X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=src%2FEncoding.h;h=6a1f89c4982a19236b4183a473e3f858243e88bf;hb=4db3e641ed6765e005343010cb90ee8af26f8f99;hp=615556dff8d4117217e0388a4964dcfaea955bc8;hpb=f1cba8ff64b369792fd49f5ddf90e8126ab476ac;p=lyx.git diff --git a/src/Encoding.h b/src/Encoding.h index 615556dff8..6a1f89c498 100644 --- a/src/Encoding.h +++ b/src/Encoding.h @@ -44,17 +44,19 @@ class Encoding { public: /// Which LaTeX package handles this encoding? enum Package { - none, - inputenc, - CJK, - japanese + none = 1, + inputenc = 2, + CJK = 4, + japanese = 8 }; + /// Represent any of the above packages + static int const any; /// Encoding() {} /// Encoding(std::string const & n, std::string const & l, std::string const & g, std::string const & i, - bool f, Package p); + bool f, bool u, Package p); /// void init() const; /// @@ -66,7 +68,11 @@ public: /// std::string const & iconvName() const { return iconvName_; } /// - bool const & hasFixedWidth() const { return fixedwidth_; } + bool hasFixedWidth() const { return fixedwidth_; } + /// + bool unsafe() const { return unsafe_; } + /// \p c is representable in this encoding without a LaTeX macro + bool encodable(char_type c) const; /** * Convert \p c to something that LaTeX can understand. * This is either the character itself (if it is representable @@ -74,13 +80,34 @@ public: * If the character is not representable in this encoding, but no * LaTeX macro is known, a warning is given of lyxerr, and the * character is returned. + * \return the converted character and a flag indicating whether + * the command needs to be terminated by {} or a space. + */ + std::pair latexChar(char_type c) const; + /** + * Convert \p input to something that LaTeX can understand. + * This is either the string itself (if it is representable + * in this encoding), or a LaTeX macro. + * If a character is not representable in this encoding, but no + * LaTeX macro is known, a warning is given of lyxerr, and the + * character is returned in the second string of the pair and + * omitted in the first. + * \p dryrun specifies whether the string is used within source + * preview (which yields a special warning). */ - docstring latexChar(char_type c, bool for_mathed = false) const; + std::pair latexString(docstring const input, + bool dryrun = false) const; /// Which LaTeX package handles this encoding? Package package() const { return package_; } /// A list of all characters usable in this encoding std::vector symbolsList() const; private: + /** + * Do we have to output this character as LaTeX command in any case? + * This is true if the "force" flag is set. + * We need this if the inputencoding does not support a certain glyph. + */ + bool isForced(char_type c) const; /// std::string name_; /// @@ -91,11 +118,16 @@ private: std::string iconvName_; /// Is this a fixed width encoding? bool fixedwidth_; + /// Is this encoding TeX unsafe, e.g. control characters like {, } + /// and \\ may appear in high bytes? + bool unsafe_; /// typedef std::set CharSet; /// Set of UCS4 characters that we can encode (for singlebyte /// encodings only) mutable CharSet encodable_; + /// Set of UCS4 characters that we can't encode + CharSet const * forced_; /// All code points below this are encodable. This helps us to avoid /// lokup of ASCII characters in encodable_ and gives about 1 sec /// speedup on export of the Userguide. @@ -140,9 +172,14 @@ public: void read(support::FileName const & encfile, support::FileName const & symbolsfile); /// Get encoding from LyX name \p name - Encoding const * fromLyXName(std::string const & name) const; - /// Get encoding from LaTeX name \p name - Encoding const * fromLaTeXName(std::string const & name) const; + Encoding const * + fromLyXName(std::string const & name, bool allowUnsafe = false) const; + /// Get encoding from LaTeX name \p name and package \p package + Encoding const * fromLaTeXName(std::string const & name, + int const & package = Encoding::any, bool allowUnsafe = false) const; + /// Get encoding from iconv name \p name and package \p package + Encoding const * fromIconvName(std::string const & name, + int const & package = Encoding::any, bool allowUnsafe = false) const; /// const_iterator begin() const { return encodinglist.begin(); } @@ -172,6 +209,8 @@ public: static char_type transformChar(char_type c, LetterForm form); /// Is this a combining char? static bool isCombiningChar(char_type c); + /// Return the TIPA shortcut + static std::string const TIPAShortcut(char_type c); /** * Is this a known char from some language? * If \p preamble is empty and code point \p c is known to belong @@ -181,16 +220,10 @@ public: * \p c is a known character matching the preamble entry. */ static bool isKnownScriptChar(char_type const c, std::string & preamble); - /** - * Do we have to output this character as LaTeX command in any case? - * This is true if the "force" flag is set. - * We need this if the inputencoding does not support a certain glyph. - */ - static bool isForced(char_type c); /** * Do we have to display in italics this character when in mathmode? * This is true if the "mathalpha" flag is set. We use this for - * accented characters that are output as math commands. + * letters and accented characters that are output as math commands. */ static bool isMathAlpha(char_type c); /** @@ -220,26 +253,46 @@ public: /** * Initialize mathcmd, textcmd, and mathsym sets. */ - static void initUnicodeMath(Buffer const & buffer); + static void initUnicodeMath(Buffer const & buffer, bool for_master = true); /** * If \p c cannot be encoded in the given \p encoding, convert * it to something that LaTeX can understand in mathmode. + * \p needsTermination indicates whether the command needs to be + * terminated by {} or a space. * \return whether \p command is a mathmode command */ static bool latexMathChar(char_type c, bool mathmode, - Encoding const * encoding, docstring & command); + Encoding const * encoding, docstring & command, + bool & needsTermination); /** * Convert the LaTeX command in \p cmd to the corresponding unicode - * point and set \p combining to true if it is a combining symbol + * point and set \p combining to true if it is a combining symbol. + * \p needsTermination indicates whether the command needs to be + * terminated by {} or a space. */ - static char_type fromLaTeXCommand(docstring const & cmd, bool & combining); + static char_type fromLaTeXCommand(docstring const & cmd, int cmdtype, + bool & combining, bool & needsTermination, + std::set * req = 0); + /// + enum LatexCmd { + /// + MATH_CMD = 1, + /// + TEXT_CMD = 2 + }; /** * Convert the LaTeX commands in \p cmd and \return a docstring * of corresponding unicode points. The conversion stops at the * first command which could not be converted, and the remaining - * unconverted commands are returned in \p rem + * unconverted commands are returned in \p rem. + * The \p cmdtype parameter can be used to limit recognized + * commands to math or text mode commands only. + * \p needsTermination indicates whether the command needs to be + * terminated by {} or a space. */ - static docstring fromLaTeXCommand(docstring const & cmd, docstring & rem); + static docstring fromLaTeXCommand(docstring const & cmd, int cmdtype, + bool & needsTermination, docstring & rem, + std::set * req = 0); /** * Add the preamble snippet needed for the output of \p c to * \p features.