// -*- C++ -*-
-
-/*
- This is a collection of string helper functions that works
- together with string (and later also with STL String. Some of these
- would certainly benefit from a rewrite/optimization.
-*/
+/**
+ * \file lstrings.h
+ * This file is part of LyX, the document processor.
+ * Licence details can be found in the file COPYING.
+ *
+ * \author Lars Gullik Bjønnes
+ * \author Jean-Marc Lasgouttes
+ *
+ * Full author contact details are available in file CREDITS.
+ *
+ * A collection of string helper functions that works with string.
+ * Some of these would certainly benefit from a rewrite/optimization.
+ */
#ifndef LSTRINGS_H
#define LSTRINGS_H
-#ifdef __GNUG__
-#pragma interface
-#endif
+#include "support/docstring.h"
-#include <cstring>
-#include <cctype>
+#include <string>
+#include <vector>
-#include "Lsstream.h"
-#include "LString.h"
+namespace lyx {
+namespace support {
+/// Compare \p s and \p s2, ignoring the case.
+/// Does not depend on the locale.
+int compare_no_case(docstring const & s, docstring const & s2);
-///
-int compare_no_case(string const & s, string const & s2);
+/// Compare \p s and \p s2, ignoring the case of ASCII characters only.
+int compare_ascii_no_case(std::string const & s, std::string const & s2);
-///
-int compare_no_case(string const & s, string const & s2, unsigned int len);
+/// Compare \p s and \p s2, ignoring the case of ASCII characters only.
+int compare_ascii_no_case(docstring const & s, docstring const & s2);
///
-inline
-int compare(char const * a, char const * b)
-{
- return strcmp(a, b);
-}
+bool isStrInt(std::string const & str);
-///
-inline
-int compare(char const * a, char const * b, unsigned int len)
-{
- return strncmp(a, b, len);
-}
+/// does the std::string represent an unsigned integer value ?
+bool isStrUnsignedInt(std::string const & str);
///
-bool isStrInt(string const & str);
+bool isStrDbl(std::string const & str);
-///
-int strToInt(string const & str);
+/// does the string contain a digit?
+bool hasDigitASCII(docstring const & str);
-///
-bool isStrDbl(string const & str);
+bool isHexChar(char_type);
-///
-double strToDbl(string const & str);
+bool isHex(docstring const & str);
-///
+int hexToInt(docstring const & str);
+
+/// is \p str pure ascii?
+bool isAscii(docstring const & str);
+
+/// is \p str pure ascii?
+bool isAscii(std::string const & str);
+
+/**
+ * Changes the case of \p c to lowercase.
+ * Don't use this for non-ASCII characters, since it depends on the locale.
+ * This overloaded function is only implemented because the char_type variant
+ * would be used otherwise, and we assert in this function that \p c is in
+ * the ASCII range.
+ */
char lowercase(char c);
-///
+/**
+ * Changes the case of \p c to uppercase.
+ * Don't use this for non-ASCII characters, since it depends on the locale.
+ * This overloaded function is only implemented because the char_type variant
+ * would be used otherwise, and we assert in this function that \p c is in
+ * the ASCII range.
+ */
char uppercase(char c);
-///
-string const lowercase(string const &);
+/// Changes the case of \p c to lowercase.
+/// Does not depend on the locale.
+char_type lowercase(char_type c);
-///
-string const uppercase(string const &);
+/// Changes the case of \p c to uppercase.
+/// Does not depend on the locale.
+char_type uppercase(char_type c);
-/// convert T to string
-template<typename T>
-inline
-string const tostr(T const & t)
-{
- std::ostringstream ostr;
- ostr << t;
- return ostr.str().c_str();
- // We need to use the .c_str since we sometimes are using
- // our own string class and that is not compatible with
- // basic_string<char>. (of course we don't want this later)
-}
+/// Checks if the supplied character is lower-case
+bool isLowerCase(char_type ch);
+/// Checks if the supplied character is upper-case
+bool isUpperCase(char_type ch);
-///
-inline
-string const tostr(bool b)
-{
- return (b ? "true" : "false");
-}
-
-/// Does the string start with this prefix?
-bool prefixIs(string const &, char const *);
+/// same as lowercase(), but ignores locale
+std::string const ascii_lowercase(std::string const &);
+docstring const ascii_lowercase(docstring const &);
-/// Does the string end with this char?
-bool suffixIs(string const &, char);
+/// Changes the case of \p s to lowercase.
+/// Does not depend on the locale.
+docstring const lowercase(docstring const & s);
-/// Does the string end with this suffix?
-bool suffixIs(string const &, char const *);
+/// Changes the case of \p s to uppercase.
+/// Does not depend on the locale.
+docstring const uppercase(docstring const & s);
-///
-bool contains(char const * a, string const & b);
+/// Returns the superscript of \p c or \p c if no superscript exists.
+/// Does not depend on the locale.
+char_type superscript(char_type c);
-///
-bool contains(string const & a, char const * b);
+/// Returns the subscript of \p c or \p c if no subscript exists.
+/// Does not depend on the locale.
+char_type subscript(char_type c);
-///
-bool contains(string const & a, string const & b);
+/// Does str start with c?
+bool prefixIs(docstring const & str, char_type c);
-///
-bool contains(char const * a, char const * b);
+/// Does str start with pre?
+bool prefixIs(std::string const & str, std::string const & pre);
+bool prefixIs(docstring const & str, docstring const & pre);
-///
-bool containsOnly(string const &, char const *);
+/// Does the string end with this char?
+bool suffixIs(std::string const &, char);
+bool suffixIs(docstring const &, char_type);
-///
-bool containsOnly(string const &, string const &);
+/// Does the string end with this suffix?
+bool suffixIs(std::string const &, std::string const &);
+bool suffixIs(docstring const &, docstring const &);
-///
-bool containsOnly(char const *, char const *);
+/// Is b contained in a?
+inline bool contains(std::string const & a, std::string const & b)
+{
+ return a.find(b) != std::string::npos;
+}
-///
-bool containsOnly(char const *, string const &);
+inline bool contains(docstring const & a, docstring const & b)
+{
+ return a.find(b) != docstring::npos;
+}
-/// Counts how many of character c there is in a
-unsigned int countChar(string const & a, char c);
+inline bool contains(std::string const & a, char b)
+{
+ return a.find(b) != std::string::npos;
+}
+
+inline bool contains(docstring const & a, char_type b)
+{
+ return a.find(b) != docstring::npos;
+}
+
+///
+bool containsOnly(std::string const &, std::string const &);
/** Extracts a token from this string at the nth delim.
- Doesn't modify the original string. Similar to strtok.
- Example:
- #"a;bc;d".token(';', 1) == "bc";#
- #"a;bc;d".token(';', 2) == "d";#
+ Doesn't modify the original string. Similar to strtok.
+ Example:
+ \code
+ token("a;bc;d", ';', 1) == "bc";
+ token("a;bc;d", ';', 2) == "d";
+ \endcode
*/
-string const token(string const & a, char delim, int n);
+std::string const token(std::string const & a, char delim, int n);
+docstring const token(docstring const & a, char_type delim, int n);
/** Search a token in this string using the delim.
- Doesn't modify the original string. Returns -1 in case of
- failure.
- Example:
- #"a;bc;d".tokenPos(';', "bc") == 1;#
- #"a;bc;d".token(';', "d") == 2;#
+ Doesn't modify the original string. Returns -1 in case of
+ failure.
+ Example:
+ \code
+ tokenPos("a;bc;d", ';', "bc") == 1;
+ tokenPos("a;bc;d", ';', "d") == 2;
+ \endcode
*/
-int tokenPos(string const & a, char delim, string const & tok);
+int tokenPos(std::string const & a, char delim, std::string const & tok);
+int tokenPos(docstring const & a, char_type delim, docstring const & tok);
-/** Compares a string and a (simple) regular expression
- The only element allowed is "*" for any string of characters
- */
-bool regexMatch(string const & a, string const & pattern);
+/// Substitute all \a oldchar with \a newchar
+std::string const subst(std::string const & a, char oldchar, char newchar);
-/// Substitute all "oldchar"s with "newchar"
-string const subst(string const & a, char oldchar, char newchar);
+/// Substitute all \a oldchar with \a newchar
+docstring const subst(docstring const & a, char_type oldchar, char_type newchar);
-/// Substitutes all instances of oldstr with newstr
-string const subst(string const & a,
- char const * oldstr, string const & newstr);
+/// substitutes all instances of \a oldstr with \a newstr
+std::string const subst(std::string const & a,
+ std::string const & oldstr, std::string const & newstr);
-/** Strips characters off the end of a string.
- #"abccc".strip('c') = "ab".#
- */
-string const strip(string const & a, char c = ' ');
+/// substitutes all instances of \a oldstr with \a newstr
+docstring const subst(docstring const & a,
+ docstring const & oldstr, docstring const & newstr);
-/** Strips characters of the beginning of a string.
- #"cccba".frontstrip('c') = "ba"#. */
-string const frontStrip(string const & a, char c = ' ');
+/// Count all occurences of char \a chr inside \a str
+int count_char(docstring const & str, docstring::value_type chr);
-/** Strips characters off the beginning of a string.
- #"ababcdef".frontstrip("ab") = "cdef"# .*/
-string const frontStrip(string const & a, char const * p);
+/** Trims characters off the end and beginning of a string.
+ \code
+ trim("ccabccc", "c") == "ab".
+ \endcode
+*/
+docstring const trim(docstring const & a, char const * p = " ");
+
+/** Trims characters off the end and beginning of a string.
+ \code
+ trim("ccabccc", "c") == "ab".
+ \endcode
+*/
+std::string const trim(std::string const & a, char const * p = " ");
-/** Splits the string by the first delim.
- Splits the string by the first appearance of delim.
- The leading string up to delim is returned in piece (not including
- delim), while the original string is cut from after the delimiter.
- Example:
- #s1= ""; s2= "a;bc".split(s1, ';') -> s1 == "a"; s2 == "bc";#
- */
-string const split(string const & a, string & piece, char delim);
+/** Trims characters off the end of a string, removing any character
+ in p.
+ \code
+ rtrim("abcde", "dec") == "ab".
+ \endcode
+*/
+std::string const rtrim(std::string const & a, char const * p = " ");
+docstring const rtrim(docstring const & a, char const * p = " ");
+
+/** Trims characters off the beginning of a string.
+ \code
+ ("abbabcdef", "ab") = "cdef"
+ \endcode
+*/
+std::string const ltrim(std::string const & a, char const * p = " ");
+docstring const ltrim(docstring const & a, char const * p = " ");
+
+/** Splits the string given in the first argument at the first occurence
+ of the third argument, delim.
+ What precedes delim is returned in the second argument, piece; this
+ will be the whole of the string if no delimiter is found.
+ The return value is what follows delim, if anything. So the return
+ value is the null string if no delimiter is found.
+ Examples:
+ \code
+ s1= "a;bc"; s2= ""
+ ret = split(s1, s2, ';') -> ret = "bc", s2 == "a"
+ \endcode
+ */
+std::string const split(std::string const & a, std::string & piece, char delim);
+docstring const split(docstring const & a, docstring & piece, char_type delim);
/// Same as split but does not return a piece
-string const split(string const & a, char delim);
+std::string const split(std::string const & a, char delim);
/// Same as split but uses the last delim.
-string const rsplit(string const & a, string & piece, char delim);
+std::string const rsplit(std::string const & a, std::string & piece, char delim);
+docstring const rsplit(docstring const & a, char_type delim);
+
+/// Escapes non ASCII chars and other problematic characters that cause
+/// problems in latex labels.
+docstring const escape(docstring const & lab);
+
+/// Word-wraps the provided docstring, returning a line-broken string
+/// of width no wider than width, with the string broken at spaces.
+/// If the string cannot be broken appropriately, it returns something
+/// with "..." at the end, again no wider than width.
+/// We assume here that str does not contain newlines.
+/// If indent is positive, then the first line is indented that many
+/// spaces. If it is negative, then successive lines are indented, as
+/// if the first line were "outdented".
+docstring wrap(docstring const & str, int const indent = 0,
+ size_t const width = 80);
+
+/// Like the preceding, except it is intended to operate on strings
+/// that may contain embedded newlines.
+/// \param numlines Don't return more than numlines lines. If numlines
+/// is 0, we return everything.
+docstring wrapParas(docstring const & str, int const indent = 0,
+ size_t const width = 80, size_t const maxlines = 10);
+
+/// gives a vector of stringparts which have the delimiter delim
+/// If \p keepempty is true, empty strings will be pushed to the vector as well
+std::vector<std::string> const getVectorFromString(std::string const & str,
+ std::string const & delim = std::string(","),
+ bool keepempty = false);
+std::vector<docstring> const getVectorFromString(docstring const & str,
+ docstring const & delim = from_ascii(","), bool keepempty = false);
+
+/// the same vice versa
+std::string const getStringFromVector(std::vector<std::string> const & vec,
+ std::string const & delim = std::string(","));
+docstring const getStringFromVector(std::vector<docstring> const & vec,
+ docstring const & delim = from_ascii(","));
+
+/// Search \p search_token in \p str and return the position if it is
+/// found, else -1. The last item in \p str must be "".
+int findToken(char const * const str[], std::string const & search_token);
+
+/// A test string that is supposed to be translated into the gettext code
+std::string const languageTestString();
+
+template <class Arg1>
+docstring bformat(docstring const & fmt, Arg1);
+
+template <class Arg1, class Arg2>
+docstring bformat(docstring const & fmt, Arg1, Arg2);
+
+template <class Arg1, class Arg2, class Arg3>
+docstring bformat(docstring const & fmt, Arg1, Arg2, Arg3);
+
+template <class Arg1, class Arg2, class Arg3, class Arg4>
+docstring bformat(docstring const & fmt, Arg1, Arg2, Arg3, Arg4);
+
+
+template<> docstring bformat(docstring const & fmt, int arg1);
+template<> docstring bformat(docstring const & fmt, long arg1);
+template<> docstring bformat(docstring const & fmt, unsigned int arg1);
+template<> docstring bformat(docstring const & fmt, docstring arg1);
+template<> docstring bformat(docstring const & fmt, char * arg1);
+template<> docstring bformat(docstring const & fmt, docstring arg1, docstring arg2);
+template<> docstring bformat(docstring const & fmt, docstring arg1, int arg2);
+template<> docstring bformat(docstring const & fmt, char const * arg1, docstring arg2);
+template<> docstring bformat(docstring const & fmt, int arg1, int arg2);
+template<> docstring bformat(docstring const & fmt, docstring arg1, docstring arg2, docstring arg3);
+template<> docstring bformat(docstring const & fmt, docstring arg1, docstring arg2, docstring arg3, docstring arg4);
+
+
+} // namespace support
+} // namespace lyx
#endif