4 * This file is part of LyX, the document processor.
5 * Licence details can be found in the file COPYING.
9 * Full author contact details are available in file CREDITS.
19 #include "support/docstream.h"
24 enum mode_type {UNDECIDED_MODE, TEXT_MODE, MATH_MODE, MATHTEXT_MODE, TABLE_MODE};
26 mode_type asMode(mode_type oldmode, std::string const & str);
29 // These are TeX's catcodes
31 catEscape, // 0 backslash
42 catLetter, // 11 a-zA-Z
43 catOther, // 12 none of the above
46 catInvalid // 15 <delete>
51 FLAG_BRACE_LAST = 1 << 1, // last closing brace ends the parsing
52 FLAG_RIGHT = 1 << 2, // next \\right ends the parsing process
53 FLAG_END = 1 << 3, // next \\end ends the parsing process
54 FLAG_BRACK_LAST = 1 << 4, // next closing bracket ends the parsing
55 FLAG_TEXTMODE = 1 << 5, // we are in a box
56 FLAG_ITEM = 1 << 6, // read a (possibly braced token)
57 FLAG_LEAVE = 1 << 7, // leave the loop at the end
58 FLAG_SIMPLE = 1 << 8, // next $ leaves the loop
59 FLAG_EQUATION = 1 << 9, // next \] leaves the loop
60 FLAG_SIMPLE2 = 1 << 10, // next \) leaves the loop
61 FLAG_OPTION = 1 << 11, // read [...] style option
62 FLAG_BRACED = 1 << 12, // read {...} style argument
63 FLAG_CELL = 1 << 13, // read table cell
64 FLAG_TABBING = 1 << 14 // We are inside a tabbing environment
70 // Helper class for parsing
76 Token() : cs_(), cat_(catIgnore) {}
78 Token(docstring const & cs, CatCode cat) : cs_(to_utf8(cs)), cat_(cat) {}
80 /// Returns the token as string
81 std::string const & cs() const { return cs_; }
82 /// Returns the catcode of the token
83 CatCode cat() const { return cat_; }
85 char character() const { return cs_.empty() ? 0 : cs_[0]; }
86 /// Returns the token verbatim
87 std::string asInput() const;
96 std::ostream & operator<<(std::ostream & os, Token const & t);
100 * Actual parser class
102 * The parser parses every character of the inputstream into a token
103 * and classifies the token.
104 * The following transformations are done:
105 * - Consecutive spaces are combined into one single token with CatCode catSpace
106 * - Consecutive newlines are combined into one single token with CatCode catNewline
107 * - Comments and %\n combinations are parsed into one token with CatCode catComment
114 Parser(idocstream & is);
116 Parser(std::string const & s);
120 /// change the latex encoding of the input stream
121 void setEncoding(std::string const & encoding);
122 /// get the current latex encoding of the input stream
123 std::string getEncoding() const { return encoding_latex_; }
126 int lineno() const { return lineno_; }
129 /// dump contents to screen
133 typedef std::pair<bool, std::string> Arg;
135 * Get an argument enclosed by \p left and \p right.
136 * \returns wether an argument was found in \p Arg.first and the
137 * argument in \p Arg.second. \see getArg().
139 Arg getFullArg(char left, char right);
141 * Get an argument enclosed by \p left and \p right.
142 * \returns the argument (without \p left and \p right) or the empty
143 * string if the next non-space token is not \p left. Use
144 * getFullArg() if you need to know wether there was an empty
145 * argument or no argument at all.
147 std::string getArg(char left, char right);
149 * \returns getFullArg('[', ']') including the brackets or the
150 * empty string if there is no such argument.
152 std::string getFullOpt();
154 * \returns getArg('[', ']') including the brackets or the
155 * empty string if there is no such argument.
157 std::string getOpt();
159 * \returns getFullArg('[', ']') including the parentheses or the
160 * empty string if there is no such argument.
162 std::string getOptContent();
164 * the same as getOpt but without the brackets
166 std::string getFullParentheseArg();
168 * \returns the contents of the environment \p name.
169 * <tt>\begin{name}</tt> must be parsed already, <tt>\end{name}</tt>
170 * is parsed but not returned.
172 std::string const verbatimEnvironment(std::string const & name);
174 * Returns the character of the current token and increments
175 * the token position.
179 void error(std::string const & msg);
180 /// Parses one token from \p is
183 void push_back(Token const & t);
184 /// The previous token.
185 Token const prev_token() const;
186 /// The current token.
187 Token const curr_token() const;
189 Token const next_token();
190 /// Make the next token current and return that.
191 Token const get_token();
192 /// \return whether the current token starts a new paragraph
194 /// skips spaces (and comments if \p skip_comments is true)
195 void skip_spaces(bool skip_comments = false);
196 /// puts back spaces (and comments if \p skip_comments is true)
197 void unskip_spaces(bool skip_comments = false);
199 void lex(std::string const & s);
203 std::string verbatim_item();
205 std::string verbatimOption();
206 /// resets the parser to initial state
209 void setCatCode(char c, CatCode cat);
211 CatCode getCatCode(char c) const;
217 std::vector<Token> tokens_;
221 idocstringstream * iss_;
224 /// latex name of the current encoding
225 std::string encoding_latex_;