tomasz.sowa/pikotools
1
0
Fork 0
Code Issues 1 Pull Requests Releases Activity

added to utf8:

Browse Source 
bool UTF8ToWide(const char * utf8, size_t utf8_len, WTextStream & res, bool clear = true, int mode = 1);
bool UTF8ToWide(const char * utf8,                  WTextStream & res, bool clear = true, int mode = 1);
bool UTF8ToWide(const std::string & utf8,           WTextStream & res, bool clear = true, int mode = 1);
bool UTF8ToWide(std::istream & utf8,                WTextStream & res, bool clear = true, int mode = 1);

void WideToUTF8(PT::WTextStream & buffer,                       std::string & utf8,  bool clear = true, int mode = 1);
void WideToUTF8(PT::WTextStream & buffer,                       std::ostream & utf8, int mode = 1);

these functions need some testing yet




git-svn-id: svn://ttmath.org/publicrep/pikotools/trunk@1156 e52654a7-88a9-db11-a3e9-0013d4bc506e
This commit is contained in:
Tomasz Sowa
2018-11-24 13:30:45 +00:00
parent 8b3ae14d9a
commit da4ed9c1f8
8 changed files with 215 additions and 74 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
utf8/Makefile.dep
View File

@@ -1,3 +1,6 @@
# DO NOT DELETE
utf8.o: utf8.h
utf8.o: utf8.h ../textstream/textstream.h ../space/space.h
utf8.o: ../textstream/types.h ../date/date.h ../convert/convert.h
utf8.o: ../convert/inttostr.h ../convert/strtoint.h ../convert/text.h
utf8.o: ../convert/misc.h ../membuffer/membuffer.h ../textstream/types.h

210
utf8/utf8.cpp
View File

@@ -5,7 +5,7 @@
*/
/*
* Copyright (c) 2010-2014, Tomasz Sowa
* Copyright (c) 2010-2018, Tomasz Sowa
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
@@ -38,6 +38,7 @@
#include "utf8.h"
namespace PT
{
@@ -264,6 +265,7 @@ static void IntToWide(int c, std::wstring & res)
{
if( sizeof(wchar_t)==2 && c>0xffff )
{
// UTF16 surrogate pairs
c -= 0x10000;
res += static_cast<wchar_t>(((c >> 10) & 0x3FF) + 0xD800);
res += static_cast<wchar_t>((c & 0x3FF) + 0xDC00);
@@ -275,8 +277,61 @@ static void IntToWide(int c, std::wstring & res)
}
static void IntToWide(int c, WTextStream & res)
{
if( sizeof(wchar_t)==2 && c>0xffff )
{
// UTF16 surrogate pairs
c -= 0x10000;
res << static_cast<wchar_t>(((c >> 10) & 0x3FF) + 0xD800);
res << static_cast<wchar_t>((c & 0x3FF) + 0xDC00);
}
else
{
res << static_cast<wchar_t>(c);
}
}
static bool UTF8ToWideGeneric(const char * utf8, size_t utf8_len, int mode, auto convert_function)
{
int z;
size_t len;
bool correct, was_error = false;
while( utf8_len > 0 )
{
if( (unsigned char)*utf8 <= 0x7f )
{
// small optimization
len = 1;
correct = true;
z = static_cast<unsigned char>(*utf8);
}
else
{
len = UTF8ToInt(utf8, utf8_len, z, correct); // the len will be different from zero
}
if( !correct )
{
if( mode == 1 )
convert_function(0xFFFD); // U+FFFD "replacement character"
was_error = true;
}
else
{
convert_function(z);
}
utf8 += len;
utf8_len -= len;
}
return !was_error;
}
/*!
this function converts an utf8 string into wide string (std::wstring)
@@ -295,48 +350,20 @@ static void IntToWide(int c, std::wstring & res)
*/
bool UTF8ToWide(const char * utf8, size_t utf8_len, std::wstring & res, bool clear, int mode)
{
int z;
size_t len;
bool correct, was_error = false;
if( clear )
res.clear();
while( utf8_len > 0 )
{
if( (unsigned char)*utf8 <= 0x7f )
{
// small optimization
len = 1;
correct = true;
z = static_cast<unsigned char>(*utf8);
}
else
{
len = UTF8ToInt(utf8, utf8_len, z, correct); // the len will be different from zero
}
bool status = UTF8ToWideGeneric(utf8, utf8_len, mode, [&res](int c) {
IntToWide(c, res);
});
if( !correct )
{
if( mode == 1 )
res += 0xFFFD; // U+FFFD "replacement character"
was_error = true;
}
else
{
IntToWide(z, res);
}
utf8 += len;
utf8_len -= len;
}
return !was_error;
return status;
}
/*!
this function converts an utf8 string into wide string (std::wstring)
@@ -425,6 +452,67 @@ return !was_error;
}
// need to be tested
bool UTF8ToWide(const char * utf8, size_t utf8_len, WTextStream & res, bool clear, int mode)
{
if( clear )
res.clear();
bool status = UTF8ToWideGeneric(utf8, utf8_len, mode, [&res](int c) {
IntToWide(c, res);
});
return status;
}
// need to be tested
bool UTF8ToWide(const char * utf8, WTextStream & res, bool clear, int mode)
{
size_t utf8_len = 0;
while( utf8[utf8_len] != 0 )
utf8_len += 1;
return UTF8ToWide(utf8, utf8_len, res, clear, mode);
}
// need to be tested
bool UTF8ToWide(const std::string & utf8, WTextStream & res, bool clear, int mode)
{
return UTF8ToWide(utf8.c_str(), utf8.size(), res, clear, mode);
}
// need to be tested
bool UTF8ToWide(std::istream & utf8, WTextStream & res, bool clear, int mode)
{
int z;
bool correct, was_error = false;
if( clear )
res.clear();
while( UTF8ToInt(utf8, z, correct) > 0 )
{
if( !correct )
{
if( mode == 1 )
res << 0xFFFD; // U+FFFD "replacement character"
was_error = true;
}
else
{
IntToWide(z, res);
}
}
return !was_error;
}
/*!
this function converts one wide character into UTF-8 sequence
@@ -1174,6 +1262,58 @@ return res;
// not tested
// FIX ME it is not using surrogate pairs from input stream
// and mode parameter
static void WideToUTF8Generic(PT::WTextStream & buffer, int mode, auto write_function)
{
char utf8_buffer[256];
std::size_t buffer_len = sizeof(utf8_buffer) / sizeof(char);
std::size_t utf8_sequence_max_length = 10;
std::size_t index = 0;
PT::WTextStream::const_iterator i = buffer.begin();
while( i != buffer.end() )
{
if( index + utf8_sequence_max_length > buffer_len )
{
write_function(utf8_buffer, index);
index = 0;
}
index += PT::IntToUTF8(*i, utf8_buffer + index, buffer_len - index);
++i;
}
if( index > 0 )
{
write_function(utf8_buffer, index);
}
}
// not tested
void WideToUTF8(PT::WTextStream & buffer, std::string & utf8, bool clear, int mode)
{
if( clear )
utf8.clear();
WideToUTF8Generic(buffer, mode, [&utf8](const char * utf8_buffer, std::size_t buffer_len){
utf8.append(utf8_buffer, buffer_len);
});
}
// not tested
void WideToUTF8(PT::WTextStream & buffer, std::ostream & utf8, int mode)
{
WideToUTF8Generic(buffer, mode, [&utf8](const char * utf8_buffer, std::size_t buffer_len){
utf8.write(utf8_buffer, buffer_len);
});
}
} // namespace

19
utf8/utf8.h
View File

@@ -5,7 +5,7 @@
*/
/*
* Copyright (c) 2010-2014, Tomasz Sowa
* Copyright (c) 2010-2018, Tomasz Sowa
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
@@ -40,11 +40,14 @@
#include <fstream>
#include <string>
#include "textstream/textstream.h"
namespace PT
{
/*!
UTF-8, a transformation format of ISO 10646
http://tools.ietf.org/html/rfc3629
@@ -83,6 +86,16 @@ bool UTF8ToWide(const std::string & utf8, std::wstring & res, bool cle
bool UTF8ToWide(std::istream & utf8, std::wstring & res, bool clear = true, int mode = 1);
/*!
converting UTF-8 string to a WTextStream stream
(need to be tested)
*/
bool UTF8ToWide(const char * utf8, size_t utf8_len, WTextStream & res, bool clear = true, int mode = 1);
bool UTF8ToWide(const char * utf8, WTextStream & res, bool clear = true, int mode = 1);
bool UTF8ToWide(const std::string & utf8, WTextStream & res, bool clear = true, int mode = 1);
bool UTF8ToWide(std::istream & utf8, WTextStream & res, bool clear = true, int mode = 1);
/*!
converting one int character to UTF-8
*/
@@ -97,18 +110,22 @@ size_t IntToUTF8(int z, std::ostream & utf8);
bool WideToUTF8(const wchar_t * wide_string, size_t string_len, std::string & utf8, bool clear = true, int mode = 1);
bool WideToUTF8(const wchar_t * wide_string, std::string & utf8, bool clear = true, int mode = 1);
bool WideToUTF8(const std::wstring & wide_string, std::string & utf8, bool clear = true, int mode = 1);
void WideToUTF8(PT::WTextStream & buffer, std::string & utf8, bool clear = true, int mode = 1);// not tested
bool WideToUTF8(const wchar_t * wide_string, size_t string_len, std::ostream & utf8, int mode = 1);
bool WideToUTF8(const wchar_t * wide_string, std::ostream & utf8, int mode = 1);
bool WideToUTF8(const std::wstring & wide_string, std::ostream & utf8, int mode = 1);
void WideToUTF8(PT::WTextStream & buffer, std::ostream & utf8, int mode = 1);// not tested
bool WideToUTF8(const wchar_t * wide_string, size_t string_len, char * utf8, size_t utf8_len, size_t & utf8_written, int mode = 1);
bool WideToUTF8(const wchar_t * wide_string, char * utf8, size_t utf8_len, size_t & utf8_written, int mode = 1);
bool WideToUTF8(const std::wstring & wide_string, char * utf8, size_t utf8_len, size_t & utf8_written, int mode = 1);
// implement void WideToUTF8(PT::WTextStream & buffer, char * utf8, size_t utf8_len, size_t & utf8_written, int mode = 1);
bool WideToUTF8(const wchar_t * wide_string, size_t string_len, char * utf8, size_t utf8_len, int mode = 1);
bool WideToUTF8(const wchar_t * wide_string, char * utf8, size_t utf8_len, int mode = 1);
bool WideToUTF8(const std::wstring & wide_string, char * utf8, size_t utf8_len, int mode = 1);
// implement void WideToUTF8(PT::WTextStream & buffer, char * utf8, size_t utf8_len, int mode = 1);