#define ENCODING_H
#include "support/docstring.h"
+#include "support/types.h"
#include <map>
#include <set>
+#include <vector>
namespace lyx {
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
+ };
///
Encoding() {}
///
Encoding(std::string const & n, std::string const & l,
- std::string const & i);
+ std::string const & g, std::string const & i,
+ bool f, 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 const & hasFixedWidth() const { return fixedwidth_; }
/**
* Convert \p c to something that LaTeX can understand.
* This is either the character itself (if it is representable
* LaTeX macro is known, a warning is given of lyxerr, and the
* character is returned.
*/
- docstring const latexChar(char_type c) const;
+ docstring latexChar(char_type c, bool for_mathed = false) const;
+ /// Which LaTeX package handles this encoding?
+ Package package() const { return package_; }
+ /// A list of all characters usable in this encoding
+ std::vector<char_type> symbolsList() const;
private:
///
- 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_;
///
typedef std::set<char_type> CharSet;
/// Set of UCS4 characters that we can encode (for singlebyte
/// encodings only)
- CharSet encodable_;
+ mutable CharSet encodable_;
/// 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.
- char_type start_encodable_;
+ mutable char_type start_encodable_;
+ /// Which LaTeX package handles this encoding?
+ Package package_;
+ /**
+ * If this is true the stored information about the encoding covers
+ * all encodable characters. We set this to false initially so that
+ * we only need to query iconv for the actually used encodings.
+ * This is needed especially for the multibyte encodings, if we
+ * complete all encoding info on startup it takes 2-3 minutes.
+ */
+ mutable bool complete_;
};
class Encodings {
/// \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;
+ Encoding const * fromLyXName(std::string const & name) const;
/// Get encoding from LaTeX name \p name
- Encoding const * getFromLaTeXName(std::string const & name) const;
+ Encoding const * fromLaTeXName(std::string const & name) const;
///
const_iterator begin() const { return encodinglist.begin(); }
const_iterator end() const { return encodinglist.end(); }
///
- enum Letter_Form {
+ enum LetterForm {
///
FORM_ISOLATED,
///
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);
+ /**
+ * 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 output this character as LaTeX command in any case?
+ * This is true if the "forced" flag is set.
+ * We need this if the inputencoding does not support a certain glyph.
+ */
+ static bool isForced(char_type c);
+ /**
+ * If \p c cannot be encoded in the given \p encoding, convert
+ * it to something that LaTeX can understand in math mode.
+ * \return whether \p command is a math mode command
+ */
+ static bool latexMathChar(char_type c, Encoding const * encoding, docstring & command);
+
+ /**
+ * Convert the LaTeX command in \p cmd to the corresponding unicode
+ * point and set \p combining to true if it is a combining symbol
+ */
+ static char_type fromLaTeXCommand(docstring const & cmd, bool & combining);
+ /**
+ * 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
+ */
+ static docstring fromLaTeXCommand(docstring const & cmd, docstring & rem);
/**
* Add the preamble snippet needed for the output of \p c to
* \p features.
* 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:
///
projects / lyx.git / blobdiff