string encoded as UTF-8.
/// @param pos The unicode character index.
/// @returns A pointer to the unicode character. It returns a null if pos is out of range.
const char* char_ptr(const char* text_utf8, unsigned pos);
const char* char_ptr(const ::std::string& text_utf8, unsigned pos);
/// Get the unicode character by a specified character index.
/// @param text_utf8 A string encoded as UTF-8.
/// @param pos The unicode character index.
/// @param len A unsigned pointer to receive the number of bytes it takes in UTF-8 encoded. If len is a nullptr, it is ignored.
/// @returns A unicode character. '\0' if pos is out of range.
wchar_t char_at(const char* text_utf8, unsigned pos, unsigned * len);
wchar_t char_at(const ::std::string& text_utf8, unsigned pos, unsigned * len);
}
enum class unicode
{
utf8, utf16, utf32

177
source/charset.cpp
View File

@ -30,6 +30,183 @@
namespace nana
{
namespace utf
{
const char* char_ptr(const char* text, unsigned pos)
{
auto ustr = reinterpret_cast<const unsigned char*>(text);
auto const end = ustr + std::strlen(text);
for (unsigned i = 0; i != pos; ++i)
{
const auto uch = *ustr;
if (uch < 0x80)
{
++ustr;
continue;
}
if (uch < 0xC0)
return nullptr;
if ((uch < 0xE0) && (ustr + 1 < end))
ustr += 2;
else if (uch < 0xF0 && (ustr + 2 <= end))
ustr += 3;
else if (uch < 0x1F && (ustr + 3 <= end))
ustr += 4;
else
return nullptr;
}
return reinterpret_cast<const char*>(ustr);
}
const char* char_ptr(const std::string& text_utf8, unsigned pos)
{
auto ustr = reinterpret_cast<const unsigned char*>(text_utf8.c_str());
auto const end = ustr + text_utf8.size();
for (unsigned i = 0; i != pos; ++i)
{
const auto uch = *ustr;
if (uch < 0x80)
{
++ustr;
continue;
}
if (uch < 0xC0)
return nullptr;
if ((uch < 0xE0) && (ustr + 1 < end))
ustr += 2;
else if (uch < 0xF0 && (ustr + 2 <= end))
ustr += 3;
else if (uch < 0x1F && (ustr + 3 <= end))
ustr += 4;
else
return nullptr;
}
return reinterpret_cast<const char*>(ustr);
}
wchar_t char_at(const char* text_utf8, unsigned pos, unsigned * len)
{
if (!text_utf8)
return 0;
if (pos)
{
text_utf8 = char_ptr(text_utf8, pos);
if (!text_utf8)
return 0;
}
const wchar_t uch = *reinterpret_cast<const unsigned char*>(text_utf8);
if (uch < 0x80)
{
if (len)
*len = 1;
return *text_utf8;
}
if (uch < 0xC0)
{
if (len)
*len = 0;
return 0;
}
const auto end = text_utf8 + std::strlen(text_utf8);
if (uch < 0xE0 && (text_utf8 + 1 <= end))
{
if (len)
*len = 2;
return (wchar_t(uch & 0x1F) << 6) | (reinterpret_cast<const unsigned char*>(text_utf8)[1] & 0x3F);
}
else if (uch < 0xF0 && (text_utf8 + 2 <= end))
{
if (len)
*len = 3;
return ((((uch & 0xF) << 6) | (reinterpret_cast<const unsigned char*>(text_utf8)[1] & 0x3F)) << 6) | (reinterpret_cast<const unsigned char*>(text_utf8)[2] & 0x3F);
}
else if (uch < 0x1F && (text_utf8 + 3 <= end))
{
if (len)
*len = 4;
return ((((((uch & 0x7) << 6) | (reinterpret_cast<const unsigned char*>(text_utf8)[1] & 0x3F)) << 6) | (reinterpret_cast<const unsigned char*>(text_utf8)[2] & 0x3F)) << 6) | (reinterpret_cast<const unsigned char*>(text_utf8)[3] & 0x3F);
}
if (len)
*len = 0;
return 0;
}
wchar_t char_at(const ::std::string& text_utf8, unsigned pos, unsigned * len)
{
const char* ptr;
if (pos)
{
ptr = char_ptr(text_utf8, pos);
if (!ptr)
return 0;
}
else
ptr = text_utf8.c_str();
const wchar_t uch = *reinterpret_cast<const unsigned char*>(ptr);
if (uch < 0x80)
{
if (len)
*len = 1;
return *ptr;
}
if (uch < 0xC0)
{
if (len)
*len = 0;
return 0;
}
const auto end = text_utf8.c_str() + text_utf8.size();
if (uch < 0xE0 && (ptr + 1 <= end))
{
if (len)
*len = 2;
return (wchar_t(uch & 0x1F) << 6) | (reinterpret_cast<const unsigned char*>(ptr)[1] & 0x3F);
}
else if (uch < 0xF0 && (ptr + 2 <= end))
{
if (len)
*len = 3;
return ((((uch & 0xF) << 6) | (reinterpret_cast<const unsigned char*>(ptr)[1] & 0x3F)) << 6) | (reinterpret_cast<const unsigned char*>(ptr)[2] & 0x3F);
}
else if (uch < 0x1F && (ptr + 3 <= end))
{
if (len)
*len = 4;
return ((((((uch & 0x7) << 6) | (reinterpret_cast<const unsigned char*>(ptr)[1] & 0x3F)) << 6) | (reinterpret_cast<const unsigned char*>(ptr)[2] & 0x3F)) << 6) | (reinterpret_cast<const unsigned char*>(ptr)[3] & 0x3F);
}
if (len)
*len = 0;
return 0;
}
}
namespace detail
{
class locale_initializer

2
source/gui/programming_interface.cpp
View File

@ -348,7 +348,7 @@ namespace API
text.erase(pos, 1);
if(shortkey == 0 && pos < text.length())
{
shortkey = text.at(pos);
shortkey = utf::char_at(text.c_str() + pos, 0, nullptr);
if(shortkey == '&') //This indicates the text contains "&&", it means the symbol have to be ignored.
shortkey = 0;
else if(skpos)

4
source/gui/widgets/button.cpp
View File

@ -258,7 +258,9 @@ namespace nana{ namespace drawerbase
if(shortkey)
{
unsigned off_w = (shortkey_pos ? graph.text_extent_size(mbstr.c_str(), static_cast<unsigned>(shortkey_pos)).width : 0);
nana::size shortkey_size = graph.text_extent_size(to_wstring(mbstr.c_str() + shortkey_pos), 1);
wchar_t keystr[2] = {nana::utf::char_at(mbstr.c_str() + shortkey_pos, 0, 0), 0};
auto shortkey_size = graph.text_extent_size(keystr, 1);
unsigned ascent, descent, inleading;
graph.text_metrics(ascent, descent, inleading);