X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=src%2FEncoding.h;h=6a1f89c4982a19236b4183a473e3f858243e88bf;hb=0ce10160daadc904ab1221eaeef0d005694f0f0c;hp=b8cf8c59d9fe929fe8e6e727eff7c19d22548ad5;hpb=80b94c6bf50dfb5c96ad1ca6003bac93d844e6ce;p=lyx.git diff --git a/src/Encoding.h b/src/Encoding.h index b8cf8c59d9..6a1f89c498 100644 --- a/src/Encoding.h +++ b/src/Encoding.h @@ -4,7 +4,7 @@ * 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 * \author Jean-Marc Lasgouttes * * Full author contact details are available in file CREDITS. @@ -14,39 +14,65 @@ #define ENCODING_H #include "support/docstring.h" +#include "support/types.h" #include #include +#include namespace lyx { namespace support { class FileName; } +class Buffer; class LaTeXFeatures; +class EncodingException : public std::exception { +public: + EncodingException(char_type c); + virtual ~EncodingException() throw() {} + virtual const char * what() const throw(); + + char_type failed_char; + int par_id; + pos_type pos; +}; + /// class Encoding { public: /// Which LaTeX package handles this encoding? enum Package { - none, - inputenc, - CJK + 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 & i, bool f, Package p); + std::string const & g, std::string const & i, + bool f, bool u, Package p); /// void init() const; /// - std::string const & name() const { return Name_; } + std::string const & name() const { return name_; } /// - std::string const & latexName() const { return LatexName_; } + std::string const & latexName() const { return latexName_; } + /// + std::string const & guiName() const { return guiName_; } /// std::string const & iconvName() const { return iconvName_; } + /// + 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 @@ -54,24 +80,54 @@ 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. */ - docstring const latexChar(char_type c) const; + 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). + */ + 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_; + std::string name_; /// - std::string LatexName_; + std::string latexName_; + /// + std::string guiName_; /// 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. @@ -90,6 +146,12 @@ private: class Encodings { public: + /// + typedef std::set MathCommandSet; + /// + typedef std::set TextCommandSet; + /// + typedef std::set MathSymbolSet; /// typedef std::map EncodingList; /// iterator to iterate over all encodings. @@ -108,11 +170,16 @@ public: /// \param encfile encodings definition file /// \param symbolsfile unicode->LaTeX mapping file void read(support::FileName const & encfile, - support::FileName const & symbolsfile); + support::FileName const & symbolsfile); /// Get encoding from LyX name \p name - Encoding const * getFromLyXName(std::string const & name) const; - /// Get encoding from LaTeX name \p name - Encoding const * getFromLaTeXName(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(); } @@ -120,7 +187,7 @@ public: const_iterator end() const { return encodinglist.end(); } /// - enum Letter_Form { + enum LetterForm { /// FORM_ISOLATED, /// @@ -131,17 +198,101 @@ public: FORM_MEDIAL }; /// - static bool isComposeChar_hebrew(char_type c); + static bool isHebrewComposeChar(char_type c); /// - static bool isComposeChar_arabic(char_type c); + static bool isArabicComposeChar(char_type c); /// - static bool is_arabic_special(char_type c); + static bool isArabicSpecialChar(char_type c); /// - static bool is_arabic(char_type c); + static bool isArabicChar(char_type c); /// - static char_type transformChar(char_type c, Letter_Form form); + 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 + * to a supported script, true is returned and \p preamble is set + * to the corresponding entry in the unicodesymbols file. + * If \p preamble is not empty, a check is made whether code point + * \p c is a known character matching the preamble entry. + */ + static bool isKnownScriptChar(char_type const c, std::string & preamble); + /** + * 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 + * letters and accented characters that are output as math commands. + */ + static bool isMathAlpha(char_type c); + /** + * Register \p c as a mathmode command. + */ + static void addMathCmd(char_type c) { mathcmd.insert(c); } + /** + * Register \p c as a textmode command. + */ + static void addTextCmd(char_type c) { textcmd.insert(c); } + /** + * Register \p c as a mathmode symbol. + */ + static void addMathSym(char_type c) { mathsym.insert(c); } + /** + * Tell whether \p c is registered as a mathmode command. + */ + static bool isMathCmd(char_type c) { return mathcmd.count(c); } + /** + * Tell whether \p c is registered as a textmode command. + */ + static bool isTextCmd(char_type c) { return textcmd.count(c); } + /** + * Tell whether \p c is registered as a mathmode symbol. + */ + static bool isMathSym(char_type c) { return mathsym.count(c); } + /** + * Initialize mathcmd, textcmd, and mathsym sets. + */ + 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, + 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. + * \p needsTermination indicates whether the command needs to be + * terminated by {} or a space. + */ + 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. + * 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, int cmdtype, + bool & needsTermination, docstring & rem, + std::set * req = 0); /** * Add the preamble snippet needed for the output of \p c to * \p features. @@ -149,11 +300,17 @@ public: * package only maps the code point \p c to a command, it does not * make this command available. */ - static void validate(char_type c, LaTeXFeatures & features); + static void validate(char_type c, LaTeXFeatures & features, bool for_mathed = false); private: /// EncodingList encodinglist; + /// + static MathCommandSet mathcmd; + /// + static TextCommandSet textcmd; + /// + static MathSymbolSet mathsym; }; extern Encodings encodings;