lyx_mirror/src/support/unicode.C

/**
 * \file unicode.C
 * This file is part of LyX, the document processor.
 * Licence details can be found in the file COPYING.
 *
 * \author Lars Gullik Bjønnes
 *
 * Full author contact details are available in file CREDITS.
 *
 * A collection of unicode conversion functions, using iconv.
 */

#include <config.h>

#include "unicode.h"

#include "debug.h"

#include <iconv.h>

#include <cerrno>
#include <iomanip>
#include <map>

using std::endl;

namespace lyx {

#ifdef WORDS_BIGENDIAN
	char const * ucs4_codeset = "UCS-4BE";
	char const * ucs2_codeset = "UCS-2BE";
#else
	char const * ucs4_codeset = "UCS-4LE";
	char const * ucs2_codeset = "UCS-2LE";
#endif

static const iconv_t invalid_cd = (iconv_t)(-1);


struct IconvProcessor::Private {
	Private(): cd(invalid_cd) {}
	iconv_t cd;
};


IconvProcessor::IconvProcessor(char const * tocode,
		char const * fromcode): tocode_(tocode), fromcode_(fromcode),
		pimpl_(new IconvProcessor::Private)
{
}


IconvProcessor::~IconvProcessor()
{
	if (iconv_close(pimpl_->cd) == -1) {
		lyxerr << "Error returned from iconv_close("
			<< errno << ")" << endl;
	}
	delete pimpl_;
}


bool IconvProcessor::init()
{
	if (pimpl_->cd != invalid_cd)
		return true;

	pimpl_->cd = iconv_open(tocode_.c_str(), fromcode_.c_str());
	if (pimpl_->cd != invalid_cd)
		return true;

	lyxerr << "Error returned from iconv_open" << endl;
	switch (errno) {
		case EINVAL:
			lyxerr << "EINVAL The conversion from " << fromcode_
				<< " to " << tocode_
				<< " is not supported by the implementation."
				<< endl;
			break;
		default:
			lyxerr << "\tSome other error: " << errno << endl;
			break;
	}
	return false;
}


int IconvProcessor::convert(char const * buf, size_t buflen,
		char * outbuf, size_t maxoutsize)
{
	if (buflen == 0)
		return 0;

	if (pimpl_->cd == invalid_cd) {
		if (!init())
			return -1;
	}

	char ICONV_CONST * inbuf = const_cast<char ICONV_CONST *>(buf);
	size_t inbytesleft = buflen;
	size_t outbytesleft = maxoutsize;

	int res = iconv(pimpl_->cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);

	//lyxerr << std::dec;
	//lyxerr << "Inbytesleft: " << inbytesleft << endl;
	//lyxerr << "Outbytesleft: " << outbytesleft << endl;

	if (res != -1)
		// Everything went well.
		return maxoutsize - outbytesleft;

	// There are some errors in the conversion
	lyxerr << "Error returned from iconv" << endl;
	switch (errno) {
		case E2BIG:
			lyxerr << "E2BIG  There is not sufficient room at *outbuf." << endl;
			break;
		case EILSEQ:
			lyxerr << "EILSEQ An invalid multibyte sequence"
				<< " has been encountered in the input.\n"
				<< "When converting from " << fromcode_
				<< " to " << tocode_ << ".\n";
			lyxerr << "Input: " << std::hex;
			for (size_t i = 0; i < buflen; ++i) {
				boost::uint32_t const b = buf[i];
				lyxerr << "0x" << b << " ";
			}
			lyxerr << endl;
			break;
		case EINVAL:
			lyxerr << "EINVAL An incomplete multibyte sequence"
				<< " has been encountered in the input.\n"
				<< "When converting from " << fromcode_
				<< " to " << tocode_ << ".\n";
			lyxerr << "Input: " << std::hex;
			for (size_t i = 0; i < buflen; ++i) {
				boost::uint32_t const b = buf[i];
				lyxerr << "0x" << b << " ";
			}
			lyxerr << endl;
			break;
		default:
			lyxerr << "\tSome other error: " << errno << endl;
			break;
	}
	// We got an error so we close down the conversion engine
	if (iconv_close(pimpl_->cd) == -1) {
		lyxerr << "Error returned from iconv_close("
			<< errno << ")" << endl;
	}
	pimpl_->cd = invalid_cd;
	return -1;
}


namespace {


template<typename RetType, typename InType>
std::vector<RetType>
iconv_convert(IconvProcessor & processor,
	      InType const * buf,
	      size_t buflen)
{
	if (buflen == 0)
		return std::vector<RetType>();

	char const * inbuf = reinterpret_cast<char const *>(buf);
	size_t inbytesleft = buflen * sizeof(InType);

	size_t const outsize = 32768;
	static char out[outsize];
	char * outbuf = out;

	int bytes = processor.convert(inbuf, inbytesleft, outbuf, outsize);

	RetType const * tmp = reinterpret_cast<RetType const *>(out);
	return std::vector<RetType>(tmp, tmp + bytes / sizeof(RetType));
}

} // anon namespace


std::vector<lyx::char_type> utf8_to_ucs4(std::vector<char> const & utf8str)
{
	if (utf8str.empty())
		return std::vector<lyx::char_type>();

	return utf8_to_ucs4(&utf8str[0], utf8str.size());
}


std::vector<lyx::char_type>
utf8_to_ucs4(char const * utf8str, size_t ls)
{
	static IconvProcessor processor(ucs4_codeset, "UTF-8");
	return iconv_convert<lyx::char_type>(processor, utf8str, ls);
}


lyx::char_type
ucs2_to_ucs4(unsigned short c)
{
	return ucs2_to_ucs4(&c, 1)[0];
}


std::vector<lyx::char_type>
ucs2_to_ucs4(std::vector<unsigned short> const & ucs2str)
{
	if (ucs2str.empty())
		return std::vector<lyx::char_type>();

	return ucs2_to_ucs4(&ucs2str[0], ucs2str.size());
}


std::vector<lyx::char_type>
ucs2_to_ucs4(unsigned short const * ucs2str, size_t ls)
{
	static IconvProcessor processor(ucs4_codeset, ucs2_codeset);
	return iconv_convert<lyx::char_type>(processor, ucs2str, ls);
}


unsigned short
ucs4_to_ucs2(lyx::char_type c)
{
	return ucs4_to_ucs2(&c, 1)[0];
}


std::vector<unsigned short>
ucs4_to_ucs2(std::vector<lyx::char_type> const & ucs4str)
{
	if (ucs4str.empty())
		return std::vector<unsigned short>();

	return ucs4_to_ucs2(&ucs4str[0], ucs4str.size());
}


std::vector<unsigned short>
ucs4_to_ucs2(lyx::char_type const * s, size_t ls)
{
	static IconvProcessor processor(ucs2_codeset, ucs4_codeset);
	return iconv_convert<unsigned short>(processor, s, ls);
}


std::vector<char>
ucs4_to_utf8(lyx::char_type c)
{
	static IconvProcessor processor("UTF-8", ucs4_codeset);
	return iconv_convert<char>(processor, &c, 1);
}


std::vector<char>
ucs4_to_utf8(std::vector<lyx::char_type> const & ucs4str)
{
	if (ucs4str.empty())
		return std::vector<char>();

	return ucs4_to_utf8(&ucs4str[0], ucs4str.size());
}


std::vector<char>
ucs4_to_utf8(lyx::char_type const * ucs4str, size_t ls)
{
	static IconvProcessor processor("UTF-8", ucs4_codeset);
	return iconv_convert<char>(processor, ucs4str, ls);
}


std::vector<lyx::char_type>
eightbit_to_ucs4(char const * s, size_t ls, std::string const & encoding)
{
	static std::map<std::string, IconvProcessor> processors;
	if (processors.find(encoding) == processors.end()) {
		IconvProcessor processor(ucs4_codeset, encoding.c_str());
		processors.insert(std::make_pair(encoding, processor));
	}
	return iconv_convert<char_type>(processors[encoding], s, ls);
}


std::vector<char>
ucs4_to_eightbit(lyx::char_type const * ucs4str, size_t ls, std::string const & encoding)
{
	static std::map<std::string, IconvProcessor> processors;
	if (processors.find(encoding) == processors.end()) {
		IconvProcessor processor(encoding.c_str(), ucs4_codeset);
		processors.insert(std::make_pair(encoding, processor));
	}
	return iconv_convert<char>(processors[encoding], ucs4str, ls);
}

} // namespace lyx