X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=src%2Fsupport%2Funicode.h;h=6e83180c10692edcbe03fcba4af10fb7a6d6c407;hb=c48091f33a773732fa6c789927e5833e44108d9d;hp=6af70074eb0c32bbeb38c42c3d8c80046314d828;hpb=59fdb31af390b57fba34b54b6f846e5de8706175;p=lyx.git diff --git a/src/support/unicode.h b/src/support/unicode.h index 6af70074eb..6e83180c10 100644 --- a/src/support/unicode.h +++ b/src/support/unicode.h @@ -3,7 +3,7 @@ * This file is part of LyX, the document processor. * Licence details can be found in the file COPYING. * - * \author Lars Gullik Bjønnes + * \author Lars Gullik Bjønnes * * Full author contact details are available in file CREDITS. * @@ -13,52 +13,90 @@ #ifndef LYX_SUPPORT_UNICODE_H #define LYX_SUPPORT_UNICODE_H -#include "support/types.h" +#include "support/strfwd.h" +#include #include -// utf8_to_ucs4 + +namespace lyx { + +class IconvProcessor +{ +public: + IconvProcessor(char const * tocode = "", char const * fromcode = ""); + /// copy constructor needed because of pimpl_ + IconvProcessor(IconvProcessor const &); + /// assignment operator needed because of pimpl_ + void operator=(IconvProcessor const &); + /// destructor + ~IconvProcessor(); + + /// convert any data from \c fromcode to \c tocode unicode format. + /// \return the number of bytes of the converted output buffer. + int convert(char const * in_buffer, size_t in_size, + char * out_buffer, size_t max_out_size); + + /// source encoding + std::string from() const; + /// target encoding + std::string to() const; + +private: + /// open iconv. + /// \return true if the processor is ready to use. + bool init(); + /// hide internals + struct Impl; + Impl * pimpl_; +}; // A single codepoint conversion for utf8_to_ucs4 does not make // sense, so that function is left out. -std::vector -utf8_to_ucs4(std::vector const & utf8str); +std::vector utf8_to_ucs4(std::vector const & utf8str); -std::vector -utf8_to_ucs4(char const * utf8str, size_t ls); +std::vector utf8_to_ucs4(char const * utf8str, size_t ls); -// ucs2_to_ucs4 +// utf16_to_ucs4 -lyx::char_type -ucs2_to_ucs4(unsigned short c); +std::vector utf16_to_ucs4(unsigned short const * s, size_t ls); -std::vector -ucs2_to_ucs4(std::vector const & ucs2str); +// ucs4_to_utf16 -std::vector -ucs2_to_ucs4(unsigned short const * ucs2str, size_t ls); +std::vector ucs4_to_utf16(char_type const * s, size_t ls); -// ucs4_to_ucs2 +// ucs4_to_utf8 -unsigned short -ucs4_to_ucs2(lyx::char_type c); +std::vector ucs4_to_utf8(char_type c); -std::vector -ucs4_to_ucs2(std::vector const & ucs4str); +std::vector ucs4_to_utf8(std::vector const & ucs4str); -std::vector -ucs4_to_ucs2(lyx::char_type const * s, size_t ls); +std::vector ucs4_to_utf8(char_type const * ucs4str, size_t ls); -// ucs4_to_utf8 +/// convert \p s from encoding \p encoding to ucs4. +/// \p encoding must be a valid iconv 8bit encoding +std::vector +eightbit_to_ucs4(char const * s, size_t ls, std::string const & encoding); + +/// convert \p s from ucs4 to encoding \p encoding. +/// \p encoding must be a valid iconv 8bit encoding +std::vector ucs4_to_eightbit(char_type const * ucs4str, + size_t ls, std::string const & encoding); + +/// convert ucs4 character \p c to encoding \p encoding. +/// \p encoding must be a valid iconv 8bit encoding +char ucs4_to_eightbit(char_type c, std::string const & encoding); + +/// +void ucs4_to_multibytes(char_type ucs4, std::vector & out, + std::string const & encoding); -std::vector -ucs4_to_utf8(lyx::char_type c); +extern char const * ucs4_codeset; -std::vector -ucs4_to_utf8(std::vector const & ucs4str); +/// How many bytes does one UCS4 code point use at most in encoding \p encoding? +int max_encoded_bytes(std::string const & encoding); -std::vector -ucs4_to_utf8(lyx::char_type const * ucs4str, size_t ls); +} // namespace lyx #endif