-/* This file is part of
- * ======================================================
- *
- * LyX, The Document Processor
- *
- * Copyright 1995 Matthias Ettrich
- * Copyright 1995-2000 The LyX Team.
+/**
+ * \file lstrings.C
+ * 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.
+ */
#include <config.h>
-#ifdef __GNUG__
-#pragma implementation
-#endif
+#include "support/lstrings.h"
+#include "support/lyxlib.h"
+#include "support/tostr.h"
-#include <algorithm>
+#include "debug.h"
+
+#include <boost/tokenizer.hpp>
+#include <boost/assert.hpp>
+#include <boost/format.hpp>
#include <cctype>
#include <cstdlib>
-#include "LString.h"
-#include "lstrings.h"
-#include "LRegex.h"
-#include "LAssert.h"
+#include <algorithm>
+#include <sstream>
-using std::count;
using std::transform;
+using std::string;
+using std::vector;
+
#ifndef CXX_GLOBAL_CSTD
+using std::atof;
+using std::isdigit;
using std::tolower;
using std::toupper;
#endif
+namespace lyx {
+namespace support {
+
int compare_no_case(string const & s, string const & s2)
{
string::const_iterator p = s.begin();
}
+namespace {
+ int ascii_tolower(int c) {
+ if (c >= 'A' && c <= 'Z')
+ return c - 'A' + 'a';
+ return c;
+ }
+}
+
+
+int compare_ascii_no_case(string const & s, string const & s2)
+{
+ string::const_iterator p = s.begin();
+ string::const_iterator p2 = s2.begin();
+
+ while (p != s.end() && p2 != s2.end()) {
+ int const lc1 = ascii_tolower(*p);
+ int const lc2 = ascii_tolower(*p2);
+ if (lc1 != lc2)
+ return (lc1 < lc2) ? -1 : 1;
+ ++p;
+ ++p2;
+ }
+
+ if (s.size() == s2.size())
+ return 0;
+ if (s.size() < s2.size())
+ return -1;
+ return 1;
+}
+
+
int compare_no_case(string const & s, string const & s2, unsigned int len)
{
string::const_iterator p = s.begin();
bool isStrInt(string const & str)
{
if (str.empty()) return false;
-
+
// Remove leading and trailing white space chars.
- string const tmpstr = frontStrip(strip(str, ' '), ' ');
+ string const tmpstr = trim(str);
if (tmpstr.empty()) return false;
-
+
string::const_iterator cit = tmpstr.begin();
if ((*cit) == '-') ++cit;
string::const_iterator end = tmpstr.end();
bool isStrUnsignedInt(string const & str)
{
if (str.empty()) return false;
-
+
// Remove leading and trailing white space chars.
- string const tmpstr = frontStrip(strip(str, ' '), ' ');
+ string const tmpstr = trim(str);
if (tmpstr.empty()) return false;
-
+
string::const_iterator cit = tmpstr.begin();
string::const_iterator end = tmpstr.end();
for (; cit != end; ++cit) {
{
if (isStrInt(str)) {
// Remove leading and trailing white space chars.
- string const tmpstr = frontStrip(strip(str, ' '), ' ');
+ string const tmpstr = trim(str);
// Do the conversion proper.
- return lyx::atoi(tmpstr);
+ return atoi(tmpstr);
} else {
return 0;
}
{
if (isStrUnsignedInt(str)) {
// Remove leading and trailing white space chars.
- string const tmpstr = frontStrip(strip(str, ' '), ' ');
+ string const tmpstr = trim(str);
// Do the conversion proper.
- return lyx::atoi(tmpstr);
+ return atoi(tmpstr);
} else {
return 0;
}
bool isStrDbl(string const & str)
{
if (str.empty()) return false;
-
+
// Remove leading and trailing white space chars.
- string const tmpstr = frontStrip(strip(str, ' '), ' ');
+ string const tmpstr = trim(str);
if (tmpstr.empty()) return false;
// if (1 < tmpstr.count('.')) return false;
{
if (isStrDbl(str)) {
// Remove leading and trailing white space chars.
- string const tmpstr = frontStrip(strip(str, ' '), ' ');
+ string const tmpstr = trim(str);
// Do the conversion proper.
return ::atof(tmpstr.c_str());
} else {
}
-char lowercase(char c)
-{
- return char( tolower(c) );
+char lowercase(char c)
+{
+ return char(tolower(c));
}
-char uppercase(char c)
-{
- return char( toupper(c) );
+char uppercase(char c)
+{
+ return char(toupper(c));
}
+namespace {
+
+// since we cannot use std::tolower and std::toupper directly in the
+// calls to std::transform yet, we use these helper clases. (Lgb)
+
+struct local_lowercase {
+ char operator()(char c) const {
+ return tolower(c);
+ }
+};
+
+struct local_uppercase {
+ char operator()(char c) const {
+ return toupper(c);
+ }
+};
+
+struct local_ascii_lowercase {
+ char operator()(char c) const {
+ return ascii_tolower(c);
+ }
+};
+
+} // end of anon namespace
+
string const lowercase(string const & a)
{
string tmp(a);
-#if 1
- string::iterator result = tmp.begin();
- string::iterator end = tmp.end();
- for (string::iterator first = tmp.begin();
- first != end; ++first, ++result) {
- *result = lowercase(*first);
- }
-#else
- // We want to use this one. (Lgb)
- transform(tmp.begin(), tmp.end(), tmp.begin(), lowercase);
-#endif
+ transform(tmp.begin(), tmp.end(), tmp.begin(), local_lowercase());
return tmp;
}
-
string const uppercase(string const & a)
{
string tmp(a);
-#if 1
- string::iterator result = tmp.begin();
- string::iterator end = tmp.end();
- for (string::iterator first = tmp.begin();
- first != end; ++first, ++result) {
- *result = uppercase(*first);
- }
-#else
- // We want to use this one. (Lgb)
- transform(tmp.begin(), tmp.end(), tmp.begin(), uppercase);
-#endif
+ transform(tmp.begin(), tmp.end(), tmp.begin(), local_uppercase());
return tmp;
}
-bool prefixIs(string const & a, char const * pre)
+string const ascii_lowercase(string const & a)
{
- Assert(pre);
-
- size_t const l = strlen(pre);
- string::size_type const alen = a.length();
-
- if (l > alen || a.empty())
- return false;
- else {
-#if !defined(USE_INCLUDED_STRING) && !defined(STD_STRING_IS_GOOD)
- // Delete this code when the compilers get a bit better.
- return ::strncmp(a.c_str(), pre, l) == 0;
-#else
- // This is the code that we really want to use
- // but until gcc ships with a basic_string that
- // implements std::string correctly we have to
- // use the code above.
- return a.compare(0, l, pre, l) == 0;
-#endif
- }
+ string tmp(a);
+ transform(tmp.begin(), tmp.end(), tmp.begin(),
+ local_ascii_lowercase());
+ return tmp;
}
{
string::size_type const prelen = pre.length();
string::size_type const alen = a.length();
-
+
if (prelen > alen || a.empty())
return false;
else {
-#if !defined(USE_INCLUDED_STRING) && !defined(STD_STRING_IS_GOOD)
- return ::strncmp(a.c_str(), pre.c_str(), prelen) == 0;
-#else
+#if defined(STD_STRING_IS_GOOD)
return a.compare(0, prelen, pre) == 0;
+#else
+ return ::strncmp(a.c_str(), pre.c_str(), prelen) == 0;
#endif
}
}
}
-bool suffixIs(string const & a, char const * suf)
-{
- Assert(suf);
-
- size_t const suflen = strlen(suf);
- string::size_type const alen = a.length();
-
- if (suflen > alen)
- return false;
- else {
-#if !defined(USE_INCLUDED_STRING) && !defined(STD_STRING_IS_GOOD)
- // Delete this code when the compilers get a bit better.
- string tmp(a, alen - suflen);
- return ::strncmp(tmp.c_str(), suf, suflen) == 0;
-#else
- // This is the code that we really want to use
- // but until gcc ships with a basic_string that
- // implements std::string correctly we have to
- // use the code above.
- return a.compare(alen - suflen, suflen, suf) == 0;
-#endif
- }
-}
-
-
bool suffixIs(string const & a, string const & suf)
{
string::size_type const suflen = suf.length();
string::size_type const alen = a.length();
-
+
if (suflen > alen) {
return false;
} else {
}
-bool contains(char const * a, string const & b)
-{
- Assert(a);
- string const at(a);
- return contains(at, b);
-}
-
-
-bool contains(string const & a, char const * b)
-{
- Assert(b);
- string const bt(b);
- return contains(a, bt);
-}
-
-
-bool contains(string const & a, string const & b)
-{
- if (a.empty())
- return false;
- return a.find(b) != string::npos;
-}
-
-
-bool contains(string const & a, char b)
-{
- if (a.empty())
- return false;
- return a.find(b) != string::npos;
-}
-
-
-bool contains(char const * a, char const * b)
-{
- Assert(a && b);
- string const at(a);
- string const bt(b);
- return contains(at, bt);
-}
-
-
-bool containsOnly(string const & s, char const * cset)
-{
- Assert(cset);
-
- return s.find_first_not_of(cset) == string::npos;
-}
-
-
bool containsOnly(string const & s, string const & cset)
{
return s.find_first_not_of(cset) == string::npos;
}
-bool containsOnly(char const * s, char const * cset)
-{
- Assert(s && cset);
-
- return string(s).find_first_not_of(cset) == string::npos;
-}
-
-
-bool containsOnly(char const * s, string const & cset)
-{
- Assert(s);
-
- return string(s).find_first_not_of(cset) == string::npos;
-}
-
-
-string::size_type countChar(string const & a, char c)
-{
-#ifdef HAVE_STD_COUNT
- return count(a.begin(), a.end(), c);
-#else
- unsigned int n = 0;
- count(a.begin(), a.end(), c, n);
- return n;
-#endif
-}
-
-
// ale970405+lasgoutt-970425
// rewritten to use new string (Lgb)
string const token(string const & a, char delim, int n)
{
if (a.empty()) return string();
-
+
string::size_type k = 0;
string::size_type i = 0;
}
-bool regexMatch(string const & a, string const & pattern)
-{
- // We massage the pattern a bit so that the usual
- // shell pattern we all are used to will work.
- // One nice thing about using a real regex is that
- // things like "*.*[^~]" will work also.
- // build the regex string.
- string regex(pattern);
- regex = subst(regex, ".", "\\.");
- regex = subst(regex, "*", ".*");
- LRegex reg(regex);
- return reg.exact_match(a);
-}
-
-
string const subst(string const & a, char oldchar, char newchar)
{
string tmp(a);
}
-string const subst(string const & a,
- char const * oldstr, string const & newstr)
-{
- Assert(oldstr);
-
- string lstr(a);
- string::size_type i = 0;
- string::size_type olen = strlen(oldstr);
- while((i = lstr.find(oldstr, i)) != string::npos) {
- lstr.replace(i, olen, newstr);
- i += newstr.length(); // We need to be sure that we dont
- // use the same i over and over again.
- }
- return lstr;
-}
-
-
string const subst(string const & a,
string const & oldstr, string const & newstr)
{
- string lstr(a);
+ string lstr = a;
string::size_type i = 0;
string::size_type const olen = oldstr.length();
- while((i = lstr.find(oldstr, i)) != string::npos) {
+ while ((i = lstr.find(oldstr, i)) != string::npos) {
lstr.replace(i, olen, newstr);
i += newstr.length(); // We need to be sure that we dont
// use the same i over and over again.
}
-string const strip(string const & a, char c)
+string const trim(string const & a, char const * p)
{
- if (a.empty()) return a;
- string tmp(a);
- string::size_type i = tmp.find_last_not_of(c);
- if (i == a.length() - 1) return tmp; // no c's at end of a
- if (i != string::npos)
- tmp.erase(i + 1, string::npos);
- else
- tmp.erase(); // only c in the whole string
- return tmp;
+ BOOST_ASSERT(p);
+
+ if (a.empty() || !*p)
+ return a;
+
+ string::size_type r = a.find_last_not_of(p);
+ string::size_type l = a.find_first_not_of(p);
+
+ // Is this the minimal test? (lgb)
+ if (r == string::npos && l == string::npos)
+ return string();
+
+ return a.substr(l, r - l + 1);
}
-string const frontStrip(string const & a, char const * p)
+string const rtrim(string const & a, char const * p)
{
- Assert(p);
-
- if (a.empty() || !*p) return a;
- string tmp(a);
- string::size_type i = tmp.find_first_not_of(p);
- if (i > 0)
- tmp.erase(0, i);
- return tmp;
+ BOOST_ASSERT(p);
+
+ if (a.empty() || !*p)
+ return a;
+
+ string::size_type r = a.find_last_not_of(p);
+
+ // Is this test really needed? (Lgb)
+ if (r == string::npos)
+ return string();
+
+ return a.substr(0, r + 1);
}
-string const frontStrip(string const & a, char c)
+string const ltrim(string const & a, char const * p)
{
- if (a.empty()) return a;
- string tmp(a);
- string::size_type i = tmp.find_first_not_of(c);
- if (i > 0)
- tmp.erase(0, i);
- return tmp;
+ BOOST_ASSERT(p);
+
+ if (a.empty() || !*p)
+ return a;
+
+ string::size_type l = a.find_first_not_of(p);
+
+ if (l == string::npos)
+ return string();
+
+ return a.substr(l, string::npos);
}
if (i != string::npos) { // delimiter was found
piece = a.substr(0, i);
tmp = a.substr(i + 1);
- } else { // delimter was not found
+ } else { // delimiter was not found
piece.erase();
}
return tmp;
}
+
+
+// This function escapes 8-bit characters and other problematic
+// characters that cause problems in latex labels.
+string const escape(string const & lab)
+{
+ char hexdigit[16] = { '0', '1', '2', '3', '4', '5', '6', '7',
+ '8', '9', 'A', 'B', 'C', 'D', 'E', 'F' };
+ string enc;
+ for (string::size_type i = 0; i < lab.length(); ++i) {
+ unsigned char c= lab[i];
+ if (c >= 128 || c == '=' || c == '%') {
+ enc += '=';
+ enc += hexdigit[c>>4];
+ enc += hexdigit[c & 15];
+ } else {
+ enc += c;
+ }
+ }
+ return enc;
+}
+
+
+/// gives a vector of stringparts which have the delimiter delim
+vector<string> const getVectorFromString(string const & str,
+ string const & delim)
+{
+// Lars would like this code to go, but for now his replacement (below)
+// doesn't fullfil the same function. I have, therefore, reactivated the
+// old code for now. Angus 11 Nov 2002.
+#if 1
+ vector<string> vec;
+ if (str.empty())
+ return vec;
+ string keys(rtrim(str));
+ for(;;) {
+ string::size_type const idx = keys.find(delim);
+ if (idx == string::npos) {
+ vec.push_back(ltrim(keys));
+ break;
+ }
+ string const key = trim(keys.substr(0, idx));
+ if (!key.empty())
+ vec.push_back(key);
+ string::size_type const start = idx + delim.size();
+ keys = keys.substr(start);
+ }
+ return vec;
+#else
+ boost::char_separator<char> sep(delim.c_str());
+ boost::tokenizer<boost::char_separator<char> > tokens(str, sep);
+#ifndef USE_INCLUDED_STRING
+ return vector<string>(tokens.begin(), tokens.end());
+#else
+ vector<string> vec;
+ using boost::tokenizer;
+ using boost::char_separator;
+
+ tokenizer<char_separator<char> >::iterator it = tokens.begin();
+ tokenizer<char_separator<char> >::iterator end = tokens.end();
+ for (; it != end; ++it) {
+ vec.push_back(STRCONV((*it)));
+ }
+ return vec;
+#endif
+#endif
+}
+
+
+// the same vice versa
+string const getStringFromVector(vector<string> const & vec,
+ string const & delim)
+{
+ string str;
+ int i = 0;
+ for (vector<string>::const_iterator it = vec.begin();
+ it != vec.end(); ++it) {
+ string item = trim(*it);
+ if (item.empty())
+ continue;
+ if (i++ > 0)
+ str += delim;
+ str += item;
+ }
+ return str;
+}
+
+
+#if USE_BOOST_FORMAT
+
+string bformat(string const & fmt, string const & arg1)
+{
+ return (boost::format(fmt) % arg1).str();
+}
+
+
+string bformat(string const & fmt, string const & arg1, string const & arg2)
+{
+ return (boost::format(fmt) % arg1 % arg2).str();
+}
+
+
+string bformat(string const & fmt, int arg1, int arg2)
+{
+ return (boost::format(fmt) % arg1 % arg2).str();
+}
+
+
+string bformat(string const & fmt, string const & arg1, string const & arg2,
+ string const & arg3)
+{
+ return (boost::format(fmt) % arg1 % arg2 % arg3).str();
+}
+
+
+string bformat(string const & fmt, string const & arg1, string const & arg2,
+ string const & arg3, string const & arg4)
+{
+ return (boost::format(fmt) % arg1 % arg2 % arg3 % arg4).str();
+}
+
+
+string bformat(string const & fmt, string const & arg1, string const & arg2,
+ string const & arg3, string const & arg4, string const & arg5)
+{
+ return (boost::format(fmt) % arg1 % arg2 % arg3 % arg4 % arg5).str();
+}
+
+#else
+
+string bformat(string const & fmt, string const & arg1)
+{
+ BOOST_ASSERT(contains(fmt, "%1$s"));
+ string const str = subst(fmt, "%1$s", arg1);
+ return subst(str, "%%", "%");
+}
+
+
+string bformat(string const & fmt, string const & arg1, string const & arg2)
+{
+ BOOST_ASSERT(contains(fmt, "%1$s"));
+ BOOST_ASSERT(contains(fmt, "%2$s"));
+ string str = subst(fmt, "%1$s", arg1);
+ str = subst(str, "%2$s", arg2);
+ return subst(str, "%%", "%");
+}
+
+
+string bformat(string const & fmt, int arg1, int arg2)
+{
+ BOOST_ASSERT(contains(fmt, "%1$d"));
+ BOOST_ASSERT(contains(fmt, "%2$d"));
+ string str = subst(fmt, "%1$d", tostr(arg1));
+ str = subst(str, "%2$d", tostr(arg2));
+ return subst(str, "%%", "%");
+}
+
+
+string bformat(string const & fmt, string const & arg1, string const & arg2,
+ string const & arg3)
+{
+ BOOST_ASSERT(contains(fmt, "%1$s"));
+ BOOST_ASSERT(contains(fmt, "%2$s"));
+ BOOST_ASSERT(contains(fmt, "%3$s"));
+ string str = subst(fmt, "%1$s", arg1);
+ str = subst(str, "%2$s", arg2);
+ str = subst(str, "%3$s", arg3);
+ return subst(str, "%%", "%");
+}
+
+
+string bformat(string const & fmt, string const & arg1, string const & arg2,
+ string const & arg3, string const & arg4)
+{
+ BOOST_ASSERT(contains(fmt, "%1$s"));
+ BOOST_ASSERT(contains(fmt, "%2$s"));
+ BOOST_ASSERT(contains(fmt, "%3$s"));
+ BOOST_ASSERT(contains(fmt, "%4$s"));
+ string str = subst(fmt, "%1$s", arg1);
+ str = subst(str, "%2$s", arg2);
+ str = subst(str, "%3$s", arg3);
+ str = subst(str, "%4$s", arg4);
+ return subst(str, "%%", "%");
+}
+
+
+string bformat(string const & fmt, string const & arg1, string const & arg2,
+ string const & arg3, string const & arg4, string const & arg5)
+{
+ BOOST_ASSERT(contains(fmt, "%1$s"));
+ BOOST_ASSERT(contains(fmt, "%2$s"));
+ BOOST_ASSERT(contains(fmt, "%3$s"));
+ BOOST_ASSERT(contains(fmt, "%4$s"));
+ BOOST_ASSERT(contains(fmt, "%5$s"));
+ string str = subst(fmt, "%1$s", arg1);
+ str = subst(str, "%2$s", arg2);
+ str = subst(str, "%3$s", arg3);
+ str = subst(str, "%4$s", arg4);
+ str = subst(str, "%5$s", arg5);
+ return subst(str, "%%", "%");
+}
+
+#endif
+
+} // namespace support
+} // namespace lyx