4 * This file is part of LyX, the document processor.
5 * Licence details can be found in the file COPYING.
7 * \author Lars Gullik Bjønnes
8 * \author Jean-Marc Lasgouttes
10 * Full author contact details are available in file CREDITS.
12 * A collection of string helper functions that works with string.
13 * Some of these would certainly benefit from a rewrite/optimization.
19 #include "support/docstring.h"
27 /// Compare \p s and \p s2, ignoring the case.
28 /// Does not depend on the locale.
29 int compare_no_case(docstring const & s, docstring const & s2);
31 /// Compare \p s and \p s2, ignoring the case of ASCII characters only.
32 int compare_ascii_no_case(std::string const & s, std::string const & s2);
34 /// Compare \p s and \p s2, ignoring the case of ASCII characters only.
35 int compare_ascii_no_case(docstring const & s, docstring const & s2);
38 bool isStrInt(std::string const & str);
40 /// does the std::string represent an unsigned integer value ?
41 bool isStrUnsignedInt(std::string const & str);
44 bool isStrDbl(std::string const & str);
46 /// does the string contain a digit?
47 bool hasDigitASCII(docstring const & str);
49 bool isHexChar(char_type);
51 bool isHex(docstring const & str);
53 int hexToInt(docstring const & str);
55 /// is \p str pure ascii?
56 bool isAscii(docstring const & str);
58 /// is \p str pure ascii?
59 bool isAscii(std::string const & str);
62 * Changes the case of \p c to lowercase.
63 * Don't use this for non-ASCII characters, since it depends on the locale.
64 * This overloaded function is only implemented because the char_type variant
65 * would be used otherwise, and we assert in this function that \p c is in
68 char lowercase(char c);
71 * Changes the case of \p c to uppercase.
72 * Don't use this for non-ASCII characters, since it depends on the locale.
73 * This overloaded function is only implemented because the char_type variant
74 * would be used otherwise, and we assert in this function that \p c is in
77 char uppercase(char c);
79 /// Changes the case of \p c to lowercase.
80 /// Does not depend on the locale.
81 char_type lowercase(char_type c);
83 /// Changes the case of \p c to uppercase.
84 /// Does not depend on the locale.
85 char_type uppercase(char_type c);
87 /// Checks if the supplied character is lower-case
88 bool isLowerCase(char_type ch);
90 /// Checks if the supplied character is upper-case
91 bool isUpperCase(char_type ch);
93 /// same as lowercase(), but ignores locale
94 std::string const ascii_lowercase(std::string const &);
95 docstring const ascii_lowercase(docstring const &);
97 /// Changes the case of \p s to lowercase.
98 /// Does not depend on the locale.
99 docstring const lowercase(docstring const & s);
101 /// Changes the case of \p s to uppercase.
102 /// Does not depend on the locale.
103 docstring const uppercase(docstring const & s);
105 /// Returns the superscript of \p c or \p c if no superscript exists.
106 /// Does not depend on the locale.
107 char_type superscript(char_type c);
109 /// Returns the subscript of \p c or \p c if no subscript exists.
110 /// Does not depend on the locale.
111 char_type subscript(char_type c);
113 /// Does str start with c?
114 bool prefixIs(docstring const & str, char_type c);
116 /// Does str start with pre?
117 bool prefixIs(std::string const & str, std::string const & pre);
118 bool prefixIs(docstring const & str, docstring const & pre);
120 /// Does the string end with this char?
121 bool suffixIs(std::string const &, char);
122 bool suffixIs(docstring const &, char_type);
124 /// Does the string end with this suffix?
125 bool suffixIs(std::string const &, std::string const &);
126 bool suffixIs(docstring const &, docstring const &);
128 /// Is b contained in a?
129 inline bool contains(std::string const & a, std::string const & b)
131 return a.find(b) != std::string::npos;
134 inline bool contains(docstring const & a, docstring const & b)
136 return a.find(b) != docstring::npos;
139 inline bool contains(std::string const & a, char b)
141 return a.find(b) != std::string::npos;
144 inline bool contains(docstring const & a, char_type b)
146 return a.find(b) != docstring::npos;
150 bool containsOnly(std::string const &, std::string const &);
152 /** Extracts a token from this string at the nth delim.
153 Doesn't modify the original string. Similar to strtok.
156 token("a;bc;d", ';', 1) == "bc";
157 token("a;bc;d", ';', 2) == "d";
160 std::string const token(std::string const & a, char delim, int n);
162 docstring const token(docstring const & a, char_type delim, int n);
164 /** Search a token in this string using the delim.
165 Doesn't modify the original string. Returns -1 in case of
169 tokenPos("a;bc;d", ';', "bc") == 1;
170 tokenPos("a;bc;d", ';', "d") == 2;
173 int tokenPos(std::string const & a, char delim, std::string const & tok);
174 int tokenPos(docstring const & a, char_type delim, docstring const & tok);
177 /// Substitute all \a oldchar with \a newchar
178 std::string const subst(std::string const & a, char oldchar, char newchar);
180 /// Substitute all \a oldchar with \a newchar
181 docstring const subst(docstring const & a, char_type oldchar, char_type newchar);
183 /// substitutes all instances of \a oldstr with \a newstr
184 std::string const subst(std::string const & a,
185 std::string const & oldstr, std::string const & newstr);
187 /// substitutes all instances of \a oldstr with \a newstr
188 docstring const subst(docstring const & a,
189 docstring const & oldstr, docstring const & newstr);
191 /// Count all occurences of char \a chr inside \a str
192 int count_char(docstring const & str, docstring::value_type chr);
194 /** Trims characters off the end and beginning of a string.
196 trim("ccabccc", "c") == "ab".
199 docstring const trim(docstring const & a, char const * p = " ");
201 /** Trims characters off the end and beginning of a string.
203 trim("ccabccc", "c") == "ab".
206 std::string const trim(std::string const & a, char const * p = " ");
208 /** Trims characters off the end of a string, removing any character
211 rtrim("abcde", "dec") == "ab".
214 std::string const rtrim(std::string const & a, char const * p = " ");
215 docstring const rtrim(docstring const & a, char const * p = " ");
217 /** Trims characters off the beginning of a string.
219 ("abbabcdef", "ab") = "cdef"
222 std::string const ltrim(std::string const & a, char const * p = " ");
223 docstring const ltrim(docstring const & a, char const * p = " ");
225 /** Splits the string given in the first argument at the first occurence
226 of the third argument, delim.
227 What precedes delim is returned in the second argument, piece; this
228 will be the whole of the string if no delimiter is found.
229 The return value is what follows delim, if anything. So the return
230 value is the null string if no delimiter is found.
231 'a' and 'piece' must be different variables.
235 ret = split(s1, s2, ';') -> ret = "bc", s2 == "a"
238 std::string const split(std::string const & a, std::string & piece, char delim);
239 docstring const split(docstring const & a, docstring & piece, char_type delim);
241 /// Same as split but does not return a piece
242 std::string const split(std::string const & a, char delim);
244 /// Same as split but uses the last delim.
245 std::string const rsplit(std::string const & a, std::string & piece, char delim);
246 docstring const rsplit(docstring const & a, docstring & piece, char_type delim);
247 docstring const rsplit(docstring const & a, char_type delim);
249 /// Escapes non ASCII chars and other problematic characters that cause
250 /// problems in latex labels.
251 docstring const escape(docstring const & lab);
253 /// Word-wraps the provided docstring, returning a line-broken string
254 /// of width no wider than width, with the string broken at spaces.
255 /// If the string cannot be broken appropriately, it returns something
256 /// with "..." at the end, again no wider than width.
257 /// We assume here that str does not contain newlines.
258 /// If indent is positive, then the first line is indented that many
259 /// spaces. If it is negative, then successive lines are indented, as
260 /// if the first line were "outdented".
261 docstring wrap(docstring const & str, int const indent = 0,
262 size_t const width = 80);
264 /// Like the preceding, except it is intended to operate on strings
265 /// that may contain embedded newlines.
266 /// \param numlines Don't return more than numlines lines. If numlines
267 /// is 0, we return everything.
268 docstring wrapParas(docstring const & str, int const indent = 0,
269 size_t const width = 80, size_t const maxlines = 10);
271 /// gives a vector of stringparts which have the delimiter delim
272 /// If \p keepempty is true, empty strings will be pushed to the vector as well
273 std::vector<std::string> const getVectorFromString(std::string const & str,
274 std::string const & delim = std::string(","),
275 bool keepempty = false);
276 std::vector<docstring> const getVectorFromString(docstring const & str,
277 docstring const & delim = from_ascii(","), bool keepempty = false);
279 /// the same vice versa
280 std::string const getStringFromVector(std::vector<std::string> const & vec,
281 std::string const & delim = std::string(","));
282 docstring const getStringFromVector(std::vector<docstring> const & vec,
283 docstring const & delim = from_ascii(","));
285 /// Search \p search_token in \p str and return the position if it is
286 /// found, else -1. The last item in \p str must be "".
287 int findToken(char const * const str[], std::string const & search_token);
289 template <class Arg1>
290 docstring bformat(docstring const & fmt, Arg1);
292 template <class Arg1, class Arg2>
293 docstring bformat(docstring const & fmt, Arg1, Arg2);
295 template <class Arg1, class Arg2, class Arg3>
296 docstring bformat(docstring const & fmt, Arg1, Arg2, Arg3);
298 template <class Arg1, class Arg2, class Arg3, class Arg4>
299 docstring bformat(docstring const & fmt, Arg1, Arg2, Arg3, Arg4);
302 template<> docstring bformat(docstring const & fmt, int arg1);
303 template<> docstring bformat(docstring const & fmt, long arg1);
304 template<> docstring bformat(docstring const & fmt, unsigned int arg1);
305 template<> docstring bformat(docstring const & fmt, docstring arg1);
306 template<> docstring bformat(docstring const & fmt, char * arg1);
307 template<> docstring bformat(docstring const & fmt, docstring arg1, docstring arg2);
308 template<> docstring bformat(docstring const & fmt, docstring arg1, int arg2);
309 template<> docstring bformat(docstring const & fmt, char const * arg1, docstring arg2);
310 template<> docstring bformat(docstring const & fmt, int arg1, int arg2);
311 template<> docstring bformat(docstring const & fmt, docstring arg1, docstring arg2, docstring arg3);
312 template<> docstring bformat(docstring const & fmt, docstring arg1, docstring arg2, docstring arg3, docstring arg4);
315 } // namespace support