X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=src%2Fsupport%2Flstrings.cpp;h=656f82ee1119b7010d0f41f5b350304077efec60;hb=36e5864a8264b428ee7a59a09b25830b49d9c5af;hp=fef11cf1d4de92c1294d913d37cb590d84f9b825;hpb=d81c4751df924b10d2b9373e1383cc1442cfa74a;p=lyx.git diff --git a/src/support/lstrings.cpp b/src/support/lstrings.cpp index fef11cf1d4..656f82ee11 100644 --- a/src/support/lstrings.cpp +++ b/src/support/lstrings.cpp @@ -16,13 +16,10 @@ #include "support/convert.h" #include "support/qstring_helpers.h" -#include "support/textutils.h" -#include #include "support/lassert.h" #include -#include #include #include @@ -35,7 +32,7 @@ namespace lyx { // without #include "support/docstring" there. docstring const & empty_docstring() { - static docstring s; + static const docstring s; return s; } @@ -43,10 +40,11 @@ docstring const & empty_docstring() // without #include string const & empty_string() { - static string s; + static const string s; return s; } +namespace { /** * Convert a QChar into a UCS4 character. * This is a hack (it does only make sense for the common part of the UCS4 @@ -54,13 +52,12 @@ string const & empty_string() * This does only exist because of performance reasons (a real conversion * using iconv is too slow on windows). */ -static inline char_type qchar_to_ucs4(QChar const & qchar) +inline char_type qchar_to_ucs4(QChar const & qchar) { - LASSERT(is_utf16(static_cast(qchar.unicode())), /**/); + LASSERT(is_utf16(static_cast(qchar.unicode())), return '?'); return static_cast(qchar.unicode()); } - /** * Convert a UCS4 character into a QChar. * This is a hack (it does only make sense for the common part of the UCS4 @@ -68,17 +65,15 @@ static inline char_type qchar_to_ucs4(QChar const & qchar) * This does only exist because of performance reasons (a real conversion * using iconv is too slow on windows). */ -static inline QChar const ucs4_to_qchar(char_type const ucs4) +inline QChar const ucs4_to_qchar(char_type const ucs4) { - LASSERT(is_utf16(ucs4), /**/); + LASSERT(is_utf16(ucs4), return QChar('?')); return QChar(static_cast(ucs4)); } - -namespace { - /// Maximum valid UCS4 code point - char_type const ucs4_max = 0x10ffff; -} +/// Maximum valid UCS4 code point +char_type const ucs4_max = 0x10ffff; +} // anon namespace bool isLetterChar(char_type c) @@ -117,6 +112,11 @@ bool isPrintable(char_type c) // assume that all non-utf16 characters are printable return true; } + // Not yet recognized by QChar::isPrint() + // See https://bugreports.qt-project.org/browse/QTBUG-12144 + // LATIN CAPITAL LETTER SHARP S + else if (c == 0x1e9e) + return true; return ucs4_to_qchar(c).isPrint(); } @@ -148,13 +148,13 @@ bool isSpace(char_type c) } -bool isDigit(char_type c) +bool isNumber(char_type c) { if (!is_utf16(c)) - // assume that no non-utf16 character is a digit + // assume that no non-utf16 character is a numeral // c outside the UCS4 range is catched as well return false; - return ucs4_to_qchar(c).isDigit(); + return ucs4_to_qchar(c).isNumber(); } @@ -163,6 +163,19 @@ bool isDigitASCII(char_type c) return '0' <= c && c <= '9'; } + +bool isAlnumASCII(char_type c) +{ + return isAlphaASCII(c) || isDigitASCII(c); +} + + +bool isASCII(char_type c) +{ + return c < 0x80; +} + + namespace support { int compare_no_case(docstring const & s, docstring const & s2) @@ -259,7 +272,7 @@ bool isStrInt(string const & str) string::const_iterator end = tmpstr.end(); for (; cit != end; ++cit) - if (!isdigit((*cit))) + if (!isDigitASCII(*cit)) return false; return true; @@ -279,7 +292,7 @@ bool isStrUnsignedInt(string const & str) string::const_iterator cit = tmpstr.begin(); string::const_iterator end = tmpstr.end(); for (; cit != end; ++cit) - if (!isdigit((*cit))) + if (!isDigitASCII(*cit)) return false; return true; @@ -303,7 +316,7 @@ bool isStrDbl(string const & str) ++cit; string::const_iterator end = tmpstr.end(); for (; cit != end; ++cit) { - if (!isdigit(*cit) && *cit != '.') + if (!isDigitASCII(*cit) && *cit != '.') return false; if ('.' == (*cit)) { if (found_dot) @@ -315,7 +328,18 @@ bool isStrDbl(string const & str) } -static bool isHexChar(char_type c) +bool hasDigitASCII(docstring const & str) +{ + docstring::const_iterator cit = str.begin(); + docstring::const_iterator const end = str.end(); + for (; cit != end; ++cit) + if (isDigitASCII(*cit)) + return true; + return false; +} + + +bool isHexChar(char_type c) { return c == '0' || c == '1' || @@ -385,14 +409,14 @@ bool isAscii(string const & str) char lowercase(char c) { - LASSERT(static_cast(c) < 0x80, /**/); + LASSERT(isASCII(c), return '?'); return char(tolower(c)); } char uppercase(char c) { - LASSERT(static_cast(c) < 0x80, /**/); + LASSERT(isASCII(c), return '?'); return char(toupper(c)); } @@ -415,6 +439,16 @@ char_type uppercase(char_type c) } +bool isLowerCase(char_type ch) { + return lowercase(ch) == ch; +} + + +bool isUpperCase(char_type ch) { + return uppercase(ch) == ch; +} + + namespace { // since we cannot use tolower and toupper directly in the @@ -422,19 +456,13 @@ namespace { struct local_lowercase { char_type operator()(char_type c) const { - if (!is_utf16(c)) - // We don't know how to lowercase a non-utf16 char - return c; - return qchar_to_ucs4(ucs4_to_qchar(c).toLower()); + return lowercase(c); } }; struct local_uppercase { char_type operator()(char_type c) const { - if (!is_utf16(c)) - // We don't know how to uppercase a non-utf16 char - return c; - return qchar_to_ucs4(ucs4_to_qchar(c).toUpper()); + return uppercase(c); } }; @@ -444,6 +472,7 @@ template struct local_ascii_lowercase { } // end of anon namespace + docstring const lowercase(docstring const & a) { docstring tmp(a); @@ -452,6 +481,16 @@ docstring const lowercase(docstring const & a) } +/* Uncomment here and in lstrings.h if you should need this. +string const lowercase(string const & a) +{ + string tmp(a); + transform(tmp.begin(), tmp.end(), tmp.begin(), local_lowercase()); + return tmp; +} +*/ + + docstring const uppercase(docstring const & a) { docstring tmp(a); @@ -478,6 +517,132 @@ docstring const ascii_lowercase(docstring const & a) } +char_type superscript(char_type c) +{ + switch (c) { + case '2': return 0x00b2; + case '3': return 0x00b3; + case '1': return 0x00b9; + case '0': return 0x2070; + case 'i': return 0x2071; + case '4': return 0x2074; + case '5': return 0x2075; + case '6': return 0x2076; + case '7': return 0x2077; + case '8': return 0x2078; + case '9': return 0x2079; + case '+': return 0x207a; + case '-': return 0x207b; + case '=': return 0x207c; + case '(': return 0x207d; + case ')': return 0x207e; + case 'n': return 0x207f; + case 'h': return 0x02b0; + case 0x0266: return 0x02b1; // LATIN SMALL LETTER H WITH HOOK + case 'j': return 0x02b2; + case 'r': return 0x02b3; + case 0x0279: return 0x02b4; // LATIN SMALL LETTER TURNED R + case 0x027b: return 0x02b5; // LATIN SMALL LETTER TURNED R WITH HOOK + case 0x0281: return 0x02b6; // LATIN SMALL LETTER CAPITAL INVERTED R + case 'w': return 0x02b7; + case 'y': return 0x02b8; +// case 0x0294: return 0x02c0; // LATIN LETTER GLOTTAL STOP) +// case 0x0295: return 0x02c1; // LATIN LETTER PHARYNGEAL VOICED FRICATIVE + // (= LATIN LETTER REVERSED GLOTTAL STOP) + case 'l': return 0x02e1; + case 's': return 0x02e2; + case 'x': return 0x02e3; +// case 0x0295: return 0x02e4; // LATIN SMALL LETTER REVERSED GLOTTAL STOP + case 'A': return 0x1d2c; + case 0x00c6: return 0x1d2d; // LATIN CAPITAL LETTER AE + case 'B': return 0x1d2e; + case 'D': return 0x1d30; + case 'E': return 0x1d31; + case 'G': return 0x1d33; + case 'H': return 0x1d34; + case 'I': return 0x1d35; + case 'J': return 0x1d36; + case 'K': return 0x1d37; + case 'L': return 0x1d38; + case 'M': return 0x1d39; + case 'N': return 0x1d3a; + case 'O': return 0x1d3c; + case 'P': return 0x1d3e; + case 'R': return 0x1d3f; + case 'T': return 0x1d40; + case 'U': return 0x1d41; + case 'W': return 0x1d42; + case 'a': return 0x1d43; + case 0x0250: return 0x1d44; // LATIN SMALL LETTER TURNED A + case 0x0251: return 0x1d45; // LATIN SMALL LETTER ALPHA + case 'b': return 0x1d47; + case 'd': return 0x1d48; + case 'e': return 0x1d49; + case 0x0259: return 0x1d4a; // LATIN SMALL LETTER SCHWA + case 0x025b: return 0x1d4b; // LATIN SMALL LETTER OPEN E + case 0x1d08: return 0x1d4c; // LATIN SMALL LETTER TURNED OPEN E + case 'g': return 0x1d4d; + case 0x1d09: return 0x1d4e; // LATIN SMALL LETTER TURNED I + case 'k': return 0x1d4f; + case 'm': return 0x1d50; + case 0x014b: return 0x1d51; // LATIN SMALL LETTER ENG + case 'o': return 0x1d52; + case 0x0254: return 0x1d53; // LATIN SMALL LETTER OPEN O + case 0x1d16: return 0x1d54; // LATIN SMALL LETTER TOP HALF O + case 0x1d17: return 0x1d55; // LATIN SMALL LETTER BOTTOM HALF O + case 'p': return 0x1d56; + case 't': return 0x1d57; + case 'u': return 0x1d58; + case 0x1d1d: return 0x1d59; // LATIN SMALL LETTER SIDEWAYS U + case 0x1d1f: return 0x1d5a; // LATIN SMALL LETTER SIDEWAYS TURNED M + case 'v': return 0x1d5b; + case 0x03b2: return 0x1d5d; // GREEK SMALL LETTER BETA + case 0x03b3: return 0x1d5e; // GREEK SMALL LETTER GAMMA + case 0x03b4: return 0x1d5f; // GREEK SMALL LETTER DELTA + case 0x03c6: return 0x1d60; // GREEK SMALL LETTER PHI + case 0x03c7: return 0x1d61; // GREEK SMALL LETTER CHI + } + return c; +} + + +char_type subscript(char_type c) +{ + switch (c) { + case 'i': return 0x1d62; + case 'r': return 0x1d63; + case 'u': return 0x1d64; + case 'v': return 0x1d65; + case 0x03b2: return 0x1d66; // GREEK SMALL LETTER BETA + case 0x03b3: return 0x1d67; // GREEK SMALL LETTER GAMMA + case 0x03c1: return 0x1d68; // GREEK SMALL LETTER RHO + case 0x03c6: return 0x1d69; // GREEK SMALL LETTER PHI + case 0x03c7: return 0x1d6a; // GREEK SMALL LETTER CHI + case '0': return 0x2080; + case '1': return 0x2081; + case '2': return 0x2082; + case '3': return 0x2083; + case '4': return 0x2084; + case '5': return 0x2085; + case '6': return 0x2086; + case '7': return 0x2087; + case '8': return 0x2088; + case '9': return 0x2089; + case '+': return 0x208a; + case '-': return 0x208b; + case '=': return 0x208c; + case '(': return 0x208d; + case ')': return 0x208e; + case 'a': return 0x2090; + case 'e': return 0x2091; + case 'o': return 0x2092; + case 'x': return 0x2093; + case 0x0259: return 0x2093; // LATIN SMALL LETTER SCHWA + } + return c; +} + + bool prefixIs(docstring const & a, char_type c) { if (a.empty()) @@ -649,6 +814,7 @@ basic_string const subst_char(basic_string const & a, return tmp; } + /// Substitute all \a oldchar with \a newchar docstring const subst_char(docstring const & a, docstring::value_type oldchar, docstring::value_type newchar) @@ -668,7 +834,7 @@ template inline String const subst_string(String const & a, String const & oldstr, String const & newstr) { - LASSERT(!oldstr.empty(), /**/); + LASSERT(!oldstr.empty(), return a); String lstr = a; size_t i = 0; size_t const olen = oldstr.length(); @@ -680,10 +846,11 @@ String const subst_string(String const & a, return lstr; } + docstring const subst_string(docstring const & a, docstring const & oldstr, docstring const & newstr) { - LASSERT(!oldstr.empty(), /**/); + LASSERT(!oldstr.empty(), return a); docstring lstr = a; size_t i = 0; size_t const olen = oldstr.length(); @@ -725,9 +892,34 @@ docstring const subst(docstring const & a, } +int count_char(string const & str, char chr) +{ + int count = 0; + string::const_iterator lit = str.begin(); + string::const_iterator end = str.end(); + for (; lit != end; ++lit) + if ((*lit) == chr) + count++; + return count; +} + + +/// Count all occurences of char \a chr inside \a str +int count_char(docstring const & str, docstring::value_type chr) +{ + int count = 0; + docstring::const_iterator lit = str.begin(); + docstring::const_iterator end = str.end(); + for (; lit != end; ++lit) + if ((*lit) == chr) + count++; + return count; +} + + docstring const trim(docstring const & a, char const * p) { - LASSERT(p, /**/); + LASSERT(p, return a); if (a.empty() || !*p) return a; @@ -746,7 +938,7 @@ docstring const trim(docstring const & a, char const * p) string const trim(string const & a, char const * p) { - LASSERT(p, /**/); + LASSERT(p, return a); if (a.empty() || !*p) return a; @@ -764,7 +956,7 @@ string const trim(string const & a, char const * p) string const rtrim(string const & a, char const * p) { - LASSERT(p, /**/); + LASSERT(p, return a); if (a.empty() || !*p) return a; @@ -781,7 +973,7 @@ string const rtrim(string const & a, char const * p) docstring const rtrim(docstring const & a, char const * p) { - LASSERT(p, /**/); + LASSERT(p, return a); if (a.empty() || !*p) return a; @@ -798,7 +990,7 @@ docstring const rtrim(docstring const & a, char const * p) string const ltrim(string const & a, char const * p) { - LASSERT(p, /**/); + LASSERT(p, return a); if (a.empty() || !*p) return a; size_t l = a.find_first_not_of(p); @@ -810,7 +1002,7 @@ string const ltrim(string const & a, char const * p) docstring const ltrim(docstring const & a, char const * p) { - LASSERT(p, /**/); + LASSERT(p, return a); if (a.empty() || !*p) return a; size_t l = a.find_first_not_of(from_ascii(p)); @@ -899,6 +1091,20 @@ string const rsplit(string const & a, string & piece, char delim) } +docstring const rsplit(docstring const & a, docstring & piece, char_type delim) +{ + docstring tmp; + size_t i = a.rfind(delim); + if (i != string::npos) { // delimiter was found + piece = a.substr(0, i); + tmp = a.substr(i + 1); + } else { // delimiter was not found + piece.erase(); + } + return tmp; +} + + docstring const rsplit(docstring const & a, char_type delim) { docstring tmp; @@ -916,13 +1122,14 @@ docstring const escape(docstring const & lab) docstring enc; for (size_t i = 0; i < lab.length(); ++i) { char_type c = lab[i]; - if (c >= 128 || c == '=' || c == '%') { + if (c >= 128 || c == '=' || c == '%' || c == '#' || c == '$' + || c == '}' || c == '{' || c == ']' || c == '[' || c == '&') { // Although char_type is a 32 bit type we know that // UCS4 occupies only 21 bits, so we don't need to // encode bigger values. Test for 2^24 because we // can encode that with the 6 hex digits that are // needed for 21 bits anyway. - LASSERT(c < (1 << 24), /**/); + LASSERT(c < (1 << 24), continue); enc += '='; enc += hexdigit[(c>>20) & 15]; enc += hexdigit[(c>>16) & 15]; @@ -990,6 +1197,7 @@ docstring wrap(docstring const & str, int const ind, size_t const width) docstring wrapParas(docstring const & str, int const indent, size_t const width, size_t const maxlines) { + docstring const dots = from_ascii("..."); if (str.empty()) return docstring(); @@ -1004,9 +1212,21 @@ docstring wrapParas(docstring const & str, int const indent, if (nlines == 0) continue; size_t const curlines = retval.size(); - if (maxlines > 0 && curlines + nlines >= maxlines) { - tmp.resize(maxlines - curlines - 1); - tmp.push_back(from_ascii("...")); + if (maxlines > 0 && curlines + nlines > maxlines) { + tmp.resize(maxlines - curlines); + docstring last = tmp.back(); + size_t const lsize = last.size(); + if (lsize > width - 3) { + size_t const i = last.find_last_of(' ', width - 3); + if (i == docstring::npos || i <= size_t(indent)) + // no space found + last = last.substr(0, lsize - 3) + dots; + else + last = last.substr(0, i) + dots; + } else + last += dots; + tmp.pop_back(); + tmp.push_back(last); } retval.insert(retval.end(), tmp.begin(), tmp.end()); if (maxlines > 0 && retval.size() >= maxlines) @@ -1117,7 +1337,7 @@ int findToken(char const * const str[], string const & search_token) template<> docstring bformat(docstring const & fmt, int arg1) { - LASSERT(contains(fmt, from_ascii("%1$d")), /**/); + LATTEST(contains(fmt, from_ascii("%1$d"))); docstring const str = subst(fmt, from_ascii("%1$d"), convert(arg1)); return subst(str, from_ascii("%%"), from_ascii("%")); } @@ -1126,7 +1346,7 @@ docstring bformat(docstring const & fmt, int arg1) template<> docstring bformat(docstring const & fmt, long arg1) { - LASSERT(contains(fmt, from_ascii("%1$d")), /**/); + LATTEST(contains(fmt, from_ascii("%1$d"))); docstring const str = subst(fmt, from_ascii("%1$d"), convert(arg1)); return subst(str, from_ascii("%%"), from_ascii("%")); } @@ -1135,7 +1355,7 @@ docstring bformat(docstring const & fmt, long arg1) template<> docstring bformat(docstring const & fmt, unsigned int arg1) { - LASSERT(contains(fmt, from_ascii("%1$d")), /**/); + LATTEST(contains(fmt, from_ascii("%1$d"))); docstring const str = subst(fmt, from_ascii("%1$d"), convert(arg1)); return subst(str, from_ascii("%%"), from_ascii("%")); } @@ -1144,7 +1364,7 @@ docstring bformat(docstring const & fmt, unsigned int arg1) template<> docstring bformat(docstring const & fmt, docstring arg1) { - LASSERT(contains(fmt, from_ascii("%1$s")), /**/); + LATTEST(contains(fmt, from_ascii("%1$s"))); docstring const str = subst(fmt, from_ascii("%1$s"), arg1); return subst(str, from_ascii("%%"), from_ascii("%")); } @@ -1153,7 +1373,7 @@ docstring bformat(docstring const & fmt, docstring arg1) template<> docstring bformat(docstring const & fmt, char * arg1) { - LASSERT(contains(fmt, from_ascii("%1$s")), /**/); + LATTEST(contains(fmt, from_ascii("%1$s"))); docstring const str = subst(fmt, from_ascii("%1$s"), from_ascii(arg1)); return subst(str, from_ascii("%%"), from_ascii("%")); } @@ -1162,8 +1382,8 @@ docstring bformat(docstring const & fmt, char * arg1) template<> docstring bformat(docstring const & fmt, docstring arg1, docstring arg2) { - LASSERT(contains(fmt, from_ascii("%1$s")), /**/); - LASSERT(contains(fmt, from_ascii("%2$s")), /**/); + LATTEST(contains(fmt, from_ascii("%1$s"))); + LATTEST(contains(fmt, from_ascii("%2$s"))); docstring str = subst(fmt, from_ascii("%1$s"), arg1); str = subst(str, from_ascii("%2$s"), arg2); return subst(str, from_ascii("%%"), from_ascii("%")); @@ -1173,8 +1393,8 @@ docstring bformat(docstring const & fmt, docstring arg1, docstring arg2) template<> docstring bformat(docstring const & fmt, docstring arg1, int arg2) { - LASSERT(contains(fmt, from_ascii("%1$s")), /**/); - LASSERT(contains(fmt, from_ascii("%2$d")), /**/); + LATTEST(contains(fmt, from_ascii("%1$s"))); + LATTEST(contains(fmt, from_ascii("%2$d"))); docstring str = subst(fmt, from_ascii("%1$s"), arg1); str = subst(str, from_ascii("%2$d"), convert(arg2)); return subst(str, from_ascii("%%"), from_ascii("%")); @@ -1184,8 +1404,8 @@ docstring bformat(docstring const & fmt, docstring arg1, int arg2) template<> docstring bformat(docstring const & fmt, char const * arg1, docstring arg2) { - LASSERT(contains(fmt, from_ascii("%1$s")), /**/); - LASSERT(contains(fmt, from_ascii("%2$s")), /**/); + LATTEST(contains(fmt, from_ascii("%1$s"))); + LATTEST(contains(fmt, from_ascii("%2$s"))); docstring str = subst(fmt, from_ascii("%1$s"), from_ascii(arg1)); str = subst(fmt, from_ascii("%2$s"), arg2); return subst(str, from_ascii("%%"), from_ascii("%")); @@ -1195,8 +1415,8 @@ docstring bformat(docstring const & fmt, char const * arg1, docstring arg2) template<> docstring bformat(docstring const & fmt, int arg1, int arg2) { - LASSERT(contains(fmt, from_ascii("%1$d")), /**/); - LASSERT(contains(fmt, from_ascii("%2$d")), /**/); + LATTEST(contains(fmt, from_ascii("%1$d"))); + LATTEST(contains(fmt, from_ascii("%2$d"))); docstring str = subst(fmt, from_ascii("%1$d"), convert(arg1)); str = subst(str, from_ascii("%2$d"), convert(arg2)); return subst(str, from_ascii("%%"), from_ascii("%")); @@ -1206,9 +1426,9 @@ docstring bformat(docstring const & fmt, int arg1, int arg2) template<> docstring bformat(docstring const & fmt, docstring arg1, docstring arg2, docstring arg3) { - LASSERT(contains(fmt, from_ascii("%1$s")), /**/); - LASSERT(contains(fmt, from_ascii("%2$s")), /**/); - LASSERT(contains(fmt, from_ascii("%3$s")), /**/); + LATTEST(contains(fmt, from_ascii("%1$s"))); + LATTEST(contains(fmt, from_ascii("%2$s"))); + LATTEST(contains(fmt, from_ascii("%3$s"))); docstring str = subst(fmt, from_ascii("%1$s"), arg1); str = subst(str, from_ascii("%2$s"), arg2); str = subst(str, from_ascii("%3$s"), arg3); @@ -1220,10 +1440,10 @@ template<> docstring bformat(docstring const & fmt, docstring arg1, docstring arg2, docstring arg3, docstring arg4) { - LASSERT(contains(fmt, from_ascii("%1$s")), /**/); - LASSERT(contains(fmt, from_ascii("%2$s")), /**/); - LASSERT(contains(fmt, from_ascii("%3$s")), /**/); - LASSERT(contains(fmt, from_ascii("%4$s")), /**/); + LATTEST(contains(fmt, from_ascii("%1$s"))); + LATTEST(contains(fmt, from_ascii("%2$s"))); + LATTEST(contains(fmt, from_ascii("%3$s"))); + LATTEST(contains(fmt, from_ascii("%4$s"))); docstring str = subst(fmt, from_ascii("%1$s"), arg1); str = subst(str, from_ascii("%2$s"), arg2); str = subst(str, from_ascii("%3$s"), arg3);