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"
28 /// Compare \p s and \p s2, ignoring the case.
29 /// Does not depend on the locale.
30 int compare_no_case(docstring const & s, docstring const & s2);
32 /// Compare \p s and \p s2, ignoring the case of ASCII characters only.
33 int compare_ascii_no_case(std::string const & s, std::string const & s2);
35 /// Compare \p s and \p s2, ignoring the case of ASCII characters only.
36 int compare_ascii_no_case(docstring const & s, docstring const & s2);
39 bool isStrInt(std::string const & str);
41 /// does the std::string represent an unsigned integer value ?
42 bool isStrUnsignedInt(std::string const & str);
45 bool isStrDbl(std::string const & str);
47 /// does the string contain a digit?
48 bool hasDigit(docstring const & str);
50 bool isHex(docstring const & str);
52 int hexToInt(docstring const & str);
54 /// is \p str pure ascii?
55 bool isAscii(docstring const & str);
57 /// is \p str pure ascii?
58 bool isAscii(std::string const & str);
61 * Changes the case of \p c to lowercase.
62 * Don't use this for non-ASCII characters, since it depends on the locale.
63 * This overloaded function is only implemented because the char_type variant
64 * would be used otherwise, and we assert in this function that \p c is in
67 char lowercase(char c);
70 * Changes the case of \p c to uppercase.
71 * Don't use this for non-ASCII characters, since it depends on the locale.
72 * This overloaded function is only implemented because the char_type variant
73 * would be used otherwise, and we assert in this function that \p c is in
76 char uppercase(char c);
78 /// Changes the case of \p c to lowercase.
79 /// Does not depend on the locale.
80 char_type lowercase(char_type c);
82 /// Changes the case of \p c to uppercase.
83 /// Does not depend on the locale.
84 char_type uppercase(char_type c);
86 /// same as lowercase(), but ignores locale
87 std::string const ascii_lowercase(std::string const &);
88 docstring const ascii_lowercase(docstring const &);
90 /// Changes the case of \p s to lowercase.
91 /// Does not depend on the locale.
92 docstring const lowercase(docstring const & s);
94 /// Changes the case of \p s to uppercase.
95 /// Does not depend on the locale.
96 docstring const uppercase(docstring const & s);
98 /// Does str start with c?
99 bool prefixIs(docstring const & str, char_type c);
101 /// Does str start with pre?
102 bool prefixIs(std::string const & str, std::string const & pre);
103 bool prefixIs(docstring const & str, docstring const & pre);
105 /// Does the string end with this char?
106 bool suffixIs(std::string const &, char);
107 bool suffixIs(docstring const &, char_type);
109 /// Does the string end with this suffix?
110 bool suffixIs(std::string const &, std::string const &);
111 bool suffixIs(docstring const &, docstring const &);
113 /// Is b contained in a?
114 inline bool contains(std::string const & a, std::string const & b)
116 return a.find(b) != std::string::npos;
119 inline bool contains(docstring const & a, docstring const & b)
121 return a.find(b) != docstring::npos;
124 inline bool contains(std::string const & a, char b)
126 return a.find(b) != std::string::npos;
129 inline bool contains(docstring const & a, char_type b)
131 return a.find(b) != docstring::npos;
135 bool containsOnly(std::string const &, std::string const &);
137 /** Extracts a token from this string at the nth delim.
138 Doesn't modify the original string. Similar to strtok.
141 token("a;bc;d", ';', 1) == "bc";
142 token("a;bc;d", ';', 2) == "d";
145 std::string const token(std::string const & a, char delim, int n);
147 docstring const token(docstring const & a, char_type delim, int n);
149 /** Search a token in this string using the delim.
150 Doesn't modify the original string. Returns -1 in case of
154 tokenPos("a;bc;d", ';', "bc") == 1;
155 tokenPos("a;bc;d", ';', "d") == 2;
158 int tokenPos(std::string const & a, char delim, std::string const & tok);
159 int tokenPos(docstring const & a, char_type delim, docstring const & tok);
162 /// Substitute all \a oldchar with \a newchar
163 std::string const subst(std::string const & a, char oldchar, char newchar);
165 /// Substitute all \a oldchar with \a newchar
166 docstring const subst(docstring const & a, char_type oldchar, char_type newchar);
168 /// substitutes all instances of \a oldstr with \a newstr
169 std::string const subst(std::string const & a,
170 std::string const & oldstr, std::string const & newstr);
172 /// substitutes all instances of \a oldstr with \a newstr
173 docstring const subst(docstring const & a,
174 docstring const & oldstr, docstring const & newstr);
176 /// Count all occurences of char \a chr inside \a str
177 int count_char(docstring const & str, docstring::value_type chr);
179 /** Trims characters off the end and beginning of a string.
181 trim("ccabccc", "c") == "ab".
184 docstring const trim(docstring const & a, char const * p = " ");
186 /** Trims characters off the end and beginning of a string.
188 trim("ccabccc", "c") == "ab".
191 std::string const trim(std::string const & a, char const * p = " ");
193 /** Trims characters off the end of a string, removing any character
196 rtrim("abcde", "dec") == "ab".
199 std::string const rtrim(std::string const & a, char const * p = " ");
200 docstring const rtrim(docstring const & a, char const * p = " ");
202 /** Trims characters off the beginning of a string.
204 ("abbabcdef", "ab") = "cdef"
207 std::string const ltrim(std::string const & a, char const * p = " ");
208 docstring const ltrim(docstring const & a, char const * p = " ");
210 /** Splits the string given in the first argument at the first occurence
211 of the third argument, delim.
212 What precedes delim is returned in the second argument, piece; this
213 will be the whole of the string if no delimiter is found.
214 The return value is what follows delim, if anything. So the return
215 value is the null string if no delimiter is found.
219 ret = split(s1, s2, ';') -> ret = "bc", s2 == "a"
222 std::string const split(std::string const & a, std::string & piece, char delim);
223 docstring const split(docstring const & a, docstring & piece, char_type delim);
225 /// Same as split but does not return a piece
226 std::string const split(std::string const & a, char delim);
228 /// Same as split but uses the last delim.
229 std::string const rsplit(std::string const & a, std::string & piece, char delim);
230 docstring const rsplit(docstring const & a, char_type delim);
232 /// Escapes non ASCII chars and other problematic characters that cause
233 /// problems in latex labels.
234 docstring const escape(docstring const & lab);
236 /// Word-wraps the provided docstring, returning a line-broken string
237 /// of width no wider than width, with the string broken at spaces.
238 /// If the string cannot be broken appropriately, it returns something
239 /// with "..." at the end, again no wider than width.
240 /// We assume here that str does not contain newlines.
241 /// If indent is positive, then the first line is indented that many
242 /// spaces. If it is negative, then successive lines are indented, as
243 /// if the first line were "outdented".
244 docstring wrap(docstring const & str, int const indent = 0,
245 size_t const width = 80);
247 /// Like the preceding, except it is intended to operate on strings
248 /// that may contain embedded newlines.
249 /// \param numlines Don't return more than numlines lines. If numlines
250 /// is 0, we return everything.
251 docstring wrapParas(docstring const & str, int const indent = 0,
252 size_t const width = 80, size_t const maxlines = 10);
254 /// gives a vector of stringparts which have the delimiter delim
255 /// If \p keepempty is true, empty strings will be pushed to the vector as well
256 std::vector<std::string> const getVectorFromString(std::string const & str,
257 std::string const & delim = std::string(","),
258 bool keepempty = false);
259 std::vector<docstring> const getVectorFromString(docstring const & str,
260 docstring const & delim = from_ascii(","), bool keepempty = false);
262 /// the same vice versa
263 std::string const getStringFromVector(std::vector<std::string> const & vec,
264 std::string const & delim = std::string(","));
265 docstring const getStringFromVector(std::vector<docstring> const & vec,
266 docstring const & delim = from_ascii(","));
268 /// Search \p search_token in \p str and return the position if it is
269 /// found, else -1. The last item in \p str must be "".
270 int findToken(char const * const str[], std::string const & search_token);
272 template <class Arg1>
273 docstring bformat(docstring const & fmt, Arg1);
275 template <class Arg1, class Arg2>
276 docstring bformat(docstring const & fmt, Arg1, Arg2);
278 template <class Arg1, class Arg2, class Arg3>
279 docstring bformat(docstring const & fmt, Arg1, Arg2, Arg3);
281 template <class Arg1, class Arg2, class Arg3, class Arg4>
282 docstring bformat(docstring const & fmt, Arg1, Arg2, Arg3, Arg4);
285 template<> docstring bformat(docstring const & fmt, int arg1);
286 template<> docstring bformat(docstring const & fmt, long arg1);
287 template<> docstring bformat(docstring const & fmt, unsigned int arg1);
288 template<> docstring bformat(docstring const & fmt, docstring arg1);
289 template<> docstring bformat(docstring const & fmt, char * arg1);
290 template<> docstring bformat(docstring const & fmt, docstring arg1, docstring arg2);
291 template<> docstring bformat(docstring const & fmt, docstring arg1, int arg2);
292 template<> docstring bformat(docstring const & fmt, char const * arg1, docstring arg2);
293 template<> docstring bformat(docstring const & fmt, int arg1, int arg2);
294 template<> docstring bformat(docstring const & fmt, docstring arg1, docstring arg2, docstring arg3);
295 template<> docstring bformat(docstring const & fmt, docstring arg1, docstring arg2, docstring arg3, docstring arg4);
298 } // namespace support